Vv 546 58 989 Volume 3: ‘> Prevention and UBL Interventions in Youth Suicide Report of the Secretary’ Task Force on Youth Suicide PL Aes | U.S. DEPARTMENT OF HEALTH & HUMAN SERVICES Public Health Service Alcohol, Drug Abuse, and Mental Health Administration Volume 3: rein nd Report of the Youth Suicide Secretary’ Task Force on Youth Suicide Edited by: Marcia R. Feinleib January 1989 EF NS TC A KS, U.S. DEPARTMENT OF HEALTH & HUMAN SERVICES Public Health Service Alcohol, Drug Abuse, and Mental Health Administration The members of the Secretary's Task Force on Youth Suicide wish to acknowledge the extraordinary effort of the Executive Secretary, Ms. Eugenia P. Broumas. Suggested citation: Alcohol, Drug Abuse, and Mental Health Administration. Report of the Secretary's Task Force on Youth Suicide. Volume 3: Prevention and Interventions in Youth Suicide. DHHS Pub. No. (ADM)89-1623. Washington, D.C.: Supt. of Docs., U.S. Govt. Print. Off., 1989. 3-ii CONTENTS: Volume 3 Members of the Secretary's Task Force on Youthsuicide . . . ....... Members of the Work Group on Prevention and Interventions . . . . . . . Summary of the National Conference on Prevention and Interventions in Youth Suicide . . . . .................... Commissioned Papers: 1. Robert Felner: Primary Prevention: A Consideration of General Principles and Findings for the Prevention of Youth Suicide . ......... 2. David Shaffer, K. Bacon: A Critical Review of Preventive Intervention Efforts in Suicide, with Particular Reference to Youth Suicide . . . .... .. 3. Betsy S. Comstock, Jane T. Simmons, Jack L. Franklin: Overview of Prevention Efforts in Adolescent Suicide . . .... .. 4. Betsy S. Comstock, Jane T. Simmons, Jack L. Franklin: Community Response to Adolescent Suicide Clusters . . . . . . . . 5. Jane T. Simmons, Betsy S. Comstock, Jack L. Franklin: Prevention/Intervention Programs for Suicidal Adolescents . . . . . 6. Jack L. Franklin, Betsy S. Comstock, Jane T. Simmons, Mark Mason: Characteristics of Suicide Prevention/Intervention Programs: AnalysisofaSurvey . ........ 7. Robert E. Litman: Psychological Autopsies of Youth Suicide . . . ........... 8. Paul Gibson: Gay Male and Lesbian Youth Suicide . . . .............. 9. Curtis Mitchell: Issues for SUNVIVOIS . . . «oo i i eee 10. James W. Thompson: Prevention of Adolescent Suicide Among American Indian and Alaska Native Peoples . . . ................... 11. Elena S. Yu, Ching-Fu Chang, William T. Liu, Marilyn Fernandez: Suicide Prevention and Intervention Among Asian Youth . . . . .. 12. F.M. Baker: Black Youth Suicide: Literature Review with a Focus on Prevention 13. Jack C. Smith, James A. Mercy, Mark L. Rosenberg: Hispanic Suicide in the Southwest, 1980-82 . . ........... 14. Barbara P. Wyatt: The Role of Volunteer Workers in Suicide Prevention Centers . . . . 15. Bryan Tanney: Preventing Suicide by Improving the Competency of Caregivers . . 16. Richard G. Katzoff: The Samaritans and the Prevention of Youth Suicide . ....... 3-iii 17. 18. 19. 20. 21. 22. 23. 3-iv John E. Meeks: Evaluation and Management of Suicidal Risk in Chemically Dependent Adolescents . . . . .. ...................... 232 Susan J. Blumenthal, David J. Kupfer: Overview of Early Detection and Treatment Strategies for Suicidal Behaviorin Young People . . . . . .......... 239 Paul D. Trautman: Specific Treatment Modalities for Adolescent Suicide Attempters . . . . . . . .. . .. 253 Iris M. Bolton: Perspectives of Youth on Preventive Intervention Strategies . . . . . .......... 264 Alan L. Berman: Mass Media and Youth Suicide Prevention . . . . .................... 276 Pamela C. Cantor: Intervention Strategies: Environmental Risk Reduction for Youth Suicide . . . . . . . . 285 Barry D. Garfinkel: School-Based Prevention Programs. . . . . . ...................... 294 MEMBERS OF THE SECRETARY'S TASK FORCE ON YOUTH SUICIDE Shervert Frazier, M.D. Chairman Formerly, Director National Institute of Mental Health Carolyn Doppelt Gray Acting Deputy Assistant Secretary Office of Human Development Services M. Gene Handelsman (served until June 1986) Formerly, Deputy Assistant Secretary Office of Human Development Services Robert B. Helms, Ph.D. Acting Assistant Secretary for Planning and Evaluation Office of the Secretary Jerome H. Jaffe, M.D. (served until June 1986) Director, Addiction Research Center National Institute on Drug Abuse Stephanie Lee-Miller Assistant Secretary for Public Affairs Office of the Secretary Markku Linnoila, M.D. Clinical Director, Division of Intramural Clinical and Biological Research National Institute on Alcohol Abuse and Alcoholism Dodie T. Livingston Commissioner, Administration for Children, Youth, and Families Office of Human Development Services Robert G. Niven, M.D. (served until January 1986) Formerly, Director, National Institute on Alcohol Abuse and Alcoholism Everett R. Rhoades, M.D. Director, Indian Health Service Health Resources and Services Administration Mark L. Rosenberg, M.D., M.P.P. Assistant Director for Science Division of Injury Epidemiology and Control Center for Environmental Health and Injury Control Centers for Disease Control Charles R. Schuster, Ph.D. Director National Institute on Drug Abuse Robert L. Trachtenberg Deputy Administrator Alcohol, Drug Abuse, and Mental Health Administration 3-1 STAFF Eugenia P. Broumas, Coordinator, Task Force Activities Special Assistant to the Deputy Director National Institute of Mental Health Heather A. Pack, Policy Coordinator Executive Secretariat Office of the Secretary ALTERNATES Dorynne Czechowicz, M.D. Assistant Director for Medical and Professional Affairs Office of Science National Institute on Drug Abuse Jack Durell, M.D. (served until July 1986) Formerly, Associate Director for Science National Institute on Drug Abuse Paget Wilson Hinch Associate Commissioner, Family and Youth Services Bureau Administration for Children, Youth, and Families Office of Human Development Services Chuck Kline Deputy Assistant Secretary for Public Affairs - News Office of the Secretary Arnold R. Tompkins Deputy Assistant Secretary for Social Services Policy Office of the Assistant Secretary for Planning and Evaluation Office of the Secretary 3-2 WORK GROUP ON PREVENTION AND INTERVENTIONS IN YOUTH SUICIDE Jack Durell, M.D., Cochairperson Formerly, Associate Director for Science National Institute on Drug Abuse Dodie T. Livingston, Cochairperson Commissioner, Administration for Children, Youth, and Families Office of Human Development Services Alan L. Berman, Ph.D. Professor of Psychology, American University Washington, D.C. Iris M. Bolton, M.A. Executive Director, Link Counseling Center Atlanta, Georgia Pamela C. Cantor, Ph.D. Executive Director, National Committee on Youth Suicide Prevention Norwood, Massachusetts Robert E. Litman, M.D. Codirector and Chief Psychiatrist, Los Angeles Suicide Prevention Center Los Angeles, California Michael L. Peck, Ph.D. Cochair of the California State Youth School Suicide Prevention Group Los Angeles Suicide Prevention Center Los Angeles, California Seymour Perlin, M.D. Professor of Psychiatry and Behavioral Sciences George Washington University Medical Center Washington, D.C. Barbara P. Wyatt Arlington, Virginia 3-3 Resource Staff Lynne Heneson Special Assistant to the Associate Commissioner Family and Youth Services Bureau Administration for Children, Youth, and Families Office of Human Development Services Paget Wilson Hinch Associate Commissioner, Family and Youth Services Bureau Administration for Children, Youth, and Families Office of Human Development Services Reactor Panel Seymour Perlin, M.D. Professor of Psychiatry and Behavioral Sciences Department of Psychiatry George Washington University Medical Center Washington, D.C. Ronald W. Maris, Ph.D. Professor of Sociology and Preventive Medicine Director, Center for the Study of Suicide University of South Carolina Columbia, South Carolina Jerome A. Motto, M.D. Professor of Psychiatry University of California at San Francisco School of Medicine San Francisco, California Charlotte P. Ross President and Executive Director Youth Suicide National Center Washington, D.C. Morton Silverman, M.D. Formerly, Associate Administrator for Prevention Alcohol, Drug Abuse, and Mental Health Administration Rockville, Maryland 3-4 NATIONAL CONFERENCE ON PREVENTION AND INTERVENTIONS IN YOUTH SUICIDE INTRODUCTION Since the mid-1950s, fundamental and im- portant changes occurred in suicide patterns in the United States. Suicide rates among older persons decreased while rates for young persons between ages 15 to 24 nearly tripled. Between 1970 and 1985, ap- proximately 75,000 young people took their own lives. Beginning in 1980, more than half of all suicides occurred among persons less than 40 years old. The taking of one’s own life is now the second leading cause of death among those ages 15 to 24 and is one of the leading causes of premature death--200,000 potential years of life are lost annually be- cause of suicide. White males account for the preponderance of all suicides and more males are using guns to commit suicide. Suicide among the young was recognized as a public health issue of national importance when the nation’s health priorities were reor- dered in 1979. The Surgeon General’s report, HealthyPeople called for reducing the suicide rate among persons 15 to 24 from 12.4 per 100,000 (in 1978) to 11 per 100,000 by 1990. The Secretary’s Task Force on Youth Suicide was established as a response to the public demand for action to end these tragic events. It was charged with the responsibility for coordinating suicide activities among various Federal agencies, Congress, State and local governments, private agencies, and professional organizations. Its major func- tions are to assess and consolidate informa- tion on suicide, to provide a forum for communication among health care providers, educators, social service profes- sionals, and families; and to recommend and initiate activities to address the problem. The ultimate goal of this task force is to for- mulate a national plan comprised of research activities, educational efforts, and health ser- vices involving the public and private sector in efforts to reduce youth suicide. NATIONAL CONFERENCES The Task Force on Youth Suicide sponsored a series of national conferences which served as forums for exchanging the most up-to-date information on risk factors and preventive strategies. The first of the conferences made clear that while each suicide is different, a complex interplay of characteristics, or risk factors, contribute to suicide in general and youth suicide in particular; these contribut- ing factors are complex, often interrelated, and only partially understood. Indicators of risk include, but are not limited to, the presence of psychiatric disorders such as depression or schizophrenia, parental loss and family disruption, being abused or neglected, being a friend or family member of a suicide victim, having genetic or biochemical factors (such as elevated serotonin or S-HIAA levels), sexual identity problems, being a runaway, having a family with a history of substance abuse, having an unwanted pregnancy, suffering a humiliation or perceived humiliation, and having a propensity toward impulsive and aggressive behavior. Alcohol and drug abuse often complicates and sometimes precipitates suicidal behavior. Cultural pressures and Report of the Secretary’s Task Force on Youth Suicide socioeconomic variables also contribute to suicide among minority youth. Conference on Prevention and Interventions in Youth Suicide In June 1986, a national conference was held in Oakland, California to review the current state of knowledge in suicide prevention ac- tivities and intervention strategies. Par- ticipants in the conference included persons in many disciplines who work with troubled youth: researchers who study suicide; professionals in mental health, medicine, education, and social work; representatives of national and professional organizations, representatives of community-based service programs, including volunteers, civic and religious leaders, parents, and others who work at the front lines of prevention programs. The papers and research studies presented at the conference addressed intervention strategies on several levels--primary and secondary prevention, community-level health and social services, the role of volun- teers in suicide prevention, school-based education, interventions for special popula- tions, early detection and treatment for suicidal adolescents, and federally supported research and demonstration centers. Issues relating to the effectiveness of prevention ac- tivities were thoroughly aired. The value of the conference was expanded by frequent open discussion periods and review panels that critiqued the presentations. Many recommendations for intervention and evaluation approaches made by repre- sentatives of public and private interests have been integrated into the task force’s recom- mendations to the Secretary; many more sug- gestions are included in the commissioned papers contained in this volume. For many years, research into suicide con- centrated on the well established relation- ship between suicide and psychiatric disorders in adults, usually white males over the age of 40. Intervention and prevention efforts generally involved the detection and treatment of mental illness, most commonly depression. Recent research, however, sug- gests that among the young, the patterns of suicide differ from the traditional picture; only a portion of young people who commit suicide are known to have a diagnosable mental disorder. Many investigators now believe that treatment modes used for adults (e.g., treating depression) will not be effec- tive with young people and that conclusions drawn from research on adults cannot be generalized to youth. From the conference on risk factors, we learned that a multiplicity of factors con- tribute to a young person’s decision to end his or her life, and from the conference on prevention and interventions that no single therapeutic model is ideal to combat the problem; multiple interventions are neces- sary. Clearly, we need to know a great deal more about the usefulness and effectiveness of interventions. Good epidemiologic data onsuicides will help in identifying risk factors and in planning prevention and intervention approaches targeted to specific needs in the - populations at risk. We need to evaluate carefully the services now in use and to develop well-planned interventions that in- clude rigorous analyses and interpretation of results. The participants in the conference believe that the precursors to suicide can be treated and that many potential suicides can be redirected toward alternate, life-sustaining choices. Success will require the combined efforts of all sectors of society: parents, peers, and caring people; and professionals in the fields of health, education, social and mental health services, collaborating to prevent a broad range of self-destructive behaviors in youth. CONFERENCE SUMMARY Suicide is a rare phenomenon, affecting about 1 out of 10,000 people. Most of our knowledge about the causes and prevention of suicide is based on relatively few cases. Research projects studying young people National Conference on Prevention and Interventions are, with a few exceptions, poorly designed, lack adequate comparison groups or realistic outcome measures, and often involve too few subjects to be considered statistically sig- nificant. A strategy for prevention of youthful suicidal behavior must reconcile two points of view. One view sees completed suicides as the cul- mination of processes that begin with some problem early in life-educational, be- havioral, family, psychosocial, physical-and progress along a continuum of suicidal thoughts, attempts, and finally successful suicide. Many of these young people are highly vulnerable to the stresses of life and they are unable or refuse to adapt to life con- ditions. Inability to cope with these problems may be predictive of a number of self destruc- tive behaviors, including suicide threats, at- tempts, and completions. The other point of view characterizes at- tempted suicides and completed suicides as distinct but overlapping entities. An es- timated 25 to 40 percent of those who com- plete suicide are known to have made a previous attempt. A prevention effort for the former entails a broad-based primary intervention at early points along the continuum, while the latter group requires an approach tailored to a smaller group of individuals whose charac- teristics or specific problems place them at high risk for suicide. A prevention program targeted only to those with demonstrable risk factors may miss reaching some potentially suicidal individuals in the larger population; on the other hand, a broad-based effort re- quires a large cohort, but one in which low risk individuals may be strengthened while helping those who need to be helped. Primary prevention in youth suicide The concept of primary prevention dictates that preventive efforts, grounded in a sound knowledge base, be administered before signs of a condition or problem develop. For suicide, this refers to any intervention that reduces the possibility of suicide by an adolescent. Because the pathways that led to suicide are so varied, the issue of when to apply primary prevention may become a problem. For example, should efforts be in- stituted when a specific disorder identified with suicide appears? Or, should efforts begin when risk factors appear, such as early trauma, sexual identity problems, drug problems, or negative social behavior? Dr. Felner (see paper, this volume) avers that youth suicide is part of a developmental pat- tern of general emotional and behavioral problems related to stress, including depres- sion, acting out problems, risk-taking and other self-destructive behaviors such as al- cohol and substance abuse, confounded by the intentional, perhaps impulsive, use of lethal means. A broad-based primary prevention model as proposed by Dr. Felner involves two strategies. One strategy would be devoted to modifying social systems in which youth func- tion in ways that make the youngsters’ en- vironment less difficult to adapt to. Making the school environment less anxiety-produc- ing, for example, might affect the mental health and well-being of individuals such that stress is reduced and life opportunities are in- creased. The second strategy seeks to help youths develop skills that enable them feel better about themselves and less anxious about the future. Enhancing young persons’ problemsolving, decisionmaking, and coping skills through educational programs and sup- port networks are examples that might equip youngsters to function better in their en- vironment. While broad-based prevention programs re- quire exposing a very large population of children to a "treatment," reducing suicide is not the only expected outcome. Numerous beneficial effects are possible with this ap- proach that may generate better mental health even among persons at low risk for suicide including enhancing self-esteem, reducing school failure, increasing the sense of control young people have over their fu- ture, and reducing depression and a range of other health-related risky behaviors in Report of the Secretary’s Task Force on Youth Suicide adolescents. The effects of primary prevention are hard to isolate from other variables that may in- fluence a potential suicide. Few data, no es- tablished models or sound evidence suggest that a broad-based primary prevention ap- proach would be effective in reducing suicide among young people. Before proceeding with such a costly effort, models confirming the benefits of such programs need to be tested and evaluated. Overview of prevention activities Opinions of scientists in the field vary regard- ing the relationship between suicide at- tempters and completers. Data on unsuccessful suicide attempts are skimpy and uncertain, and it is not clear to what extent the psychological problems of attempters resemble those of completers. Nevertheless, it seems prudent to regard any suicidal be- havior as presaging completion. In his review, therefore, Dr. Shaffer addressed a wide range of primary and secondary preven- tion activities (See Shaffer, this volume). The primary prevention activities critiqued by Dr. Shaffer include providing psychiatric care to vulnerable groups; increasing sen- sitivity of school personnel to the characteris- tics of the suicide-prone child, and providing skills for teachers, counselors, and other pupils to use when they identify a child at risk; providing information on suicidal behaviors to school children; encouraging students to talk about their suicidal thoughts; suggesting referrals to students suspected to be at risk; and intervening early and treating conditions which are known to predispose toward suicide. Secondary prevention encompasses ac- tivities directed toward preventing comple- tion of suicide among persons who have already threatened or attempted suicide. Approaches include psychological treatment for suicidal individuals, providing emergency crisis intervention at times of maximal stress (including crisis centers and telephone hot- lines for counseling and referral), and ongo- ing treatment after the crisis has passed. 3-8 Identifying suicidal youth through psychological autopsies The psychological autopsy is a method used by investigators to obtain more information about suicide victims with the goal of iden- tifying a set of conditions or warning signs which may predict other suicides. (See Lit- man, this volume.) Through interviews with family members, friends, teachers, and other contacts, the researchers attempt to reconstruct the lifestyle, symptoms and be- haviors, personal and occupational histories, and medical records of deceased persons. Psychological autopsies of adults show that depression and alcoholism account for most suicides. Analyses of younger victims, however, show that while adolescent suicides are preceded by psychological maladjust- ment, fewer young suicide victims suffered from depression; most had a combination of affective and antisocial, aggressive be- haviors. The largest group of young suicides (mostly males) are those with conduct or per- sonality disorders, often mixed with drug use. These include impulsive or antisocial young people, many of whom had gotten into some kind of trouble. Another group (usually females) suffer from depression. A third group of suicides consist of youngsters who are compulsive, hard striving perfectionists, socially inhibited and prone to extreme anxiety in the face of any social or academic challenge. A proportion of youngsters do not appear to have diagnosable psychologi- cal disorders and their emotional or psychological problems are sometimes un- recognized and untreated. The researchers further found that adoles- cents who were presuicidal differed from one another in behaviors, psychological diag- noses, and responses to environmental stres- ses. School problems and conduct disorders were common as were social withdrawal and friendlessness. Some were high achievers, some were low achievers. The breakup of a relationship was a traumatic factor strongly contributing to suicide. Many suicide victims had been exposed to suicide previously National Conference on Prevention and Interventions through suicidal siblings, friends, parents, or other relatives. Many had thought about, threatened, or made a previous suicide at- tempt. What all had in common were periods of hopelessness and thoughts of death as a solu- tion to their problems. Most of the youngsters gave clues to their suicidal inten- tions, to a peer, a family member, or a profes- sional person, either verbally or by their behavior. These clues might be construed as "reaching out for help." Clues, however, were often recognizable only in hindsight be- cause adolescents tended to camouflage them well. About half of the young suicides had recent, but brief contact with the mental health system. In fact, a major problem with young people at risk for suicide is getting them into a therapeutic contact and keeping them there. Families and therapists tend to ignore or deny clues to suicide, thus making it even more difficult for a child in trouble to enter treatment. Role of alcohol and substance use Few clinicians doubt the close association of alcohol and heavy drug use with suicide. An estimated one-half of all suicides are as- sociated with alcohol use. The effects of chronic drug use increase the likelihood of depression, despair, and feelings of hopeless- ness. Dr. Meeks points out that a great deal of depression found in chemically dependent individuals is a result of addiction rather than its cause (see Meeks, this volume). Chronic use of cocaine, for example, produces depression and dysphoria. These drug ef- fects are enhanced in adolescents who are developing emotionally and physically. Adolescents heavily involved in drugs are particularly susceptible to increasing feelings of guilt resulting from the loss of judgment and self control, alienation from families, ac- cumulating failures, and personally unaccep- table behaviors which the adolescent had to perform in order to get drugs. Although these feelings may be denied at first, con- tinued drug use coupled with the feelings of uselessness, failure, and confusion, eventual- ly may overwhelm the young person and lead to suicidal behavior as a way out of psychological pain. In addition to depression, other frequently observed factors make chemically dependent youngsters more vulnerable to suicide: a strong family history of alcoholism or other drug abuse, and a recent loss or separation (most commonly parental separation or divorce). In treating suicidal adolescent drug users, the crucial aspects in preventing suicide are recognizing when the adolescent is in crisis (e.g., stating a desire to die) and providing care and protection at the time. These patients are difficult to treat partly because adolescents and their parents resist treat- ment and try to make the process as difficult as possible. In addition, the underlying problems that led the youngster to use drugs in the first place are difficult to change on a permanent basis. They require long-term treatment and continued alertness to the recurrence of suicidal risk. A team approach treatment--involving family, friends, and peers--seems to work well for chemically de- pendent adolescents. MINORITY AND GAY YOUTH Some young people may respond to external pressures with which they cannot cope by ex- hibiting self-destructive behaviors, the most extreme of which is taking one’s own life. Suicide victims who are homosexual, belong to minority groups, and children who are un- duly influenced by violence in the media may fit this model. Many significant life events ultimately deter- mine one’s behavior, personality, and coping styles. Adolescence is a turbulent period filled with many complex physical and psychosocial developmental problems which make the transition from childhood to adul- thood difficult. A major developmental task for adolescents is to establish a stable iden- tity. Adolescents are beginning to develop sexually and trying to understand their sexual identity. Those who have homosexual ten- Report of the Secretary’s Task Force on Youth Suicide dencies are confused about their feelings and face a tremendous internal struggle to under- stand and accept themselves. Stresses related to being a member of a minority group in the United States may com- plicate the adolescent maturing process. En- vironmental factors and conflicts a young person encounters in reconciling minority cultures with the dominant American culture have been postulated as contributing factors to suicides among black, Native American, Hispanic, and Asian youth. Gay youth Gay and lesbian youth are two to three times more likely to attempt suicide than other young people (See Gibson, this volume). Gay youth face a hostile and condemning en- vironment, verbal and physical abuse, and rejection and isolation from families and peers (an estimated 25% of young gay males are forced to leave home because of conflicts over their sexual identity). The traumatic consequences of these external pressures make gay, lesbian, bisexual, and transsexual youth more vulnerable than other youth to a variety of psychosocial problems and self- destructive behavior, including substance abuse, chronic depression, relationship con- flicts, and school failure, each of which are risk factors for suicidal feelings and behavior. Help for these adolescents needs to derive from all levels of a society that stigmatizes and discriminates against gays and lesbians. For example, mental health and youth ser- vice agencies can provide acceptance and support for young homosexuals, train their personnel on gay issues, and provide ap- propriate gay adult role models; schools can protect gay youth from abuse from their peers and provide accurate information about homosexuality in health curricula; families should accept their child and work toward educating themselves about the development and nature of homosexuality. Minority youth Data show that the highest suicide rates 3-10 among minorities (Native Americans, blacks and Hispanics) occur in the younger age groups, a pattern that differs from whites among whom suicide rates increase with age. Strong cultural traditions and social support systems among minority groups are believed to play a role in protecting older age groups from suicide. Because minority groups are made up of many different cultural entities, one must be cautious about using aggregate data to generalize about a segment of a larger group. For example, Hispanics include people with Mexican, Cuban, Puerto Rican, and other Latin American heritages; Native Americans include members of more than 500 federally recognized tribes, many of whom have dif- ferent languages, customs, and cultural tradi- tions. Native Americans Dr. Thompson (see report, this volume) points out that a true picture of suicide among Native American youth may not be accurate from the aggregated data furnished by the Indian Health Service (IHS). Suicide rates vary considerably among individual American Indian tribes and data from a few tribes cannot be generalized to all Indians. Several other reasons contribute to the un- reliability of American Indian suicide data; few data are available on American Indians who live outside IHS service areas and off reservations. American Indians living in urban areas may not be correctly identified as Indians on death certificates; deaths may not be reported as suicides partly because of reluctance to bring adverse publicity to an American Indian community; and small changes in raw numbers of suicide may look very large in terms of changes in rates. The IHS, nevertheless, reports an average rate of 27.9 suicides per 100,000 Native Americans of ages 15 to 24 during 1981 to 1983; the suicide rate for all Americans 15 to 24 was 12.2 during the same time period. The base population, however, used by the IHS to calculate these rates has changed over time, thereby presenting problems in observ- National Conference on Prevention and Interventions ing suicide trends. Suicide patterns among Native Americans differ from those of their white peers in that the peak rate of suicide occurs between ages 15 through 24; white suicide rates are higher after age 35. Preventive interventions for Native Americans should focus on those com- munities which can be demonstrated by epidemiologically sound research to have a problem needing a specialized response. A better approach is needed for suicide data collection and data comparisons. Addressing the enormous cultural conflicts between the white and Indian cultures is a necessary preventive strategy. Other efforts, such as primary prevention and early recognition and treatment of the social and psychiatric conditions which lead to self-destructive be- haviors, must be planned in conjunction with the individual tribes themselves with un- obtrusive measures and with cultural sen- sitivity. Blacks Dr. Baker (see report, this volume) reviewed black suicide rates and characteristics of black suicide attempters in an effort to dis- cern reasons for the large increase in black suicide rates for 15 to 24 year olds--from 4.9/100,000 in 1950 to 11.1 in 1981. Except in specific instances, the suicide rate for blacks is roughly half as great as among whites for both males and females. Black male suicides outnumber black females by about 4 to 1. Many scientists have suggested that the lower suicide rates reflected a strong support system in traditional black culture, rein- forced by the black church and social and fraternal organizations, but that marked sociocultural changes in black families and black communities have caused those institu- tions to lose their appeal to many black people. Dr. Baker discusses several theories that have been advanced to explain the rise of suicide among blacks, but only the theories which emphasize interpersonal conflicts, familial discord, financial concerns, and the impact of poverty and racism on the in- dividual and his family seem to hold up as specific etiologies of suicide attempts and completions among blacks. Primary preventive strategies, Dr. Baker sug- gests, should focus on helping black youth understand the sources of their stress and identify effective action. For example, im- proving families’ knowledge of symptoms of mental illness and alcohol and drug abuse will allow a family to seek help before a destruc- tive episode occurs. Secondary prevention should focus on evaluation and crisis inter- vention for suicide attempters and their families, and, to prevent further suicide at- tempts, provide a family with alternatives for help in the event of a future crisis. Hispanics Suicide data on Hispanic youth were ob- tained from five southwestern States where more than 60 percent of all Hispanics in the United States reside (see Smith, this volume). The suicide rate for Hispanics (mostly Mexican-American in this geographic area), is lower than the rate for non-Hispanic whites but higher than rates for blacks. Young Hispanic males, however, in the 15 to 19 year age group, have a slightly higher suicide rate than non-Hispanic white males in the same age group. Contrary to the patterns observed among non-Hispanic whites for whom the suicide rate increases with age, the highest suicide rates for Hispanics occur in the 20 to 24 year age group. The ratio of male to female suicides is 4.3 to 1 for Hispanics. The lower overall suicide rate among Hispanics likely reflects the strength of Hispanic cultural traditions in which close family ties along with the desire not to dishonor one’s family through suicide decrease the risk of social isolation. The ex- tent to which these cultural traditions con- tinue to be held within Hispanic communities may influence the future incidence of suicide. As younger Hispanics become assimilated 3-11 Report of the Secretary's Task Force on Youth Suicide into American culture, Hispanic traditions may lose their power to influence behavior. Hispanic youths, caught between traditional values and their experiences in the larger so- cial order, coupled with the marginal socioeconomic status of this group, may ex- perience stresses that explain the younger age distribution pattern of Hispanic suicide victims. Asian-Americans Very few studies have examined suicide among Asian American youth. Available in- formation indicates that Chinese, Japanese, and Filipino male suicide rates are generally lower than those of American males except in the oldest age groups. Dr. Yu analyzed the sparse suicide data for Asian American (see this volume). By calculating proportional mortality rates, Dr. Yu showed that, within Chinese and Japanese-American popula- tions, suicides have risen dramatically be- tween 1970 and 1980. Possible explanations may be rooted in the social problems faced by young Asian Americans in their struggle to excel and establish themselves in American society. Influence of the media on suicide Associal factor often cited as having an effect on suicide is the popular media. Several re- search projects have suggested that violence on television leads to imitative aggressive be- havior by children and teenagers. Other studies suggest that televised news stories or fictional portrayals of suicide con- tribute to suicidal behaviors among young people with imitative behavior of suicides. One study reported an increase in teenage suicides during the week following news or feature stories about suicide. Another group analyzed the effect of four fictional television programs about suicide that were broadcast in the New York City area. Much advance publicity generated public awareness of youth suicide and some areas provided telephone hotline numbers and information about local crisis services. The rates of both 3-12 completed and attempted suicides rose in the two weeks following the telecasts. In his review, Dr. Berman (see this volume) believes that television’s influence on suicides is equivocal. Broadcasts portraying violence or suicide (including news reports of celebrity suicides) might influence the method of suicide in persons already predisposed to killing themselves, but are un- likely to entice nonsuicidal youngsters. As a significant part of the sociocultural milieu in which children are raised, television and other mass media have the potential to profoundly alter the message environment to which children react. The best use of the media for prevention may be in prosocial education in early childhood. Cooperation between media representatives and suicidologists might be useful in establishing guidelines for news reports and fictional presentations of suicides. Public informa- tion campaigns for suicide prevention should be guided by principles, learned from other health promotion campaigns, regarding dis- semination, targeting, timing, frequency, and duration of messages. A media campaign should be reinforced by supplementary ef- forts in the home, school, or other settings where interpersonal communication is promoted. Careful evaluations must be in- corporated into a public information cam- paign to establish rational bases for future campaigns. Religious beliefs and family structure Religion and the family are social contexts in which people are physically, emotionally, and psychologically bonded. Religious committ- ment and strong family ties, in general, provide protection from suicide by promot- ing shared values, strong social interaction and supportive connections with other people. The greater the intensity of people’s ties and connections with each other, the less chance there will be of suicide. National Conference on Prevention and Interventions Criminal and juvenile justice Each year about a half million juveniles are put in adult jails. The risk of suicide for these young persons is particularly high. Jails are frightening, intimidating environments, especially for youngsters detained for the first time. Although many facilities have in- corporated special precautions to minimize the opportunity for suicide and prevent detainees from taking their lives, the suicide rate for juveniles in jails is five times higher than the national rate. Most of the juveniles arrested are coming out of drug or alcohol intoxication, runaways, children fleeing from abuse or neglect, or are retarded, disturbed, mentally ill, or hand- icapped individuals. Approximately 10 to 15 percent of young people are jailed for violence offenses. For all of them, the first few hours of confinement are the most dangerous. Young people who have been arrested should be evaluated by a mental health professional to determine whether incar- ceration or hospitalization is appropriate. If placed in jail, careful observation, separation of juveniles from adults, and removal of per- sonal items that can be used as a means for hanging might reduce the chance of suicide. COMMUNITY EFFORTS IN SUICIDE PREVENTION AND INTERVENTION Various suicide prevention programs emerged during the 1960s and expanded rapidly during the 1970s, but little attention was paid, until relatively recently, to the specific needs of young people. Very few reports appeared in the health or social scientific literature addressing detection and treatment of suicidal behavior in the young. Among the earliest prevention concepts were the suicide prevention centers, which were consortia of psychologists, social workers, psychiatrists, and trained volunteers combining suicide research with treatment of troubled individuals. Although supported initially with Federal funds, many suicide prevention centers evolved into locally funded, community-based services. About 1,000 suicide prevention programs or crisis programs, exclusive of community mental health programs, are in operation in the United States today; about 200 are specifical- ly called suicide prevention centers. Most of the program names, however, e.g., crisis cen- ter, telephone hotline, reflect a broader, crisis intervention purpose than suicide prevention. They offer counseling, caring voices and listeners (by telephone or in per- son), and other crisis services such as short- term therapy delivered by trained para-professionals. These programs generally focus their efforts on adults, but with the increasing importance of youth suicide, some, but not the majority, of these programs have established components specifically directed at adolescents (see Comstock, Simmons, Franklin, this volume). Centers usually have referral networks--a consulting staff of health professionals, ac- cess to other community services such as law enforcement, social service or mental health agencies, and emergency medical personnel, to serve as back-up resources. After a teen’s suicide, the surviving parents and siblings experience significant stress and dysfunction. Many centers, as well as private therapists, offer grief counseling to help family members, friends, and peers to deal with the pain, guilt, anger, and other emo- tions following a suicide. Counseling sur- vivors often includes examining the emotional and mental problems experienced by the young suicide victim for the purpose of making it more difficult for survivors to identify with the dead person. Such therapy helps to lessen the chances of suicide by bereaved persons. (See Mitchell, this volume.) Role of volunteers Eighty percent of the suicide prevention centers in the United States operate with nonprofessional volunteers as their primary 3-13 Report of the Secretary’s Task Force on Youth Suicide staff. In fact, the centers are one of the few instances in which trained, lay volunteers provide clinical services that had traditional- ly been provided by professionals. The crisis worker’s role is to establish a rapport with a caller, listen to a person’s description of his problems, and work with him in setting a course of action. (Some researchers have reported that people with professional train- ing are not demonstrably more effective than lay volunteers as crisis intervention workers.) Since the early 1980s, operating standards for suicide prevention centers and training and performance evaluation criteria for volun- teer crisis workers promulgated by the American Association of Suicidology have made progress in alleviating public mistrust and professional skepticism about crisis in- tervention techniques and volunteer crisis workers. (See Wyatt, this volume.) Whether health professionals or lay volunteers serve as crisis workers, special training as well as in- service training is necessary to maintain the unique and special skills required for crisis in- tervention and suicide prevention. Caring and intelligent young people can also be involved in suicide prevention strategies (see Bolton, this volume). Young people ex- perience many losses and stresses (including body changes, moving away from friends and support groups, and living up to parental and social expectations) and need help in ac- knowledging and understanding their feel- ings and coping with sorrow and anger. With proper help and guidance, young people can take charge of their own lives, handle crisis and solve problems and feel valued and worthwhile. An independent suicide prevention or- ‘ganization is The Samaritans, founded in England in 1953. (See Katzoff, this volume.) Samaritans have 275 branches worldwide and 14 branches in the United States. The branches provide walk-in and 24-hour telephone crisis services for lonely, suicidal individuals. Volunteers offer "befriending" by listening without judging, offering un- wanted advice, or intervening without being 3-14 asked. With the caller’s permission, volun- teers can call in professional consultants. Several branches provide outreach programs such as developing school curricula on suicide awareness, distributing literature, and providing speakers for schools and universities. Despite the variety of programs and com- munity agencies available to assist young people in dealing with stressful situations, to obtain help, young people must be able to identify the appropriate community agencies (e.g. emergency medical centers, community mental health agencies, child abuse services, crisis hotlines or help centers). A high level of community awareness must be achieved and sustained over time in order to facilitate knowledge about resources available within the community which can provide services to people under various forms of stress. Effectiveness Of the wide range of interventions that ad- minister to young people at risk for suicide or who have attempted suicide, little is known about their effectiveness--whether the inter- vention can prevent suicide or suicidal risk. The conference presenters emphasized that no evaluation studies of preventive activities targeted exclusively to young people ap- peared in the scientific literature. For ex- ample, no data demonstrate that the numerous community-based suicide preven- tion programs for young people (telephone hotlines, school-based suicide education programs, peer support groups, counseling of runaways, and similar attempts) are effec- tive in preventing suicides. The same applies to psychiatric or psychological therapy and to other services that include suicide preven- tion as part of their mission, such as mental health centers, clinics, or counseling agen- cies. EARLY DETECTION AND TREATMENT Awide range of psychological, sociological or National Conference on Prevention and Interventions psychiatric theories attempt to explain suicide among the young. Adolescence is a time when youngsters experience important physical and emotional changes, feel new desires, develop a sense of identity, and break the dependent bonds formed in childhood with parents. For some youngsters, this is a time of profound sadness, stress, and loss, causing serious mental and emotional adjust- ment problems. Understanding the be- haviors and life events that precede suicide are essential to designing models for detect- ing a potential suicide, preventing the act, and treating the individual. Detection Prevention would be most efficient if we could identify some common characteristics which allow individuals who have a high probability of later suicide to be identified and brought into a prevention program. Drs. Blumenthal and Kupfer (see paper, this volume) have proposed a three-level model for detecting potentially suicidal behavior. The first level represents a detection strategy in which high risk groups are identified and "red-flagged" for tracking and educational purposes. While level I includes individuals who are not in immediate danger of suicide, they have certain risk factors such as being children of substance abusing parents, or children who have experienced extreme stress such as divorce or the recent death of a parent. Level II deals with young persons with major behavioral symptoms who do not meet criteria for a psychiatric disorder, but in whom assessment and intervention may be required. Children with emotional difficul- ties, learning disabilities, extreme aggressive- ness, runaways, and those with severe self-esteem problems fall into this group. Level III represents the detection of a psychiatric disorder of sufficient severity to require assessment and intervention by men- tal health professionals. The authors also propose a theoretical model of suicidal behavior consisting of five over- lapping "risk domains" or groups of risk fac- tors for suicide. Used together with the three detection awareness levels, the overlap model of suicide risk may be usefully applied to treatment, clinical investigation, educa- tion, and clinical intervention. The five risk domains are: 1. Psychiatric diagnosis of a patient. 2. Personality traits that relate to suicide such as aggression, impulsiveness, hope- lessness, and borderline personality disor- der. 3. Psychosocial factors, such as the strength of a person’s social supports, number of negative life events, and presence of chronic medical illness. 4. Genetic and family factors that predispose an individual to suicide. 5. Neurochemical and biochemical variables which may indicate a biological vul- nerability to suicide. The authors stress that physicians-- pediatricians, internists, obstetricians, and others outside the mental health field--need to be aware of, recognize, and document suicidal risk behavior and psychosocial stres- ses. While health care professionals deal with stressful health issues such as chronic diseases or unwanted pregnancies, they often are unaware of the additional risk imposed by other factors in the model. Clinicians, there- fore, should be educated to diagnose psychiatric syndromes and suicidal behavior, and to intervene and refer when appropriate. Treatment of adolescent suicide attempters Suicide attempters often have a number of coexisting problems--mood and conduct dis- turbances, drug and alcohol abuse, aggres- sion--which are similar to those of other psychiatric patients. Suicide attempters are avery diverse group and it is difficult to know which problems will improve in therapy, but in general, suicidal behavior may not change in the long run. In his review of treatment strategies for suicide attempters, Dr. Trautman (see paper, 3-15 Report of the Secretary’s Task Force on Youth Suicide this volume) found that no specific treatment approaches--behavioral, psychotherapeutic, or psychopharamocologic--are superior to some other treatment or to no treatment at all. Or, are there specific treatments that are applicable only to suicide attempters? Once in the mental health service system after a suicide attempt, adolescents are dif- ficult to manage and retain in treatment. Most drop out early in the course of therapy. Dr. Trautman estimates that 40 percent of adolescent suicide attempters do not have a psychological evaluation and only 20 percent complete a brief therapy program of three months. In addition, parents are often resis- tant to their child’s therapy, may deny the need for continuing treatment, and refuse to participate in the child’s treatment them- selves. Many suicide attempters have immediate problems of a brief nature which are often quickly resolved. Therefore, brief, crisis- oriented treatment and followup, on an out- patient basis, makes sense for many patients. Longer treatment is necessary for more severely disturbed patients. Good therapy for adolescents is active, explanatory, teaches problem solving and other social and be- havioral skills, and uses outside resources (cognitive-behavioral therapy meets these needs). Because parent-child conflict is the most common immediate precipitating factor of suicidal behavior, family involvement, with the goals of decreasing destructive family in- teractions and increasing communication among family members, is an essential com- ponent of successful management of suicidal adolescents. New approaches, however, need to be developed to educate families about the therapy process, reach out to those who will not or cannot come to a treatment setting, and attract minority and low-income families to come for needed therapy. SCHOOL-BASED PROGRAMS FOR SUICIDE PREVENTION School-based intervention programs are be- coming increasingly common, primarily be- 3-16 cause schools offer a good opportunity for reaching the largest number of young people. Schools are the most accessible place to make an early identification of troubled, or potentially suicidal youth. Many school in- itiatives in suicide were instituted in response to local legislation or community pressure following a wave of suicides within a par- ticular school or school district. The types of programs offered in schools and the populations they serve vary greatly. Dr. Garfinkel’s review of school programs (see paper, this volume) emphasizes that success- ful school programs should integrate an un- derstanding of risk factors for youth suicide, behavioral characteristics and clinical symptoms of suicidal individuals, and various psychosocial stressors experienced by suicidal adolescents. He proposes several components that he believes are critical in developing effective school-based preven- tion programs: * Early identification and screening by teachers and other school personnel, which includes recognizing certain be- havior patterns and stressful life events that suicidal adolescents experience. * Comprehensive psychological testing and psychiatric assessment of students identified as needing further evaluation. e Crisis intervention and management. Other individuals--for example, coaches, clergy, social workers--who may be able to provide help, should collaborate in therapy as part of a suicide prevention team, which should be present in every school. The team should act as an advo- cate for any youngster suspected of being at risk for suicide. * Programs immediately following a suicide of a young person in the com- munity. These efforts aim at preventing imitation and deemphasizing feelings of guilt, responsibility and anger from over- whelming the survivors. ® Educational programs for students, teachers, and administrative school per- National Conference on Prevention and Interventions sonnel to develop sensitivity and aware- ness of youth suicide. School programs that deal with raising awareness of the student body to suicide and its preven- tion include discussions led by trained professionals that encourage students to discuss suicidal thoughts, talk about feel- ings for friends lost to suicide, and discuss how friends can help when a troubled youngster is identified. Many school programs do not deal directly with suicide, but are designed to help youth function better in their en- vironment by developing skills to cope with stressful life events, communicate more clearly, recognize depression in themselves and their peers, and feel bet- ter about themselves. ¢ Community linkage and networking. Any given school’s suicide prevention team should link with other school dis- tricts and community social service and mental health resources in order to provide information or special resources in facilitating referrals for treatment and followup for young people at risk, or after a crisis or suicide attempt. Net- working further includes coordinating community needs for education programs, and resolving media issues dealing with public coverage of suicides. No one has been able to demonstrate that school programs directed to students or school personnel are effective in reducing suicide. In fact, school suicide prevention programs have generated controversy in some communities. Some parents fear that open discussion will introduce the idea of suicide to teenagers who were not suicidal, and some school officials believe that the many demands on the school system and limited funding for special initiatives preclude suicide prevention programs. Others, however, believe that numerous beneficial effects are possible. For example, open discussion of suicide might facilitate disclosure of some student’s preoccupations with suicide, which in turn will lead to inter- ventions to reduce the risk of suicide. Im- proving coping skills might aid in raising self- esteem, reducing school failure, and reducing depression and self-destructive behaviors, thereby generating better mental health even among persons at low risk for suicide. In the long run, we must work toward the rigorous evaluation of in-school suicide prevention programs on a large enough scale to provide statistically significant results of their effectiveness. RISK REDUCTION The cost of suicide in terms of mortality, the effects on lives saved, and the costs of health care are great. The numerous factors as- sociated with suicide are far reaching and deeply rooted in the problems of society, family, and the biological makeup of the in- dividual. Each of these issues suggests a specific set of interventions. Dr. Cantor (see paper, this volume) discusses a variety of in- terventions aimed at reducing risk of suicide through changing the environment in which young people function. She concludes that the interventions most likely to have the greatest impact on youth suicide are: decreasing the cultural pervasiveness of violence; limiting the availability of lethal agents such as firearms, medications, and drugs and alcohol; and instituting education- al programs for youth, parents, and the public. Training others who come into con- tact with young people--school personnel, primary care health professionals, youth group leaders--to be aware of the warning signs of a disturbed youngster offers a way to bring young people into the helping system early enough to avoid feelings of hopeless- ness which can precede suicidal behavior. Suicides are rare events that are difficult to predict, and effective interventions have not been identified even for the groups at highest risk--suicide attempters and psychiatric patients. Screening large high-risk popula- tions is very expensive and catches relatively few suicides. Limiting screening to smaller high-risk groups yields even fewer suicides 3-17 Report of the Secretary’s Task Force on Youth Suicide such that the overall reduction in suicide is minimal. Introducing risk reduction measures enables young people at highest risk to be identified so that intensive and specific therapeutic interventions may be provided. CONCLUDING NOTE A considerable amount of energy and good- will, human sensitivity and kindness has gone into suicide prevention activities. The costs are, for the most part, very high and little evidence demonstrates their effectiveness. Even though full knowledge of the etiology of suicide is not in our grasp and research on preventive strategies is not yet complete, the time for action is now. To postpone attempts at preventive interventions until answers are provided by experimental programs would be to ignore common sense and clinical ex- perience. We must continue with new ideas and fresh strategies, trying new approaches until evaluative studies point the way. Fur- ther, interventions must integrate the diverse interests in the field, public and private, and involve a wide variety of support systems within the youngster’s environment--family, school, business and industry, health care professionals, and social and religious institu- tions. As new data emerge, the strength of the scientific base of suicide prevention will expand. The following major topics were convened in the papers presented at the conference on prevention and interventions: 1. Better statistical data on suicide and suicide attempts by persons between ages 15 and 24. Suicide, to an unknown extent, is universally understated as a cause of death in vital statistics. This underreport- ing results from difficulties in establishing suicidal intent, practical considerations (such as the loss of insurance benefits), and the social stigma associated with suicide. Problems with determining eth- nicity of a decedent may cause under- reporting of deaths in minority groups. 2. Epidemiologic analysis of suicide patterns 3-18 to facilitate identification of risk factors, high risk groups, and trends in suicide. Epidemiologic and empirical evidence is needed to lay the groundwork for a scien- tific understanding of youth suicide. Research should focus on the nature, ex- tent, and consequences of drug and al- cohol abuse among youth, as well as the influence of mental illnesses such as depression, on risk and as precursors of suicide. . Newstrategies for primary prevention and treatment. There are few specific models of primary prevention programs and little or no information on the effectiveness of such programs on suicide. Development of new techniques in primary prevention should be encouraged and tested ap- propriately. Treatment techniques similar to those used to treat depression in adults are not, in general, applicable to young people. The mental health community must develop ways of identifying the early signs and behaviors related to suicidal in- tent and design specific interventions for those at varying degrees of risk. The ap- propriateness and effectiveness of in- dividual and group therapy must be better understood. . Research into community level interven- tion efforts. While the nonmedical com- munity has responded to suicides by establishing crisis services, such as telephone hotlines or drop-in clinics, more research is needed to investigate whether these approaches are effective and how they may be made more effective. These services need to be publicized in such a way that teenagers can identify an ap- propriate community agency (or suicide prevention hotline) to assist in coping with a stressful situation. Specialized training in suicide prevention should be provided to persons who give help to young people at risk for suicide. . Understanding the special conditions of minorities. Attention needs to be given to the unique needs of gay, lesbian, black, Hispanic, Native American, and Asian National Conference on Prevention and Interventions youth, who may perceive a different and sometimes hostile world. . Education of the public. The general public and especially those who are in con- tact with youngsters, such as parents, teachers, and other gatekeepers (includ- ing the broadcast and print media) should become aware of the warning signs and cir- cumstances that may lead to suicide. Young people should become aware that they can receive help in dealing with their problems. Special care must be taken to ensure that discussion of suicide does not become a stimulus rather than a deterrent. . School-based programs for prevention. Schools are one place for identifying youth atrisk. School personnel, working with, or being trained by professionals, should develop screening methods for identifying children who may be experiencing stresses and personal problems. Prevention and intervention techniques and curricula for educating and counseling young people at risk and their peers are also needed. Al- though effectiveness has not been proved, teaching youngsters psychological strategies such as skills for coping with stressful life events, problemsolving, decisionmaking, confrontational skills, communication skills, and building self-es- teem can be helpful for all young people. Schools should develop networks with community and professional groups such that teenagers with problems can be referred appropriately for treatment. 8. Developing services to detect and treat potentially suicidal young persons. Better methods of detection (triage techniques) in hospital emergency rooms should be developed to detect whether self-inflicted injuries are indeed suicide attempts. Protocols should be developed that provide for consistent diagnosis and treat- ment of young people suspected of having made a suicide attempt. In addition, methods should be developed to retain young people at high risk for suicide in a treatment regimen. Primary care physicians should take careful histories re- lated to personal stress, substance abuse and psychological coping skills. 3-19 COMMISSIONED PAPERS PRIMARY PREVENTION: A CONSIDERATION OF GENERAL PRINCIPLES AND FINDINGS FOR THE PREVENTION OF YOUTH SUICIDE Robert D. Felner, Ph.D., Professor of Byenolog University of Illinois at Champaign/Urbana, Champaign, Illinois Morton Silverman, M.D., Associate Administrator for Prevention, Alcohol, Drug Abuse, and Mental Health Administration, Rockville, Maryland The task that we were asked to address is one that we found both daunting and important: to provide a brief overview of the current state of knowledge in the area of primary prevention, discuss what we know about cur- rent research strategies, and link it to what we know about adolescent suicide, with a particular emphasis on the implications for future interventions and research. To begin, we need a common definition of primary prevention and then, what we mean by primary prevention of suicide among children and adolescents. The essential com- ponents of the most widely agreed upon definition of primary prevention are: 1. Primary prevention seeks to reduce the in- cidence of new cases of a disorder in a population, as well as the prevalence of that disorder; 2. A key distinguishing feature of primary prevention, when contrasted to secondary or tertiary prevention, is the timing of the intervention. That is, primary prevention efforts are by definition "before the fact" in their application, i.e., before signs of the disorder are present; and 3. Primary prevention activities are "inten- tional" and based in sound generative and executive knowledge bases (Cowen, 1983). We have learned these principles from prior intervention efforts and well articulated con- ceptual frameworks or theoretical models. Now let us see what happens when we apply these issues to youth suicide. Being less certain of our expertise regarding suicide, per se, than we were about adoles- cence and prevention generally, we turned to the work of some colleagues, particularly those who, by their inclusion on this program, we could identify as experts in the area of suicide. Our first hope was that their work would tell us just when we could first identify/ categorize someone as suicidal (so we would know whom to target and if we prevented it). We also hoped the work of others would provide us with a thoughtful perspective on the current state of our knowledge concern- ing the causes of suicide so that we could then tie all this up in a neat set of suggestions for prevention. What we found was a field where the "answers", to our questions were highly ambiguous and, indeed, the data and models available to support whatever tenta- tive answers at which we might arrive, was oftenin the formative stages of development. The state of the knowledge base specific to suicide is well articulated by Dr. Maris; he summarizes the literature thusly: "the fact remains--and this may come as a surprise to 3-23 Report of the Secretary's Task Force on Youth Suicide most readers,--that few interdisciplinary sur- veys of suicidal behavior based on systematic samples have ever been done...thus one of the major problems in understanding self- destructive behaviors is that the data base for [generating] such potential explanations is conspicuously absent" (Maris, 1981, p. 6). He goes on to note the lack of sophistication in the designs employed even in those studies that have been carried out (e.g., lack of ade- quate comparison groups). Although this volume is dated 1981, our own reading of the research which has occurred in the interven- ing time seems to indicate that little has changed. While there have been several notable exceptions--for example, the data reported by Dr. Maris in his volume and else- where (Maris, 1985), and efforts by Drs. Motto and Garfinkel (Motto, 1985; Motto, Heilbron & Juster, 1985; Garfinkel, Froese & Hood, 1982) that focus more generally on "risk" factors--the cumulative weight of evidence necessary to address our concerns seemed lacking. Indeed, if one thing seems clear it is that whatever we knew about suicide in general, we knew somewhat less about youth suicide specifically. Confronted by this state of affairs, it did not seem that we could make specific recommen- dations about the primary prevention of youth suicide because two of the basic re- quirements, i.e., an adequate generative and executive knowledge base and the presence of sound conceptual models, did not seem to be met. We soon realized there were certain issues, such as the three identified earlier, that are basic to the mounting of all prevention programs and might be extremely helpful in moving us toward developing adequate ways of thinking about the prevention of youth suicide from the perspective of a "preven- tionist." Given the current state of the litera- ture and debate about suicide, its causes, and its prevention, this literature does not allow us to answer several basic questions which need addressing if we are to mount effective prevention efforts. These include: 3-24 1. When are efforts aimed at the prevention of youth suicide primary prevention rather than secondary or tertiary prevention? 2. Who are the groups to whom interven- tions should be targeted? 3. Where do we intervene, i.e., at the system or individual level? 4. What are we trying to prevent? As shall become clear, all of these issues in- terlock. Thus, conceptual or empirical slip- page in one area may have a snowball effect. Let us turn to these issues now in the context of the youth suicide literature. The question, "when is an effort termed primary prevention?" seems to be one which, by the focus of the suicide literature, if not at- tended to, could become a source of unin- tended conceptual confusion and dead-end efforts. A careful examination of the suicide literature reveals that an incredibly broad array of factors are seen as placing youth and others "at risk" for self-destruction. Maris (1982, p. 5) tells us that "under the best of conditions life is short, periodically painful, fickle, often lonely and anxiety generating" . "only if the human condition were dramatically changed would suicide change much" (ibid., p. 6) and, finally, "suicide derives from one’s inability or refusal to ac- cept the terms of the human condition" (ibid., p- 8). The general point is that each of us may be more or less vulnerable to the stresses and strains of daily life and, under certain condi- tions of heightened vulnerability, may choose a suicidal alternative as the solution for adapting to these human conditions. By con- trast, a number of other authors, including Drs. Garfinkel and Motto, as well as Dr. Maris in some of his other work, point to more specific factors such as parental divorce, a family history of mental illness, in- creased stress and alienation among the young, being other than heterosexual in sexual orientation, the presence of depres- sive symptoms or substance abuse, prior psychiatric hospitalization, parental employ- ment history, the occurrence of non-lethal R.D.Felner: Primary Prevention... attempts or a "suicidal career pattern”, and the availability of lethal means, as all being risk or etiologic factors, most of these being empirically derived from epidemiological survey research (Garfinkel, Froese, & Hood, 1982; Greuling & DeBlassie, 1980; Maris, 1981; Maris, 1985; Motto, 1980; Motto, Heilbron, & Juster, 1985; Peck & Litman, 1973). What becomes immediately apparent from this non-comprehensive set of risk factors is that, depending on which ones you elect to include on your list or to emphasize, the answers to: "To whom do we target our preventive efforts?" "What should be the focus of the intervention?" and "When should it occur?" are all quite different. The next step is to ask whether conceptual models or issues existed in the prevention literature which could help us think about these issues as they pertain to youth suicide. Having thus redefined the questions for ourselves, we can now answer these questions a bit more clear- ly. Assuming that primary prevention is targeted toward people who are not yet showing signs of disorder, the range of risk factors iden- tified by suicidologists coupled with ques- tions yet to be resolved, such as whether unsuccessful attempts are at all comparable tolethal ones (Garfinkel, et al. 1982), present real problems for the design of preventive in- terventions. Inherent in deciding the target of an inter- vention is that the timing of prevention is critical. Given the risk factors noted and a literal interpretation of the "before the fact" nature of primary prevention, it could be ar- gued that any intervention to reduce the pos- sibility of suicide by an adolescent who has not yet successfully taken his/her own life would qualify as primary prevention. Here, the reasoning goes, since one has not yet committed suicide one does not yet have the disorder. This position may seem a bit ex- treme and certainly a caricature of what a good prevention definition would be but for the fact that it is a position reflected in much of the risk instrument development research (Maris, 1981; Motto, et al. 1985). In these works, there is much talk of how a suicidal career, prior suicide attempts, or major symptoms of depression and psychiatric dis- order may be "risk" factors. It is clear that what we are really talking about is identify- ing predictors of a future lethal attempt, not precursors of suicide that are truly discon- tinuous with the disorder. The former is, of course, an important goal. However, iden- tifying too many risk factors may be more paralytic than enabling for prevention ef- forts--particularly for the development of conceptual clarity to guide us in establishing a sound knowledge base for the prevention of youth suicide. At this point, a key debate in the prevention literature becomes salient if we are to decide how and when to move from "risk factors" to programs. We need to be clear on how we answer the questions: a) Do we attempt to tailor primary prevention programs to the prevention of a specific disorder, or b) do we develop programs which are effective in al- leviating a number of conditions that are an- tecedent to a range of emotional and physical problems, including, but not limited to the target problem? The "specific disorder prevention" model rests heavily in a classic medical-public health paradigm which views diseases as caused by specific conditions that interact with in- dividual vulnerabilities, again, specifiable. In contrast, the antecedent condition model ar- gues that at least for a wide range of emotion- al and behavioral disorders, particularly those related to stress and other elements of the normal life-course, the specific etiology model is not appropriate (Goldstein, 1985). Since many of the conditions that seem to predict youth suicide (for example, early trauma, sexual deviance, drug and alcohol problems, and negative social interactions (Maris, 1985), predict other predictors of lethal attempts (such as depression and non- lethal "gestures"), and are themselves predicted by these latter predictors, we see that a "specific etiology" strategy for preven- tion may not fit the problem. On the other 3-25 Report of the Secretary’s Task Force on Youth Suicide hand, as Maris (1985) has pointed out, the overwhelming majority of adolescents do not kill themselves. Further, we note that the overwhelming majority of those adolescents and youth who display the previously iden- tified predictors and others like them, do not kill themselves. When viewed this way, it ap- pears that a specific etiology strategy may be more relevant. To be sure, a research strategy that focuses on factors relating to differential vulnerabilities appears ap- propriate to both strategies. Resolving this dilemma at the program implementation level is more difficult. To some extent "they are both right--and also wrong." This is an issue that the working groups and the con- ference must resolve if we are to progress. We do have some suggestions, however, as to directions that may be helpful in structuring the discussion. We need to distinguish between predispos- ing conditions, precipitating conditions, necessary circumstances, and causal factors. Let us work backwards from the simplest issue, causal factors. As should be clear by now, the causes and pathways to youth suicide are multifactorial- a set of conditions on which my colleagues will elaborate further, and to which we will return shortly. Hence, to search for a specific etiology or specific set of causal factors seems somewhat futile and based more on our desire to emulate medical treatment of dis- ease than on our understanding of the phenomena with which we are concerned. Indeed, even the medical establishment has embraced the concept of health promotion targeted at alleviating broadband risk factors as a major weapon. For us to include such a strategy under the rubric of prevention is neither inappropriate nor incorrect. To the extent that specific causal factors can be identified, we are generally able to do so only on a case-by-case basis. But, by this time, we believe we are far past anything that may be construed as primary prevention. Although we may intervene in specific causal factors and call this effort prevention, in that it may result in the individual's retreating from a 3-26 lethal attempt, in every other sense the inter- vention would be labelled by the medical es- tablishment as heroic care and late intervention. Further, if we wait until specific causal or predictive factors are clear- ly present, the timing and targeting of the in- terventions can be determined only at the level of the individual case. That is not to say that such efforts are not critical to reducing youth suicide, but they are not primary prevention and attempting to label them as such will simply make the already murky con- ceptual waters even less penetrable. Necessary conditions also seem easy to deal with. The youth needs both the means and the opportunity to engage in a lethal attempt. The availability of firearms, access to motor vehicles, "inviting" high places, and certain drugs may increase the probability of a suc- cessful attempt. At this point, we should also recognize that just as a knowledge of what dosage, height, or weapon is required for a lethal attempt, a lack of such knowledge fre- quently may move what was meant to be an attention- seeking gesture into the realm of a lethal attempt. Education and access are critical, at both the individual level, e.g., in the home, and through policy efforts, such as gun control and efforts to reduce teenage drinking or substance abuse and driving. We hope our working groups will attend to these issues in developing their recommendations. In attempting to deal with the more general antecedents of youth suicide--predisposing conditions and precipitating conditions-- there appear to be more opportunities for true primary prevention, at least when con- trasted to the specific etiology approach. As noted above, we may be required to decide first whether youth suicide is a specific phenomenon or part of more general be- havioral or emotional problems such as depression, acting-out problems, risk-taking behaviors, and other self-injurious behaviors such as alcohol and substance abuse. It may be that we decide that suicide is both, i.e., that it has a large degree of shared variance with these other conditions as well as its own unique attributes, especially the intentional R.D.Felner: Primary Prevention... use of lethal methods. Certainly, any clinician who has spent more than a few hours with adolescent clients and has seen their depression and sadness following the break-up with a boyfriend or girlfriend, or after a fight with their parents, knows the fre- quency with which they voice the wish to die out of revenge or for dramatic effect, is not low. Thus, if we find that the antecedents of some youth suicides are much the same as conditions predisposing to depression or ac- ting-out behaviors, it should not surprise us. Indeed, one only has to remember adoles- cence and the verbalizations or judgments of our own friends to be surprised that more youth do not actually kill themselves. These observations make us realize that we should focus more attention on why more youth do not engage in suicidal behaviors, i.e., what conditions make them less vulnerable. To direct our inquiry and develop a greater un- derstanding of the antecedents to youth suicide requires a perspective on prevention that will guide the questions we ask and will systematize the answers we obtain. Elsewhere (Felner, Farber, & Primavera, 1983; Felner, Ginter, & Primavera, 1982; Felner & Lorion, 1985), the senior author of this paper and his colleagues as well as others (e.g., Lorion, 1983; Seidman, 1986) have begun to elaborate a model that we believe helps to meet these needs and is especially appropriate for the prevention of youth suicide given what we have discussed thus far. The model defines preventive interventions within a developmental framework and will allow us to view the full range of levels of preventive interventions for youth suicide. Within this model, a preventive intervention involves systematically altering the processes related to 1) the development of adaptation and well-being and 2) the evolution of dys- function. The goals, quite clearly, enhance the former processes and reduce the latter processes that are experienced by children and adolescents. Further, given the em- phasis accorded to the processes underlying the evolution of a specific disorder or set of adaptive difficulties, a developmental model that is transactional in nature (such as that of Sameroff and Chandler, 1975), and that em- phasizes the necessity of understanding the ecological context in which the child or adolescent is attempting to adapt, is clearly the paradigm of choice for moving toward more specific variables of concern. Such a transactional-ecological perspective em- phasizes that dynamic transactions between individual and environmental factors lead to health or disorder, and specifying "a path" is not an outcome we should seek. Rather, specifying ways of understanding the relevant processes are of paramount impor- tance. Moreover, to paraphrase Sarason and Doris (1979), it is worth emphasizing from a transactional perspective, that the individual and his environment can never be under- stood separately...from a transactional perspective the direction of the developmen- tal influences is always reciprocal. A brief example of how this approach may be combined with our notions of predisposing and precipitating factors to understand youth suicide may be helpful here. One of the "hot" media issues of the past few years has been "cluster" suicides among youths in the same school or town. The predisposing conditions for the youths who follow the first suicide may be depression or other psychiatric problems, low family support and/or a high degree of alienation from friends and family. All of these conditions both contribute to and are contributed by the emotional problems present. Children’s feelings of not belonging in school or having a restricted future be- cause of doing poorly in school may have similar impact. These latter conditions might result from the social climate of the school, its structure, or recent school transitions im- posed by the system. Thus, individual, fami- ly system, and broader social system (e.g., peer/school) factors may all contribute to predisposing the adolescent to be vulnerable at this time. Nonetheless, these adolescents in general, have not yet considered suicide seriously or attempted it as a solution to their adaptive difficulties. Indeed, they may have attained relative equilibrium, however un- satisfactory, in their coping efforts. 3-27 Report of the Secretary’s Task Force on Youth Suicide Into this pot we put not only the model of another adolescent demonstrating that suicide may be an alternative coping strategy, but the response of the social system to that event. For a lonely, highly stressed adoles- cent, with the particularly strong needs for identity and acceptance that characterize this age, the overwhelming attention and grief that the system and other persons pay to the suicide, may be sufficiently attractive and satisfying to many of the adolescent’s im- mediate needs at a level that results in dis- equilibrium and, in turn, precipitates an attempt. It is this change in equilibrium which seems to discriminate between those who make a lethal attempt from those who do not, and for whom it is so difficult to specify a distal causal pathway. In this situa- tion, more proximal factors seem more salient. If access to lethal means is relatively easy (e.g., if guns and ammunition are avail- able in the home), the probability increases that a momentary, impulsive act may occur. Similarly, anything which enhances the at- tractiveness of suicide as a coping strategy may also tip the scales (e.g., another incident with resulting publicity, a fight with one’s parents or significant other, etc.). That en- vironmental factors may contribute here is obvious from the pattern that seems to indi- cate that cluster suicides tend to be less com- mon in large systems (e.g., schools, towns) where anonymity is greater, where there may be less publicity, and the possible pay-off, in terms of the adolescent’s own developmen- tal issues, seems less certain or clear. This ex- ample illustrates the very complex interplay of developmental and environmental cir- cumstances and the individual’s own limited range of coping ability. Further, we see that: -(1) both predisposing and/or precipitating conditions may be necessary for many youth suicides to occur and; (2) even when both sets of conditions are present, in most instances, suicide does not occur. Perhaps even more ironic is that children experiencing these conditions may be more "at risk" in systems in which we might assume lower risk; i.e., small, cohesive ones. Similar examples may be developed for the high rate of suicides among 3-28 children who seem to be doing very well academically or socially (Maris, 1985) as well as those who are more obvious risks for the full array of psychological and self-injurious difficulties that plague adolescents. The above may make it seem as if primary prevention programs targeted to youth suicide will have little pay-off and that broad scope suicide prevention programs may ac- tually influence the actions of only a small group of children. We would like to argue that such views are short-sighted but, given base rates, perhaps natural conclusions will result from specific etiology or specific out- come-targeted prevention programs. What we mean is, if we develop broad-based prevention programs for the prevention of youth suicide, with our only goal being the reduction of actual cases of suicide, the very limited resources available for medical and mental health programs may force us to con- clude that what little funding is directed toward prevention of youth suicide may be better spent. What strikes us as particularly ironic is that although other federal agencies such as the Department of Defense and NASA take pains to convince us that we get far more for our dollar, especially our R & D dollar, than the targeted "product", we in human services, at least at this time, are going the other way. We fail to see the harm in reducing the in- cidence of school failure, adolescent depres- sion, non-lethal suicide attempts, and the rest of the range of health-risk behaviors that adolescents engage in, while we also attempt to reduce the suicide rate. Indeed, if we fol- low the model we have advanced above, such multiple-outcome effects seem expected and, if the programs are effective, un- avoidable. Documenting the positive as well as negative, but unintended or "ancillary" consequences of suicide prevention programs should not be discouraged. In- deed, studying the full range of outcomes as- sociated with such programs may provide important clues on how to maximize our ef- fectiveness in the prevention of youth suicide per se. R.D.Felner: Primary Prevention... If one implements the preceding transaction- al model with the goal of reducing antece- dent or predisposing conditions and influencing the individual’s threshold of vul- nerability to these conditions in addition to preventing youth suicide per se, then the lack of information specific to the effective prevention of youth suicide becomes less of a barrier to action. Programs which repre- sent both levels of primary prevention, as outlined by Cowen (1985), may be effective as well as essential for dealing with youth suicide. The first level is system-focused with the emphasis on understanding and modify- ing the multiple social systems that affect the mental health and well-being of individuals in ways that reduce stress and increase life opportunities. Examples that have been powerful influences on the general mental health and well-being of adolescents are a) policy interventions which increase the sense of control youth feel they have over their fu- tures, b) modifications of major ecosystems in which adolescents must function, such as schools, and/or c) efforts at grass-roots level interventions, such as the much publicized group of New York city students receiving free college educations from an alumnus of their grade school. Similarly, second-level primary prevention programs, which are more person-focused, may also be effective in achieving our goals. Rather than influencing the individual's threshold of vulnerability by making systems to which a person must adapt less difficult or reducing the levels of challenge that the en- vironment poses so that the adolescent’s pre- existing competencies may ensure well-being, second-level programs seek to enhance the problem-solving and coping skills of the individual more directly. Educa- tion programs, skill-building curricula, and resource/support networks may accomplish these goals. Both strategies may be applied to either general populations of youth or youth in specific high-risk circumstances, e.g., school transitions, parental divorce, households with a parent with serious emotional distur- bance. The elaboration of the specific sys- tems, competencies and vulnerabilities that need addressing, and the definition of the predisposing and/or precipitating conditions that may be appropriately "co-targeted" with youth suicide, we leave to our colleagues. For the general purposes of prevention, if we adopt a developmental-transactional- ecological perspective, several points need to be addressed further: 1. When and how do we identify a child or adolescent who is the appropriate concern of a primary prevention program for youth suicide? 2. Are antecedent conditions the ap- propriate and/or necessary targets of such programs? Are adolescent suicide and the search for specific etiologies the only con- ditions with which we should concern our- selves? 3. If we choose broad-based preventive ap- proaches, as we propose, how can we draw on our understanding of general develop- mental data and models, as well as preven- tion programs that have been developed for less focused outcomes, such as as the problem-solving approach of Spivak and Shure and the senior author’s own transi- tion program efforts, to implement and evaluate effective primary prevention ef- forts. REFERENCES 1. Cowen, E.L., (1985). Person-Centered Ap- proaches to Primary Prevention in Mental Health: Situa- tion-Focused and Competence- Enhancement. American Journal of Community Psychology, 13, 31-48. 2. Felner, R.D., Farber, S.S., & Primavera, J. (1983). Transitions and stressful life events: A framework for preventive efforts. In: R.D. Felner, L.A. Jason, J.N. Morit- sugu, & S.S. Farber (Eds.), Preventive Psychology: Theory, Research and Practice (pp.199-215). New York: Per- gamon Press. 3. Felner, R.D., Ginter, M.A,, & Primavera, J. (1982). Primary prevention during school transitions: Social sup- port and environmental structure. American Journal of Community Psychology, 10, 277-290. 4. Felner, R.D. & Lorton, R.P. (1985). Clinical child psychology and prevention: Toward a workable and satis- tying marriage. In: J.M. Tuma (Ed.), Proceedings: Con- ference on Training Clinical Child Psychologists. 3-29 Report of the Secretary's Task Force on Youth Suicide {pp-93 98. Baton Rouge: Section on Clinical child ychology, American Psychological Association. 5. Garfinkel, B.D., Froese, A., & Hood, J. (1982). Suicide attempts and adolescents. American Journal of Psychiatry, 139, 1257-1261. 6. Goldstein, M. (1985). Comments on the possibility of primary prevention in mental health. In: R.L. Hough, P.A. Gongla, V.B. Brown, & S.E. Goldston (Eds.), Psychiatric Epidemiology and Prevention; The Possibilities (pp. 65- 70). Washington, D.C., National Institute of Mental Health. 7. Greuling, JW. & DeBlassie, R.R. (1980). Adoles- cent suicide. Adolescence, 15, 589-601. 8. Lorion, RP. (1983). Evaluating preventive inter- ventions: Guidelines for the serious social change agent. In: R.D. Felner, L.A. Jason, J.N. Moritsugu, & S.S. Farber. Eds.)., Preventive Psychology: Theory, Research and ractice. New York: Pergamon Press. 9. Maris, R. (1981). Pathways to Suicide: A Survey of Self-Destructive Behaviors. Baltimore: The Johns Hopkins University Press. 10. Maris, R. (1982). Rational suicide: An impovershed self-transformation. Suicide and Life Threatening Be- havior, 12, 4-16. 11. Maris, R. (1985). The adolescent suicide problem. Suicide and Life Threatening Behavior, 15, 91-109. 12. Motto, J.A., Heilbron, D.C. & Juster, R.P. (1985). Development of a clinical instrument to estimate suicide risk. American Journal of Psychiatry, 142, 680-686. 13. Peck, M. & Litman, R. (1973). Current trends in youthful suicide. Tribuna Medica. 14. Sameroff, A.J. & Chandler, M.J. (1975). Reproduc- tive risks and the continuum of Sate saking Saustany: In: F.D. Horowitz, E.M. Hetherington, S. Scarr-Salapatek, & G. Steigal (Eds.), Review of Child Development Research. Chicago: University of Chicago Press. 15. Sarason, S.B. & Doris, J. (1979). Educational Handicap, Public PSicy and Social History: A Broadened Perspective on Mental Retardation. New York: Free Press. 16. Seidman, E. (1980). Toward a framework for primary prevention research. Proceedings of the Preven- tion Research Methodology Workshop: Washington, D.C., National Institute of Mental Health. 3-30 A CRITICAL REVIEW OF PREVENTIVE INTERVENTION EFFORTS IN SUICIDE, WITH PARTICULAR REFERENCE TO YOUTH SUICIDE D. Shaffer, M.B., B.S., F.R.C.P., F.R.C. Psych., Director, Division of Child Psychiatry, Professor of Psychiatry and Pediatrics, Columbia University, New York, New York K Bacon, Ph.D., Clinical Psychologist, New York State Psychiatric Institute, New York, New York INTRODUCTION The Purpose of this Review The principal goal of this review is to identify studies which have used reasonable methodologies to evaluate the success, or otherwise, of prevention activities in youth suicide. We have, however, identified no such studies. Given the quantity of preven- tive activity that is being conducted, this is a matter of concern. Rather than simply con- cluding that the lack of research is critical in this area, we have extended this inquiry to summarize what is known about the value of suicide prevention activities for other age groups. Many activities intended for adults may also be appropriate for teenagers; we hope this review will provide some guidelines for those concerned with preventing teen suicide. General reviews of the suicide prevention literature which the reader may find valuable, include Motto et al., 1974; McGee, 1974; Stelmachers, 1976; Auerbach and Kilmann, 1977; Stein and Lambert, 1984. Defining Prevention The classification of suicide prevention fol- lows the classification of suicidal behaviors. If suicidal thoughts, attempts, and comple- tions are on a psychological or behavioral continuum, as common sense would suggest, then primary prevention efforts should be broadly directed, aiming to reduce all suicide morbidity including threats and attempts, for all are signs of suicide potential. On the other hand if attempts and completions are separate but overlapping entities as has been by proposed by Stengel and Cook (1958), Neuringer (1962) and others, and if only a minority of those who attempt or threaten suicide really want to die, then the focus of "primary prevention" should be that sub- group of attempters or threateners who bear a "high risk" profile for later completion. This question is sufficiently important that it isworth considering the evidence for the con- tinuous versus the separate theories of suicide and attempted suicide. Demographic Differences Between Suicide and Attempted Suicide. The case for the two disorder hypothesis relies on both the manifest ambivalence of many survivors and the marked demographic differences between suicide attempters and 3-31 Report of the Secretary’s Task Force on Youth Suicide completers which have been shown in most studies (Dublin, 1963; Sainsbury, 1955; Ken- nedy et al., 1974). Suicide attempts are more common in females than in males whereas suicide, especially in the young is more com- mon in males than females. In our current New York study, the ratio of males to females for under 19 year olds is approximately 4:1 for completed suicides. National statistics for 1981 reveal a male:female ratio of 4.35:1 among 15-24 year olds, and 3.6:1 for the total population of all ages. Although these statis- tics rely upon medical examiners’ determina- tions, there is evidence that more aggressive case finding methods do not materially alter the ratio (Kennedy et al., 1974). Another demographic index which is believed to discriminate between attempts and completions is AGE. Studies under- taken 2 to 3 decades ago indicated that the incidence of suicide attempts peaked in the teens and early twenties (Kennedy et al., 1974) and then declined, whereas completed suicide became increasingly common with advancing age. This age discrepancy has now diminished, at least for males, as completed suicide is now more common in younger than in older males and no evidence suggests that the previously identified pattern of high in- cidence of suicide attempts among the young has changed. While the epidemiological data are undoub- tedly accurate, one cannot infer from them that the outcome of suicidal behavior, i.e., whether it is successful or unsuccessful, defines two different conditions. This is be- cause both age and sex are related to method preference which in turn, is closely tied to outcome. Age-related method preference is not well documented, but what evidence exists sug- gests younger children and teenagers over- dose with less lethal drugs than adults. Morgan et al. (1975) found that teens most commonly overdosed with over-the-counter analgesics, whereas older patients favored more dangerous psychoactive drugs (usually 3-32 obtained by legitimate prescription from a physician). This difference might be ex- pected to result in a smaller proportion of suicides attributable to overdoses in the young, which is precisely what is found (Centers for Disease Control, 1985). Females are more likely to take an overdose (Weissman, 1974), males to use firearms or to hang themselves. Given current medical skills, overdose is generally an ineffective way to commit suicide. It could be argued (and usually is) that these differences in method preference arise because suicidal females are less motivated to die and knowingly choose less lethal methods. The difference in lethal intention is cited as further evidence of a two disorder hypothesis. It may be, however, that the sexes are similar in their (generally ambivalent) motivation to die, and that choice of method is a sex-related behavior. That is, the sex differences be- tween completed and attempted suicide may not reflect different degrees of intention, but rather that when the sexes feel suicidal they do different things about it. These different things, at least in North America and Western Europe, have a high probability of leading to death in boys and a low one in girls. Interestingly, a report of consecutive suicides in India (Sathyavati, 1975) shows no sex dif- ferential, suicides being as common in teenage girls as in boys. One could speculate that this occurs because resuscitation methods are less effective in that country. The gender association of method could be a sex-specific behavior preference without psychopathological significance; or, it could be mediated by difference in psychopathol- ogy in suicidal males and females. Some evidence supports this. In our New York study we find high rates of aggressive and an- tisocial behavior and relatively low rates of pure major depressive disorder in boys who have suicided, but the reverse in girls. One cannot infer from this difference, however, that intention to die is different in the two conditions. D.Shaffer: A Critical Review of Preventive Intervention.. Direct Evidence for Continuity Between Suicidal Thoughts and Behaviors. Paykel et al. (1974) in their New Haven study of a household probability sample of adults, posed questions about different degrees of suicidal ideation and behavior. These were strongly interrelated in a hierarchical fashion. Almost all subjects with more severe symptoms had experienced those that were less severe. More interesting for the present argument is that the correlates of dif - ferent levels of severity were similar. In- dividuals who wished they were dead resembled those who had actually made a suicide attempt (the two extremes of the con- tinuum) with respect to both demographics and associated symptoms. Pfeffer et al. (1984) reports similar findings in school children. Those who had thought of dying were as deviant as those who had made a suicide attempt and showed a similar profile of associated symptoms. There is also evidence for overlap between suicide attempts and completions. Most retrospective psychological autopsy studies show prior attempt rates of between 30 per- cent and 50 percent (Shaffer, 1974; Kennedy et al., 1974; Robins et al., 1959; Dorpat & Ripley, 1960; Barraclough et al., 1970). Con- versely, followup studies of attempters show suicide rates 50-60 times that in the general population. Poor Predictive Specificity of Attempter Characteristics. If attempters and completers are drawn from the same population, we would expect dif- ficulty in predicting future completions among suicide attempters. Although not ex- tensively studied, the evidence supports this prediction. Motto’s (1984) 5 to 15 year fol- lowup of teenagers admitted to hospitals after an attempt or with serious depression showed that although certain factors were proportionally more common in those who would go on to suicide, the same factors were numerically many times more common in at- tempters who would not, i.e., the base rate in non-completers was high and there were no pathognomonic features for later comple- tions. The same has been found in followup studies of adults where, not only demographic characteristics, but the extent to which the suicide attempt could be judged to be serious (i.e., isolation during the commission of the attempt and its medical seriousness) were not predictive of later death (Greer and Lee, 1967). Discriminant function studies which identify differentiating characteristics in suicides and attempters (Pallis et al., 1982) draw upon different population bases, and it is not clear whether the apparent inde- pendence is a function of different condi- tions or different population frames. Suicide Accelerators Affect Both Deaths and Attempts. We cite evidence below (Gould and Shaffer, 1986) that certain television programs which dramatize the plight of the suicidal teenager serve to increase both suicidal deaths and suicide attempts. COMMENTS: Given these uncertainties it seems wisest to adopt a conservative ap- proach and regard any suicidal behavior as presaging completion. Primary prevention would prevent the initial occurrence of suicidal ideation or behavior; secondary prevention would prevent non-lethal suicidal behavior from progressing to death. We have adopted this approach in organiz- ing this paper and have grouped as primary preventions: 1. Altering the set towards suicide in unaf- fected individuals by, for example, provid- ing information on suicide behaviors in classes to normal non-disturbed school- children, or in special services for sur- vivors. 2. Early identification and treatment of con- ditions which are known to predispose towards suicidal behavior, before suicide is contemplated. 3-33 Report of the Secretary’s Task Force on Youth Suicide Secondary preventions should reduce the potential for completion among those who have already threatened or attempted suicide, through: 1. Removal of the means for committing suicide. 2. Emergency crisis interventions at times of maximal distress. 3.0Ongoing treatment after the crisis has passed. PRIMARY PREVENTION Preventing Suicidal Behavior in Vulnerable Groups General Psychiatric Care We have argued above that the feature most likely to be shared by suicides is a history of mental illness. It follows, other things being equal, that the introduction of psychiatric services to a community should reduce the burden of mental illness and with it the suicide rate. This has not been found to be the case. Neilson and Videbech (1973) examined the impact on suicide rates (all ages) of the intro- duction of a psychiatric service to the island of Samso off the coast of Denmark. There were no differences in suicide rates during the 5 years before and after the introduction of the service. Similarly Walk (1967) ex- amined suicide rates in the county of Sussex in Great Britain before and after the intro- duction of a community service and found no effect on suicide rates. COMMENTS: These studies are often cited as evidence that psychiatric treatment does not reduce suicide morbidity. However, neither of the studies had a control and so it is possible that apparently stable rates were occurring at a time of a more general rate in- crease. This is unlikely to be true in the Walk study however, because at the time, suicide rates were declining in Great Britain. More importantly the studies were conducted before the widespread use of antidepressants 3-34 and lithium and therefore, do not reflect the impact of current effective antidepressant therapies. High Risk Groups Prevention would be most efficient if we could identify individuals who have both a high probability of later suicide and some common characteristics which allow them to be centrally identified and taken into preven- tion programs. If we believe that suicide arises because of a set of social circumstances or life conditions, and that to commit suicide is in any way a “reasonable” response to untenable life con- ditions, then the high risk group would most likely be accessible to social rather than men- tal health agencies. However, there is good evidence that no matter how understandable asuicide may be to an outsider, it is almost al- ways a sign of psychopathology. Primary prevention is, therefore, most appropriately an activity for the mental health professional. Brown (1979) summarized this view succinct- ly"...although psychiatric disorder may not be sufficient cause for suicide in current Western cultures, it is a necessary one..." The evidence derives from a number of sources: a. Psychological autopsy studies on repre- sentative groups of suicides (Dorpat and Ripley, 1960; Robins, 1959; Barraclough et al., 1974; Shaffer, 1974) have found very few suicides to be free of psychiatric symptoms. b. There have been similar findings among suicide attempters (Morgan et al., 1975; Birtchnell and Alarcon, 1971; Silver et al., 1971). c. A majority of suicides have had contact with a mental health professional before their death. d. Followup studies of formerly hospitalized psychiatric patients indicate that they have significantly higher suicide rates than non- patients (Temoche, 1964; Pokorny, 1964, 1983). D.Shaffer: A Critical Review of Preventive Intervention.. e. Super-normal control groups, created by screening out individuals with psychopathology have very low suicide rates (Winokur and Tsuang, 1975). It follows that the most appropriate group for preventive interventions are individuals with current or previous psychiatric disorders--a dauntingly large group. However psychological autopsy studies carried out on adults have found that a rather narrow range of associated psychiatric disorders (affective disorder and alcoholism) account for most suicides. Similarly, diagnostic analyses of child and teenage suicides carry the promise of defin- ing more specific groups. Shaffer (1974) found that a small proportion of suicides (predominantly girls) are depressed and that a large group show a combination of affective and antisocial behaviors. The epidemiologi- cally-based study we are currently carrying out in the New York metropolitan area will provide detailed DSM III related diagnostic information on approximately 150 teenage and child suicides and will, we suspect, con- firm the findings of the other studies. We also hope that it will allow us to define the suicide group more precisely by seeing whether other characteristics such as a fami- ly history of suicide, and specific family con- stellations and social circumstances, are more common in suicides. The highest risk group, however, appears to be individuals, who have made a prior suicide attempt. Suicide Attempters or Depressives It is generally assumed that the diagnostic group with the highest risk for later suicide are individuals suffering from an affective disorder. Lists of warning signs generally in- clude the symptoms of depression and find- ings are generally consistent in adults relating suicide to depression. Temoche et al. (1964) and Pokorny et al. (1964) both found that suicide was significantly more common in previously hospitalized patients who had received a psychiatric diagnosis of depressive psychosis. Robins (1959), reviewing studies which mainly dated from before the widespread use of lithium and antidepres- sants computed an overall 15 percent suicide rate for manic depressives. However, our work with adolescents (Shaffer, 1974) sug- gests that only a minority of suicides show a picture of uncomplicated depression and that the largest diagnostic group comprises youngsters with both aggressive and antiso- cial symptoms and depression. Depression may not be a very specific group for later suicide. Major Depressive Disorder (MDD) has an estimated one year prevalence of 2,000/100,000 in adolescence (Andersonet al., 1985; Weissman et al., 1985; Kashani et al., 1983). In our current New York study we have found that only 25 per- cent of suicides meet criteria for Major Depressive Disorder giving a one year in- cidence of 3/100,000. In one year the ratio of depressed teenagers to depressed suicides would be approximately 660:1 (higher for females). This high risk group would not only identify many false positives, but also non- specificin failing to identify approximately 75 percent of suicidal deaths. Attempted suicide would seem to be a better bet, although not strictly within the domain of primary prevention. We have already referred to studies that found that a sig- nificant proportion of suicides made a pre- vious known suicide attempt, i.e., the attributable risk among suicides for prior at- tempted suicides is high. The relative risk among attempted suicides for later suicide is, perhaps, a more important statistic. If there is an effective way of aborting the natural his- tory of suicide attempters which terminates in suicide, this ratio will indicate the mag- nitude of the task. One way of finding this out is to examine the relative frequency of suicide and suicide at- tempts in an unselected population. We have found no studies which have generated age and sex-specific attempt and suicide rates for the same area. Rough and ready calcula- tions can be done and we have tabulated data from studies into the incidence of suicide at- tempts. Paykel (1974) found a one-year prevalence of suicide attempts of 3-35 Report of the Secretary’s Task Force on Youth Suicide 600/100,000 (The prevalence of suicidal ideation was 9000/100,000). Johnson et al. (1973), in a survey in London, Ontario found an attempt rate of 750 to 1,500/100,000. Given an overall suicide rate of 12/100,000 in adolescents this would put the ratio of at- tempts to deaths at 50 to 120:1. These figures are not corrected for age or multiple at- tempts and it is expected that the ratio would vary in different age and sex groups, being higher in teenage girls. Nevertheless these ratios are a good deal lower than those for major depressive disorder (MDD). A better strategy is to identify all of the known attempts within a given geographical area. Studies approaching this goal, which have included studies of non-accidental drug overdoses by Morgan et al. (1975) in the British city of Bristol and Daly et al. (1986) in the Irish city of Cork, generated age-specific attempt rates. Neither study evaluated suicide attempts treated by non-clinic based general practitioners and both confined their study to overdoses. The rates are therefore likely to be an underestimate of the true rate. However, if the same attempt rates prevailed in the United States it might appear that be- tween 30 and 50 attempts at suicide for every completed male suicide, and between 150 and 300 suicide attempts for every completed female suicide. These would provide quite reasonable rates for focusing on an at-risk population. Suicides, however, are not drawn uniquely from the pool of known suicide attempters. Studies in different age groups and in dif- ferent countries are surprisingly consistent in showing that only between 25 and 40 percent of suicides have made a previously known suicide attempt. These rates were found by Shaffer both in his British study of children under age 15 (1974) and in his U.S. study of predominantly older teenagers. Similar rates were reported for adult suicides in the U.S. (Dorpat and Ripley, 1960; Robins et al., 1959) in England (Barraclough et al., 1974) and in Scotland (Kennedy et al., 1974). (There is as yet no information about whether the proportion of deaths at- 3-36 tributable to prior attempts varies with sex or ethnicity.) The ratios of attempts to completions should, therefore, be modified and minimum es- timates would range from around 60:1 for older male teenagers to approximately 600:1 for younger female teenagers. These are for- midable ratios and certainly indicate that if we can find out how best to treat these patients we should put most emphasis on male adolescents generally and older ones most specifically. Case Finding Effective prevention requires that vul- nerable groups be identified and apprised of the preventive intervention. Although many adult suicides are known to be in contact with a treating physician shortly before their death (see below), this finding may not apply to children and teenagers. Preventive efforts for this age group have characteristically con- centrated on identifying potentially vul- nerable cases in schools or through the public media. Case Finding Through Physicians It appears that many adult suicides contact their physicians shortly before their death, suggesting that preventive identification channeled through primary health providers may be effective. Barraclough et al. (1974), in their British study of 100 consecutive suicide completers found that just under 50 percent had visited their physician during the week before their suicide. Murphy et al. (1975) and Motto and Greene (1958), in their study of previously hospitalized suicide attempters and depres- sives found that a high proportion of suicides (all ages, predominantly adult) had visited a physician shortly before their suicide, 17 per- cent in the month before death. The dif- ference could be due to the ex-patients’ insight into the psychiatric nature of their condition, or to the different nature of the health services in Britain (with minimal char- ges and universal enrollment with a primary D.Shaffer: A Critical Review of Preventive Intervention.. physician) compared to the United States. The same is true in the case of attempted suicides. Motto and Greene (1958) reported that 60 percent of suicide attempters had consulted a physician during the 6 months before their death. Johnson et al. (1973), in a followup of 878 attempter cases, found that 55 percent had seen their physician during the month before their death. Morgan et al. (1975) found that 21 percent of a sample of consecutive suicide attempts seen at an emergency room in Bristol had contacted their physician during the week before their attempt and more than half had done so within three months of their death. Studies are also consistent in showing that the physicians contacted are not aware of their patient’s suicide potential. In Murphy’s series, two-thirds of the suicides had a prior history of suicide threats or attempts, yet only 40 percent of their physicians knew this. A high proportion of the attempters in Johnson’s study had made a previous suicide attempt, but again, only 20 percent of the af- fected physicians knew of this. Motto (1969) has suggested that physicians’ own anxiety about suicide or about handling a potentially suicidal situation inhibit them from making appropriate inquiries. Johnson suggested that physicians tend to view suicide attempts in the same light as alcoholism, as a repetitive self-induced disorder for which they can do little in the face of absence of motivation by the patient. We do not know if the same is true for teenagers, but clearly it would be advisable for physicians to inquire routinely whether any depressed or suicidal patient has made a previous attempt, has ever been hospitalized for a psychiatric condition, or is a heavy user of alcohol or drugs, all factors which appear to increase the probability of suicide. Case Finding Through School-based Programs Description of School Programs School-based intervention programs are be- coming increasingly common, often being es- tablished by local or State legislation, as a result of community pressure, following a wave of suicides. Most aim to: a. Increase the sensitivity of responsible in- dividuals within the school to the features of the suicide-prone child. Methods in- clude lectures, videotaped interviews with teenagers who have made a previous suicide attempt, small discussion groups, and the distribution of lists of "early warn- ing signs". b. Provide information about special resour- ces where pupils suspected of being at risk may be referred for treatment or help. c. Provide some training in the behavioral skills that teachers, counsellors or other pupils can use when they identify a child at risk. The goal is to establish a relationship of trust and support with the child at risk, encourage open communication of trou- bling thoughts, and make it easier for the child at risk to accept a referral to a specialist resource. d.Lower the constraints children and teenagers have about discussing suicidal thoughts and preoccupations, thus en- couraging self disclosure. This may be done by raising the subject in open group discussions, sometimes with known dis- turbed youngsters in attendance (Ross and Motto, 1984). These discussions may take place in large assemblies, in smaller groups of 10 to 20 students, or in the context of a regular class. Such programs are variously addressed to groups of school personnel, parents, and/or students. Some school districts have ex- pressed reluctance to offer "suicide educa- tion" to students and have, instead, developed programs or curricula dealing with the more general topic of adolescent stress. Evaluation of School Programs One of the most striking features of this review is the almost complete absence of any systematic evaluation of in-school programs. Ottens (1984) describes responses to a ques- tionaire distributed before and after a 4 hour 3-37 Report of the Secretary’s Task Force on Youth Suicide course to 31 college counselors and students. Post-course scores on questions about ap- propriate responses to a hypothetical crisis situation more closely approximated the responses that were taught in the class. Cor- relations are described but no statistics are presented and there is no evidence that the paper and pencil responses are in any way representative of real skills. In fact, this seems unlikely as student responses were ap- parently similar to those of experienced therapists. Descriptive material prepared by the program in Fairfax County, Virginia, notes that during its first year of operation there were 5 deaths and during the second year only 3. It acknowledges that this change could be due to coincidence. The publica- tion states that the administrators believed the program was going well and stated, anec- dotally, that since it started, school-based counselors had experienced an increase in referrals and school staff had become more sophisticated in identifying pupils in trouble. However, as Stein and Lambert (1984) have demonstrated, counselors’ self-evaluations (Apsler and Hoople, 1976; Getz et al., 1975) of their own usefulness and efficacy tend to be optimistic, are prone to bias, and cannot be accepted as reasonable evidence of ef- ficacy. Ross (1980), describing the San Mateo coun- ty program, reports (anecdotally) that the in- troduction of the program led to an increase in referrals to a suicide prevention program. Workers in this field often comment on the absence of systematic evaluations, but they usually state, with some justification, that staff who are excellent at intervening may not have the necessary skills to do a sound evaluation. Others comment that the problem is too urgent to await the results of any research evaluation. Comments on School Programs Given the dearth of systematic evaluation of these important programs it is reasonable to comment from general principles. a. Early warning signs: Most are brief lists 3-38 which emphasize the features of a recent onset of depression, i.e., changes in mood, decreasing sociability, decline of school performance, increasing irritability, and some specific behaviors such as making suicide threats and giving away possessions. They have presumably been based on either the collective experience of professionals who have investigated individual cases of suicide, or on stereotypes. They are not derived empirically from representative samples, quite simply because no systematic descriptions of the natural history of changes prior to death exist for teenagers. We hope that when the New York study has been completed that we will be able to provide this type of information. However, at first sight, the traditional warning signs do not fit the large number of cases we have seen who have longstanding behavior and academic problems and were often frequent users of drugs and alcohol. Only a minority present a picture of recent onset of a major depressive disorder. Given that lists of signs which do not em- phasize the chronically disturbed youngster may be ineffective because they do not iden- tify the cases at greatest risk, might the lists be harmful? As stated above, epidemiologi- cal studies indicate a one year prevalence of depression in teenagers of 1 to 3 percent (Weissman et al., 1985; Anderson, 1985; Kashani et al., 1983). The one year incidence of teen suicide is around 12/100,000, i.e. in one year, depression affects 1,000 to 3,000 times more teenagers than suicide. Linking the common problem of depression to the uncommon one of suicide may serve to ex- pedite referrals and increase compliance with treatment, but it may also introduce the notion ofsuicide as an appropriate response to teenagers who were not suicidal. We em- phasize that there is no evidence either way on this point, but we believe it is an impor- tant question that needs to be answered. b. Promoting discussion of suicide in class: Enough evidence suggests that young people imitate actual and fantasized suicide to war- rant concern about this technique. The D.Shaffer: A Critical Review of Preventive Intervention.. evidence includes: 1. Phillips’ (1979) demonstration that prominent display of the news of a suicide leads to an increase in suicidal deaths during the period immediately following the display. (Only two weeks ago we par- ticipated in a case conference on an eight year old child who attempted to stab his ab- domen with a kitchen knife, the day after a prominent politician committed suicide in this way.) In a recent communication Phil- lips indicated that most of the excess suicides which occur in this context are of younger people. 2.Gould and Shaffer’s study of television docudramas (see below) has demonstrated that these broadcasts have the effect of in- creasing teen suicide attempts and deaths. 3. Kreitman’s (1970) observation that young attempters had many more close contacts with those who had made a suicide attempt than non-suicidal psychiatric controls. 4. The occurrence of suicide clusters which are very probably the result of imitation. This evidence raises the possibility that class- room discussions will, as intended, raise awareness of the topic and introduce, de novo, suicide in the range of contemplated behaviors for the teenage pupil, i.e., that it will "put ideas in their head". This risk must be weighed against the intended benefit, i.e., that it will facilitate disclosure of some pupils’ pre-existing suicidal preoccupations and that this will, in turn, lead to intervention which will reduce the risk of suicide. This important issue remains unanswered. It is clearly a mat- ter of some urgency that quality research be undertaken to determine whether the effects are benign or beneficial. Case Finding by Television Attempts have been made to harness the im- pact of television to promote awareness of suicidal situations and to encourage ap- propriate referral. Holding (1974; 1975) ex- amined the impact in Edinburgh of an 11-part weekly television series, "The Befrienders," which illustrated the predica- ment of a suicidal individual who was then helped by the Samaritans. During the calender year in which the programs were shown, referrals to the Samaritans increased 140 percent. However there was no change in the number of attempted suicides treated by hospitals in the city. The effect on suicide deaths was examined by tabulating both suicides and undetermined deaths (a group usually considered to consist largely of suicides (Holding and Barraclough; 1978)) during the ten weeks after and the ten weeks prior to the series and for the same period of time in 4 previous years. In each of the previous years the number of suicides had declined during those weeks. However, during the year under examination the rate remained stable. There was no reduction in the number of suicides during the broadcast of the television series. Gould and Shaffer (1986) have examined the impact of dramatized television presenta- tions of suicide on a youthful audience. Over a period of approximately 4 months, the major U.S. networks broadcast 4 dramatiza- tions of either a young person’s suicide or the reaction to a suicide in a parent. The programs were broadcast with advance publicity clearly stating that they were in- tended to make the public aware of the problem of youth suicide. To a varying de- gree they were coordinated with community programs. In some cases this took the form of advance distribution of informational material indicating where treatment for the suicidal adolescent was available or material for teachers, parents, and teens outlining the clinical features which may be present in the teenager who may try suicide. In some cases the local affiliate arranged for a hotline num- ber to be flashed onto the screen at varying times during the program. The incidence of completed suicides among teenagers aged 19 or under was examined in the States of Connecticut, New Jersey and part of New York State in 14 day blocks during the 4 months when the programs were shown, and for two one-month periods 3-39 Report of the Secretary's Task Force on Youth Suicide before the first program and after the last. Comparisons were made between the death rate during the 14 days before and after each program, and between an overall expected rate and observed rate after each show. Similar comparisons were made of the num- ber of attempted suicides treated at 6 large hospitals in the New York City area. Suicidal deaths increased significantly during the ten days following three of the programs but there were no deaths after one of them. A similar effect was noted on attempted suicides. It seemed unlikely that the increased number of referrals for suicide attempts was due sole- ly to increased awareness or a decreased referral threshold for parents or teenagers (which might have been expected to increase the proportion of minor attempts during the after period) because the severity of attempts after the programs was similar to those before the programs and because the effect was on both attempts and suicides. It also seemed unlikely that the increased number of deaths was due to a bringing for- ward of suicides which might have occurred anyway. If this had been the case, one would have expected to note a reduction in the fre- quency of deaths at some point after the program, but this was not seen. COMMENTS: Television programs are ef- fective in publicizing the availability of ser- vices. However, they do not reduce the number of suicide attempts and may, in fact, increase them. They also appear to have a provocative effect on suicide deaths. The discrepant findings from one of the programs studied by Gould and Shaffer, while not statistically significant, is intriguing because it holds out the possibility that the dramitaza- tion may have special features which prevent it from having an unwanted effect and which might even have a preventive effect. These features could lie in the associated com- munity work or in some of the contents of the dramatization; this is clearly an important area for future research. Finally, the fact that the "Befrienders" series resulted in an in- 3-40 creased number of referrals to a crisis service without reducing suicide morbidity, suggests that crisis services on the model of the Samaritans are ineffective in reducing suicide morbidity. Interventions after a Suicide Parent Survivor Groups After a teen’s suicide, the surviving parents and siblings experience significant distress and dysfunction. There is also evidence of in- creased suicidal morbidity in the surviving families (Murphy et al., 1964; Augenbraum and Neuringer, 1972), so that postvention with survivors may have a preventive func- tion. The effects on subsequent suicides may be difficult to assess. In the only prospective study we know--a five year followup of the families of 100 successful suicides--Shepherd and Barraclough (1974) found no examples of suicidal behavior. In practice, "postvention" and prevention ac- tivities are frequently carried out by the same units. Survivors of suicides are an important source of manpower for suicide prevention projects and may initiate or give other sup- port to these activities. We found no studies which specifically set out to examine the impact of survivors’ groups on the subsequent suicidal behavior of survivors. In a series of publications, Videka-Sherman (1982a; 1982b; Videka- Sherman and Leiberman, 1985) examined changes in coping responses and mood in recently bereaved parents who attended meetings organized by "Compassionate Friends," a self-help group which provides support to parents who have lost a child by any type of sudden death. The number of los- ses through suicide, if any, is not indicated. - However, in the absence of other data, the findings of the study are reported: 2,422 parents on the register of several chap- ters of Compassionate Friends were sur- veyed by mail on two occasions to ascertain coping responses to their child’s death and to elicit depressive and other psychiatric symptoms. The response rate was low; only D.Shaffer: A Critical Review of Preventive Intervention. 667 (28%) answered the first survey and only 391 of these (17% of the original sample) provided additional followup information. The social demographic distribution of the responders indicated that they were predominantly white and upper middle class. There was no bereaved non-referred control group and comparisons were made between those who attended several meetings and those who either did not choose to attend or who dropped out of the program at an early stage. Depression scores were not in- fluenced by attendance, dropping equally in both those who attended and who dropped out of group sessions. The coping style most likely to be associated with high depression scores was an obsessive preoccupation with the memory of the child. The coping styles associated with best adaptation were immers- ing oneself in another activity, or having a re- placement child. Attendance at the group did not influence the development of either of these styles but did enhance altruistic ac- tivity which regardless of its beneficial effects for society did not, in itself, appear therapeutic. Rogers et al. (1982) reported an uncon- trolled study of a Survivors Support Program coordinated professionally but administered by volunteers. Groups met in 8 sequential 2 hour didactic sessions followed by 4 biweek- ly discussion groups. The attendees were divided evenly between spouses of suicides and parents of young suicides. When asked to identify their current problems almost all indicated that they felt guilty, detached from the event, and abandoned. Many idealized the deceased. They scored high on the somatization, phobic, and obsessive compul- sive scales of the SCL-90. At followup several weeks after terminating the program, 33/37 cases were contacted. Most of the par- ticipants reported that the program had been helpful and showed a decline in SCL-90 scores. COMMENTS: Only limited conclusions can be drawn from these studies. Without a con- trol group it is not possible to know whether the improvement in SCL-90 scores reflects the natural history of mourning or the effects of intervention. Videka-Sherman’s studies did not specifically examine suicide survivors. It examined a self-selected population and did not randomly assign eligible survivors to group or no group conditions. It therefore suffers from all of the drawbacks of any in- ference about the effects of an intervention that is based on a comparison between com- pliers and non-compliers. SECONDARY PREVENTION Reducing the Lethality of Suicidal Behaviors With the sole exception of suicide by hang- ing, which in 1981 accounted for 40 percent of all suicides among U.S. boys under age 14, but for only about 20 percent of male suicides in older age groups, the methods used by young children and adolescents are very similar to those used by older individuals of the same sex (CDC, 1985). Preferred methods also vary in different countries and appear to be generally stable across time. It is, therefore, reasonable to expect that an at- tempt to reduce accessibility to, or to im- prove the treatment of the outcome of any common method, should reduce the suicide rate in general and have impact on teenagers. The British experience is an example of how reducing access to the means of suicide can have a significant effect on reducing the suicide rate. Starting in 1957, the mean carb- on monoxide content of domestic gas was reduced from 12 percent to 2 percent through the introduction of natural gas and modifications in the conversion process from coal. The process was completed by 1970. Prior to these changes, self asphyxiation with domestic cooking gas accounted for more than 40 percent of all British suicides and for an even higher proportion of male suicides (Hassall and Trethowan, 1972; Kreitman, 1976). During the period of gas content change, British suicide rates from carbon monoxide asphyxiation declined precipitously, account- ing for fewer than 10 percent of all suicides 3-41 Report of the Secretary’s Task Force on Youth Suicide by 1971. Furthermore, the overall suicide rate declined by 26 percent and analysis of death by different methods showed that al- most all of this reduction could by attributed to a fall in deaths from domestic gas asphyxia- tion. What appears to have happened was that the suicidal population, denied access to a universally available, non-deforming, non- violent method, did not then turn to other more violent (and more lethal) methods, but instead chose another non-violent method which was similarly, readily available--self- poisoning. The incidence of suicide attempts from overdoses increases markedly during this period (Johns, 1977). The impact of this change of method on suicide deaths appears to have been dampened because over the same period self-poisoning became progres- sively less lethal, in part because of the sub- stitution of the less dangerous benzodiazepine drugs for the highly toxic barbiturates, and in part because of improved methods of resuscitation. Most significantly, however, British rates, in contrast to those in all other countries, have remained at the new lower level (Farberow, 1985). The detoxification of domestic cooking gas also occurred in other countries in Europe, specifically the Netherlands, where it was not associated with any reduction in rate. In these other countries, however, the base rate of self asphyxiation from domestic gas before its composition had been changed, was not considerably lower than it had been in Britain, (%) and the expected impact was proportionately less. COMMENTS: Prevention methods that do not require the active participation of the public have traditionally been the most effec- tive (e.g., changing the water supply in South London at the time of the great London cholera epidemics). It appears to have good potential for prevention in suicide as well. The importance of improved methods for treating suicide attempts may be the reason for the sex-specific changes in the suicide rate in the United States. Suicides have been in- creasing only for males. It may be that at a 3-42 time when suicide morbidity (i.e., the number of suicidal behaviors including both attempts and completions) is increasing in both sexes (Weissman, 1974; O’Brien, 1977), the impact has only been felt by males who favor methods for which treatments have not im- proved. The increase in female attempts is compensated for by improved treatment methods for self-poisoning, the preferred method. Care During a Crisis Individuals who have already attempted suicide or expressed suicidal thoughts or wishes, seem to be an optimal target for suicide prevention efforts. 1. They are at high risk. Most post-mortem studies show that between 30 percent and 50 percent of suicides have made a prior at- tempt; the suicide rate in followup studies varies from just under 1 percent to nearly 10 percent, over 100 times the risk carried by the general population. 2. They are potentially easy to identify for it seems that most visit an emergency room after their attempt (Kennedy, Kreitman and Ovenstone, 1974), and many will be admitted to a hospital. 3. Directing a preventive effort to individuals who have already demonstrated suicidal behavior avoids concerns that the "idea" of suicide will be introduced to a naive lis- tener. A preferred intervention with the suicidal in- dividual has long been the crisis center. The rationale for suicide crisis care has been ar- ticulated by Schneidman and Farberow (1957) viz; 1. suicidal behavior is often associated with a crisis; 2. suicide is contemplated with psychological ambivalence--wishes to die exist simul- taneously with wishes to be rescued and saved; 3. humans have a basic need to express them- D.Shaffer: A Critical Review of Preventive Intervention.. selves and to communicate with others; 4. the suicidal individual’s ambivalence about dying stems from a psychiatric illness in which the suicide represents a partially un- satisfactory means of achieving "fantasies of....surcease, revenge, atonement, ecstasy, rescue and rebirth..." This con- fusion leads to an oblique communication or signal or "cry for help" whichis best iden- tified by those with special training (Lit- man et al., 1965). In practice, telephone crisis services offer several advantages. They are convenient and accessible and thus offer an individual in crisis the opportunity for discussion and sup- port without having to travel or wait for an appointment. Their anonymity may be reas- suring and may allow callers to say shocking or embarrassing things which they could not otherwise do in a face-to-face interview. The first crisis center, "The Antisuicide Bureau," was started in 1906 in London by the Salvation Army. In the same year, the National Save a Life League was established in New York City. Shortly after World War II, the Neuropsychiatric Institute in Vienna established a counseling center run by volun- teers. Six years later in London, the Samaritans was started by the Reverend Chad Varah (Varah, 1973 and Fox, 1976). Twenty-two years after its establishment, the Samaritans had 165 branches in Great Britain alone and received over 1 million calls a year. It is staffed by volunteer "listeners" and insists on strict confidentiality. Its inter- actions, characterized as acts of "befriend- ing", are predominantly non-directive. The influential Los Angeles Suicide Preven- tion Center (LASPC) was established in 1958, initially concerned with evaluation and rehabilitation of hospitalized survivors of suicide attempts (Litman et al, 1961). In 1961, it broadened its activities to include community outreach, and a short while later, a 24-hour telephone hotline, thus becoming the prototype of American crisis centers. It has regularly sponsored research projects, many of which are referred to in this review. Its early goals and operations have been described by Helig et al. (1968) and Litman et al. (1971). There was a rapid proliferation of crisis ser- vices modelled on this program during the late 1960’s and the early 1970’s. By 1974, nearly all metropolitan areas in the United States had such a center and many had two or more (Miller et al., 1979). Differences and Similarities Between Crisis Centers Telephone crisis services have certain characteristics in common; they have the capacity to offer immediate emotional sup- port; they are available outside of usual of- fice hours; they provide the opportunity for anonymity; they tend to be staffed by volun- teers; the assistance they offer is often problem rather than "diagnosis" specific and help is always short term. Within these similarities there exist differen- ces in emphasis. Some function predominantly as information or referral ser- vices, rapidly ascertaining the problem and then referring the caller to an appropriate treatment center. This service sometimes ex- tends to the volunteer making the appoint- ment and checking that it has been kept. Sometimes this type of case management is offered by multi-service agencies which link the caller with the most appropriate unit of the service. When appropriate, calls may be passed directly to a duty psychiatrist or social worker. At the extreme of the intervention spectrum, there are crisis services which primarily offers a psychological environment which the person in crisis may find supportive and which encourages callers to drop in (par- ticularly true of the Samaritans). Centers vary in the stress they place on con- fidentiality. The Samaritans generally offer total confidentiality (Hirsch, 1981), whereas, many services in the United States are willing to intervene very actively (including sum- moning the police) in order to avert a suicide. The "befriending" process of the Samaritans has been likened to Rogerian psychotherapy 3-43 Report of the Secretary's Task Force on Youth Suicide with its emphasis on acceptance and warmth. This is shared by many Centers in the United States. Ross (1980), a leader in the suicide prevention movement, states that the "most important objective in responding to suicidal youth is to open the lines of communica- tion...accomplished by showing concern, in- terest and understanding in a non-judgmental manner". It has been sug- gested that the anonymity centers provide may be especially helpful to callers who find a discussion of their problems embarrassing. Additionally, patients who are concerned with issues of control and power may be more comfortable with telephone counseling as they have the option of hanging up. Volunteers are usually, but not always, super- vised by social workers or other mental health professionals. These mental health professionals are also available for consult- ation. This would generally not be the case with Samaritan services. A number of programs target a specific population such as college students (Ottens, 1984) and at least one (Glatt et al., 1986) has a telephone situated on a bridge that is renowned as a place for fatal suicide leaps. The differences between centers are some- times subtle and are implicit rather than stated. This makes research difficult and re- quires that rather general operational criteria be adopted by researchers. For ex- ample, Bridge et al. (1977) designated any entity a suicide prevention center if: a) there was an identifiable person in the community responsible for the service; b) if it provided 24-hour telephone or emergency service coverage; and, c) if it advertised its existence. Impact of Crisis Centers on Mortality A number of cross sectional studies have compared the rates in areas with and without crisis centers, or in areas before and after the introduction of a crisis center. Two early studies (Litman and Farberow, 1969; Ringel, 1969) reported a drop in the suicide rate in Los Angeles and Vienna respectively, after a service was introduced at 3-44 atime when the rates in California and in the rest of Austria were reported to be increas- ing. However, suicide rates vary with the demographic composition of a population. Rates are associated with sex, age and eth- nicity. The demographic profile of a given area, and hence its potential suicide rate, are all susceptible to change and simple correla- tional studies of this kind are inadequate. Account of these factors must be taken and appropriate control areas need to be studied. It should be said here that with one exception (Bagley, see below), no methodologically adequate study has been able to demonstrate an impact of suicide prevention centers on the number of deaths from suicide. One of the first studies to use a control population, was carried out by Weiner (1969) to assess the impact of the LASPC. Com- parisons were made between the suicide rate during the 6 years prior to the introduction of the hotline services at the LASPC and the 6 years afterwards and between two major California metropolitan areas that had ser- vices (Los Angeles and San Francisco) and two that did not (San Diego and San Bernar- dino county). However, changes in these rates were not corrected for demographic differences between the cities studied. The study noted that there was a significant in- crease in the suicide rate after the introduc- tion of the hotline service in Los Angeles, but this increase does not seem to have been sys- tematically related to the presence of suicide prevention centers, for there were similar in- creases in San Francisco which had a service and San Diego which did not, and a fall in San Bernardino county, which did not have a ser- vice. These fluctuations in rate are common and cannot be interpreted without correc- tions for changing socio-demographic profiles. An additional confounding factor in this study was that the study period covered the development of a close collaboration be- tween the Center and the medical examiner and this may have resulted in a broader definition of suicide (Litman and Farberow, 1969) and with it, an increase in coroner’s D.Shaffer: A Critical Review of Preventive Intervention.. determinations. Lester (1973), examined the suicide rate in a number of major metropolitan areas in the United States, comparing rates in cities before 1967 and after 1969. He compared cities where a suicide prevention center had been established with rates in cities where no center existed. An analysis of covariance was used to control for the size of the city. No differences were found, but the study did not control for changes in reporting procedure or for differences in demographic make-up. The sample of cities was small and the dura- tion of surveillance short, given the low in- cidence of suicide. In a methodologically rigorous study, Bridge and colleagues at Duke University (1977) compared the incidence of suicide in coun- ties with and without suicide prevention centers in all 100 counties of North Carolina. They used a multivariate approach to ac- count for a number of possibly confounding variables at the same time. These included duration of existence of a center, and a large number of socio-demographic variables. The mean duration of existence of a center was 2.8 years. No changes in reporting pro- cedures occurred during the time under study. The highest incidence of suicide was in communities characterized by a high proportion of older, white, married persons; suicide centers were more often located in areas with different demographic charac- teristics. Their results suggested that com- pared to the influence of demographic variables, suicide centers have a minimal ef- fect on rate. They also found only trivial in- teractions between the presence of a center and community characteristics such as age distribution, type of "cause of death" deter- mining system, and population density of an area, i.e., there was no evidence that hotlines were more effective in certain communities than in others. The British study by Bagley (1968) is the one that was most widely quoted as supporting the efficacy of suicide prevention centers. It was noted that suicide rates in that country were in decline and that the period of decline coincided with the growth of the Samaritan movement. In fact this was not accurate: the decline in British suicide rates (which was al- most certainly due to the introduction of non-lethal domestic cooking gas to substitute for coal gas--see above) halted in 1971 al- though the number of Samaritan branches and clients continued to rise until 1975 (Brown, 1979). In this study, Bagley used both empirical and a priori techniques to identify control communities. The empirical match was based on the two most important factors derived from a principal components analysis. The a priori match was for popula- tion over age 65, percentage of females, and social class index. (These factors accounted for 35 percent of the variance.) He com- pared cities with and without centers match- ing those with centers to control cities identified through the two methods. Bagley found that 15 Samaritan boroughs ex- perienced a fall of 6 percent whereas control boroughs experienced a rise of 20 percent (empirical) or 7 percent (a priori). Research scientists from the Medical Research Coun- cil Suicide Research Unit attempted to repli- cate Bagley’s findings (Barraclough et al., 1977; Jennings et al., 1978). They employed methodological improvements including using a wider variety of matches, examining more geographical areas, and using matches which accounted for more of the suicide rate variance. On the same target boroughs, they used 4 coordinates to do the empirical match- ing instead of 2, thus accounting for 65 per- cent of the variance instead of Bagley’s 43 percent. Further, they broadened the search for matchable boroughs and used a different predictive rate match. To accomplish this, they choose boroughs with similar rates before the establishment of a Samaritan cen- ter and also matched for proportion of single person households. Both of these methods accounted for significantly more of the variance than those adopted by Bagley. They examined suicide rates for 6 years prior to the establishment of a center and 6 years after its opening. It was not possible to replicate Bagley’s findings; no difference was found between Samaritan and control towns. They 3-45 Report of the Secretary's Task Force on Youth Suicide also noted that the rates of suicide decrease did not parallel the increase in Samaritan usage (Barraclough et al., 1977). Bagley (1977) responded to this critical exercise by stating that the difficulties in evaluating the impact of services were too great and that there was no reasonable way to demonstrate their efficacy. In the United States, Miller et al. (1984) elaborated on the effects of suicide preven- tion services on suicide rates. The period of study began in 1968, when most cities did NOT have suicide prevention centers and ran until 1973 when most cities did have such centers. During this 6-year period, they ex- amined the effects of suicide prevention ser- vices on age-, race-, and sex-specific population groups. After going through a lengthyseries of procedures to verify the date of a center’s introduction, they compared suicide rates in 25 locations that had no cen- ter prior to 1979 but which then introduced and maintained one until at least 1973, with 50 counties which experienced no change in the number of crisis centers during that time. Age-, race-, and sex-specific rates were ex- amined for all years for all centers. Dif- ference scores were calculated by covarying on the base rate. It was reasoned that if crisis centers serve predominantly younger women, then, any impact of a service could be expected among that group. They found a small but significant reduction in suicide rate (1.75/100,000) in white females after the introduction of a service, but no evidence of an impact in other population groups. Their examination was repeated on a second set of data at a different time period and their find- ings were replicated. A variant to the crisis service provision method of studying this problem can be seen in the study by Chowdhury et al. (1973), who randomly assigned suicide attempt repeaters to routine out-patient care or to an enhanced service which also provided an emergency telephone service and a walk-in facility. The latter group received home visits if they failed to keep an appointment. The groups did not differ in reattempt rates nor on any measure 3-46 of mental state at the end of a six month fol- lowup period. COMMENTS: The disappointing impact of crisis centers on suicide mortality needs to be explained. To explore this question further, we have further analyzed the literature on who uses and does not use centers, whether they are suicidal and whether any particular types of cases seem resistant to their impact. Who Uses Crisis Centers? Descriptions of adult callers (Sawyer, 1972; Murphy et al., 1969) and teenage counseling services (King, 1977; Slem and Cotler, 1973; Morgan and King, 1975) indicate that U.S. suicide prevention centers are predominant- ly used by females. A disproportionate num- ber are under age 30 and they show the same ethnic distribution of the area in which the center is based. They do not, therefore, reflect the special demographics of suicide completion in which males predominate and blacks are underrepresented. It also seems that many, non-suicidal in- dividuals in crisis use these services. They may be lonely, isolated people. This is not in itself incompatible with the goals of a suicide service unless it diverts resources from suicidal callers. Hirsch (1981) monitored 100 calls each at the LASPC and at the London branch of the Samaritans. About 40 percent of the Los An- geles center’s calls did not concern suicide. The incidence of non-suicide related calls is higher in Europe. In Helsinki, Aalberg (1971) found that only 25 percent of calls concerned suicidal ideation or suicidality. The same proportion of suicidal calls was found among callers to the Samaritans (Hirsch, 1981). A considerable proportion of non-suicidal calls to the Samaritans were characterized as "sex" calls. Studies of Teenagers Very little published work evaluates the im- pact of hotline or crisis services on teenagers. This section will identify teenage usage rates of general hotline services and review the D.Shaffer: A Critical Review of Preventive Intervention.. findings of the single evaluation study of a hotline service designed specifically for teenagers. In an early report of 1,607 consecutive telephone callers to the LASPC (Litman et al., 1965), 5 percent of the callers were under age 20. Greer and Anderson (1979) inter- viewed 90 percent of 364 consecutive cases of attempted suicide in a busy hospital in South London of whom 19 percent were under age 19. More than 70 percent of the total group had knowledge of the Samaritans, but that proportion was far smaller among teenagers. King (1977) sur- veyed 3,000 college students who had passed their freshman year and reported that 3 per- cent had called a service, two-thirds of these were for personal counseling rather than to report disturbing behavior in others. This 3 percent utilization rate compared favorably with the proportion of students who used the student mental health service. Only 8 per- cent of the surveyed callers were currently in some form of therapy, indicating that the hot- line was reaching a population not served by other community agencies. Slem and Cotler (1973) studied the impact of a hotline service for teenagers (not specifi- cally oriented towards suicide prevention) in an upper middle class community in subur- ban Detroit. The service had been intro- duced through advertising in newspapers and on school and community bulletin boards, and widely distributed business cards. At an unspecified later time, 1763 students in a local high school were surveyed to find out whether they knew of, or had used, the ser- vice. The answers indicated that the hotline was acknowledged as a community service of which they were aware with the same fre- quency as the YMCA and high school coun- selling services. 98 percent recognized the name of the service from a list of community services and 5.6 percent had used it. When asked to rank preferred sources of help for problems, users ranked the service higher than non-users. Both groups listed friends as being the most important source of help. Not very much information was available about the users except that approximately two- thirds were female and that users ranked help from parents as being potentially less valu- able than non-users, perhaps indicating a less satisfactory home background. There was a relatively low response rate among former hotline users about whether they had found the service useful, but two-thirds of the responders confirmed that their contact had been useful. These studies, while indicating that a hotline service can obtain satisfactory community recognition, are inadequate for our purpose because neither specify the proportion of calls that pertained to suicide and neither ex- amined the impact on psychiatric morbidity generally, or suicide morbidity specifically. Studies of Adults —Suicide and Crisis Service Users; What Proportion of Callers Are Suicidal? Litman et al. (1965), in an early report from the LASPC, noted the following: 45 percent of the callers were either currently receiving or had previously received psychiatric treat- ment. 50 percent talked about suicide during their call; 40 percent had made a previous suicide attempt of which 22 percent were within the preceding week. Only 10 percent of calls were unrelated to "suicide poten- tiality" (a rather loosely defined concept). Usage patterns may have changed because Hirsch (1981) in a survey of 100 calls at the LASPC noted that 40 were not related to suicide. Evidence for the suicide potential of hotline users has been gathered from studies which have looked at the subsequent suicide rates in callers. In interpreting studies of this kind it should be remembered that several factors tend to lead to an underestimate of later suicides among users. These include: a. Many studies match callers’ names with death certificate data collected from the same administrative area (e.g. county) as the service, and will miss people who have died in other locations. 3-47 Report of the Secretary’s Task Force on Youth Suicide b. A sizable proportion of calls are made anonymously and cannot be linked to death certificates or clinic records. There is some evidence (Tabachnik and Klug- man, 1965; Nelson et al., 1975) that anonymous callers are more likely to be living on their own, which would place them in an especially high risk group. It is also important to note that subsequent suicides among crisis center callers cannot be used to infer information about the efficacy of a center because no comparison can be made with the suicidal individuals who do not call the center. Given those caveats, the studies that are available tend to confirm that crisis center users are deviant and carry a much higher risk of later suicide than a nor- mal control population. In uncontrolled, followup studies of a ran- dom sample of LASCP callers, Litman (1970) and Wold and Litman, (1973) noted that between 1 percent and 2 percent of callers had committed suicide within 2 years of their initial contact. Sawyer et al. (1972) reported a study which drew comparisons with the rate in the same geographic area but which did not correct for age and sex. They found that 0.6 percent of the approximately 11,000 callers to the Cleveland Suicide Prevention Center had committed suicide within 4 years of their call. This figure represents a rate of 288/100,000 or approximately 25 times the expected death rate (uncorrected for age and sex). Three-quarters of the suicides had been referred to the center by others, compared with one-quarter of the group overall. The median interval between time of contact and - suicide was 4 months. Only 6 percent of all suicides in the city of Cleveland had been in touch with the Suicide Prevention Center at some time before death. In the absence of age- and sex-matched controls, these rates are difficult to interpret but appear some- what lower than the 2.5 percent to S percent suicide rate noted by Greer and Lee (1967) in their 2.5 year followup of serious suicide attempters treated in general hospitals. The lower death rate found in such crude com- 3-48 parisons cannot be used to infer the efficacy of the prevention centers; it may simply reflect different demographic composition of callers to different centers. Barraclough and Shea (1970) found death (suicide and other causes) rates, corrected for age, sex, location of call, and death, of in- dividuals who had called the Samaritans in six British counties to be 32 times the expected rate during the first year after the call. This rate fell to 7 times the expected rate 3 years after the call. 30 percent of the deaths oc- curred within the first month, 71 percent within a year and 90 percent within 2 years after the call. The death rates were inter- mediate between that of former mental hospital patients and currently depressed patients, but were considerably less than those among former psychiatric in-patients who had been admitted following a previous suicide attempt (Temoche et al., 1964). There were marked differences between dif- ferent centers, some having a lower than ex- pected death rate, others a lower initial (first year) rate, but acomparable or higher second year rate (suggesting that suicide had been deferred). These differences could have reflected the quality of interventions or a dif- ferent clinical base. COMMENTS: Suicide crisis centers attract potentially suicidal individuals. —Who Will Go On To Suicide? Several studies (Ovenstone and Kreitman, 1974; Wold and Litman, 1973; Wilkins, 1970; McKenna et al.,, 1975) have found that suicidal patients who make use of crisis telephone services fall into two groups: a chronically suicidal group and an acutely stressed group without a history of prior at- tempts. Litman et al. (1965) predicted that crisis centers would be most helpful to the suicidal individual who is isolated and friend- less or one who has suffered the loss of an im- portant person through death or rejection. Crisis centers were predicted to be least help- ful to suicidal individuals with chronically dis- organized behavior, or long standing dysphoric or psychotic states. D.Shaffer: A Critical Review of Preventive Intervention.. This is, in fact, what was found by Wold and Litman (1973) in their detailed followup of a random 1 out of 10 sample of suicide preven- tion center callers. Among those who sub- sequently committed suicide, most had a chronic history of psychiatric disturbance and several previous episodes of suicidal be- havior. The crises which had led to their original call were different from those which ultimately preceded their death. They had gone on to experience, and in all likelihood, generate, additional crises. These findings were supported by Wilkins (1970) in a death certificate match of ap- proximately 1,300 callers. Suicides were more likely than non-suiciders to be unmar- ried, to have made a previous attempt, and to have received previous psychiatric treat- ment. COMMENTS: There is evidence that chronically disturbed callers who have made previous attempts and had previous psychiatric treatment are an especially high- risk group for later suicide and that crisis management is inappropriate for them. —How Suicidal Users Compare To Suicidal Non-Users Differences between attempters who have used hotlines previously and those who have not, have been reported in several studies. Barraclough and Shea (1970) found that 4 percent of a consecutive series of adult suicides had used the Samaritans. Wold (1970) compared the characteristics of 26,000 LASPC contacters with a group of 42 suicides and noted that 75 percent of the cen- ter contacters were women, compared with 36 percent of the completers. Center contac- ters were, on the average, 9 years younger than completers. A disproportionate num- ber of center contacters were less than age 30. Motto (1971), who studied 575 individuals consecutively admitted to a psychiatric in- patient unit for treatment of either a depres- sive or a suicidal state, found that 11 percent had used suicide prevention centers. More than 50 percent of these felt that they had been helped by the contact, 10 percent said they had been made worse. The most com- monly stated reason for not calling was that they had been unaware of the centers’ exist- ence. Greer and Anderson (1979) inter- viewed 90 percent of 364 consecutive cases of attempted suicide in a busy hospital in South London, 19 percent of whom were under age 19. Approximately 14 percent of these attempters had had some contact with the Samaritans in the past, but very few had done so just prior to their recent suicide at- tempt. Overall, just over 70 percent of the group had knowledge of the Samaritans; that proportion was far smaller among teenagers. Among those of all ages who knew of the crisis service, the most commonly stated reasons for not calling were: a. it did not occur to the caller, b. they wanted relief from their distress or wanted to die, c. they thought that the crisis center would be unable to help. Half of the group that felt that the Samaritans could not help had prior con- tact with the Samaritans. Greer and Weinstein (1979) studied suicidal patients who were receiving mental health treatment, comparing those who had first contacted a hotline with those who were identified by a mobile emergency team which made outreach endeavors to families or in- dividuals in crisis. Hotline patients had a lower suicide potential score on certain standard measures and were less likely to re- quire admission after being seen. The find- ings from this study may tell us as much about the seriousness of cases identified by a mobile team as they do about the mildness of disor- ders seen in patients who call hotlines. COMMENTS: In general, this is a sparse literature but an important one. There is a lot of evidence that hotlines do not have im- pact on a community’s suicide rates (see below). It is clear that their utilization rate by suicide attempters is low (the highest rate 3-49 Report of the Secretary’s Task Force on Youth Suicide in the studies cited being 14%) and although in Greer and Anderson’s study there were a proportion of attempters who reported having been disappointed at the intervention they had received previously, most of those who did not call did not think about it or did not know of the hotline’s existence. Lack of knowledge is a special problem with teenagers. Other Problems With Crisis Service Techniques Low Compliance Rates After Triage A number of studies document the low rate of compliance with care after emergency room triage interventions have been provided to suicide attempters or to hotline callers. Only one study has been carried out with adolescents. Litt et al. (1983) studied 27 adolescents seen in an emergency room. All were offered further appointments but only 33 percent kept them. Failure to keep an ap- pointment was similar in groups referred from an emergency room or from an in- patient ward, but was more common in those who had made a previous attempt. The num- bers in this study were too small to permit adequate statistical analysis to which a variety of interrelated factors might have con- tributed. Additionally, there was no ade- quate examination of the clinicians’ technique which has been shown to be im- portant in studies among adult attempters. A similar low "show" rate has been reported for adults. Chameides et al. (1973) found a compliance rate of 35 percent, and Paykel et al. (1974) reported a compliance rate of 44 percent with out-patient referrals made in the emergency room. Compliers with the out-patient referral did not differ from non- compliers with respect to clinical characteris- tics. Furthermore, many of those who do keep their first, or first few, assigned appoint- ments will fail to maintain contact with the center to which they are referred, and will drop out of their treatment program prema- turely (Kogan, 1957b; Jacobsen et al., 1965). Factors contributing to referral failure have been studied by Knesper (1982) in nearly 300 3-50 emergency room cases managed by 15 dif- ferent clinicians. Failure was found to be in- dependent of patient characteristics, at least so far as suicide intentionality was concerned, but an "outlier analysis" showed significant clinician variation. The fact that some clinicians can persuade most of their patients to attend a later appointment while others can persuade very few, suggests that clinician behavior is important. The fact that referral failure is not an index of seriousness of disor- der was found specifically by Paykel et al. (1974) who noted that attempters who com- plied with their referral and attempters who did not comply with a referral did not differ in clinical characteristics including the seriousness of their attempt. Chameides and Yamamoto (1973) found that many of those who fail to comply will see some other men- tal health professional during the year after their attempt. The same appears to hold true for other referral situations. In a study of the failure of hotline callers to comply with suggested appointments, Lester (1970) found that the percentage of shows after a telephone call to a crisis service ranged from 29 to 56 percent with some seasonal variation and con- siderable variation with individual volun- teers. Approximately half of the 20 different personnel involved had a success rate of less than 40 percent, whereas the other half had a success rate of between 50 percent and 80 percent. What contributes to clinician/volunteer failure? Knesper (1982) noted that a clinician who would spring the question of admission on a patient suddenly and without warning at the end of an examination had a very low rate of success in making referrals to an in-patient unit. Slaiku et al. (1975) found no significant cor- relations between compliance and the hotline volunteer’s conversational charac- teristics in particular, whether they made specific reference to words like suicide or in- stead used euphemisms in referring to such matters. They noted a higher show rate when attendance for an appointment was initiated D.Shaffer: A Critical Review of Preventive Intervention.. by the caller. Several studies suggest that compliance can be improved if the volunteer or clinician makes an actual appointment for the patient/caller rather than simply providing them with a name and number to call. Kogan (1975a) recorded a 37 percent compliance rate for attempters seen in an emergency room when the patient was provided a name and telephone number compared with 82 percent when an appointment was made during triage. Rogawski and Edmundson (1971), using a more stringent index of compliance (2 kept appointments), found that only 30 percent of those given a name and telephone number kept their appointment, but that 55 percent did so when an appointment was made for them. However, neither were random as- signment studies and there may have been other selection factors which contributed to being chosen for the more active interven- tion and to later compliance. There is also some evidence that compliance may be improved if referral is made to a specific clinical service rather than to a local generic service. Welu (1977) contrasted compliance to a specific, new outreach program with 57 cases seen before the estab- lishment of the new program who had been referred to their community mental health service. He found that 90 percent of the cases attended the new program compared with only 54 percent of the cases referred to the original program. No details are given about whether elements other than the novelty of the program played any part in at- taining this unusually high compliance rate. Those who attended the novel program made significantly fewer attempts. Sudak et al. (1977) reported compliance rates at the Cleveland Suicide Prevention Center. Approximately two-thirds of the referrals to this center are females. A center professional routinely makes an appoint- ment for the individual who has been triaged and will then followup to see if the appoint- ment has been kept. The overall compliance rate was 60 percent with higher rates being reported for patients who were already in treatment with another therapist. Similar rates were reported for those who had made a recent suicide attempt as for those who called for other reasons. COMMENTS: Low compliance with recommendations is clearly a pervasive problem and should be monitored routinely by any crisis service. Some variation with hel- pers will occur, but evidence is sufficient that active procedures result in a significantly im- proved compliance; they should become standard and expected. Conveying Inappropriate Information Bleach and Claiborn (1974) and Apsler and Hodas (1975) simulated real callers and found that about 15 of 96 volunteer- answered calls generated inappropriate in- formation from the volunteer. Volunteers tended to give callers a range of referral sites without doing any editing or attempting to find a best fit for the callers’ problems. Interference WithOtherTreatments It has been argued that the existence of readi- ly accessible crisis services could complicate other therapeutic interventions. Certainly many callers are receiving treatment else- where. King (1977), in a study of college stu- dent crisis service users, found that 8 percent of callers, were currently in some form of therapy. Litman et al. (1965) noted that about 20 percent of 1,607 consecutive callers to the LASPC were currently in some form of therapy. Hirsch (1981) noted that many of the calls to the LASPC dealt with complaints about therapists. COMMENTS: The undermining of other effective forms of treatment is likely to be a problem if a center has a strong theoretical bias, but there is no evidence of a negative impact through this mechanism. Popularity With Users Slem and Cotler (1973) in their assessment of high school users, reported that 68 percent had had a good experience with crisis ser- 3-51 Report of the Secretary’s Task Force on Youth Suicide vices. The findings of this study must be in- terpreted cautiously because the followup rate was relatively low (58%) and the num- ber of suicidal users was not specified. King’s (1977) study of college student users indicated that the majority of girls found the counseling services helpful, but fewer than half of the male students did so. The dif- ference between the sexes was statistically significant. However, between 20 and 33 percent of males and between 10 and 20 per- cent of females reported that the experience of hotline usage made their problem worse. Satisfaction among users who called because of suicidal ideation or attempt was markedly less for males than it was for females. Females who received counseling from a male listener on the whole reported greater satisfaction with the help received. Similar- ly, males who received help from a female lis- tener reported more satisfaction than males who spoke to male listeners. Getz et al. (1975) found that patients with like problems, e.g., problems with their parents, felt more positively about the crisis intervention than callers who had serious mental illness or drug problems. COMMENTS: It is difficult to know whether the reports listed above are parochial, i.e. apply only to the center which has been studied, or have a more general ap- plication. Judging from the number of repeat calls reported by most centers, there must be a reasonable level of satisfaction, but this may not in itself be related to efficacy. Volunteers vs. Professionals; The Impact of Training Bleach and Claiborn (1974) and Genthner (1974) used students to simulate clients and rated empathy of volunteers working in crisis centers. Using standardized rating scales, both found that most volunteers were functioning at low levels of warmth and em- pathy. Hirsch (1981), in an essentially anecdotal comparison of volunteers and professionals, suggests that volunteers show more warmth, 3-52 empathy and patience but are less skilled than professionals in eliciting relevant past history and in being able to integrate infor- mation from the volunteer. This has been at least partly confirmed by comparisons be- tween trained and untrained volunteers. Knickerbocker and McGee, (1973) found greater warmth and empathy in untrained volunteers. It was not clear, however, whether the more experienced volunteers had received specific training in empathy and warmth and where this has been the case-- training in empathetic response--improve- ments appear to occur over time (France, 1975; Kalafat et al., 1979). Differences in trained and untrained workers in these skills may be moot because the literature on psychotherapy outcome shows only poor agreement between therapist characteristics and good outcome (see Stein and Lambert for details). Another relevant dimension is that of per- missive vs. directive. Knowles (1979) and Mcarthy and Berman (1979) noted a tenden- cy for untrained volunteers to be very direc- tive and to offer advice, often prematurely and on the basis of inadequate information. Ottens (1984) developed a program initially used to train key faculty, residence hall coor- dinators, and other staff at Cornell Univer- sity. The program focuses on how to take a proactive, and directive approach to crisis management, familiarity with available resources and how to use the resources, and how to interact with the crisis victim. The program was evaluated by designing a set of situational vignettes with multiple choice answers designed to depict possible inter- venor actions. The validation criteria were established by obtaining responses on the same questions from Crisis Center staff. Al- though it is stated that there were changes in the rank ordering of several of the different items, data are not provided. Statistical values are not given nor are the statistical procedures described. Elkins and Cohen (1982) found little im- provement in hotline volunteers after 5 months of training, but those who received D.Shaffer: A Critical Review of Preventive Intervention.. pre-job training did appreciably better than others. The less dogmatic, the more sensitive and skilled the person was likely to be. The relationship between knowledge of suicide lethality and ability to deal with suicidal individuals was examined in a group of nursing students by Inman et al. (1984). There seemed to be little relationship be- tween the lethality knowledge and skill re- quired for effective management of suicidal patients. COMMENTS: Suicide crisis services are used by seriously ill individuals who have a high suicide potential. Despite their high usage rates and the high proportion of calls which pertain to suicide, an overwhelming majority of attempters do not call these ser- vices. The literature does not allow us to conclude whether their failure to effect death rates--except marginally on young white women--(See Miller, 1984) is due to the failure of their technique to meet the needs of a residual suicidal population or to their failure to attract an appropriate popula- tion. On the one hand, information on the nature of callers who suicide suggests a mis- match of technique with recurrent suicide at- tempters. Data from studies which compared users and non-users suggest low utilization rates. However the evidence is drawn from a variety of studies of different populations at different times in different countries. The issue remains an important one and should, perhaps, be a specific focus for further research. It is important to determine whether the findings relating suicide repetitions can be applied to adolescents. Despite their youth, teenagers frequently have a history of repeated attempts and attempt repetition has been found in at least one study to be a predictor of later suicide (Otto et al., 1972). It seems that a fruitful exercise for the future would be to investigate how best to increase and sustain knowledge about the availability of a hotline service to a vulnerable popula- tion group. This can be done (see Slem and Cotler’s work demonstrating a high rate of recognition in their suburban community). It may also be true that the skills of telephone answerers in a crisis service decay when the service is used predominantly by non-suicidal callers. Research is needed to see whether more narrowly demarcating the caller population improves the quality of advice or information that volunteers give. Finally, there is evidence that hotline volun- teers may have defective mastery of informa- tion, may be deficient in empathy and may use inappropriate techniques to ensure com- pliance with referral recommendations. Research indicates that, although experience does not ameliorate these problems, training may. Clearly, regular evaluation of these ele- ments of a crisis intervention service, using techniques that have already been developed, should be a routine for estab- lished centers. After the Crisis There have been no satisfactory studies in which suicidal teenagers have been random- ly assigned to differing systematic treatments with outcomes observed in a controlled fashion. (See Trautman and Shaffer, 1984) The bulk of this section, therefore, relates to studies of adults. Clearly, the expected efficacy of any psychiatric intervention is dependent on its success in attracting disturbed patients, on their compliance with treatment being recommended and on the effectiveness of the treatment recommended. The literature contains relatively few adequately designed studies for this purpose. Most are quasi- naturalistic studies in which individuals who comply with treatment recommendations are compared with those who do not, or in which outcome for those who were treated routine- ly before a new program was instituted are compared with those who enter a new program. Such studies present considerable problems in interpretation. There is no con- trol over the type of treatment that is offered. It will vary in type and quality with different practitioners. Non-compliers are a poor con- trol group because they may be either more 3-53 Report of the Secretary's Task Force on Youth Suicide (i.e., don’t adhere to schedules, etc.) or less (i.e., not severely disturbed so don’t see the need to attend) deviant than compliers. Dif- ferences between compliers and non-com- pliers may either negate or enhance the apparent effects of treatment. Before and after studies may also present difficulties be- cause the opening of a new service will tend to attract a different category of patients than those served before the existence of a new program. Naturalistic Studies Greer and Bagley (1971) contrasted suicide attempters who, due to staff oversight, had been discharged from an emergency room without a further appointment, and those who were given and complied with an ap- pointment to attend a psychiatric clinic. Non-referred cases had significantly higher suicide treatment rates than those who were seen by a psychiatrist. There were inter- mediate findings when those who had had more than 2 treatment visits were compared with those who had made only one visit. The seriousness of the initial attempt did not predict reattempt. These findings have not been replicated and itis not clear whether the non-attenders were denied additional appointments or whether they also included some cases who were given a telephone number to call and who failed to do so and were thus classified as non-com- pliers. If the untreated control group did in- clude cases of this sort then the poor results might reflect some selection factor which was predictive of both non-compliance and repetition. On the other hand, if they were all systematically excluded, then the findings would suggest an effect. Other things being equal, oversight would more likely occur with less seriously disturbed patients. Ettlinger (1975), in Denmark, instituted a new service for suicide attempters which en- couraged unfettered access to mental health professionals, daytime hotline and walk-in clinics, frequent home visits made at the patients request, close consultation with other hospitals to which the patient might be 3-54 admitted, and proactive outreach for a one- year period. The subsequent five/six year suicide rate for 670 consecutive admissions was examined and compared with the death rate for 681 attempters who had been ad- mitted to the hospital before the service had been started. The new service appears to have been popular and was used freely. Despite this, no differences were found in subsequent suicide rates or social adjustment between the two groups. The study is some- what flawed by the low retrieval rate for the control group but that would not have been expected to effect knowledge about later suicides. Welu (1977) contrasted suicide repetitions among 63 patients seen in an innovative out- reach program with 57 cases seen before the new program was introduced and who had been referred to a local community mental health center. Cases assigned to the new program were more likely to attend and made significantly fewer attempts during the 4- month followup period. Kennedy (1972) reports differences in repetition rates of 204 suicide attempters: 142 were selected for short-term admission to a suicide crisis unit, 672 were referred to a psychiatrist for out-patient after care, and 56 received no after-care. Repetition rates were significantly lower for those who were admitted for a short-term stay, but there were no differences in reattempt rates between the group that received long-term psychiatric treatment (which was often started several weeks after the attempt) and those who received no treatment. These findings per- sisted even after corrections were made to account for previous suicide attempts, a fac- tor which is thought to be a strong predictor of further repetition. The authors inferred from these findings that crisis management is important. Random Assignment Studies Chowdhury and Kreitman (1973) randomly assigned repeat attempters to routine out- patient care or an enhanced service which in- cluded emergency telephone access and a D.Shaffer: A Critical Review of Preventive Intervention.. walk-in facility. Patients also received home visits if they failed to keep an appointment. The groups did not differ in their reattempt rates or on any measure of mental state. However, the experimental group ex- perienced fewer social problems (housing difficulties, unemployment, collection of benefits, etc.) at the end of the evaluation period than the controls. One cannot argue from this study that psychiatric care was not helpful as it was received by both groups. Motto (1976) and Motto et al. (1981) iden- tified a sample of 3005 hospitalized patients at "high risk" for suicide. All were offered after care. Of these, 862 declined. These were then randomly assigned to a group which received intermittent telephone con- tact at decreasing intervals over a S-year period, or to a group which received no fur- ther contact. During the first 2 years, suicide rates were twice as common in the non-con- tacted group as in those who received con- tact. During the remainder of the followup period the rates converged. This study is dif- ficult to interpret because of the way that non-compliance with treatment is handled. There was a relatively poor followup rate among those selected to receive contact (243/417 contact subjects either refused to receive contact or else could not be con- tacted). It would have been interesting to know what their death rate was, but in the study descriptions which have thus far ap- peared in print, they are not differentiated from the remainder of the contact group. Gibbons (1978) and Gibbons et al. (1980) randomly assigned 200 cases each to a course of intensive, but time-limited (3 months), task-centered case work, and to routine treatment (some cases were followed up by a psychiatrist, others by a general practitioner, etc.). Cases with high suicide intent were precluded from the random assignment and were all designated to receive the intensive approach. No differences were found be- tween the two experimental groups but the high risk "excluded" group had a significantly higher repetition rate. There were a sig- nificant number of drop outs from the treat- ment groups and the report does not specify the repetition rates for the partially treated groups. Only one random assignment study has been identified which assigned patients to dif- ferent types of psychological treatment. Liberman and Eckman (1981) randomly as- signed a small group of attempters to either 32 hours of behavior therapy (social skills training, anxiety management, and contin- gency contracting) or to insight-oriented psychotherapy. The groups did not differ with respect to repetition of suicide attempts but the behavior therapy groups were generally less symptomatic and less preoc- cupied with suicidal ideation and threats. COMMENTS: Although none of the studies are adequate methodologically, none present clear and consistent evidence that suicide repetitions can be prevented by whatever array of interventions may have been offered in these different settings. The very high relapse rate of patients with a his- tory of chronic personality disturbance and previous suicide attempts in Chowdhury and Kreitman’s study, even though they were provided with optimal and varied manage- ment, is particularly relevant in the light of Litman’s observations that this is the group of crisis service contacters who are most prone to ultimately commit suicide. The ab- sence of psychopharmacological studies is especially striking and highlights a research priority area. CONCLUSIONS The prediction of rare events from common ones is a dispiriting process which has been commented on by several reviewers (Rosen, 1954). Predictions are plagued with low specificity (high false positive rates) which might be acceptable if the interventions were either inexpensive or efficacious or both. That is far from the case, however, in suicide. The general wisdom is that preventive inter- ventions should focus on suicide attempters or on depressed patients. With respect to the latter, Temoche et al. 3-55 Report of the Secretary's Task Force on Youth Suicide (1964) undertook a rather dispiriting analysis of what the impact might be of an optimally efficient prevention scheme focusing on patients with a serious psychiatric illness which had required hospitalization. Using Massachusetts data, they matched psychiatric hospitalization records with suicidal death notifications. Suicide rates were low during the patients stay in hospital, but were very high in comparison with the general population after their discharge. Most deaths occurred during the first year after discharge and the group with the highest suicide rates were those that had received an in-patient diagnosis of depres- sive psychosis. Assuming that optimal protection (which might involve prolonged institutional care) was provided for the highest risk group for the period of maximum risk (one year after discharge) the (very ex- pensive) intervention would only reduce the suicide rate by 4 percent. They conclude that any more effective program would have to broaden the net of patients taken into care (thus radically increasing the cost), or would have to improve the efficiency of any predic- tors so that a smaller but more specific high risk group could be identified. This exercise did not take account of the very low com- pliance rate for treatment in this group of patients. Prevention based on the effective treatment of suicide attempters has not been subject to the same analysis, but given the lower ratio of attempts to completions we would expect that it would be more cost-effective than the management of depression if we could iden- tify an effective intervention. While the literature which deals predominantly with psycho-social interventions fails to provide us with any indication that these interven- tions are effective, we cannot conclude from this that suicide attempts are untreatable. The literature, at least as we have surveyed it, is strikingly deficient in the area of psychopharmacology which, given its efficacy in the general area of affective illness, is sure- ly the area where most hope should be directed and which is most deserving of re- search support. 3-56 Recent work on biological predictors of suicide repetition and completion (Asberg et al, 1976; Stanley, 1984) has not been dis- cussed in this review but it offers the prospect of increasing the specificity and thus reduc- ing the cost of preventive interventions. A considerable amount of energy and good- will, human sensitivity, and kindness has gone into the conventional suicide prevention ac- tivities but there is little evidence that they have been effective. Can we accept the bit- ter logic of research or should the findings be qualified with the customary apology that not all benefits can be researched? Certainly there is room for more research on the benefits of hotline calls and crisis centers for problems other than suicide, but there is pre- cious little encouragement for the suicide preventer. We all want crisis services to work, but if they do not, we should have the fortitude to discontinue them. To postpone attempts until answers are provided by experimental programs would be to ignore the evidence of common sense and clinical experiences (Rogers et al., 1982). BIBLIOGRAPHY: -+eeeveereeeermenarnienennnnes 1. Aalberg, V. SOS-Service, the suicide prevention center in Helsinki. In: K. Achte and J. Lonnqvist, (eds). Suicide research: proceedings of the seminars of suicide research by Yrjo Jahnsson Foundation, 1974-1977. Hel- sinki: Psychiatria Fennica; 1977: 67-68. 2. Achte, K. Present status and evaluation of suicide prevention and crisis intervention services in Europe. Men- tal Health and Society; 1976; 3: 169-174, 1976. 3. American Academy of Pediatrics. ABC's television movie "Surviving" depicts the tragedy of teen suicide-- medical group prescribes caution in viewing. News Release; February 1, 1985. 4. American Academy of Child Psychiatry. Medical organizations list teen suicide warning signs. News Release; October 25, 1984. 5. Anderson, J.C., Williams, S., McGee, R. and Silva, P. The Prevalence of DSM-IIl Disorders in a Large Sample of Preadolescent Children. Presented at the 32nd Annual Meeting of the American Academy of Child Psychiatry; Oc- tober, 1985; San Antonio, TX. 6. Apsler, R. and Hodas, M. Evaluating hotlines with simulated calls. Crisis Intervention; 1975; 6. 7. Apsler, R. and Hoople, H. Evaluation of crisis inter- vention services with anonymous clients. American Jour- nal of Community Psychology; 1976; 2: 1-14. 8. Asberg, M,, Traskman, L. and Thoren, P. 5-HIAA in the cerebrospinal fluid: A biochemical suicide predictor. Archives of General Psychiatry; 1976; 33: 1193-1197. 9. Auerbach, S.M. Crisis intervention: A review of out- come research. Psychological Bulletin; 1977; 84: 1189- D.Shaffer: A Critical Review of Preventive Intervention.. 1217. 10. Augenbraum, B. and Neuringer, C. Survivors of Suicide. In: A. Cain, eo) Helping survivors with the impact of suicide. Springfield, IL: C.C. Thomas; 1972. 11. Bagley, C. The evaluation of a suicide prevention scheme by an ecological method. Social Science and Medicine; 1968; 2: 1-14. 12. Bagley, C. An evaluation of suicide prevention agencies. Suicide and Life Threatening Behavior; 1971; 1: 245-259. 13. Bagley, C.R. Suicide prevention by the Samaritans. Lancet; 1977; 2: 348-349. 14. Bandura, A. Social Learning Theory. New Jersey: Prentice Hall; 1977. 15. Bandura, A. Vicarious and self-reinforcement processes. In: R. Glaser, (ed). The Nature of Reinforce- ment. New York: Acedemic Press; 1971. 16. Baron, J.N. and Reiss, P.C. Do mass media events cause suicides and homicides? Stanford University, un- published; 1984. 17. Barraclough, B.M. Differences between national suicide rates. British Journal of Psychiatry; 1973; 122: 95- 96. 18. Barraclough, B.M., Bunch, J., Nelson, B,, et al. A hundred cases of suicide. British Journal of Psychiatry; 1974; 125: 355-373. 19. Barraclough, B. and Shea, M. Suicide and Samaritan clients. Lancet; 1970; 2: 868-870. 20. Barraclough, B.M. and Jennings, C. Suicide prevention by the samaritans: A controlled study of effec- tiveness. The Lancet; 1977: 237-239. 21. Barter, J.T., Swabeck, D.O. and Todd, D. Adoles- cent suicide attempts. Archives of General Psychiatry; 1968; 19: 523-527. 22. Bern, D.J. and Allen, A. On predicting some of the people some of the time: The search for cross-situational consistencies in behavior. Psychological Review; 1974; 81: 506-520. 23. Birtchnell, J. and Alarcon, J. Depression and at- tempted suicide. British Journal of Psychiatry; 1971; 118: 289-296. 24. Bleach, G. and Claiborn, W.L. Initial evaluations of hotline telephone crisis centers. Community Mental Health Journal; 1974; 10: 387-394. 25. Blonston, G. Suicide among Arapahoe youths tied to "cultural identity crisis". Hartford Courant; ober 10, 1985. 26. Bogard, H.M. Followup study of suicidal patients seen in emergency room consultation. American Journal of Psychiatry; 1970; 126: 1017-1220. 27. Bollen, K.A. and Phillips, D.P. Imitative Suicides: A national study of the effects of television news stories. American Sociological Review; 1982; 47: 802-809. 28. Bollen, K.A. and Phillips, D.P. Suicidal motor vehicle fatalities in Detroit: A replication. Americal Journal of Sociology; 1981; 2: 404-412. 29. Bridge, T.P., Potkin, S.G., Zung, W.W.K. and Soldo, B.J. Suicide prevention centers: Ecological study of effec- tiveness. The Journal of Nervous and Mental Disease; 1977; 164: 18-24. 30. Brown, J.H. Suicide in Britain: More attempts, fewer deaths, lessons for public policy. Archives of General Psychiatry; 1979; 36: 1119-1124. 31. Centers for Disease Control. Suicide Surveillance. Atlanta, GA: U.S. Department of Health and Human Ser- vices; 1985. 32. Chameides, W.A. and Yamamoto, M.D. Referral failures: A one-year followup. American Journal of Psychiatry; 1973; 130: 1157-1158. 33. Choquet, M., Facy, F. and Davidson, F. Suicide and attempted suicide among adolescents in France. In: R. Farmer and S. Hirsch, (ed). The suicide syndrome. Lon- don: Croon Helm; 1980. 34. Chowdhury, N., Hicks, R.C. and Kreitman, N. Evaluation of an aftercare service for parsuicide (“at- tempted suicide") patients. Social Psychiatry; 1973; 8: 67- 81. 35. Cutter, F. The relation of new samaritan clients and volunteers to high risk people in England and Wales (1965- 3n Suicide and Life Threatening Behavior; 1979; 9: 245-250. 36. Daly, M., Conway, M. and Kelleher, M.J. Social determinants of self-poisoning. British Journal of Psychiatry; 1986; 148: 406-413. 37. Day, G. Samaritans versus suicide (letter). British Medical Journal; 1979; 2: 935. 38. Devries, A.G. Model for the prediction of suicidal behavior. Psychological Reports; 1968; 22: 1285-1302. 39. Diekstra, R.W.F. Social and interpersonal factors in suicidal behavior. In: R.F.W. Diedstra and K.J.M. Van de Loo, (eds). The cost of crisis. Assen: Van Gorcum; 1972. 40. Doan, M. As "cluster suicides" take toll of teenagers. U.S. News and World Report; November 12, 1985; 97: 49. 41. Dorpat, T.L. and Ripley, H.S. A study of suicide in the Seattle area. Comprehensive Psychiatry; 1960; 1: 349- 359. 42. Dublin, L. Suicide: A Sociological and Statistical Study. New York: The Ronald Press Company; 1963. 43. Eastwood, M.R,, Brill, B.A. and Brown, J.H. Suicide prevention centres. Canadian Psychiatric Association Journal; 1976; 21: 571-575. 44. Elkins, R.L. Jr. and Cohen, C.R. A comparison of the effects of prejob training and job experiences on non- protessiopal telephone crisis counselors. Suicide and Life hreatening Behavior; 1982; 12: 84-89. 45. Ettinger, R. Evaluation of suicide prevention after attempted suicide. Acta Psychiatrica Scandinavica; 1975; Supp 260: 5-135. 46. Facy, F., Choquet, M. and Lechvallier, Y. Resear- che d'une typologie des adolescents suicidants. Social Psychiatry; 1979; 14: 75-84. 47. Fairfax County Public Schools. Adolescent Suicide Prevention Program: A Guide for Schools and Com- munities. Fairfax, VA; 1985. 48. Fenton, F.R. and Mann, A.M. The effectiveness of suicide prevention programs. Journal of the American Academy of Psychiatry and Neurology; 1976; 1: 5-11. 49. Fox, J., Manitowabi, D. and Ward, J.A. An Indian community with a high suicide rate--5 years after. Canadian Journal of Psychiatry; 1984; 29: 425-427. 50. Fox, R. Suicidology: Contemporary Develop- ments. New York: Grune and Stratton; 1976. 51. France, K. Evaluation of lay volunteer crisis telephone workers. American Journal of Community Psychology; 1975; 3: 197-220. 52. Ganzeboom, H.G.B. and de Haan, D. as cited in Phillips, D. The impact of mass media violence on U.S. homicides. American Sociological Review; 1983; 48: 560- 568. 53. Garzotto, N., Burglass, D., Holding, T.A. and Kreit- man, N. Aspects of suicide and parasuicide. Acta Psychiatrica Scandinavica; 1977; 56: 204-214. 54. Gelman, D. and Gangelhoff, B.K. Teen-age suicide in the sunbelt. Newsweek; 1983; 15: 70-74. 55. Genthner, R. Evaluating the functioning of com- munity-based hotlines. Professional Psychology; 1974; 5: 409-414. 3-57 Report of the Secretary’s Task Force on Youth Suicide 56. Getz, W.L., Fujita, B.N. and Allen, D. The use of araprofessionals in crisis intervention: Evaluation of an nnovative program. American Journal of Community Psychology; 1975; 3: 135-144. 57. Gibbons, J.S., Butler, J., Urwin, P., et al. Evaluation of a social work service for self-poisoning patients. British Journal of Psychiatry; 1973; 8: 67-81. 58. Gibbons, J.S., Elliot, J., Urwin, P. and Gibbons, J.L. Evaluation of a social work service for self-poisoning patients. British Journal of Psychiatry; 1978; 133: 111-118. 59. Gibbons, J.S. Management of self-poisoning: So- cial work intervention. In: R. Farmer and S. Hirsch, eds), The suicide syndrome. London: Croon Helm; 1980 60. Glasser, M., Amdur, M.J. and Backstrand, J. The impact of psychotherapists and primary physicians on suicide and other violent deaths in a rural area. Canadian Journal of Psychiatry; 1985; 30: 195-202. 61. Glatt, K.M., Sherwood, D.W. and Amission, T.J. Telephone helplines at a suicide site. Hospital and Com- munity Psychiatry; 1986; 37: 178-180. 62. Gordon, R.H. Social class bias of suicide preven- tion volunteers. American Journal of Community Psychol- ogy; 1974; 2: 393-398. 63. Gould, M.S. and Shaffer, D. The impact of suicide in tlelvision movies: Evidence of imitation. New England Journal of Medicine; 1986; 315: 690-694. 64. Green, S. and Lee, H. Subsequent pregress of potentially lethal attempted suicides. Acta Psychiatrica Scandinavica; 1972; 1: 310. 65. Greer, F.L. and Strasberg-Weinstein, R. Suicide prevention center outreach: Callers and noncallers com- pared. Psychological Reports; 1979; 44: 387-393. 66. Greer, S. and Bagley, C. Effect of psychiatric inter- vention in attempted suicide: A controlled study. British Medical Journal; 1971; 1: 310-312. 67. Greer, S. and Anderson, M. Samaritan contact among 325 parasuicide patients. British Journal of Psychiatry; 1979; 135: 263-268. 68. Hassall, C. and Bagley, C. Suicide in Birmingham. British Medical Journal; 1972; 1: 310. 69. Hassall, C. and Trethowan, W.H. Suicide in Birmin- gham. British Medical Journal; 1972; 1: 717-718. 70. Hawton, K. Domiciliary and outpatient treatment following deliberate self-poisoning. In: R. Farmer and S.R. Hirsch, (eds). The suicide syndrome. London: Croon Helm; 1979: 246-258. 71. Helig, S., Faberow, N., Litman, R. and Schneid- man, E. The role of nonprofesional volunteers in a suicide prevention center. Community Mental Health Journal; 1968; 4: 287-295. 72. Henderson, A.S., Hartigan, J., Davidson, J., et al. A typology of parasuicide. British Journal of Psychiatry; 1977; 133: 631-641. 73. Hirsch, S. A Critique of Volunteer-Staffed Suicide Prevention Centres. Canadian Journal of Psychiatry; 1981; 26: 406-410. 74. Hirsch, S.R., Walsh, C. and Draper, R. Parasuicide: A review of treatment interventions. Journal of Affective Disorders; 1982; 4: 299-311. 75. Hitchcock, J. and Wolford, J.A. Alternatives to the suicide prevention approach to mental health. Archives of General Psychiatry; 1970; 22: 547-549. 76. Holding, T.A. The B.B.C. "Befriender: series and its effect. British Journal of Psychiatry; 1974; 124: 470-472. 77. Holding, T.A. Suicide and "The Befrienders". British Medical Journal; 1975; 3: 751-753. 78. Holding, T.A. and Barraclough, B.M. Undeter- mined deaths: Suicide or accident? British Journal of Psychiatry; 1978; 133: 542-549. 3-58 79. Inman, D.J., Bascue, L.O. Kahn, W.J. and Shaw, P.A. The relationship between suicide knowledge and suicide interviewing skill. Death Education; 1984; 8: 179- 184. 80. Jacobson, G.F., etal. The scope and practice of an early-access brief treatment psychiatric center. American Journal of Psychiatry; 1965; 121: 1176-1182. 81. Joan, P. Preventing Teenage Suicide: The Living Alternative Handbook. New York: Human Science Press; 1986. 82. Johns, M.W. Self-poisoning with barbituates in England and Wales during 1959-1974. British Medical Journal; 1977; 2: 1128-1130. 83. Johnson, F.G., Ferrence, R. and Whitehead, P.C. Self-injury: Identification and intervention. Canadian Psychiatry Association Journal; 1973; 18: 101-105. 84. Kalafat, J., Boroto, D.R. and France, K. Relation- ships among experience level and value orientation and the performance of paraprofessional telephone coun- selors. American Journal of Community Psychology; 1979; 5: 167-179. 85. Kashani, J.H. and Ray, J.S. Depressive Related Symptoms among Pre-school Children. Child Psychiatry and Human Development; 1983; 13: 233-238. 86. Kashani, J.H., McGee, R.O., Clarkson, S.E., Ander- son, J.C., Walton, L.A., Williams, S., Silva, P.A., Robins, A.J., Cytryn, L. and McKnew, D.H. Depression in a Sample of 9-year-old Children: Prevalence and Associated Charac- teristics. Archives of General Psychiatry; 1983; 40: 1217- 1223. 87. Kennedy, P. Efficacy of a regional poisoning treat- ment center in preventing further suicidal behavior. British Medical Journal; 1972; 4: 255-257. 88. Kennedy, P. and Kreitman, N. An epidemiological survey of parasuicide (attempted suicide) in general prac- tice. British Journal of Psychiatry; 1973; 123: 23-24. 89. Kennedy, P., Kreitman, N. and Ovenstone, |.M.K. The Prevalence of Suicide and Parasuicide ("Attempted Suicide") in Edinburgh. British Journal of Psychiatry; 1974; 124: 36-41. 90. Kessler, R.C. and Stipe. H. The impact of fictional television suicide stories on U.S. fatalities: a replication. American Journal of Sociology; 1984; 90: 151-167. 91. King, G.D. An evaluation of the effectiveness of a telephone counselling center. American Journal of Com- munity Psychology; 1977; 5: 75-83. 92. Klerman, G.L. and Paykel, E.S. Depressive pattern, social background and hospitalization. Journal of Nervous Mental Disorders; 1970; 150: 466-478. 93. Knesper, D.J. A study of referral failures for poten- tially suicidal patients: A method of medical care evalua- tion. Hospital and Community Psychiatry; 1982; 33: 49-52. 94. Knickerbocker, D.A. Clinical effectiveness of non- professional and professional telephone workers in a crisis intervention center. In: D. Lester and G. Brockopp, (eds). Telephone therapy and crisis intervention. Springfield, IL: C.C. Thomas; 1973. 95. Knickerbocker, D.A. and McGee, R.K. clinical effec- tiveness of nonprofessional and professional telephone workers in a crisis intervention center. In: Lester, D. and Brockopp, G.W., feds] Crisis intervention and counseling by telephone. Springfield, IL: Thomas; 1973: 298-309. 96. Knowles, D. On the tendency for volunteer helpers to give advice. Journal of Counseling Psychology; 1979; 26: 352-354. 97. Kogan, L.S. The short-term case in a family agen- cy. Social Casework; 1957b; 38: 296-302. 98. Krause, J. Suicidal behavior in New South Wales. British Journal of Psychiatry; 1975; 126: 313-318. 99. Kreitman, N., Smith, P. and Tan, E.S. Attempted suicide in social networks. British Journal of Preventive So- D.Shaffer: A Critical Review of Preventive Intervention.. cial Medicine; 1969; 23: 116-123. 100. Kreitman, N., Smith, P. and Tan, E.S. Attempted suicide as language: An empirical study. British Journal of Psychiatry; 1970; 116: 465-473. 101. Kreitman, N. The coal gas story: United Kingdom suicide rates, 1960-71. British Journal of Preventive and Social Medicine; 1976; 30: 86-93. 102. Lester, D. Prevention of suicide. Journal of the American Medical Association; 1973; 225-992. 103. Lester, D. Steps toward the evaluation of a suicide prevention center: Part one. Crisis Intervention; 1970; 2: 42-45. 104. Liberman, R. and Eckman, T. Behavior therapy vs. insight oriented therapy for repeated suicide attempters. Archives of General Psychiatry; 1981; 38: 1126-1130. 105. Litman, R.E. and Faberow, N.L. Evaluating the ef- fectiveness of suicide prevention. Proceedings of the Fifth International Conference for Suicide Prevention. London; 1969: 246-250. 106. Litman, R.E., Shneidman, E.S. and Faberow, N.L. Suicide prevention centers. American Journal of Psychiatry; 1961; 117: 1084-1087. 107. Litman, R.E., Faberow, N.L., Shneidman, E.S., Heilig, S.M. and Kramer, J.A. Suicide prevention telephone service. Journal of the American Medical Association; 1965; 192: 107-111. 108. Litman, R.E. Suicide prevention center patients: A follow-up study. Bulletin of Suicidology; 1970: 12-17. 109. Litman, R.E. Suicide prevention: Evaluating effec- tiveness. Suicide and Life Threatening Behavior; 1971; 1: 156-162. 110. Litt, I.F., Cuskey, W.R. and Rudd, S. Emergency room evaluation of the adolescent who attempts suicide: Compliance with follow-up. Society for Adolescent Medicine; 1983; 4: 106-108. 111. Littmann, S.K. Suicide epidemics and newspaper reporting. Suicide and Life Threatening Behavior; 1985; 15: 43-50. 112. McCain, M. Suicide at an early age. Boston Globe; March 25, 1984. 113. McCarthy, B.W. and Berman, A.L. A student operatd crisis center. Personnel and Guidance; 1971; 49: 523-528. 114. McCord, J. A thirty year follow-up of treatment ef- fects. American Psychologist; 1978; 33: 284-289. 115. McGee, RK. Crisis Intervention in the Community. Baltimore, MD: University Park Press; 1974. 116. McKenna, J., Nelson, G., Chatterson, J., Koperno, M. and Brown, J.H. Chronically and acutely suicidal per- sons one month after contact with a crisis intervention cen- ter. Canadian Psychiatric Association Journal; 1975; 26: 451-454, 117. Meacham, M. and Acey, K.T. Considerations in evaluating a crisis outreach service. Crisis Intervention; 1974; 5: 25-35. 118. Miles, C.P. Conditions predosposing to suicide: A review. Journal of Mental Disorders; 1977; 164: 231-246. 119. Miller, H.L., Coombs, D.W., Leeper, J.D. and Bar- ton, S.N. An analysis of the effects of suicide prevention facilities on suicide rates in the United States. American Journal of Public Health; 1984; 74: 340-343. 120. Miller, H.L., Coombs, D.W., Mukherjee, D. and Bar- ton, S.N. Suicide prevention services in America. Alaska Journal of Medical Science; 1979; 16: 26-31. 121. Montgomery, S.A., Montgomery, D.B., Rani, S.J., Shaw, P. and McAuley, R. Maintenance therapy in repeat suicidal behavior--A placebo controlled trial. Proceedings of the 10th International Congress of Suicide Prevention and Crisis Intervention; 1979; Ottawa. 122. Morgan, H.G., Burns-Cox, C.J., Pocock, H.J. and Pottle, S. Deliberate self-harm: Clinical and socioecomonic characteristics of 368 patients. British Journal of Psychiatry; 1975; 127: 564-574. 123. Morgan, J.P. and King, G.D. The selection and evaluation of the volunteer paraprofessional telephone counselor: A validity study. American Journal of ‘Com- munity Psychology; 1975; 3: 237-249. 124. Motto, J.A., Heilbron, D.C., Juster, R.P. and Bostrom, A.G. Communication as a suicide prevention program. Depression et Suicide; 1981: 148-154. 125. Motto, J.A. and Heilbron, D.C. Development and validation of scales for estimation of suicide risk. In: E.S. Shneidman, (ed). Suiidelogy, contemporary develop- ments. New York: Grune and Stratton; 1976: 169-199. 126. Motto, J.A. Evaluation of a suicide prevention cen- ter by sampling the population at risk. Suicide and Life Threatening Behavior; 1971; 1: 18-22, 127. Motto, J.A. Preliminary field-testing of a risk es- timator for suicide. Suicide and Life Threatening Behavior; 1985; 15: 139-150. 128. Motto, J.A. The psychopathology of direct self- destruction: A clinical model approach. In: K. Achte and J. Loenngpvist, (eds). Psychopathology of direct and indirect self-destruction: Psychiatria Fennica Supplementum: 47- 57. 129. Motto, J.A., Brooks, R.M., Ross, C.P. and Allen, N.H. Standards for suicide prevention and crisis centers. New York: Behavior Publications; 1974. 130. Motto, J.A. and Greene, C. Suicide and the medi- cal community. American Medical Association Archives of Neurology and Psychiatry; 1958; 80: 776-781. 131. Motto, J.A. Suicide and suggestibility-the role of the press. American Journal of Psychiatry; 1967; 124: 252- 256. 132. Motto, J.A. Suicide prevention for Highs persons who refuse treatment. Suicide and Life Threatening Be- havior; 1976; 6: 223-230. 133. Motto, J.A. Toward suicide prevention in medical practice. Journal of the American Medical Association; 1969: 210. 134. Murphy, G.E. The clinical identification of suicidal risk. In: H.L.P. Resnick and D.J. Lettieri, (eds). The predic- tion of suicide. Bowie, MD: Charles Press; 1974: 109-118. 135. Murghy, G.E. The physicians responsibility for suicide: 1. An error of commission, and 2. Errors of omis- sion. Annual Internal Medicine; 1975; 82: 301-309. 136. Murphy, G.G., Wetzel, R.N. and Swallow, C.S. Who calls the suicide prevention center? American Journal of Psychiatry; 1969; 126: 314-324. 137. Neilson, J. and Videbech, T. Suicide frequency before and after introduction of community psychiatry in a Danish island. British Journal of Psychiatry; 1973; 123: 35-399. 138. Neimeyer, R.A. and Macinnes, W.D. Assessing paraprofessional competence with the suicide interven- tion response inventory. Journal of Counseling Psychol- ogy; 1981; 28: 176-179. 139. Neimeyer, R.A. and Oppenheimer, B. Concurrent and predictive validity of the suicide intervention response inventory. Psychological Reports; 1983; 52: 594. 140. Nelson, G., McKenna, J., Okoperno, M., Chatter- son, J. and Brown, J.H. The role of anonymity in suicidal contacts with a crisis intervention center. Canadian Psychiatry Association Journal; 1975; 20: 455-459. 141. Neuringer, G. Methodological problems in suicide research. Journal of Consulting Psychology; 1962; 26: 273-278. 142. O'Brien, J.P. Increase in suicide attempts by drug ingestion: The Boston experience. Archives of General Psychiatry; 1977; 34: 1165-1169. 3-59 Report of the Secretary’s Task Force on Youth Suicide 143. Ottens, A.J. Evaluation of a crisis training program in suicide prevention for the campus community. Crisis In- tervention; 1980; 13: 25-40. 144. Otto, U. Suicidal acts by children and adolescents: a follow-up study. Acta Psychiatrica Scandinavica, Sup- plementum, 1972; 233. 145. Overstone, |.M.K. Spectrum of suicidal behaviorsin Edinburgh. British Journal of Preventive and Social Medicine; 1973; 27: 27-35. 146. Overstone, |.M.K. and Kreitman, N. Two syndromes of suicide. British Journal of Psychiatry; 1974; 124: 336-345. 147. Pallis, D.J., Barraclough, B.M., Levey, A.B., Jenkins, J.S. and Sainsbury, P. Estimating suicide risk among attempted suicides. British Journal of Psychiatry; 1982; 141: 37-44. 148. Pallis, D.J. and Sainsbury, P. The value of assess- ing intent in attempted suicide. Psychological Medicine; 1976; 6: 487-492. 149. Paykel, E.S., Myers, J.K,, Lindenthal, J.J. and Tan- ner, J. Suicidal feelings in the general population: A pleveisnce study. British Journal of Psychiatry; 1974; 124: 150. Paykel, E.S., Hallowell, C., Dressler, D.M., Shapiro, D.L. and Weissman, M.M. Treatment of suicide at- tempters: A descriptive study. Archives of General Psychiatry; 1974; 31: 487-491. 151. Pfeffer, C.R., Zuckerman, S., Plutchik, R. and Miz- ruchi, M.S. Suicidal behavior in normal school children: A comparison with child prychatie inpatients. Journal of the American Academy of Child Psychiatry; 1984; 21: 564- 569. 152. Phillips, D.P. Airplane accidents, murder, and the mass media: towards a theory of imitation and suggestion. Social Forces; 1980; 58: 1001-1024. 153. Phillips, D.P. The impact of fictional television stories on U.S. adult fatalities: new evidence on the effect of the mass media on violence. American Journal of Sociology; 1982; 87: 1340-1359. 154. Phillips, D. The influence of suggestion on suicide: Substantive and theoretical implication of the Werther ef- fect. American Sociological Review; 1974; 39: 340-354. 155. Phillips, D.P. Suicide, Motor Vehicle Fatalities, and the Mass Media: Evidence toward a theory of suggestion. American Journal of Sociology; 1979; 84: 1150-1174. 156. Phillips, D. Teenage and adult temporal fluctua- tions in suicide and auto fatalities. In: H.S. Sudak, A.B. Ford, N.B. Rushforth, (eds). Suicide in the Young. Boston, MA: John Wright PSG, Inc.; 1984. 157. Pitts, F.N. Jr. and Winokur, G.D. Affective disorder. lll: Diagnostic correlates and incidence of suicide. Journal of Nervous and Mental Disorders; 1964; 139: 176-181. 158. Pokorny, A.D. Prediction of suicide in psychiatric patients. Archives of General Psychiatry; 1983; 40: 249- 257. 159. Pokorny, A.D. Suicide rates in various psychiatric disorders. Journal of Nervous and Mental Disorders; 1964; 139: 499-506 160. Ringel, E. Suicide prevention in Vienna. In: H.L.P. Resnick, (ed). Suicide behaviours. Boston, MA: Little, Brown and Company; 1968: 381-390. 161. Robbins, D. and Conroy, R.C. A cluster of adoles- cent suicide attempts: Is suicide contagious? Journal of Adolescent Health Care; 1983; 3: 253-255. 162. Robins, E., Gassner, S., Kayes, J., Wilkinson, R.H. Jr. and Murphy, G.E. The communication os suicide in- tent: A study of 134 consecutive cases of successful (com- pleted) suicide. American Journal of Psychiatry; 1959; 115: 724-733. 163. Rogawski, A.B. and Edmundson, B. Factors affect- ing the outcome of psychiatry interagency referral. 3-60 American Journal of Psychiatry; 1971; 127: 925-934. 164. Rogers, J., Sheldon, A., Barwick, C., Letofsky, K. and Lancee, W. Help for families of suicide: Survivors sup- port program. Canadian Journal of Psychology; 1982; 27: 444-448. 165. Rosen, A. Detection of suicidal patients: An ex- ample of some limitations in the prediction of infrequent events. Journal of Consulting Psychology; 1954; 18: 397- 403. 166. Ross, C.P. and Motto, J.A. Group counseling for suicidal adolescents. In: H.S. Sudak, A.B. Ford and N.B. Rushforth, (eds). Suicide in the Young. Boston, MA: John Wright PSG, Inc.; 1984: 367-392. 167. Ross, C.P. Mobilizing schools for suicide preven- tion. Suicide and Life Threatening Behavior; 1980; 10: 239- 243. 168. Rudestrom, K.E. Physical and physiological responses to suicide in the family. Journal of Consulting and Clinical Psychology; 1977; 45: 162-170. 169. Sainsbury, P. Suicide in London, Maudsley Monograph No. 1. London: Chapman and Hall. 170. Sathyavathi, K. Suicide among children in Ban- galore. Indian Journal of Pediatrics; 1975; 42: 149-157. 171. Sawyer, J.B., Sudak, H.S. and Hall, S.R. A follow- up study of 53 suicides known to a suicide prevention cen- ter. Suicide and Life Threatening Behavior; 1972; 2: 228-238. 172. Schneidman, E.S. and Farberow, N.L. Clues to Suicide. New York: Blakison; 1957. 173. Schneidman, E.S. and Farberow, N.L. Statistical comparisons between attempted and committed suicides. In: N.L. Farberow and E.S. Schneideman, (eds). The cry for help. New York: McGraw Hill Book Company; 1961. 174. Schneidman, E.S. Suicide, lethality and the psychologies autopsy. In: E.S. Schneidman and M. Or- tega, (eds). Aspects of depression. Boston, MA: Little, Brown and Company; 1969. 175. Schneidman, E. In: A.C. Cain, (ed). Survivors of suicide. Springfield, IL: Charles C. Thomas; 1972. 176. School District of La Crosse. Student Assistance Program. La Crosse, WI; 1985. 177. Shaffer, D. and Bacon, K. A critical review of prevention intervention efforts in suicide with particular reference to youth suicide. Prevention and Intervention Work Group of the HHS Task Force on Youth Suicide; June, 1986; Oakland, CA. 178. Shaffer, D. and Fisher, P. The epidemiology of suicide in children and young adolescents. Journal of the American Academy of Child Psychiatry; 1981; 20: 545-565. 179. Shaffer, D. and Garland, A. An evaluation of New Jersey youth suicide prevention programs. Manuscript in preparation; 1986. 180. Shaffer, D. and Gould, D. NIMH Grant “A Study of Completed and Attempted Suicide in Adolescents" (# MH 31918). Progress Report; 1985. 181. Shaffer, D. Suicide in childhood and early adoles- cence. Journal of Child Psychology and Psychiatry; 1974; 15: 275-291. 182. Shepherd, D.M. and Barraclough, B.M. The After- math of Suicide. British Medical Journal; 1974; 2: 600-603. 183. Silver, J.S., Cohnert, M., Beck, A.T. and Marcus, D. Relation of depression of attempted suicide and serious- ness of intent. Archives of General Psychiatry; 1971; 25: 573-576. 184. Singh, A.N. Suicide prevention. Canadian Psychiatric Association Journal; 1973; 18: 117-121. 185. Slaikeu, K.A,, Tulkin, S.R. and Speer, D.C. Process and outcome in the evaluation of telephone counseling referrals. Journal of Consulting and Clinical Psychology; D.Shaffer: A Critical Review of Preventive Intervention.. 1975; 43: 700-707. 186. Slem, C.M. and Cotler, S. Crisis phone services: Evaluation of hotline program. American Journal of Com- munity Psychology; 1973; 1: 219-227. 187. South Bergen Mental Health Center. Adolescent Ricide awareness program: Proposal. East Rutherford, ; 1985. 188. Stack, S. The effect of Suggestion on suicide: A reassessment. Annual Meeting of the American Sociologi- cal Association; 1984; San Antonio, TX. 189. Stanley, M. Cholinergic binding in the frontal cor- tex of suicide victims. American Journal of Psychiatry; 1984; 141: 11. 190. Stein, D.M. and Lambert, M.J. Telephone counsel- ing and crisis intervention: A review. American Journal of Community Psychology; 1984; 12: 101-126. 191. Stelmachers, Z.T. Current status of program evaluation efforts. Suicide and Life Threatening Behavior; 1976; 6: 67-78. 192. Stengel, E. and Cook, N.G. Attempted suicide. Lon- don: Oxford University Press; 1958. 193. Stengel, E. A survey of follow-up examinations of attempted suicides. In: J. Waldenstrom, T. Barsson and N. Ljungstedt, (eds). Suicide and attempted suicide. Stock- holm: Nordiska Bokhadelns Forlag; 1972. 194. Sudak, H.S., Sawyer, J.B., Spring, G.K. and Coak- well, C.M. He referral success rates in a crisis center. Hospital and Community Psychiatry; 1977; 28: 530-532. 195. Sudak, H.S., Ford, A.B. and Rushforth, N.B. Treat- ment: Review and Comment. In: H.S. Sudak, A.B. Ford and N.B. Rushforth, (eds). Suicide in the Young. Boston, MA: John Wright PSG, Inc.; 1984: 417-426. 196. Tabachnick, N. and Klugman, D.J. No-name--a study of anonymous suicidal telephone calls. Psychiatry; 1965; 28: 70-78. 197. Taylor, P. Cluster phenomenon of young suicides raises contagion theory. Washington Post; March 11, 1984: 15-16. 198. Temoche, A., Pugh, T.F. and McMahon, B. Suicide rates among current and former mental institution patients. Journal of Nervous and Mental Disorders; 1964; 138: 124-130. 199. Trautman, P.D. and Shaffer, D. Treatment of child and adolescent suicide attempters. In: H.S. Sudak, A.B. Ford and N.B. Rushforth, (eds). Suicide in the Young. Bos- ton, MA: John Wright PSG, Inc.; 1984: 307-323. 200. U.S. Bureau of the Census. Statistical Abstract of the United States: 1982-1983 (103rd Edition). Washington, D.C.: U.S. Bureau of the Census; 1982. 201. Varah, C. The Smaritans in the 70's. London: Con- stable and Company, Ltd.; 1973. 202. Videka-Sherman, L. Coping with the death of a child: A study over time. American Journal of Orthop- sychiatry; 1982; 52: 688-698. 203. Videka-Sherman, L. Effects of participation in a self-help group for bereaved parents: Compassionate friends. Prevention in the Human Services; 1982; 1: 69-77. 204. Videka-Sherman, L. and Liberman, M. The effects of self-help and psychotherapy intervention on child loss: The limits of recovery. American Journal of Orthop- sychiatry; 1985; 55: 70-82. 205. Walk, D. Suicide and community care. British Jour- nal of Psychiatry; 1967; 113: 1381-1391. 206. Ward, J.A. and Fox, J. A suicide epidemic on an In- dian reserve. Canadian Psychiatric Association Journal; 1977, 22: 423-426. 207. Wasserman, |.M. Imitation and suicide: A reexamination of the Werther effect. American Sociologi- cal Review; 1984; 49: 427-436. 208. Weissman, M.M. The epidemioloy of suicide at- tempts. Archives of General Psychiatry; 1974; 30: 737-746. 209. Weissman, M. and Gammon, D. Epidemiology of Childhood Depression: Some Findings from a High Risk Study. Presented at the 32nd Annual Meeting of the American Academy of Child Psychiatry; October, 1985; San Antonio, TX. 210. Welu, T.C. A follow-up program for suicide at- tempters: Evaluation of effectiveness. Suicide and Life Threatening Behavior 1977; 7: 17-30. 211. Whitehead, P.C., Johnson, F.G. and Ferrence, R.G. Measuring the incidence of self-injury: Some methodological and design considerations. American Journal of Orthopsychiatry; 1973; 43: 124-148. 212. Wilkins, J.L. Predicting suicides. American Be- havioral Science; 1970; 14: 185-201. 213. Winer, W. the effectiveness of a suicide prevention program. Mental Hygiene; 1969; 53: 357-363. 214. Winokur, G. and Tsuang, M. The lowa 500: Suicide in mania, depression and schizophrenia. American Jour- nal of Psychiatry; 1975; 132: 650-651. 215. Wold, C.I. Characteristics of 26,000 suicide preven- tion center patients. Bulletin of Suicidology; 1970: 24-48. 216. Wold, C.I. and Litman, R.E. Suicide after contact with a suicide prevention center. Archives of General Psychiatry; 1973; 28: 735-739. 3-61 OVERVIEW OF PREVENTION EFFORTS IN ADOLESCENT SUICIDE Betsy S. Comstock, M.D., Professor of Clinical Psychiatry, Baylor College of Medicine, Houston, Texas Jane T. Simmons, Ph.D., Consultant, Texas Department of Mental Health and Mental Retardation, Houston, Texas Jack L. Franklin, Ph.D., Project Director, Texas Teen Suicide Project, Houston, Texas SUMMARY The development of general suicide preven- tion efforts in the United States and beyond is reviewed, with emphasis on the paucity of attention to youth suicide until recent years. Training programs are reviewed, again with few found specific to youth. A current sur- vey of prevention and intervention programs in the United States is introduced, detailing the types of programs which could be iden- tified. Special therapy needs of adolescents are considered, and recommendations for fu- ture programs are detailed. INTRODUCTION Suicide stands out among major causes of death when accounting for deaths which might have been prevented by an act of will. Prevention efforts in communicable diseases and in disorders with parameters influenced by human choice, such as lung cancer and smoking, are considered relatively effective; suicide, however, is the choice not to live and it can be prevented only by a change in choice. As a result, prevention efforts have held the attention of those concerned with suicide for as long as suicide has been ap- proached as a public health problem. During the past decade the dilemma of increasing 3-62 rates of suicide by American youth has brought into focus the waste of life involved when the very young choose to take their lives. The purpose of this paper is to review what has been done in suicide prevention in the past and to summarize the types of suicide prevention efforts currently occur- ring in American communities, with specific reference to youth. HISTORICAL REVIEW The suicide prevention movement had its origins in the mid-1960s; there was little at- tention paid to suicide before that time. The authors studying suicide most often quoted in earlier years were Freud (1), and Durkheim (2). Freud believed that suicide was the ultimate example of introjected rage and self punishment; he also speculated on the existence of a separately operating, biologically established death instinct. Durkheim’s early studies of sociological fac- tors associated with suicide are best remem- bered for his elaboration of the anomie of the suicidal individual. Of particular interest in the context of youth suicide was the 1910 meeting of the Vienna B. Comstock: Overview of Prevention Efforts. . . Psychoanalytic Society, reported in OnSuicide (3). The meeting was planned because of the suicide death of a school boy and focused on harmful influences affecting young people. Actually, suicide as a problem was well estab- lished long before the work of these two giants. Goethe, another literary giant, brought attention to romantic suicide in his book, The Sorrows of Young Werther. 1t is said that after Goethe’s love-struck Werther’s story was popularized (4), a num- ber of imitative suicides occurred among German youths. The romanticization of suicide may have accounted for other clusters of youth suicides over the years, although this phenomenon was not actively studied until very recent times when alarm over the in- crease in absolute numbers of youth suicides stimulated research. It was not suicide among the young that stimulated the beginning of the suicide - prevention movement. During the 1960s, suicide prevention centers came into being in a number of different locations. The earliest centers generally developed out of the inter- est of mental health professionals and, for the most part, consisted of telephone answering services staffed by trained volun- teers offering crisis intervention counseling anonymously. Impetus was given to suicidology as a field of study by legislation enacted in the 1960s by the United States Congress. Early advocates for suicide as a deserving and needed field of study included Dr. Edwin Shneidman, Dr. Robert Litman, Dr. Norman Farberow, Dr. Harvey Resnik, Dr. Seymour Perlin, and others. The legislation established a Center for Studies of Suicide Prevention within the National Institute of Mental Health (NIMH). This center, headed first by Dr. Shneidman, and later by Dr. Resnick, provided a focus for suicide efforts over a number of years. Beginning in 1968, the center published the Bulletin of Suicidology. Research money distributed through its grant program gave major impetus to investigations throughout the nation. During and preceding this period in the United States, several organizations in other countries became involved in suicide preven- tion activities. The Samaritans organization originated in England and spread world- wide. Their activities are detailed in another paper in this volume. A parallel effort by Contact Teleministries world-wide ac- counted for many additional telephone answering, crisis intervention services. By 1968, sufficient interest in the field of suicide prevention had developed in the United States and a sufficient number of workers existed in the field to warrant a na- tional meeting. The first meeting was or- ganized in Chicago, Illinois under the sponsorship of the University of Chicago and became the formative meeting for the American Association of Suicidology (A.A.S.). Suicidology was a new term intro- duced by a charter member and original leader of the association, Dr. Edwin Shneid- man, and indicated the scope of the new or- ganization, that is, the study of suicide and its psychological, sociological, and clinical manifestations. Despite the academic focus indicated by the title and expressed by the early members, an immediate division developed within the organization between those who were primarily invested in suicide prevention (especially the volunteers) and those who were academicians. Over the years this divergence of interests has persist- ed within the organization, nevertheless, the intent of the original organizers has been honored in that the A.A.S. has not been divided. It remains a multidisciplinary or- ganization where research and application reinforce one another. Through an annual national meeting, the organization has fostered research and reporting on varied aspects of suicide. One of its most important contributions has been the development of standards by which suicide prevention centers can be judged and certified. The organization’s scientific journal and newslet- ter have had the desired effect of stimulating continuously increasing interest in the area. 3-63 Report of the Secretary's Task Force on Youth Suicide Youth suicide was a special focus among the early students of suicide. Review of the an- nual programs of the A.A.S. indicates that an increase in suicide among youth was noted as early as 1974, although the increase at that point was not marked enough to allow firm conclusions about the dimensions of the problem. In the decade from 1974 to 1984, the very clear increase in youth suicide was documented and led, not only to an increase in the number of publications and specialized programs in suicide prevention for youth, but also to specific federal responses. In 1979, the U.S. Public Health Service promulgated the 1990 Objectives for the Nation; among them were a number of objectives related to control of stress and violent behavior. NIMH created a staff position relating to suicide issues, announced a programmatic goal of a 10 percent reduction in youth suicide in the next decade, and aimed for identification of crisis telephone services by 60 percent of youth. Specific legislation at the State level, first in California and sub- sequently in other States, was introduced for dealing with suicide prevention programs for youth. In 1985, the Secretary of Health and Human Services established the Secretary’s Task Force on Youth Suicide. A series of meetings and reports have been generated by this task force. During the approximately two decades when suicide prevention was a specific focus of professionals in the United States, the num- ber of suicide prevention programs steadily grew throughout the country. Since 1981, new programs dealing with primary preven- tion efforts in schools have been developed. Currently in the U.S. about 1,000 suicide prevention or crisis programs have been identified exclusive of community mental health programs. Generally these programs were organized with goals broader than suicide prevention. Their names often reflect the broader crisis intervention pur- pose, such as Crisis Center, Telephone Hot- line, etc. Some, but by no means the majority, of these programs have established components specifically related to adoles- cent suicide. School programs have become 3-64 quite widespread, generally having been in- itiated only after a suicide involving a student has brought the issue of youth suicide strong- ly into focus. In some instances, school programs are provided by outside experts, in others educational psychologists employed in school systems have developed training programs for student bodies. An ongoing component of these programs is the task of dealing with students directly affected when a close friend has died from suicide. In the late 1960s and early 1970s a number of crisis programs were organized specifically for the youth population and were advertised as such. They tended to use youthful volun- teers as telephone respondents. Typically, these programs were organized by young people, training was less rigorous than for the more general suicide prevention programs, and, for the most part, these programs have disappeared from American cities. There is some appeal to the notion that a young per- son in crisis might prefer to talk to a willing peer volunteering in a telephone response service; nevertheless, the relative lack of sur- vival of these programs might suggest other- wise. In any case, the appropriate and adequate training of youthful volunteers remains problematic and has not been at- tempted by most established services. Other suicide prevention efforts have developed in connection with the emergence of emergency psychiatry as a sub-specialty. General and psychiatric hospitals have in- stalled discrete programs for dealing with emergencies brought to hospitals and with psychiatric aspects of medical treatment in general hospital emergency rooms. These programs encounter a large number of patients who have taken overdoses, and, especially in training hospitals around the country, have developed program com- ponents specific to suicide. Recognition of the need for ongoing care after emergency interventions with suicidal patients have led to the establishment of outpatient programs for suicidal patients. Several centers provide ongoing care through therapy groups homogeneous for suicide ideation. Many of B. Comstock: Overview of Prevention Efforts. . . these are located in public outpatient clinics associated with teaching hospitals, in com- munity mental health programs, and as walk- in services in a few suicide prevention centers. The author maintained a therapy group for several years for adolescent suicide attempters treated in a general hospital emergency room. Yet another type of suicide-related effort involves therapy groups for survivors of a suicide. Typically, these attract the parents of young people who killed themselves. Such groups exist as free-standing services staffed by mental health professionals. Many have developed as walk-in services in suicide prevention centers. AFFILIATIONS IN OTHER COUNTRIES Thus far, we have focused primarily on programs within the United States. In fact, the suicide prevention movement was active in Europe earlier than in the United States. In 1960, the first international suicide prevention meeting was held and the Inter- national Association for Suicide Prevention was formed with central offices in Vienna. The name of the organization was later changed to indicate the broader interests shared by member organizations. It con- tinues today as the International Association for Suicide Prevention and Crisis Intervention. This group, like the A.A.S,, holds plenary and regional meetings and publishes a scientific journal. The 25th an- niversary meeting was held in Vienna in 1985. The A.A.S. has functioned as a national member of the international association since A.A.S. was formed. SUICIDE PREVENTION TRAINING Training for mental health professionals in the area of suicide prevention has increased steadily in recent years as the visibility of suicide, especially among adolescents and youth increased. Schools of social work, nursing, psychology, and psychiatry have par- ticipated in and expanded their emphasis on training. There remains, however, a substan- tial lack of uniformity among programs across the country; some refer to suicide prevention in single lectures and case reviews, and others provide discrete cur- riculum segments containing many hours of both didactic teaching and case experience in suicide prevention. The scope of instruction in professional schools generally covers the identification of populations at risk, in- dividual case assessment techniques with respect to suicide risk, techniques in crisis in- tervention relevant to suicidal persons, and special care needs of particularly high-risk populations. Specific attention to adoles- cent suicide may not go beyond acknow- ledging that suicide in this group has shown recent increases. The dilemma for profes- sional schools is that a well-defined body of knowledge about adolescent suicide does not yet exist and few individuals have specific ex- perience or training in this area to be able to teach it well. Several specialized training programs deserve mention. The National Institute of Mental Health provided funding since 1967 for a fellowship program in suicidology at the Johns Hopkins University School of Medicine. One outcome of this program, headed by Dr. Seymour Perlin, was a hand- book (5) which provided an important resource for other training programs (4). Of interest, this volume published in 1975, did not index adolescents or youth or teens, an accurate reflection that special concern for youth had not yet emerged. Many of the cur- rent leaders in the field were trained in that program. After the close of the Johns Hop- kins program, there was no specific training opportunity for individuals motivated for an academic career in suicidology. Recently Dr. Ronald Maris developed a new fellowship program at the University of South Carolina which offers promise of renewed leadership in the field. It seems to be the general case that programs for physician’s assistants and on-site inter- veners such as emergency medical tech- 3-65 Report of the Secretary’s Task Force on Youth Suicide nicians (EMTs) and policemen are relatively unsophisticated in suicide prevention treat- ment. Model curricula have been developed for these groups and have been promulgated by the American Psychiatric Association’s Task Force on Emergency Psychiatry Care Issues (6). The situation for volunteers in suicide prevention centers is relatively better defined than that for mental health profes- sionals. Training curricula for volunteers have been developed (7) and widely promoted and standards and criteria for training of volunteers has improved the uniformity of training around the country. The scope of training for center volunteers includes didactic instruction in crisis inter- vention techniques, experiential involve- ment through role-playing, and supervised participation in crisis work on telephone lines. Volunteers often become extremely proficient in dealing with people in crisis and may have more training for this task than do mental health professionals. It remains the case however, that specific training in deal- ing with adolescents is not separately ad- dressed in most programs. Textbooks devoted in whole or in part to suicide prevention inevitably lag a number of years behind the current state of knowledge inthe field. Itis not surprising that the recent books offer relatively little in the area of adolescent suicide prevention. Several books, in the past two years, do address this area and include Youth Suicide, (Peck, et al. 8). Two scientific journals are devoted to suicide-related topics: Suicide and Life Threatening Behavior, the journal of the A.A.S., and Suicide, the journal of the Inter- national Association of Suicide Prevention and Crisis Intervention. An earlier publica- tion, Bulletin of Suicidology was published by NIMH during the years of the Center for Studies of Suicide Prevention. There have been some efforts to develop model curricula in suicide prevention. Dr. James Lomax has written curriculum recom- 3-66 mendations for psychiatry residency programs which were accepted by the or- ganization of heads of psychiatric training programs and is available from them (9,10). Dr. Alan Berman, in heading an educational committee for the American Association of Suicidology, collected curricula from training programs around the country and developed specific professional training recommenda- tions. The most effective way to upgrade educa- tional emphasis in an area is to influence the accreditation examinations in that area. Ef- forts have been made, for example, to in- crease the number of questions about suicide and its prevention in the examinations of the American Board of Psychiatry and Neurol- ogy. These examination questions have not yet focused directly on adolescent suicide, but it is hoped that they will do so in the fu- ture. The development of standards in suicide prevention efforts has been a uniquely thor- ny problem. Standards of care are needed both in specific suicide prevention programs and in any program offering mental health services. Workers have been reluctant to delineate highly specific standards because of the likelihood of related litigation in the event of a completed suicide. A.A.S. has considerable experience with standards developed more than a decade ago for suicide prevention centers (11). These were organized as minimal standards of com- petence for programs and have been used as the basis for a certification process for suicide prevention centers. This has been an out- standing effort in that it has had obvious and gratifying impact on the quality of services developed within programs and has not in- duced troublesome litigation. In contrast, the experience of hospitals in es- tablishing suicide prevention standards has been very mixed. Tremendous lack of uniformity exists among hospitals around the nation in policies directed to suicide preven- tion. Some hospitals have very strict require- ments whereas others have decided to do nothing in order to avoid the problems of B. Comstock: Overview of Prevention Efforts. . . diversity in care (12). A.A.S. has worked for a number of years in surveying hospital prac- tices and currently is developing a set of recommendations for consideration by the Joint Commission on Accreditation of Hospitals. The latter group has been reluc- tant to establish standards for care of suicidal individuals although it does require documentation of risk assessment. In con- trast, a major care provider, the Veteran’s Administration (V.A.), has had a set of nurs- ing regulations and a manual on suicidal and violent patients for several years. These standards are far more specific than are those encountered in the private sector. It is un- clear whether these have substantially raised the awareness and sensitivity of V.A. staff members to suicide issues compared to professionals in the private sector. As ex- pected, the V.A. standards did not address youth suicide as a special problem. Since adolescents requiring psychiatric hospitaliza- tion generally are segregated from the adult population, it seems likely that a professional group such as the American Society for Adolescent Psychiatry may become involved in development of specialized standards of care for this group. SUICIDE PREVENTION EFFICACY One of the most troublesome aspects for planners in the area of suicide prevention for adolescents is the lack of substantial and con- vincing data about the efficacy of existing programs. Outcome research is made dif- ficult by the mobility of the young popula- tion, the lack of adequate and accurate reporting of suicide deaths, and ethical dilemmas encountered when specific inter- vention programs are to be compared to a control population deemed equally at risk; that is, withholding intervention from populations at risk cannot be sanctioned. A few outcome studies have been greeted with enthusiasm but also with considerable methodological criticism. One study in- volved the comparison of similar towns in England where the Samaritans were and were not active, with a favorable decline in the suicide rate in the town where the Samaritans intervened through a program of befriending those identified as being at risk (13). Another piece of evidence involved the cessation of carbon-monoxide-producing coal oil as a cooking fuel in Great Britain with a corresponding drop in the suicide rate (14). These data have been particularly interesting to advocates of firearms control in the United States because they suggest that the control of a popular means of suicide may indeed in- fluence the overall frequency of death. Lester has reported a correlation across States between handgun control and suicide (15) and most important, has extended his analysis to show fewer adolescent deaths in States with stricter controls (16). No con- vincing studies, as yet, in adolescent suicide show that specific kinds of intervention other than gun control absolutely decrease the suicide rate. The shifts in the suicide rate over time are confounding variables. It ap- pears, for example, that the alarming rise during the past 15 years in adolescent suicide is now reaching a plateau and we may be ex- periencing the beginning of a gratifying drop in these deaths. The stated goal of the Department of Health and Human Services to achieve a 10 percent drop in adolescent suicide may have been fortuitously timed. That, of course, does not guarantee that any- thing efficacious has been done. The task of identifying suitable comparison groups and discretely defined intervention strategies remains for researchers in the future. PUBLIC AWARENESS In the absence of interventions of proven ef- ficacy, suicide prevention planners intuitive- ly have assumed that factual information serves the public well and that increased sen- sitivity to adolescent suicide may decrease its occurrence. This area of concern has been enjoying considerable popularity in recent years as evidenced by a number of television documentaries treating the problem of adolescent suicide in some depth. The human interest potential for such program- 3-67 Report of the Secretary's Task Force on Youth Suicide ming is very high and, in general, the quality of media productions has been considered high. The national media, when airing such documentaries, have taken responsibility for alerting community service providers and for developing expert commentary on the con- tent of such programs, particularly with the goal of guarding against suicide by suggestion to viewers and listeners. One network dis- tributed elaborate and high quality school curriculum materials in advance of airing a youth suicide documentary (17). Local school districts have become alerted to the problems of adolescent suicide and have responded with a great variety of suicide prevention programs in schools. Most of these include educational efforts to heighten the awareness within the student body of the possibility of intervention by friends when a troubled youth is identified. School cur- riculum planning is very active in this area at the present time. In various locales, depend- ing on school personnel, educational psychologists and guidance counselors make classroom presentations or invite outside professionals to teach about youth suicide. Charlotte Ross in California was a pioneer in the latter type (18). Schools also have stimu- lated parent-teacher organizations to attend to this area and presentations in their annual programming are becoming very frequent. A.AS. is currently collecting suicide preven- tion materials developed for schools and will develop specific recommendations and models for school awareness programs. In a closely related development, A.A.S. and other groups are giving attention to plans for school intervention programs when suicides occur or are threatened. PROBLEMS IN SUICIDE PREVENTION One of the major difficulties in planning for suicide prevention programs is that parameters for identifying the population at risk are so non-specific that inevitably a very large population must be dealt with. Inter- ventions focusing on public awareness ob- viously do not suffer from this problem and 3-68 this may account for the great amount of energy directed toward that effort. George Murphy (19) has clearly defined the dilemma of overinclusiveness of risk measures. For example, very high risk groups, such as suicidal manic depressive and schizophrenic patients are underrepresented in the adoles- cent population. For groups in which suicide risk can be identified by individual behavior, such as suicide attempt or threat, the actual risk is only about 5 percent and all cases iden- tified in advance account only for about 20 percent of the eventual fatalities (20). Prevention efforts, then, must be rather broadly directed and must approach in- dividual dynamic issues which in the future may be found to be associated with suicidal impulses in youth. A further problem in implementing suicide prevention efforts arises from the conflicting need to control the behavior of an identified potentially suicidal individual on the one hand, and the need to promote growth and personal responsibility on the other hand. Every therapist is or should be aware of this conflict in every situation of intervention with potentially suicidal youth. There seems to be great variation among therapists in the way this is addressed. Some go to great lengths to prevent the possibility of death, even though this provides considerable inter- ference in the progress of therapy. Others reason that some suicidal deaths inevitably will occur even in therapy and that the greater preservation of life and quality of life is assured by promoting personal respon- sibility on the part of suicidal individuals. With so much disagreement among therapists it seems unlikely that clear stand- ards for intervention techniques can be developed. There is a further conflict in the intervention models endorsed for suicide prevention ef- forts. Historically, major emphasis has been given to the crisis model where suicide is seen as a time-limited crisis in the life of an in- dividual whose pre- and post-crisis ego func- tions are at a reliable level. The task addressed in the crisis model is that of B. Comstock: Overview of Prevention Efforts. . . restraining destructive acts by the individual until the crisis is passed. This seems a defen- sible model with considerable support from existing crisis intervention services, even in the absence of rigorous research validation. In contrast, a number of suicidologists see suicide as an end-point in a suicidal life style. Both Edwin Shneidman (21) and Ronald Maris (22) have written extensively on this point of view. Indeed, most clinicians from time to time have had experience with a chronically suicidal individual. Such experiences dis- courage adherence to the crisis model. It is possible, of course, for an individual therapist working with an individual client to ac- complish some combination of these view- points, paying attention equally to the meaning over time of self destructive pat- terns in the individual and to the crisis which occurs when a suicide impulse is active. In program planning, however, such individual- ized attention may be neglected. Suicide in- terventions tend to be very time-limited with relative neglect of the important areas of referral and long-term followup. Longer term treatment of suicidal youth ap- propriately expands beyond the issue of the self-destructive behavior which typically is the cause for initiating therapy, and which is often referred to as a cry for help. Review of the dynamics of youth self-destruction in- evitably becomes a review of the multiple psychological tasks required in the passage into adulthood and of the psychopathology specific to this age group. For the purpose of this overview paper, only a few points will be made. In satisfactory maturation, the adolescent or young adult reworks, in a definitive way, the conflict between the wish to gain security from the care provided by others and the wish to gain independence and self reliance. Under the best of circumstances this conflict produces trial solutions and failures, disap- pointments, and changes of direction. The extraordinary grandiosity of mid-adoles- cence, when anything seems possible and confidence may outreach wisdom, must be revised in the light of limitations in personal abilities, social resources, and the exclusions required by progress in a particular direction. Major changes in interpersonal relationships occur, especially in families; and investments in relationships outside the family become crucial. This is the interval when, if develop- ment is satisfactory, self concept reaches a relatively stable form, including such impor- tant aspects as body image, self esteem, self motivation, and differentiation from others. Failures in this stabilization of self concept have far-reaching consequences. Youth suicide in general terms can be under- stood as a reaction to living with such failures. The compelling question in the context of recent increases in youth suicide is "why now?" The compelling need in intervention in youth suicide, in the absence of useful answers to that question, is for individualized work with individuals in distress. Crisis inter- ventions, at best, can keep a young person alive during a period of very high risk and can facilitate entry into longer-term therapy. It is mainly in the course of longer work that the individual life course can be altered. Crisis events at times are referred to as growth opportunities, and this idea has merit both in the sense that crisis-anxiety promotes development of new coping skills and in the sense that failing character defenses become rather transparent during crisis, making it relatively easy to grasp underlying dynamics. It must be remembered, however, that a protective boost through a crisis interval also may reinforce a young person’s sense of in- competence. The quality of work done in crisis resolution is critical to the final impact of the crisis, and that work may require an ex- tended therapy interval. With respect to adolescent suicide there ex- ists a troublesome lack of convincing data on the outcome of intervention techniques, on the means by which suicide ideation and urges seem to be contagious in groups of young people, and on the importance of major sociological variables such as family mobility, divorce in the family, changes in academic standards in schools, lack of youth 3-69 Report of the Secretary's Task Force on Youth Suicide optimism about future employment, etc. There is immediate need for research in all of these areas. TYPES OF PREVENTION PROGRAMS In preparation for the National Conference on Prevention and Intervention in Youth Suicide, a research task force in Houston sur- veyed programs throughout the United States which may provide services to suicidal youth. Details of the methodology and sum- mary results are presented in a separate paper (23), and a survey analyzing charac- teristics of programs, by program type, are presented in another paper (24). For the sur- vey, programs were classified as crisis phone lines, walk-in crisis clinics, hospital-based emergency programs, mental health centers with crisis components, school intervention programs, free standing crisis stabilization units with beds, and combination programs. It is noteworthy that these programs consis- tently identified a combination of prevention and intervention goals. They seem generally to subscribe to a crisis model of intervention, although most programs described extensive referral linkage in their communities. Neither their titles nor their services defined them as being organized specifically to respond to youth clients. Probably the single outstanding result of this survey was to learn that adolescent suicide has not, as yet, had direct impact on community agencies except for schools. In yet another paper, the responses of com- munities where youth suicide has had special visibility are discussed (25). Clusters of adolescent suicides are special and alarming events. They have been reported from many regions of the country, from small com- munities and large but common characteris- tics that might identify communities at risk have not emerged. Current research will contribute better understanding of this phenomenon. Interventions in youth suicide clusters generally have provided emergency training for school personnel, awareness education for students, rapid reinforcement 3-70 of the treatment community by referral net- works, and organizing professional volunteer services. As yet, no systematic examination of the effectiveness of such efforts has been performed, and no natural history of a cluster has been defined. It is not known whether the interventions undertaken have decreased or increased the suicides, although it seems that they have been effective in en- ding the clusters. Detailed case studies are needed, with particular attention to the role of the media. ADOLESCENT SUICIDE PREVENTION NEEDS In evaluating the planning for expansion of the services available in any community for the prevention of youth suicide, the follow- ing considerations seem important. First, programs within the community need to be especially visible to young people. This can be accomplished through public service announcements of entry sites, especially for suicide prevention programs. At least two types of crisis services are needed: those available by telephone contacts and those available for individuals identified in hospi- tals as suicidal youth. To this may be added crisis services for individuals identified in schools where the intervention needs go beyond the capability of school personnel. For all of these entry categories, a strong referral network is needed. Second, there is unfortunately, substantial loss of individuals who have undergone some treatment in the course of a suicidal crisis and been referred for on-going therapy. The referral network needs to be well understood within the community and should include the following components: 1) Therapists ex- perienced in family therapy, especially fami- ly therapy involving adolescents. 2) Specialized support groups of peers. (Ex- perience shows that the old-fashioned adage against dealing with suicidal people together in a group is unwarranted--such therapy seems to work.) 3) Social work assistance needs to be available to troubled youth to at- B. Comstock: Overview of Prevention Efforts. . . tend to specialized needs, including residen- tial placement, educational needs, and legal help. 4) Long term psychotherapy needs to be available. This is especially true in the public sector where it often may be difficult to obtain. 5) Both hospital and partial hospi- tal programs need to be available with staffs trained specifically in dealing with suicidal youth. Partial hospital programs are espe- cially attractive for the large number of in- dividuals who have survived through a suicide crisis, have no immediate suicide in- tent, and can both remain in contact with school and family and simultaneously can have the advantage of daily part-time hospi- tal treatment. REFERENCES 1. Freud S: Mourning and Melancholia. Standard edition of the Complete Psychological Works. London: Hogarth Press V.X14. 2. Durkheim E: Suicide, (1987), trans. J.A. Spaulding and G. Simpson. Glencoe, IL.: The Free Press 1951. 3. Friedman P (ed.): On Suicide with Particular Reference to Suicide Among Young Students. New York: Inter. Univ. Press 1967. 4. Friedenthal R: Goethe: His Life and Times. New York: World Publishing Co. 1965. 5. Perlin S: AHandbook for the Study of Suicide. New York: Oxford Univ. Press 1975. 6. Friedman RS, Barton G, Comstock BS, and Walker E: Training and Research in Emergency Psychiatry. In: Emergency Psychiatry: the Administrative Handbook. Bar- ton, Gail, Friedman, Rohn (eds.), Haworth Press (in press). 7. Training Manual, American Association of Suicidol- ogy, Denver. 8. Peck ML, Farberow NL, Litman, RE: Youth Suicide. New York: Springer 1985. 9. Lomax JW: A Proposed Curriculum on Suicide Care for Pjeivesy Residents. Suicide and Life Threaten- ing Behavior, 1 (in press). 10. The American Association of Directors of Psychiatric Residence Training Programs offers cur- riculum recommendations through its offices at the In- stitute for Living, New Haven, Connecticut, 11. Certification Manual. American Association of Suicidology. Denver. 12. Litman RE: Hospital Suicides: Law Suits and Stan- dards. Suicide and Life Threatening Behavior 24(4), 1982. 13. Bagley D: An evaluation of suicide prevention agencies. Life Threatening Behavior 1:245-259, 1971. 14. Barraclough BM, Jennings C, and Moss J: Suicide Prevention by the Samaritans. Lancet 237-238, 1977. 15. Lester D, Murrell ME: The Influence of our Control Laws on Suicide Behavior. Am J Psychiatry 140:1259, 1983. 16. Lester D: Preventive Effect of Strict Handgun Con- trol Laws on Suicide Rates. Am J Psychiatry 140:1259, 1983. 17. Columbia Broadcasting System: docudrama. Silence of the Heart. Classroom guide. 1985. 18. Ross CP: Mobilizing Schools for Suicide Preven- tion. Suicide and Life Threatening Behavior 10(4):239-243, 1980. 19. Murphy G: On Suicide Prediction and Prevention. Archives of General Psychiatry 40:343-344, 1983. 20. Comstock BS: Suicide: Emergency Issues, Chap- ter in Phenomenology and Treatment of Psychiatric Emer- goricies, B. Comstock et al. (eds.), New York: Spectrum ublications, 1984. 21. Schneidman ES: The Gifted in Suicide, Theory and Clinical Aspects. Littleton, MA. PSG Publishing Com- pany, Inc. pp. 309-322, 1979. 22. Maris R: Pathways to Suicide. The Johns Hopkins University Press. Baltimore 67-99;317-319, 1981. 23. Simmons JT: Prevention/Intervention Programs for Suicidal Adolescents. Prepared for the Secretary's Task Force on Youth Suicide. June 1986. 24. Franklin JL: Characteristics of Suicide Preven- tion/Intervention Programs: Analysis of a survey. Prepared for the Secretary's Task Force on Youth Suicide. June 1986. 25. Comstock, BS: Community Response to Adoles- cent Suicide Clusters. Prepared for the Secretary's Task Force on Youth Suicide. June 1986. ACKNOWLEDGMENT --- cooooemmmreeneeenns - The author wishes to thank Mrs. Carolyn Patterson for assisting in the preparation of this document. Address correspondence or requests to Dr. Betsy Comstock, V.A. Medical Center, Day Hospital (116A5), 2002 Holcombe Blvd., Houston, Texas 77030. 3-71 COMMUNITY RESPONSE TO ADOLESCENT SUICIDE CLUSTERS Betsy S. Comstock, M.D., Professor of Clinical Psychiatry, Baylor College of Medicine, Houston, Texas Jane T. Simmons, Ph.D., Consultant, Texas Department of Mental Health and Mental Retardation, Houston, Texas Jack L. Franklin, Ph.D., Project Director, Texas Teen Suicide Project, Houston, Texas SUMMARY The typical community reactions to disaster, as identified from the literature, parallel in- dividual reactions, including death—preoc- cupation, guilt, psychic numbing, rage, and a search for explanations. Adolescent suicide clusters are examples of disasters of limited scope but nevertheless generate fear respon- ses because of their uncertain duration and extent, and their implication that something is wrong with community quality of life. A cluster of suicides in Clear Lake, Texas is reported as a case study, and recommenda- tions for community planning are suggested. INTRODUCTION Youth suicide is increasing; knowing that it is increasing invokes many questions. Is there something about the quality of community life in recent times that accounts for the in- crease? What is the role of the family as an institution? Is the growing instability of families and the frequency of divorce direct- ly connected with the increased incidence of adolescent suicide? What are the roles of other institutions in our communities: chur- ches, schools, the justice system, the media, and mental health service providers? Are these institutions failing our young people in 3-72 some crucial way or ways that connect direct- ly with the increase in adolescent suicide? How have communities responded to adoles- cent suicides? How should they respond? A significant literature exists on community responses to natural and to man-made dis- asters from which much can be learned about the after-effects of crisis on survivors and about responses of individuals and com- munity groups not directly involved in the dis- asters (1). How these experiences and data obtained from them can be applied to adoles- cent suicide has hardly been studied. The suicide death of a young person is a remarkably personal and private event, but it has powerful impact on many others who knew the dead youth. Since the network of acquaintances of a young person tends to be large through school contacts, and because a young person’s death is so unexpected, youth suicide generates complex bereavement pat- terns and invites comparison with the bereavement tasks following a disaster. When a cluster of adolescent suicides occurs and is reported in the media, many more comparisons are invited. Lifton (2,3) has identified five survivor reactions after a dis- aster: death-preoccupation, guilt, psychic B.Comstock: Community Response to Adolescent Suicide... numbing, poorly focused rage, and a search for meaning or explanation. These seem easily transferable to the situation following adolescent suicide clusters. Green (4) reviewed studies of psychological sequelae of disasters and emphasizes the im- portance of the geographic centrality of events as determinants of outcome. For ex- ample, a plane crash does not affect the sup- port network of survivors, and is considered a peripheral disaster. A natural disaster, however, may cause death, property loss, and residential displacement within a com- munity, and thus is a central disaster. Erick- son (5) focused on the community response to disasters, studying the Buffalo Creek flood where whole communities were destroyed. He suggests that the availability of pre-exist- ing community support is a critical factor in the severity of disaster sequelae. These ob- servations seem relevant to planning com- munity responses to suicides. Suicide clusters are relatively central events in com- munities, although they do not involve property loss or residential displacement. Applying a typology developed by Berren et al. (6), a suicide cluster is a man-made crisis, which has a slow onset, affects the com- munity widely but has relatively few in- dividuals directly at risk, has a worrisome potential for reoccurrence, and has limited possibilities for control over its future im- pact. The impact of a suicide can be prolonged and fosters widespread fear in the community. Because it is difficult to know when the impact has ended, uncertainty tends to heighten anxiety. EMOTIONAL REACTIONS TO SUICIDE Cluster suicides evoke many more emotion- al reactions within communities than do in- dividual suicides. In general, these reactions follow patterns common to other kinds of crisis events within communities. What are referred to as community emotional reac- tions are not, in fact, very different from in- dividual reactions in an individual crisis. They include: initial disbelief or denial of any direct impact of the crisis, followed by an in- terval of questioning: "Why our com- munity?" "What is wrong here?" Fear develops in the community. In the case of adolescent suicide, the fear is related direct- ly to concern that the cluster of suicides may not have ended. A mentality develops in the community of waiting for the next bad news to arrive. Collective fear extends beyond the concern for the lives of individuals and the ef- fects of suicide deaths; it is also fear that something mysterious, poorly understood, and obviously lethal is afoot in the com- munity. A subsequent reaction may be one of outrage. "Too much is being made of this!" "Why don’t people just leave us alone?", etc. Following the painful reactions of confusion, fear, and outrage, a number of pathological defenses can be identified. Obviously, dif- ferent individuals in the community react in different ways and labeling reactions as pathological defenses does not imply that a community is "sick"; but the defensive nature of these reactions and their connection with shared, painful emotions seem clear. Defen- sive reactions include efforts to place blame; for example, people may believe the fault is with the school or family instability, or that drug abuse is ruining the lives of our young people. The message behind any blaming ef- fort is "we are not to blame", that is, an effort to dispel a sense of guilt which a community attaches to the suicide crisis. A different defense reaction is isolation. Communities turn inward in a crisis and become curiously resistant to interventions from elsewhere. A third defensive posture is detachment. "This does not involve us;" "Yes, I guess there is something going on over there somewhere-- we do not know anything about it." The final or resolution phase of crisis reac- tions, when viewed from a community perspective, may range from acceptance of an ongoing stigmatization in relation to the crisis, to a gradual restoration of the status existing before the crisis. Various efforts may be made to address the causes of the crisis and possible preventive actions and long-term efforts to deal with the aftermath 3-73 Report of the Secretary’s Task Force on Youth Suicide of the crisis. In the specific instance of a suicide cluster generating a community crisis, the tendency is certainly toward return to the status quo, although many preventive efforts seem appropriate. Of all the community reactions to a suicide crisis, probably the one most troublesome and most deserving of attention is the fear response. It seems to be precisely because the phenomenon of cluster suicides is so poorly understood, that in the course of such suicides a sense of great fear develops. A remarkably different course is easy to im- agine if, in fact, the causes of suicide clusters were well understood, if interventions had been researched, and the most effective in- terventions identified. In such a situation communities would marshall efforts with the confidence that they were doing something appropriate and that the situation would be contained. That clusters of suicides are a distinct phenomenon has been appreciated only recently even though they have been reported for many decades. It is understood that not every suicide cluster is like every other. When faced with multiple suicides, communities must wonder within the context of their specific community what is going on and what can be done. The way is open for all sorts of fantasies and worst-case accounts. Fear of a lethal phenomenon which is not un- derstood is a normal and predictable response. In my estimation, fear contributes very substantially to what can be identified as community reactions in cluster suicide. CLEAR LAKE TEXAS AS A CASE STUDY The Clear Lake area in southeast Texas is a circle of towns around Clear Lake, the best known of these being the city of Clear Lake, where NASA's Johnson Space Center is lo- cated. The population center is about 70 miles from the center of Houston. In October 1984, residents of Clear Lake, Houston, and the surrounding communities became aware that they were experiencing a 3-74 cluster of adolescent suicides. A sense of emergency developed and a great deal of ac- tivity was generated in an effort to identify, understand, and intervene appropriately to put a stop to the youths’ deaths. Two former students of the Clear Creek In- dependent School District, both age 19, killed themselves in August and in Septem- ber 1984. Because they were not connected directly with the student body at the time and because the deaths were separated by more than a month, no significance was attached at the time to these two deaths, other than that attached to any suicide--that is, grief for the families and sadness among those who knew of the deaths. Later, on September 28th, a very popular high school student died in an auto/bicycle accident. It was stated that his death was "mourned throughout the school district" because of the prominence of the student and the considerable sense of tragedy associated with his death. Six days later, on October 4, another former student, also 19 years old, killed himself. No crisis was sensed at this time, but within a week of that death, on the Sth, 9th, and 11th of October, three high school students killed themselves, making four adolescent suicides in the com- munity in one week and a total of six suicides in a two-month period. By the fourth suicide, the school district was convinced that there was problem; and by the time of the sixth suicide the entire community had been alerted by the media. A very considerable sense of dread developed. Following the acknowledgment by the school district and the media-generated publicity of the cluster of suicides, a great many com- munity actions occurred. The Clear Creek Independent School District called in two professionals from Houston, who had pre- vious experience in suicide, as consultants. They also contracted with the Houston Psychiatric Society to intervene with the families of the dead students and former stu- dents. The school district administration or- ganized, publicized, and held a public meeting specifically for parents of all stu- dents to review what had happened and what B.Comstock: Community Response to Adolescent Suicide... was being done. The school district held a press conference which was attended by rep- resentatives of the major national news ser- vices as well as by the local media. The school district established an in-school intervention process which included meetings, by grades, with all junior and senior high school stu- dents led by a group of clinical and education- al psychologists, some of whom were hired on a temporary basis. Meetings were held be- tween the psychologists and all the teachers and guidance counselors. An internal system was established for identifying children who seemed at risk for suicide on any basis, for counseling those children, and for facilitating referrals for treatment if that was deemed ap- propriate. The school district consultants or- ganized a treatment referral network through professional organizations related to psychotherapy. Hospital beds and profes- sionals offering treatment were identified and agreements were made to provide no- cost or low-cost therapy whenever it was in- dicated for children referred by the school district. Referrals, of course, were possible only when parents cooperated with them. Dozens of volunteer professionals par- ticipated in evaluating students and in estab- lishing therapeutic interventions. In addition to these efforts, a number of or- ganizations providing mental health services announced special services available in the Clear Lake area. These included teams from Houston International Hospital; the Family Emergency Intervention Team from the Houston Child Guidance Center, and special services provided by the Bay Area Crisis Hot- line. The Houston Crisis Intervention Ser- vice, which is linked organizationally with the Crisis Hotline, provided major leadership and provided many volunteers to deal with calls from distressed residents of the com- munity. The Houston Psychiatric Society es- tablished teams of its members who intervened directly with the families and friends of those who had killed themselves. Community leaders, drawn from the Clear Lake area and from Houston, organized a series of planning meetings with leadership from the Mental Health Association and the Houston Crisis Intervention Service, and sponsored a public forum on adolescent suicide. In Houston, a research group of rep- resentatives from the major academic institu- tions was organized to try to reach better understanding of the phenomenon in the Clear Lake area and to develop other re- search efforts related to adolescent suicide. A number of problems were encountered in the course of all of these activities. It seems most fitting to discuss the things that did not go well. Residents in the Clear Lake area felt con- siderably intruded upon by the flurry of ac- tivities resulting from the cluster of adolescent suicides. Some expressed con- cern that they felt indicted by the suggestion that there was something wrong in the com- munity. Second, a number of mental health profes- sionals in the community felt disregarded when their offers to provide assistance were not accepted. No mechanism was in place for identifying the individuals who had relevant experience and expertise in suicide beyond the usual training and experience of every mental health professional. Every mental health professional considers himself of her- self an expert in suicide, because this is a problem encountered from time to time in al- most any work setting. In the Clear Lake situation, it was considered desirable to iden- tify those professionals most qualified in specific suicide-related experience. A num- ber of children were referred to Galveston and Houston for treatment. As a result, the professionals in the Clear Lake area felt that their turf had been intruded upon. A third problem involved the lack of clarity about roles and responsibilities for the many agencies that were mandated to respond to any community crisis. Agency repre- sentatives came to Clear Lake from three counties and seven municipalities with a high level of interest and motivation to provide services. Regrettably, no clear network was established with the Clear Lake community by which the services could be administered. 3-75 Report of the Secretary’s Task Force on Youth Suicide A fourth problem, related to the third, had to do with unclear community leadership. The school district responded rapidly in estab- lishing a case-finding and referral network which seemed to function relatively well. The school administration was very clear, however, in stating that it was not a treatment entity and that it wanted more appropriate community agencies to take over that responsibility whenever possible. Clear leadership for this task actually did not emerge. In a series of planning meetings, community leaders attempted to organize themselves; these meetings for the most part were orderly and congenial but they involved the kind of status-seeking group process and jockeying for position that is inevitable when people come together who have no structure for working together. A fifth problem, related to the previous two, involved the need perceived by almost everyone for an orderly transfer of initiative from outsiders temporarily coming to the community to the community agencies and leadership already in place. Despite general agreement that this should happen, no clear plan ever was worked out by which it would happen; as a result, a number of plans initial- ly greeted with energy and enthusiasm in fact floundered; and, to my knowledge, no long- term plans have emerged for community ac- tivity centered on adolescent suicide. A sixth problem involved the media. There seems to be general agreement that media personnel in the Clear Lake suicides, in general, behaved responsibly and recorded events accurately. Nevertheless, some in- trusions were problematic. In the first week, the school grounds adjacent to the ad- ministration building were encircled by reporters waiting for students to leave the grounds, the reporters having been barred from school property. Students were as- sailed by cameras, microphones, and the as- sociated people, and were asked rapid-fire questions: "Do you know anyone who has committed suicide? Are you thinking about killing yourself? Do you know anybody who is going to kill himself? What would you do 3-76 if it were a friend of yours?" etc. It is doubt- ful that anyone could answer graciously or even sensibly to such a barrage of questions. The media continued to be problematic be- cause of the possibility of suggestion received by vulnerable or at-risk individuals. A seventh problem identified in Clear Lake was defining how extensive the problem was. The suicides occurred in a defined com- munity served by one school district. A con- siderable number of volunteers and media and civic planners who flowed into this area focused on the identified problem and problem area. Simultaneously, a number of other youth suicides were reported in neigh- boring areas. Overall, these received far less attention than did the Clear Lake group simply because they were in outlying geographic areas. There was no equally coordinated effort to deal with the so-called "outliers" which, in fact, may have been re- lated, in some as yet undefined way, to the Clear Lake cluster. The eighth problem, already referred to briefly, has to do with geographical divisions and community organization. The Clear Lake area in fact involves so many governmental groups that coordinated plan- ning in a community crisis was extremely dif- ficult. Parts of three counties were involved, and 7 different police departments were in- volved in one way or another from 7 different municipalities. Part of the area received public mental health services administered from Austin and other parts received mental health services administered from com- munity programs. The Clear Lake area cer- tainly is not unique in having so many adjacent community organizations. In such an organizationally complicated area the need for a preplanned mental health response is especially clear. On the positive side, a number of good things happened in the Clear Lake experience. At the top of the list, of course, is that no further suicides occurred after the week with the four suicides. In addition, appropriate inter- ventions were made and a number of at-risk children were successfully referred and es- B.Comstock: Community Response to Adolescent Suicide... tablished in needed therapy. A third positive outcome of the experience is that a more careful look at community response and problem identification was made possible. COMMUNITY BEHAVIOR IN CRISIS In reviewing the Clear Lake suicide cluster as a case study, a number of issues about group behavior present themselves. Often the same vocabulary and the same ideas are ap- plied to individual and to group psychology; however, some comments about the similarities and differences are in order. It is the case that groups experience shared emotions. Groups are made up of individuals with memories in common and frequently prevailing or unifying ideas and behavior can be identified. On the other hand, these com- mon themes within groups are less stable than are parallel emotions and ideas and be- havior in individuals. This is true particular- ly because the participants vary over time and groups are not themselves stably constituted. In referring to whole communities as groups, it is especially true that there is not the kind of strict connection between past and present experience that is encountered in in- dividuals. For example, in the Clear Lake area there are sub-communities which have had remarkably stable populations where parents and grandparents have been in the same homes and the shared memories and sense of community past are very strong. Other areas within the community involve a very mobile population, many members of which have very short histories in the com- munity and cannot share in the sense of longer-term memory and stability. When a crisis occurs, shock and disbelief are normal individual and group reactions. There tends to be an early effort to grasp the dimensions of what has happened. This may vary from wild over-estimation in a rather sensational way to problematic under-es- timation in the form of denial that anything very serious is going on. Specific negative reactions to crisis discussed earlier include fear, resentment, guilt, blaming, isolation, and opposition to interventions. Community planning for intervention in a suicide crisis certainly needs to take into account the presence of these reactions. The shock and disbelief and misassessment of the situation are best dealt with by the availability of fac- tual information, and for this the media serve a very needed and appreciated role. The group negative reactions may be under-es- timated. Unless these are taken into ac- count, ventilated adequately, and addressed in specific ways when they generate inter- ferences, then appropriate interventions will be stalled. One would predict that in the long term, communities as groups would be sensitized to a trauma experienced previously and would retain some continuing anxiety about that sort of trauma. Communities generally seem to be restored to pre-crisis functioning rela- tively well; however, just as in individuals we know that crisis often is the occasion for human growth, similarly, communities in crisis might be thought of as seeking a resolu- tion level where new strengths are added rather than having the community return to the status quo. In this sense, it is particular- ly important to try to retain focus on what happens to the community initiatives which stall out after a crisis interval has passed. LESSONS FROM THE CLEAR LAKE EXPERIENCE In the immediate aftermath of the Clear Lake suicide cluster several research initia- tives were taken. These have continued al- though at present none is complete. Within Houston, a task force was established which outlined a series of studies which were deemed desirable. Funding was obtained im- mediately from the Texas Department of Mental Health and Mental Retardation. Subsequently, a research contract was negotiated with the U.S. Public Health Ser- vice for studying the families both in Clear Lake and in Plano, Texas where adolescent suicide clusters occurred. A data tape cover- ing 10 years of suicide experience throughout 3-77 Report of the Secretary’s Task Force on Youth Suicide the State of Texas was obtained and funding is being sought for its analysis, with particular interest in correlating media coverage with suicide clusters. There is of course a high level of local inter- est in what has been learned about the specific situations in the Clear Lake suicides. General conclusions will be drawn from the analysis of data which we are editing. Preliminary observations indicate that the series of suicides seems to have occurred in individuals where the death was determined mainly on the basis of the individual story of the adolescent. A search for connecting links revealed some interconnections among the individuals but nothing with the strength to explain why so many deaths occurred in so short a time. Similarly there has not been any identification of social factors in the Clear Lake area which could be thought of as responsible for the suicide cluster. It seems unreasonable to identify this as a toxic com- munity. Perhaps in the future, more detailed correlations between adolescent suicide clusters and the quality of community life will reveal important connections. One thing learned in the Clear Lake ex- perience is that in spite of the problems, a great deal of positive response was ac- complished in a remarkably short time. One special point that was raised in two of the public meetings was of interest and oc- casioned some alarm. Several individuals in Clear Lake were concerned with the stigma of mental illness in the family, not in terms of its effect on social status, but in terms of its effect on employment. There was remark- able assent in the audience when one in- dividual stated that having a suicidal adolescent might cost him his job. The per- ception in this group was that the employers, meaning government directly and govern- ment subcontractors, deal very unsym- pathetically with mental illness; families, therefore, are drawn into a system of denial when things are wrong. Learning more about media impact on cluster suicides seems especially important because the search for influences on con- 3-78 tagion of suicide ideation focus naturally on the media. Cluster suicides are being studied as a recent phenomenon. In fact, they have been reported for many decades and a few have been reported over several centuries. From the existing reports, it is clear that not all cluster suicides are the same and it would be wrong to make generalizations about media involvement in reporting these suicides. The cluster of suicides in Clear Lake must have been a very different phenomenon from the cluster of suicides, for example, which surrounded the publication and popularization of The Sorrows of Young Werther by Goethe (7). This fictional ac- count of ayoung man who died for love is said to have incited many young men to kill them- selves. A related phenomenon is seen when there is extensive media coverage of the death by suicide, or presumed suicide, of very well known figures, such as those of Marilyn Monroe, Janis Joplin, Freddy Prinz, and John Belushi (8). Nothing of this sort was going on at the time of the Clear Lake suicides or at the time of the Plano suicides. In both instances, it is probably important to note that a well-loved member of the school population died a tragic accidental death. Since the media did not deal with the ac- cidental death, the media may be exonerated from any involvement in a grief-related con- tagion phenomenon. (The Clear Lake acci- dent received considerable media coverage after the suicide cluster and in connection with a campaign against drunk driving.) It remains an issue of concern that vulnerable, at-risk individuals may have their ideas about suicide made more concrete by the high level of attention paid to the deaths of school mates. Suggestion may play an important role in adolescent suicide, therefore, the media has a weighty responsibility both to report the news that needs to be reported, and if possible to guard against the phenomenon that that news can become suicidogenic. Since public awareness of the death may be a specific goal of a suicide, there is no avoiding the bind placed on the media. The best indirect solution is the conscien- tious effort on the part of reporters to deal B.Comstock: Community Response to Adolescent Suicide... with the ambivalence felt by suicidal in- dividuals, and to underscore alternatives available in the lives of people who are in emotional crisis. RECOMMENDATIONS A number of recommendations follow from the Clear Lake experience: 1. A pre-developed plan for community Response to adolescent suicide clusters is needed. 2. The crisis plan should cover much more than the mental health crisis of adolescent suicide. Every community would be wise to have such a plan for any crisis with men- tal health implications. Such a planshould be developed through existing community structures utilizing the available com- munity leadership, with a part of the plan defining ways in which others from outside the community can be useful. 3. A great deal of attention can be ap- propriately devoted to public education. This education should prepare the public to expect difficult emotional tasks in the event of a crisis. Individuals should be prepared in advance to resist blaming and guilt and the whole range of problems as- sociated with stigmatization associated with a crisis. Specifically, with respect to adolescent suicide and family stigmatiza- tion associated with mental illness, public education should extend to employers to ameliorate the fears of employees that getting help might jeopardize their jobs. 4. Media education is a major consideration. The importance of suggestion in cluster contagion is unknown; but many are con- cerned that the role of suggestion may be of major importance. Media personnel who, in the future, may cover issues of adolescent suicide, need to be trained in crisis psychology. They must recognize that when individuals are overwhelmed and may, indeed, be suicidal in the course of a crisis, there is potential not only for healing, but for growth in such experien- ces. This more positive aspect needs to be stressed during reporting. 5. Adolescents need to be educated. The preparation of the. population at risk is most critical. Adolescents are very responsive to preventive mental health in- itiatives. They are able to understand the importance of recognizing signs of trouble both in themselves and in their friends and they can be taught ways to seek and find help in the event of suicidal ideation. REFERENCES 1. Logue, JN, Melick, ME, Hansen H: Research Is- sues and Directions in the Epidemiology of Health Effects of Disasters. Epidemiologic Reviews 3:140-142, 1981. 2. Lifton JL, Olson E: The Human Meaning of Total Disaster: The Buffalo Creek Experience. Psychiatry 39:1- 18, 1976. 3. Lifton JL: Death in Life, Survivors of Hiroshima. New York, Random House 479-541, 1967. 4. Green BL: Assessing levels of psychological im- pairment following disaster. J Nervous and Mental Dis- ease, 170(9):544-552. 5. Erikson, KT: Loss of Communality at Buffalo Creek. Am J Psychiatry 133(3):302-305, 1976. 6. Berren MR, Beigel A, Ghertner, SA: A typology for the classification of disasters. Community Mental Health J. 16:103-111, 1980. 7. Friedenthal, R: Goethe: His Life and Times. New York, World Publishing Company, 1965. 8. Phillips, DP: Suicide, motor vehicle fatalities, and the mass media: Evidence toward a theory of suggestion. Am J of Sociology 84:1150-1174, 1979. ACKNOWLEDGMENT -eeeeeeerrmmmmnnnnnns “ The author wishes to thank Mrs. Carolyn Patterson for assisting in the preparation of this document. Address correspondence or requests to Dr. Betsy Comstock, V.A. Medical Center, Day Hospital (116AS5), 2002 Holcombe Blvd., Houston, Texas 77030. 3-79 PREVENTION/INTERVENTION PROGRAMS FOR SUICIDAL ADOLESCENTS Jane T. Simmons, Ph.D., Consultant, Texas Department of Mental Health and Mental Retardation, Houston, Texas Betsy S. Comstock, M.D., Professor of Clinical Psychiatry, Baylor College of Medicine, Houston, Texas Jack L. Franklin, Ph.D., Project Director, Texas Teen Suicide Project, Houston, Texas SUMMARY This study describes current efforts in adoles- cent suicide prevention and intervention. Using mail questionnaires, a variety of programs that serve suicidal adolescents were surveyed: crisis telephone services, walk-in clinics, hospital-based emergency programs, mental health centers with crisis components, school-based suicide interven- tion programs, non-hospital-based crisis stabilization units, and others such as support groups for survivors of suicide, counseling agencies, and networks. This paper describes the methodology and analyzes the findings from 396 programs. Topics investigated include: visibility of programs, reasons for program start and age of program, funding sources, services, availability and linkage to other community ‘resources, client statistics, program problems and needs, community needs, and certifica- tion status. Services for suicidal adolescents are not centralized, but are found in numerous com- munity agencies with little networking among services. Many programs do not keep adequate records to assess service outcome and utilization by suicidal adolescents. Programs responding to the survey identified 3-80 community education, school programs, staff and funding, specialized training regarding adolescent issues, and residential facilities as the program and community features most needed to serve suicidal adolescents better. Other major problems were the low visibility of programs, lack of certification and lack of written standards for suicide-related ser- vices. INTRODUCTION The research subcommittee of the Houston Task Force on Adolescent Suicide was asked by the Secretary’s Task Force on Youth Suicide to identify and describe programs throughout the country which provide suicide-related services for youth. Our goal was to describe what is currently available for suicidal adolescents and to delineate programming gaps and problems. It was im- mediately obvious that services to suicidal adolescents were not necessarily found in programs neatly labeled as "adolescent suicide programs." Rather, services to suicidal adolescents were available through a variety of crisis intervention programs such as crisis telephones, walk-in crisis clinics, and emergency programs administered through J.Simmons: Prevention/Intervention Programs. . . community mental health agencies. Other kinds of programs addressed adolescent suicide through non-crisis services. These programs included school-based educational programs, grief support groups, and task for- ces that coordinated services for suicidal adolescents. Adolescent suicide is not an isolated event. Rather, it occurs in a social-psychological en- vironment in which elements that influence the environment, indirectly at least, affect suicidal behavior. Using this line of reason- ing, programs that seek to improve the self- esteem or problem-solving skills of adolescents are in some way primary preven- tion services. For the purpose of this survey, however, we limited the kinds of programs to those that would directly intervene with suicidal adolescents or were school-based educational/intervention programs. In short, we sought programs in which the prevention of, or intervention with, suicidal adolescents was a direct service goal. METHODOLOGY The project was conducted in two phases. The first was to identify programs to be in- cluded in the survey and the second was the fielding and analysis of the survey. Program Identification To identify programs for the survey, we first asked the 42 State mental health associations (MHAs) to provide us with directories of local MHAs. The local MHASs were, in turn, sent a project identification form asking them to identify and furnish addresses of programs that served suicidal adolescents in their com- munities. Although directories identified many community mental health centers, hot- lines, and crisis services, they did not cover the full range of programs in which we were interested. We also asked local community- resource people to identify programs in their communities that provided services to suicidal adolescents. As such, we were able to identify the programs on which local com- munities currently rely. We also sent the project identification form to 264 programs listed in the 1984 Directory of Information and Referral Services (I&R) in the U.S. and Canada. If two or more I&R services in a single community were found, one was randomly selected to receive the project identification form. In cases of com- munity duplication between local MHA centers and I&R services, we sent the project identification form only to the I&R service; in large cities, we contacted both resources. One section of the project identification form asked respondents to identify others who could help us find the kinds of programs for which we were looking. These people, if in other communities, were sent the project identification form. A total of 523 forms were mailed. Five percent (n=24) were "returned to sender." Forty-six percent returned usable forms. Almost all listed at least one program; a few said their com- munities had no such services. Survey questionnaires were then sent to identified programs. They were also sent to all programs listed in the 1982 Directory of Suicide Prevention and Crisis Intervention Agencies (The American Association of Suicidology). Finally, a systematic random sample was drawn from the 198] Directory of Federally Funded Community Mental Health Centers (DHHS). In all, 118 ques- tionnaires were mailed; 396 (34%) were returned in time to be included in this analysis. Community Characteristics A wide variety of communities were repre- sented in the survey. Thirty-seven percent of the programs were located in cities with populations of 100,000 or more, 41 percent of them were in cities with populations of 15,000 to 100,000, and 16 percent were in cities with less than 25,000. Four percent were in the suburbs of large cities and 3 percent said they were located in two or more places throughout the county. The size of program catchment areas ranged from 1,000 to 8 million. Half the programs had catchment area populations of 235,000 3-81 Report of the Secretary’s Task Force on Youth Suicide or less while 8 percent (n=31) had catchment areas of 1 million or more. Half of the programs were serving popula- tions with special characteristics. Twenty- four percent were in areas with unusual racial or ethnic distributions. Table I, Distribution of Programs with Unusual Racial/Ethnic Dis- tributions, summarizes the distribution. As shown, catchment areas that had heavy con- centrations of whites, blacks, Hispanics, and multi-ethnicities were found in our sample. Twenty-seven of the catchment areas had large populations of young people while another 27 areas had large aged or retiree populations. Those with young populations were usually college towns. An unusual income distribution was the most often mentioned special characteristic of catchment populations. A full 14 percent (n=55) of the programs were in areas of high poverty and/or unemployment. On the other hand, 13 programs reported high concentra- tions of wealthy people. Another 13 programs said their populations had high concentrations of both wealthy and poor. Twenty-seven programs were in counties with military bases and six were in areas where the State or Federal Government was a large employer. Many other programs were also in areas characterized by special oc- cupations. Six areas had either a paper mill or other single factory. Ten were in high technology areas. Twenty-nine were in predominantly mining, agricultural, gaming, lumber, or fishing areas, and 7 programs had to cope with high levels of tourism. Another 12 programs responded with other special oc- cupational characteristics. Finally, 33 programs served rural or geographically iso- lated populations. RESULTS Age of Program Suicide services were, for the most part, provided by well-established programs. Forty-one percent of the programs in the sur- vey began between 1970 and 1975, the years of the greatest program development. Only 25 percent (n=98) began in the 1980s. Clear- ly, the recent resurgence of interest and con- cern in adolescent suicide has not sparked a rash of new programs. However, all of the programs specific for teen suicide began since 1980. It is interesting to note that 41 percent of the 39 programs identified as having suicide com- ponents by their titles, began in 1970 or before. Even though many programs developed between 1970 and 1975, their titles did not reveal a suicide-specific component. After 1970, seven years passed before another program in our sample, had such a title. Then, in 1984 and 1985 the remaining 16 (41 percent of all programs with "suicide" in their titles) began. In short, during the 1970s when social programs were rapidly increas- Distribution of Programs with Unusual Racial/Ethnic Distribution Characteristic Number of Programs % of All Programs Hispanic 20 5.0 Black 13 3.3 Native American/ Alaska native 7 1.8 Appalachian 3 .8 White 24 6.1 Multi-ethnic 18 45 Other 8 2.0 Total 93 23.5 Table 1. 3-82 J.Simmons: Prevention/Intervention Programs. . . ing, suicide-specific programs were not forthcoming. Instead, more general programs that included suicide related ser- vices along with other crisis and non-crisis services were beginning. These programs es- tablished in the 1970s, are still the major providers of suicide services to adolescents. Program Visibility To whom does a suicidal adolescent turn? As anyone who has tried to find a particular agency in the phone directory has ex- perienced, the title of an agency is usually needed in order to contact it. Because we were curious about how helpful titles were for "finding" a program, we analyzed program titles in terms of their key words. Incredibly few programs (n=8) had the words "teen," "youth," or "adolescent suicide" in their titles. Only 33 other programs (8%) had the word "teen" without the word "suicide." This is a critical shortcoming given that adolescents tend not to contact general services. Rather, as the child sexual abuse hotline in Knox County, Tennessee con- tended, adolescents are much more likely to contact a program that is advertised specifi- cally for them. Thus, much needs to be done to make programs more attractive to teens to encourage them to use these programs more frequently. Even titles containing the key word "suicide" were not common. Only 31 (8%) programs fell into this category. Instead, program tit- les reflected an orientation to general problems, as opposed to specific ones, and to broad populations rather than special groups. A large number of programs had "crisis" or "emergency" in their title. Almost 1/4 (n=88) of the programs had "crisis" while 6 percent (n=24) had "emergency" in their titles. Another 39 programs (10%) were called Helpline, Hotline, or Hopeline. A relatively large number of programs (19%, n=75) were titled "mental health" or "coun- seling" programs. At least one program com- mented that the term "mental health" was seen as a stigma deterring teenage utilization. Finally, 32 programs (8%) were called "CONTACT" and 66 programs (17 percent) had titles which did not convey crisis, emer- gency, suicide, or even helpline services. These programs had titles such as "Center for Human Services" and "Gateway." The visibility of a program is largely depend- ent on its advertising and 46 percent of the programs in the sample did not advertise to reach adolescents specifically. Reasons for Program Development Almost 40 percent of the programs in the sample began because of a particular interest in suicide services; in 11 percent of the cases, a specific suicide incident was the impetus. A few programs (2%), mentioned a high suicide rate in their communities. For 29 percent of the programs, however, issues other than suicide were the reason for starting the program. These included: perceived need in community for general crisis services (22%), response to drug abuse (6%), and response to street youth (1%). The latter emerged not only to assist runaways, but also to address problems spawned by the deinstitutionaliza- tion of status offenders. (See Table 2, Dis- tribution of Reasons for Program Development.) Funding Because of differences among organizations, we were not able to obtain budget informa- tion solely for suicide components for all programs. Only 75 programs (19%) reported the amount spent solely on suicide services, 174 (44%) provided the amount for all crisis and referral services, and 45 (11%) provided a total agency budget that included more than crisis services. As shown in Table 3, Description of Program Funding by Budget Type, the average budget for a suicide component of an agency is $63,667. On the average, the 1986 budgets were $10,000 higher than for 1985. Between fiscal years 1984-85 and 1985-86, funding 3-83 Report of the Secretary's Task Force on Youth Suicide devoted solely to suicide services increased proportionately higher than funding for both total crisis services and total agency services. The ratios of the average 1985-86 budget compared to the average 1984-85 budget was 1.22. The corresponding ratio for total crisis budgets was LI12 and for total agency budgets it was 1.06. Funding Sources The majority of programs did not receive federal funding, nor did they receive funding only from one source. Rather, they obtained funds from the State (58%) and a variety of local sources such as fundraising, donations, and United Way. When programs seek funds outside government sources (n=174) they look to fundraising (34%), United Way (27%), a combination of United Way and fundraising (16%), churches (4%), insurance and other third party payers (10%), and trainer/speaker fees (9%). Services The most prevalent service offered by programs was a crisis telephone, usually staffed 24-hours a day. Eighty-eight percent had this service component. Usually, even if an agency did not provide a crisis telephone service, another agency in the community did. Only 2 programs stated that a crisis telephone service was not provided in their communities. More than 6 percent of the programs also offered a walk-in crisis service while 6 percent said it was not available at all. Education and public awareness were usual- ly provided by programs; very few com- Distribution of Reasons for Program Development N % 1. Specific suicide incident 45 11.39 2. Federal mandate for community mental health emergency components 103 26.08 3. Professional interest in suicide services 146 36.96 4. Non-professional concern with suicide 8 2.02 5. Perceived need in community for crisis services in general 88 22.02 6. Expansion of existing agency services 20 5.05 7. Response to drug abuse 22 5.56 8. Response to street youth 5 1.26 9. High suicide 8 2.02 Table 2. Description of Program Funding by Budget Type, 1985-1986 (in dollars) Number of Average Mean Change Mean Change Programs Funding 84/85 to 85/86 83/84 to 84/85 Suicide Only 75 63,667 +10,033 +6,875 Total Crisis 174 132,374 + 15,907 +10,221 Total Agency 37 1,150,817 +93,686 +99,294 Table 3. 3-84 J.Simmons: Prevention/Intervention Programs. . . munities had no such service. In spite of the high proportion (85%) of agencies par- ticipating in educational activities, half of them stated that their communities needed more education and awareness of suicide or better knowledge of resources. Grief counseling, either face to face or via telephone, was available in most com- munities. Sixty-two percent of the programs offered the service, but only 12 programs specifically mentioned an SOS (Survivors of Suicide) group. Presumably, grief counsel- ing is still provided through more traditional formats such as individual and other group counseling. Followup therapy, including individual, group, or family was offered by less than half the programs in the survey sample. When crisis telephone programs were excluded from the data base, 70 percent of the remain- ing programs provided followup therapy. The rest reported that the service was avail- able elsewhere in their community. Only 3 percent said it was not provided at all. Nine percent of the programs claimed that school intervention, in any form, was not available in their communities. Neither they nor another agency provided the service. More than half of the programs furnished some form of school-based intervention, whether personnel training, crisis interven- tion, or student awareness training, and in the remaining cases, another agency provided at least one of these services. Crisis interven- tion was the most likely service to be offered, followed by student awareness training and training school personnel. School-based in- tervention was regarded as an important ser- vice for adolescents. Indeed, when asked what more their programs needed for ade- quate service to suicidal adolescents, a full third responded "school programs.” When another agency provided any of the above services, the program responding to the survey was unlikely to have any formal agreement with it. An agreement with a medical care facility was most common but only 21 percent of the programs had one. Eleven percent had an agreement with an agency providing followup therapy and 8 per- cent had one with a walk-in crisis clinic. Clearly, there is little formal networking in communities among agencies providing ser- vices to suicidal adolescents. Other Resources Respondents were also asked about the availability of, and their relationship to, other resources typically involved in suicidal inter- vention: private therapists, mental health programs, police, ambulance services, medi- cal emergency treatment, and psychiatric hospitalization. More than 25 percent of the programs reported that private therapists specifically interested in suicide were not available in their communities. Even when such therapists were available, only 6 percent of the programs had written transfer agree- ments with them and a few programs stated that it was program policy not to refer to private, for-profit, therapists. Linkage problems included: private therapists who would not accept clients who could not pay (n=10); insufficient number of private therapists (n=10); problematic client motiva- tion and follow-through with such referrals (n=3); clients having to wait too long for an appointment (n=l); and inadequate com- munication or linkage between the program and the therapist (n=5). Three more programs mentioned other problems such as clients lacking transportation to get to a therapists office. Only 8 of the 396 programs claimed their communities had no mental health program. Twenty percent of the programs in the sample were, themselves, the mental health program. Of the remaining 316 programs, 41 percent had written transfer agreements with the mental health program in their com- munity. The most frequent (n=17) com- plaint regarding linkage was the long waiting time before clients could be seen--up to six weeks in one case. Other problems included: clients unable to pay; a lack of after-hours crisis intervention care or outreach; com- munication problems between the mental 3-85 Report of the Secretary’s Task Force on Youth Suicide health program and the responding program regarding aftercare following crisis or hospitalization; and difficulty with client motivation and referral compliance. Several programs mentioned treatment barriers specific to adolescents, namely, the need to get parental consent or an agency policy not to serve anyone under 17 years old. In a few communities (n=7) ambulance ser- vices were not available. For the vast majority of the remaining programs, getting to a helping resource was the client’s respon- sibility, even when in crisis. Eighty-one per- cent of the programs had no formal agreement with an ambulance service. Four percent of the sample reported that clients in crisis also faced other problems with am- bulance services. Ambulances were reluc- tant to transport if they were unsure of reimbursement, and in some communities, they responded only to medical needs. Thus, if a client was suicidal but had not yet at- tempted, ambulances would not transport that person to a crisis intervention facility. The programs in our sample were unlikely to have formalized agreements with police-- only 13 percent did. The programs were even more critical of police than they were of am- bulance services. Seven programs men- tioned that police lacked suicide-intervention training and/or were reluctant to accept such training. Nine programs complained about the lack of police response to crises situations. One program mentioned that police were disil- lusioned about crisis intervention because of the State’s deinstitutionalization laws. Other programs referred to the failure of police to use appropriate community resources and to operational problems such as difficulty in tracing calls. Another 5 percent of the sample acknowledged problems but did not specify what they were. Less than one quarter of the programs had written agreements with hospital emergency medical care resources. Among the 353 programs with valid data, 9 percent had linkage problems with these resources. The most frequent problems mentioned involved 3-86 the actual referral process: the lack of clear guidelines for referral; the lack of a central- ized reporting mechanism; being able to refer only with police intervention; and the unwillingness of medical care resources to handle clients with psychiatric problems, even when there were medical problems. For several programs, the lack of appropriate psychiatric staff within the medical care facility was a referral barrier. Hospital psychiatric admission was available to 92 percent of the programs in the sample. Moreover, it was the resource with which programs were most likely to have written transfer agreements (36% of the 342 programs who furnished complete data). It is with this resource, however, that the programs in the sample also experienced the most problems. The lack of beds was fre- quently mentioned (n=18). Other programs referred to problems with the admission process, such as delayed admission, disagree- ment as to the appropriateness of certain referrals, refused admission to those who could not pay, and refused admission to clients with medical problems. Distance to nearest facility and other transportation is- sues posed complications for twelve programs. For some communities, the nearest facility, while available, was 50 to 100 miles away. Getting a client to these facilities was a problem, particularly for indigent clients lacking personal transportation. Program Outcome Data One section of the questionnaire asked for program outcome data, specifically: number of clients directly served in the past fiscal year; number of suicide-related clients and how many of these were adolescent; number of suicide-related clients who return to crisis within a year; and the number of documented client deaths. In responding to the question on number of suicidal clients, programs in- cluded not just attempters but also clients with suicidal thoughts and ideation. Five percent of the programs could not give the number of clients served last year. Some programs, particularly the crisis telephone J.Simmons: Prevention/Intervention Programs. . . programs, did not have records on the num- ber of individual clients; their data reflected the total number of contacts. Thus, the data for crisis telephone programs reflected the number of calls although some clients called repeatedly. Even among similar programs, there is wide variation in the number of clients or client contacts. One telephone crisis program reported more than 103,000 calls; another had less than 2,000. One school-based program intervened in 8 cases; another intervened in 50 cases. The extreme variation among programs and between program types made any further generaliza- tion regarding total number of clients inap- propriate. The most striking observation about other outcome data was that programs frequently did not gather this kind of information. Twenty seven percent of the programs stated they were unable to give the number of suicide-related cases or simply left the ques- tion blank. Information on the number of adolescent suicide-related cases was even less frequently available. Forty-five percent of the programs did not respond. Forty-two percent did not know how many suicidal clients returned in crisis within the year and another 22 percent left the question blank. Twenty-one percent of the programs did. not know the number of documented client deaths from suicide last year while 18 percent did not answer at all. Our analysis of outcome data included only those programs for which data were avail- able. For a few, suicide-related cases oc- curred infrequently. Seven percent of the programs had ten or less such cases last year. Most programs (63%), however, handled 100 or more suicide-related clients. Among the 291 programs for which data was available, 12 percent had 1,500 or more suicidal con- tacts/clients. (See Table 4 for the distribu- tion). Overall, 7 percent of all clients served were suicidal. The number of suicidal adoles- cents that programs served also varied wide- ly not only between types of programs but also among similar programs. Some programs served no suicidal adolescents last year while others served more than 600. Distribution of Number of Suicide-Related Cases Number of Suicide- % of All Programs Related Cases Number of Programs with Valid Data (n=291) 1-99 120 41 100-199 27 9 200-299 36 12 300-399 14 5 400-499 15 5 500-599 14 5 600-699 7 2 700-799 4 1 800-899 1 1 900-999 4 1 1000-1999 23 8 2000-2999 11 4 3000 + 15 5 Don’t know 43 N/A Missing 62 N/A Total 396 100% Table 4. 3-87 Report of the Secretary's Task Force on Youth Suicide Across all programs in the sample, 13 percent of the suicide-related clients were adoles- cents. Table 5 presents the distribution of the num- ber of documented suicides among clients. Very few of each program’s clientele actual- ly committed suicide. More than half (59%) of the programs that furnished data reported no client deaths from suicide last year. Thir- teen percent had only I client suicide and 18 percent had two or three. The remaining 14 percent had four or more. Overall, there were 297 suicides per 100,000 suicide-related clients served. Problems Using a combination of closed and open- ended questions, we asked respondents to identify specific program problems, especial- ly those that affected service delivery to suicidal adolescents. We also asked respon- dents what more their communities needed for adequate services to this special popula- tion. Table 6 presents the distribution of the responses to close-ended questions. Staffing and Professional Issues Staff shortage was the major complaint of 54 percent of the programs. This problem was reiterated when respondents were asked what more their programs needed for ade- quate service. Twenty seven percent again replied "staff and money." Not only the number but the quality of staff was of some concern. Of all persons employed by the programs in the sample, more than half (57%) were volunteers. When crisis telephone programs are ex- cluded from the analysis, 33 percent of all employees are volunteers. Problems with troubled volunteers was not a major issue but 10 percent of respondents stated that their staff needed more training, particularly regarding adolescent issues. More than a third of the respondents claimed Distribution of Client Deaths From Suicide Number of Documented % of Programs Client Deaths Number of Programs with Valid Data (n=239) 0 142 59.4 1 32 13.4 2 28 11.7 3 14 5.8 4 10 4.2 5 4 1.7 6 1 4 7 1 A 8 1 4 10 1 4 12 1 4 13 1 4 20+ 3 1.2 Don’t know 81 N/A Missing 76 N/A Total 396 100% Table 5. 3-88 J.Simmons: Prevention/Intervention Programs. . . that various staff and professional issues hampered their ability to serve suicidal adolescents. In addition to more staff and training, they wanted more time to work with adolescents and better networking with other professionals. Similar concerns were stated regarding community needs: better networking and case followup, better trained professionals, more professional resources in general, and more professional commitment to the problem. Community Education Programs wanted to do more community education activities but were hampered by the lack of funds and staff. In fact, 20 percent of them said this was a major need in their communities. Their concern particularly centered around the need for greater recog- nition and awareness that teen suicide exists and is a viable problem. School Programs Respondents considered schools to be criti- cal intervention arenas. Many programs were already doing some form of school in- tervention; ll percent thought this was what their programs should be doing but were not, and 16 percent replied that this was a major community need. In many communities, school systems were reluctant to have suicide-specific educational programs. One objection was the belief that talking about suicide would give teenagers "wrong ideas." Some programs also suggested that school personnel be trained regarding indices and intervention techniques. Not only the lack of qualified school personnel, but also the availability of school staff to teenagers, were problems. Often, school counselors were tied up with helping students schedule cour- ses and had little time to talk with troubled students. Specific Adolescent Programs Sixteen percent of respondents listed specific adolescent services which they believed would enhance their program’s effectiveness in serving suicidal adolescents. Those most often mentioned, in order of frequency, in- cluded: residential/in-patient treatment sup- port groups for both attempters and survivors of victims, safe-houses and other non-hospi- tal residential services, peer counseling, and family involvement in treatment. Many of these services were listed again under additional services communities needed to serve suicidal adolescents ade- quately. Almost 25 percent of respondents listed services specially attuned to adolescent needs such as: in-patient beds and residen- tial facilities, teen community centers and drop-in clinics, peer counseling, prevention programs, support groups, and long-term, Distribution of Problems Encountered by Programs Responses to Questions Yes No Missing # of Programs % N % N % Inadequate physical facility 123 31.1 225 56.8 48 12.1 Staff shortage 215 54.3 134 33.8 47 11.9 Inadequate staff training 60 15.1 288 72.7 48 12.1 Funding instability 165 41.7 183 46.2 48 12.1 Funding deficiency 171 43.2 177 44.7 48 12.1 Troubled volunteers 27 6.8 321 81.1 48 12.1 Inadequate referral resources 73 18.4 274 69.2 49 124 Table 6. 3-89 Report of the Secretary’s Task Force on Youth Suicide family-oriented treatment. Another concern was the need for more leverage to work with adolescents when parents refused counseling or help for their adolescents or when adoles- cents needed to be separated from the parent. Certification Standards and Qualifications The certification of programs and qualifica- tions of professionals is of some concern. Only 9 percent of the programs with valid data (n=367) were certified by the American Association of Suicidology and one quarter of the programs (26%) did not have written standards for suicide-related procedures. Twenty nine percent of all professional staff working directly with clients (i.e., excluding volunteers, administrative, coordinative, and secretarial staff) had undergraduate degrees or less. The lack of advanced education coupled with programs’ own requests for staff better trained in adolescent issues sug- gests that attention be given to the qualifica- tions of persons who are working with suicidal adolescents. A similar concern was repeated regarding "helping" individuals outside a program’s auspices. Several respondents expressed dis- may that self-proclaimed "experts" in suicide may do more harm than good. They gave specific examples of trainers and other per- sons who were obviously uninformed and un- prepared to deal with suicide. Illinois is one State that is attempting to establish certifica- tion standards for individuals as well as for programs. Suicide-Specific Programs: A Subanalysis From the first 271 questionnaires returned, we analyzed separately 27 interven- tion/prevention programs exclusively or predominantly devoted to suicide, as deter- mined by their titles. None were located in small towns of populations under 15,000, but neither were they solely a product of dense- ly populated areas. They were as likely to be 3-90 found in cities of 15,000 to 100,000 as in larger ones. As reasons for program development, most listed professional interest in suicide (70%) followed by a specific suicide incident (37%). Only 11 percent listed "response to high suicide rate." These programs were not overwhelmingly multi-service agencies. No more than 60 per- cent had a crisis telephone service and only 15 percent had a walk-in crisis service. The most common activity was education and public awareness (93%), followed by student awareness (67%), and direct crisis interven- tion in the school (63%). These agencies usually did not offer followup therapy--only 22 percent did--and when it was offered, it was most likely to be individual therapy, not family or group. Even these specialized groups were not for- mally linked to (i.e., have written agreements with) other traditional resources in the com- munity such as police, ambulance services, and private therapists. The resource most often linked with the suicide-related program was a mental health program. One quarter (26%) of the programs had written agreements with a mental health agency and reported no problems in working with this resource. On the other hand, while only lI percent had written agreements with hospi- tal psychiatric resources, almost a fourth reported problems, particularly, too few beds, and clients being refused admission for financial and other reasons. Funding for these programs came primarily from the State, United Way, and dona- tions/fundraising. Budgets were not large. Thirty percent of the programs operated on $7,000 or less. Another third had budgets be- tween $29,000 and $100,000. Only one program had a budget greater than $250,000. The programs varied greatly in the number of clients served. One reported 30, while several reported more than 25,000. It was not surprising that among these programs, suicidal clients comprised a larger proportion of all clients served than among other programs (13% vs. 7%). J.Simmons: Prevention/Intervention Programs. . . On the other hand, suicide-specific programs did not attract adolescents as well as the other programs. Only 7 percent of all suicidal clients were adolescents compared to 12 per- cent for the entire sample. The client death- rate, however, is smaller--245 vs. 304 per 100,000 suicidal clients. Twenty-two percent of these programs listed more community education and school programs as needed components to their ser- vices. More training, more funding, or more time to work with adolescents were program needs for more than half the sample. Very few (n=3) wished to add a specific adolescent service to their program, but 25 percent did want to have such services developed within their communities. These programs were more likely than other programs to advertise specifically to adoles- cents (70% vs. 43%). Still, 22 percent of them did not. Also, more than 80 percent of them were not certified by the American As- sociation of Suicidology and 22 percent did not have written standards for suicide-re- lated procedures. DISCUSSION Suicide prevention/intervention services for adolescents are generally not provided by agencies established solely for that purpose. Specializing in suicidal adolescents is the ex- ception rather than the rule. Services are usually available through general crisis programs, with suicide being just one type of crisis handled. Yet, there seems to be a grow- ing recognition that working with adolescent suicide requires a level of staff expertise that is currently lacking. Many of the programs in the sample want more staff training and indi- cated a need for staff trained in adolescent development and issues. They see this as a need not only for their own programs but also for other programs in their communities. For example, a respondent for a crisis telephone program indicated the need for the mental health clinic in the community to have a specialist in adolescence. Most communities included in our sample have not addressed adolescent suicide as a separate issue. Agencies, Or even programs within agencies, do not specialize in suicide. For many communities the lack of specializa- tion is partly fostered by the belief that adolescent suicide is not a problem or does not exist and survey respondents often ex- pressed dismay over the extent of community unawareness. Program personnel consider community education as a priority issue inim- proving service to suicidal adolescents. Addressing adolescent suicide as a separate issue is difficult because it is often accom- panied by a wide variety of other problems such as drug and alcohol abuse, depression, family conflict, running away, and even satanic worship. Clearly, for a community to develop a comprehensive service system for suicidal adolescents, it must be able to hand- le these other problems as well. In many communities it is not the lack of services for accompanying problems that lessens effec- tive interaction but, rather, the lack of net- working among programs in a community. The absence of networking was felt on two levels: case coordination and service delivery. We found instances of ambulances refusing to transport patients if there were no physical injury, psychiatric wards refusing to admit if there were physical injuries, and medical programs reluctant to treat if there were psychiatric/ psychological problems. Perhaps community coordination by some- one trained in adolescent issues would en- courage these programs to be receptive to adolescents in crisis. To serve suicidal adolescents better, survey respondents recommend a combination of strengthening services already in operation through funding, staffing, training, and net- working as well as the development of teen- specialized services such as teen community centers, peer counseling, and safehouses. Still, the existence of a service does not mean that teenagers will automatically use it. Several mental health agencies specifically mentioned that teenagers were unlikely to use their services. Current programs need to consider factors affecting service utilization 3-91 Report of the Secretary’s Task Force on Youth Suicide by teenagers and to make special efforts to advertise their programs directly to this population. The issue of allowing suicide preven- tion/education programs in schools remains unsettled. Respondents felt that these programs are important for reaching adoles- cents adequately, yet many school systems oppose such efforts. Some programs have developed school intervention curricula, are training students and personnel, and are providing crisis intervention. Other programs and schools might benefit by ob- serving the strategies in schools in which programs are already developed. Parents, too, are sometimes obstacles to helping suicidal adolescents. Several programs have problems in obtaining paren- tal permission to counsel adolescents. To strengthen the accessibility of services by adolescents, changes in consent laws must be addressed. The problem is compounded when parents themselves, are primary con- tributing factors to a troubled adolescent. Respondents to the survey urged com- munities not wishing to provide a com- prehensive system of service, to provide for separation of parent and adolescent and to find alternative ways to involve parents in treatment. Another dilemma that must be addressed in serving suicidal adolescents is the availability of residential treatment beds, when needed. In many communities, the need for such beds is relatively infrequent so that none are reserved for such emergencies. Thus, a suicidal adolescent may need inpatient treat- ment when a bed is not available. Survey respondents recommend that such beds be established and held in reserve, to be used when needed, regardless of frequency of use. The same concern was stated in a somewhat different form when one program director wrote that, on the basis of prevalence, other mental health problems such as schizophrenia took staffing and funding priority over adolescent suicide. In short, decision makers will have to examine priorities and prevalence issues when debat- 3-92 ing program development for adolescent suicide services at the community level. The lack of certain services for adolescents, particularly in-patient/residential treatment space, must be addressed if a more effective system for serving suicidal adolescents is to be established. These developments will probably take money; certainly, they will re- quire agency policy changes. Professionals in communities do not have to wait, however, for such major problems to be resolved. Many services are already available and their coordination would be a beginning of an im- proved delivery system for suicidal adoles- cents. ACKNOWLEDGMENT -----seeeeeeveessnneee . The authors wish to acknowledge the assis- tance of Mrs. Carolyn Patterson in assisting with questionnaire production and mailing, and in preparing this document. CHARACTERISTICS OF SUICIDE PREVENTION/INTERVENTION PROGRAMS: ANALYSIS OF A SURVEY Jack L. Franklin, Ph.D., Project Director, Texas Teen Suicide Project, Houston, Texas Betsy S. Comstock, M.D., Professor of Clinical Psychiatry, Baylor College of Medicine, Houston, Texas Jane T. Simmons, Ph.D., Consultant, Texas Department of Mental Health and Mental Retardation, Houston, Texas Mark Mason, M.S., Houston, Texas SUMMARY This paper reports the results of a survey of 395 suicide prevention/ intervention programs in the United States. Included in the sample are 152 crisis telephone services, 9 walk-in clinics, 24 hospital-based emergen- cy programs, 142 mental health centers with a crisis component, 17 school-based suicide intervention programs, 8 non-hospital-based crisis stabilization units, 22 combinations of two of the above, and 21 other programs such as survivors of suicide, counseling agencies, and networks. Crisis telephone services and mental health clinics make up 74 percent of the organizations that responded to the sur- vey. Each type of program is described in terms of location, special characteristics of the catch- ment population, services provided, number of clients served, number of suicide-related cases served, number of suicide-related adolescent cases served, number of docu- mented suicides, available resources, budgets, funding sources, and problems en- countered by the programs. Most of the programs were developed in response to professional interest in suicide services. Although program labels were often misleading, a surprisingly large number of programs offered comprehensive services directed toward suicide prevention/interven- tion. The services most often offered were education and public awareness efforts, and crisis telephones. Medical care for suicide at- tempters was least likely to be provided. Budgets ranged from under $20,000 to well over $1,000,000 and almost all programs reported multiple funding sources. Staff shortages, funding deficiencies and the in- stability of funding were problems en- countered most often by the 395 programs in our survey. Other than more funds and more staff, most programs had the resources they need to serve suicidal adolescents in the com- munity. The average number of suicide-related adolescents served last year by programs in our survey was 76, ranging from 26 served by school-based intervention programs to 375 3-93 Report of the Secretary’s Task Force on Youth Suicide suicide-related adolescents served by mental health centers. The highest rate of suicide per 1,000 suicide-related clients was 38.2 in school-based intervention programs and the lowest was in non-hospital-based programs with .29. INTRODUCTION While the community response to the in- creasing rates of adolescent suicide is thought to be massive, very little is known about the actual programs that have developed, where they are located, why they began, what services they offer, what resour- ces are available to them, what their annual budgets and funding sources are, how many clients they serve, and what their problems are. To address these issues, we first asked the local programs of the Mental Health As- sociation and the programs listed in the 1984 Directory of Information and Referral Programs in the United States and Canada to identify suicide prevention/intervention programs known to them. All programs listed in the Directory of the American As- sociation of Suicidology and a random selec- tion from the 1981 Directory of Community Mental Health Centers in the United States were added for a total of 1,181 programs. Methodological details of the survey are dis- cussed elsewhere (Simmons, Comstock, and Franklin). This paper describes the 395 programs which returned usable survey forms to us. PROGRAM CHARACTERISTICS Table 1 summarizes the types of suicide prevention/intervention programs respond- ing to the survey. Of the 395 programs, 38 percent (152) describe themselves as crisis telephone services, 2 percent (9) as walk-in crisis clinics; 6 percent (24) as hospital-based emergency programs; 36 percent (142) as mental health centers with a crisis com- ponent which includes suicide intervention; 4 percent (17) as school-based suicide inter- vention programs; and 2 percent (8) as non- hospital-based crisis stabilization units, city-sponsored community crisis services, general intervention services or comprehen- sive general crisis agencies. Six percent (22) of the programs are combinations of two of the programs listed above and S percent (21) of the programs describe themselves as sur- vivors of suicide, community mental health clinics with no emergency service, com- munity-based education and support groups, counseling agencies and networks. For dis- cussion purposes, these programs are labeled Other. Population Thirty-three percent of the 395 programs are Types of Suicide Prevention/intervention Programs Responding to Survey Percent of Type of Program Number All Programs 1. Crisis telephone service (CTS) 152 38 2. Walk-in crisis clinic (WIC) 9 2 3. Hospital-based emergency service (HBES) 24 6 4. Mental health center with crisis component (MHC) 142 36 5. School-based intervention program (SBIP) 17 4 6. Non-hospital-based program (NHBP) 8 2 7. Combinations (Comb) 22 6 8. Other (Other) 21 5 Total 395 Table 1. 3-94 J.Franklin:Characteristics of Suicide Prevention... located in central cities with populations of more than 100,000; 41 percent in cities with populations of 15,000 to 100,000; 16 percent are located in small towns with populations less than 15,000; and 7 percent are located in other areas. Table 2 summarizes the popula- tions served by the programs surveyed. Special Characteristics Fifty-three percent of the 395 programs con- sider their catchment population "special." (Table 3) Twenty-four percent list unusual racial/ethnic distributions, 18 percent report unusual age distributions and 21 percent con- sider the distribution of income unusual. Eight percent list sex distribution as unusual, 18 percent report than their catchment population has an unusual occupational dis- tribution such as a single industry town, and 20 percent report other characteristics, such as geographic isolation, which make their catchment population unusual. Walk-in clinics were the most likely to con- sider their population unusual (67%), close- ly followed by mental health centers (66%), hospital- based emergency services (63%), non-hospital-based programs (56%), com- binations (50%), crisis telephone services (44%), school-based intervention programs (35%), and other programs (33%). Populations Served by Various Types of Programs Type of Over 15,000- Program 100,000(%) 100,000(%) 15,000(%) Other(%) All Programs Combined 37 41 16 7 CTS 45 42 9 4 wIC 11 67 22 - HBES 75 21 4 -- MHC 19 45 30 6 SBIP 36 24 24 16 NHBP 56 22 11 11 Comb 36 64 -- -- Other 62 24 5 10 Table 2. Special Characteristics Reported by Programs Responding to Survey Percent Reporting Special Characteristics of Catchment Population Type of Racial/ Program % ethnic Age Income Sex Occupation Other All Programs Combined 53 24 18 21 8 18 20 CTS 44 20 18 17 1 20 16 WIC 67 22 11 33 -- 22 45 HBES 63 33 25 21 4 - 8 MHC 66 27 17 26 7 22 22 SBIP 35 17 11 29 -- 11 17 NHBP 56 22 -- 45 11 11 22 Comb 50 32 23 5 9 9 18 Other 33 9 24 19 -- 5 9 Table 3. 3-95 Report of the Secretary’s Task Force on Youth Suicide Initiation of Program Table 4 summarizes the reasons for estab- lishing the service programs surveyed. Eleven percent of the 395 programs began in response to a specific suicide incident. But the most often listed reasons for developing programs were professional interest in suicide services with no specific incident (37%), federal mandate for community men- tal health center emergency component (26%), and perceived need in the community for general services (24%). Services Programs were asked to list their services in terms of those directly provided, those provided by another agency under formal agreement with the reporting agency, those provided by another agency in the com- munity with no formal ties to the reporting agency, and those services not provided in the community. A surprisingly large number of agencies provide a wide range of suicide preven- tion/intervention services. Community men- tal health centers are the most comprehensive: 90 percent provide crisis telephone services, 96 percent provide walk- in clinics, 90 percent provide grief counsel- ing, and 96 percent provide family emergency therapy. They provide followup therapy for individuals (93%), groups (83%), and families (92%). Eighty-eight percent provide education and public awareness ser- vices and 70 percent provide school interven- tion services, including personnel training (54%), crisis intervention (75%), and stu- dent awareness training (46%). Crisis telephone services provide the client with a wider range of services than expected. Twenty-three percent of the 152 crisis telephone services provide walk-in crisis ser- vices, 2 percent provide medical care for suicide attempters, 40 percent provide grief counseling, 12 percent provide family emer- gency therapy and followup therapy, 83 per- cent provide education/public awareness services and 44 percent provide school inter- vention services, including training person- nel for school intervention (45%), providing crisis intervention to schools (41%), and providing awareness training for students (53%). All walk-in crisis clinics provide direct crisis telephone services and family emergency therapy. Most provide education/public awareness services (89%), intervention for schools (67%), and grief counseling (67%). Forty-four percent of the walk-in crisis clinics also provide direct training for school inter- vention and (33%) student awareness train- ing. Reasons for Initiating Programs Figures given in percent Type of Suicide Federal Professional Perceived Need Program Incident Mandate Interest in Community All Programs Combined 11 26 37 24 CTS 13 6 39 39 WIC 17 44 22 11 HBES - 38 33 12 MHC 5 53 32 8 SBIP 44 6 69 6 NHBP -- 11 11 33 Comb 14 5 50 14 Other 33 10 38 19 (Percentages do not add to 100 due to rounding and/or to missing data.) Table 4. 3-96 J.Franklin:Characteristics of Suicide Prevention... Hospital-based emergency services tend to be comprehensive in that they provide crisis telephone services (79%); walk-in crisis ser- vices (79%); medical care for suicide at- tempters (42%); grief counseling (63%); family emergency therapy (96%); followup therapy (71%) to individuals (67%), groups (54%), and families (67%); education/public awareness services (75%); and school inter- vention services (63%), including personnel training (50%), crisis intervention (50%), and student awareness training (46%). Only 2 (12%) of the school-based interven- tion programs provide crisis telephone ser- vices and none provide medical care for the suicide attempters. But they do provide walk-in crisis services (47%); grief counseling (53%); family emergency therapy (29%); fol- lowup therapy (29%) to individuals (47%), groups (29%), and families (24%); educa- tion/public awareness services (94%); and school intervention services (88%) including training (88%), crisis intervention (76%), and student awareness training (100%). The non-hospital-based programs provide crisis telephone services (100%); walk-in crisis services (89%); grief counseling (44%); family emergency services (78%); followup therapy (33%) for individuals (33%), group (22%), and families (44%); education/public awareness services (78%) and school inter- vention services (67%) which includes per- sonnel training (56%), crisis intervention (67%), and student awareness training (44%). All combination programs provide crisis telephone services (100%); and some provide walk-in crisis services (82%); grief counseling (59%); family emergency therapy (41%); followup therapy (36%) to in- dividuals (32%), groups (23%), and families (23%); education/public awareness services (91%); and school intervention (73%) which includes personnel training (68%), crisis in- tervention (68%), and student awareness training (86%). Programs in the "other" category provide crisis telephone services (43%); walk-in clinic services (52%); medical care for suicide attempts (5%); grief counseling (48%); fami- ly emergency therapy (41%); followup therapy (52%) to individuals (52%), groups (43%), and families (52%); education/public awareness services (81%); and school inter- vention (62%) which includes personnel training (67%), crisis intervention (52%), and student awareness training (57%). As indicated by these statistics, programs do notdiffer greatly in terms of services that they provide directly to clients. Ofthe services provided by the 395 programs, medical care for suicide attempters is the least likely, provided by only 7 percent of the programs (2% of the crisis telephone ser- vices, 42% of the hospital-based emergency clinics, 10% of the community mental health centers, and 5% of the programs in the "other" category). However, an additional 21 percent provide medical care for suicide at- tempters by formal agreement with another agency and more than half (54%) report that these services are provided by another agen- cy which has no formal ties to the reporting program. Community Resources Programs report very few deficiencies with community resources. Private therapists who are interested in suicide are available to 69 percent of the programs; only 7 percent have written agreements with private therapists and only 9 percent report problems making referrals. Some of the problems associated with referrals include the client’s inability to pay and therapists not being available at off hours. Mental health services are available to 96 percent of the programs and 33 percent have written agreements which allow clients to be transferred between the reporting program and the mental health service. Eleven per- cent report problems linking clients with mental health services; the most often cited problems are long waiting lists and lack of staff at the mental health facility. Almost all (96%) of the programs have an 3-97 Report of the Secretary’s Task Force on Youth Suicide ambulance service available to them. Only 9 percent have written agreements with am- bulance services and only 4 percent report problems. Police are also available to most (97%) programs; 13 percent have written agreements with police and 10 percent report problems working with police, mostly due to lack of police training in the area of suicide. Hospital emergency medical care is available to 97 percent of the 395 programs, 23 percent have written agreements and only 8 percent report problems linking clients to this resource. Ninety-two percent of the programs have psychiatric hospital services as an available resource and 36 percent have written agreements with a psychiatric facility. However, 19 percent of the programs report problems linking clients with a psychiatric facility such as insufficient number of beds, delays in admission, and reimbursement is- sues. Availability of community resources is not re- lated to program type and other than a general shortage of private therapists who are interested in suicide and who are avail- able to the program, few programs report problems with the availability of resources. Costs Budget Breakdown Ninety (23%) of the programs did not report budgets. Of the 305 that reported budgets, 21 percent have annual budgets of $20,000 or less, while 8 percent report budgets of more than 1 million dollars for fiscal years 1985- 1986. Crisis telephone services reported annual budgets of $20,000 or less (21%); $21,000 to $50,000 (29%); $51,000 to $75,000 (15%); $76,000 to $100,000 (10%); and $101,000 to $500,000 (24%). One program reported a budget of $501,000 to $1,000,000 and 11 per- cent of the crisis telephone service programs did not report budget information for fiscal year 1985-86. Walk-in crisis clinics report annual budgets 3-98 of $21,000 to $50,000 (38%); $51,000 to $75,000 (13%); $76,000 to $100,000 (25%); $101,000 to $500,000 (13%); and $501,000 to $1,000,000 (13%). Eleven percent did not report budgets. One hospital-based emer- gency services program has a budget of $79,000, two have budgets of $200,000 and $335,000, one reports $523,000 and another reports $600,000. Seven percent have budgets of $501,000 to $1,000,000 and 21 percent report budgets in excess of 1 million dollars. Four (21%) programs did not report budget data. Mental health centers report budgets in each of the seven categories: $20,000 or less (17%); $21,000 to $50,000 (9%); $51,000 to $75,000 (6%); $76,000 to $100,000 (4%); $101,000 to $500,000 (35%); $501,000 to $1,000,000 (7%); and more than $1,000,000 (18%). Thirty percent did not report budget data. The school-based intervention programs are small--all but one (with a $220,000 budget) report annual budgets of $20,000 or less (41% did not report). Non-hospital-based emergency programs range from $76,000 to $100,000 (20%); $101,000 to $500,000 (60%); and $501,000 to $1,000,000 (20%) (56% did not report). Combination programs report budgets of $20,000 or less (11%); $21,000 to $50,000 (17%); $51,000 to $75,000 (6%); $76,000 to $100,000 (17%); $101,000 to $500,000 (44%); and over $1,000,000 (6%). Eight percent did not report budget data. Annual budgets of "other" programs include $20,000 or less (35%); $21,000 to $50,000 (18%); $51,000 to $75,000 (18%); $101,000 to $500,000 (6%); $501,000 to $1,000,000 (12%); and over $1,000,000 (12%). Nineteen percent did not report. Funding Sources Multiple funding sources characterize the 395 programs. The most often mentioned funding source was from local sources (66%), followed by State (58%), client fees (35%), federal (25%), and foundations (17%). Crisis telephone services receive funding J.Franklin:Characteristics of Suicide Prevention... from the following sources: local (66%), State (34%), foundations (28%), federal (17%), and client fees (5%). Walk-in clinics report funding from local sources (89%), State (89%), client fees (66%), federal (56%), and foundations (11%). Non-hospi- tal- based crisis units receive funding from local sources (89%), State (67%), federal (44%), foundations (22%), and client fees (11%). Combination programs report fund- ing from local sources (82%), State (68%), federal (32%), foundations (31%), and client fees (14%). "Other" programs include fund- ing from local (38%), State (38%), client fees (19%), federal (14%), and foundation (14%) sources. Whereas local funding was the most often mentioned source of funds for the programs listed above, State funding was cited most often by mental health centers. Funding sources of mental health centers include State (82%), local (71%), client fees (69%), federal (34%), and foundations (6%). School-based intervention programs report funding from State (63%), local (63%), foun- dation (6%), federal (19%), and client fees (13%). Hospital-based emergency programs receive funds from client fees (71%) and from State (58%), local (38%), federal (21%), and foun- dations (4%). SERVICE DATA In this section each program type is described in terms of: e Number of clients served in the last fiscal year. e Number of clients who were suicide-re- lated. e Number of the suicide-related cases who were adolescents. e Number of suicide-related clients who returned in crisis within a year. eo Number of documented client deaths from suicide. ¢ Suicide rate per 1,000 suicide-related cases. Although 395 programs returned usable sur- veys, only 287 reported both the number of clients served and the number of suicide-- related clients served in the past fiscal year. Only 215 programs reported both the num- ber of suicide- related clients and the number of adolescent suicide-related clients; and 147 programs reported both the number of suicide-related clients and the number of suicide-related clients that returned within a year. More than half--211 programs-- reported both the number of suicide-related cases served last fiscal year and the number of documented client deaths from suicide last year. The percentages and rates in the fol- lowing discussion are conservative and, in all cases would be the same or larger if we in- cluded only programs that reported all data elements. Table 5 summarizes the data presented in the following section. Crisis telephone services report serving 1,682,703 contacts during the past fiscal year. Six percent of the clients were suicide-related and 8 percent of the suicide-related contacts were adolescents. Only 5 percent of the suicide-related clients returned in crisis within a year and 137 deaths from suicide were documented during the last year. The rate of documented suicides was 1.3 per 1,000 suicide-related contact. Walk-in clinics reported serving 18,059 clients during the last fiscal year. Eleven per- cent were suicide-related and 8 percent of the suicide-related cases were adolescents. About 20 percent of the suicide-related clients returned in crisis within a year and 9 of the suicide-related clients died from suicide last year. The rate of documented suicide was 4.7 per 1,000 suicide-related clients served. Hospital-based emergency programs reported serving 81,372 clients in the past fis- cal year. About 19 percent were suicide-- related and 13 percent of the suicide-related cases were adolescents. Two percent of the suicide-related cases returned in crisis within 3-99 Report of the Secretary’s Task Force on Youth Suicide ayear. Thirty-three deaths from suicide were documented during the year for a rate of 2.1 suicides per 1,000 suicide-related clients ser- ved. Mental health centers reported serving 306,596 clients during the past fiscal year. Eight percent of the clients were suicide-- related cases and 2 percent of the suicide-- related cases were adolescents. About 12 percent of the suicide-related cases returned in crisis within a year and 143 suicides were recorded during the past year. The suicide rate was 5.9 per 1,000 suicide-related clients served. School-based intervention programs reported serving 11,152 clients during the past fiscal year. One percent of those served were suicide-related and all of the suicide-re- lated cases were adolescents, as expected. Eighteen percent of the suicide-related cases returned in crisis within a year and six suicides were recorded last year. The suicide rate was 38.2 per 1,000 suicide-related clients served during the year. Non-hospital-based crisis programs served 138,300 clients during the past fiscal year. Three percent of the caseload were suicide- related clients and about 3 percent of the suicide-related clients were adolescents. Six percent of the suicide-related clients returned in crisis within a year. Only 1 death from suicide was documented last year and the rate of suicide was .22 per 1,000 suicide- related clients served. Combination programs served 135,169 clients last fiscal year. About 10 percent were suicide-related and 8 percent of the suicide-related clients were adolescents. Ten percent of the suicide-related cases returned in crisis within a year. Twenty-six suicides were recorded last year for a rate of 1.9 suicides per 1,000 suicide-related clients served. Other programs reported 49,804 clients served during the past fiscal year. About 19 percent were suicide-related clients; 9 per- cent of the suicide-related clients returned in crisis within a year. Fourteen suicides were documented last year resulting in a rate of 1.5 suicides per 1,000 suicide-related clients ser- ved. The 287 programs that reported data in this section of the survey served 2,423,155 clients during the past year. Seven percent were suicide-related cases and 9 percent of the suicide-related cases were adolescents. Six percent of the suicide-related clients returned in crisis within a year. The number Service Data Provided by Survey Respondents Numbers reported by programs for past fiscal year Suicide- Suicide Related Suicide Suicide- Rate/ Clients Clients Related Related Suicide 1000 Served (S.R.C.) Adol. Return Deaths S.R.C. CTS 1,682,703 106,116 8,891 4,732 137 1.29 WIC 18,059 1,903 159 383 9 4.73 HBES 81,372 15,392 2,017 296 33 2.14 MHC 306,596 24,340 2,999 2,887 143 5.88 SBIP 11,152 167 157 28 6 38.22 NHBP 138,300 4,608 141 271 1 .22 Comb 135,169 13,883 1,117 1,342 26 1.87 Other 49,804 9,275 852 120 14 1.51 Total 2,423,155 175,674 16,333 10,059 369 2.10 Table 5. 3-100 J.Franklin:Characteristics of Suicide Prevention... of documented suicides, 369, yields a rate of 2.1 suicides per 1,000 suicide-related cases served during the year. Special Problem Areas When asked to select all problems ex- perienced by their program from a list of problems in the survey instrument, staff shortage was selected most frequently (54%), followed by funding deficiencies (43%), funding instability (42%), inadequate physical facilities (31%), inadequate referral sources (18%), inadequate staff training (15%), and troubled volunteers (7%). Crisis telephone services selected staff shortages as the problem encountered most often, followed by funding instability, fund- ing deficiency, inadequate physical facilities, inadequate referral resources, troubled volunteers, and inadequate staff training. Walk-in crisis clinics selected problems of in- adequate physical facility, funding instability and funding deficiency as most often en- countered; followed by staff shortages and in- adequate staff training, troubled volunteers and inadequate referral resources. Hospital-based emergency services listed in- adequate physical facility as the most fre- quently encountered problem, followed by inadequate referral resources, staff shortages, funding deficiency, inadequate staff training, funding instability and troubled volunteers. Mental health centers face problems of staff shortages most often. Other problems in order of most frequently encountered in- clude funding deficiency, funding instability, inadequate physical facility, inadequate staff training, inadequate referral resources, and troubled volunteers. School-based intervention programs en- countered funding instability most often, fol- lowed by staff shortages, funding deficiency, inadequate physical facility, inadequate staff training and inadequate referral resources. Non-hospital-based services encountered funding instability most often, followed by funding deficiency, inadequate referral resources, staff shortages, inadequate staff training, and inadequate physical facility. Combination programs listed the following problems in order of frequency: staff shortages, funding deficiency, funding in- stability, inadequate physical facility, inade- quate referral resources, inadequate staff training, and troubled volunteers. Programs in the "Other" category listed fund- ing deficiency as the problem encountered most often, followed by staff shortages, fund- ing instability, inadequate physical facility, short training, inadequate referral resources, and troubled volunteers. Several programs (35) mentioned that they were unable to reach populations at risk due to inadequate funding, lack of outreach resources, successful suicides not as likely to use services as attempters, stigma of mental illness, and problems of getting parents invol- ved. Program Needs When asked what more they needed to provide adequate services to suicidal adoles- cents, relatively few needs were identified by the 395 programs other than more funds and more staff. In the area of community/school- related needs, programs mentioned the need to provide programs in schools, more out- reach programs, greater community aware- ness, more publicity for programs, and more advertisements directed toward adolescents. The needs most often stated relating to professional issues were for more staff and more funds. Training of staff, networking, more space, and more staff time were also listed in that order, as problems. Needs that are associated with specific ser- vices to adolescents include residential treat- ment facilities, support groups, safe houses, peer intervention services, and walk-in clinics, in that order. Many more problems were identified by the 395 programs in response to the question, "What more does your community need for 3-101 Report of the Secretary’s Task Force on Youth Suicide adequate service to suicidal adolescents?" Leading the list was community education followed by more school programs, greater recognition of the problem, better coopera- tion of schools, more awareness in schools, better knowledge of available resources, and financial help. Professional issues included more resources, better networking and better trained person- nel. Specific services for adolescents that were mentioned as community needs were more beds and more services such as out- patient services, alternative methods of deal- ing with families when parents are problems, peer counseling, long-term family-oriented treatment, and support groups. ACKNOWLEDGMENT ----eeeeeeeeeeeeeenennn . The authors wish to acknowledge the assis- tance of Mrs. Carolyn Patterson in preparing this document. 3-102 PSYCHOLOGICAL AUTOPSIES OF YOUTH SUICIDES Robert E. Litman, M.D., Co-Director, Suicide Prevention Center, Los Angeles, California SUMMARY Psychological autopsies of youth suicides in- dicate that about half of them had a relative- ly recent contact with the mental health system viewed broadly to include various therapists and counselors. Mostly, the inter- actions were focused on evaluation and brief support. Families and therapists both tended to ignore and deny clues to suicide. Since the teenage subjects also use denial extensively, it takes special efforts to bring suicidal youngsters into the helping system and hold them there. Continuity of care is recom- mended. A team approach (as contrasted with one-on-one psychotherapy) might ease the therapeutic load of contending with com- plex and multi-dimensional pre-suicidal states. Other features noted from psychological autopsies include: ¢ Adolescent drug abuse seems to be close- ly associated with adolescent suicide, especially in older (17-19) males. e A suicide in the family is a major stress event, leaving survivors at risk for suicide themselves. ¢ Bereavement counseling is important in preventing further suicides. School problems and conduct disorders are common in pre-suicidal adolescents. e School counseling was important in help- ing some control cases avoid self-destruc- tive acts. My task is to review reports of psychological autopsies of suicides among young people in order to clarify the possible role of preven- tion and/or treatment activities in these cases. Psychological autopsies are retrospec- tive biographies of deceased persons based on interviews with family members, friends, teachers and physicians. The lifestyles, symptoms and behaviors, personal and oc- cupational histories, and medical records are reviewed by a death investigation team. As we reconstruct the lives of persons who are now deceased, we think of the subjects as having been in "pre-suicidal” states. The in- vestigations reveal that these "pre-suicidal” subjects do not make up a homogeneous population. Instead, they tend to differ in various characteristics and behaviors, they represent different psychological and psychiatric diagnoses, and they have en- countered different types of environmental stresses. A majority had communicated something about their discomfort to some- one else, a peer, family member or profes- sional person. Many of the subjects might have revealed further clues to suicide if they had been questioned specifically about suicidal thoughts. Some of the pre-suicidal adolescents were diagnosed as "depressed." Others were described as having "conduct disorders." Some were high achievers and some were low achievers. Some were physically impaired, others were successful athletes. What all of them had in common were periods of hope- 3-103 Report of the Secretary’s Task Force on Youth Suicide lessness and thoughts of death as a solution to their problems. I think of such "pre-suicidal" individuals as having existed in a psycho-social "suicide zone" which is populated by many people, of whom only a minority kill themselves. In a given period of time, say a year, only 1 per- cent of the people in the "suicide zone" ac- tually commit suicide. That does not mean that the other 99 percent are "false positives" in the sense that treating them is unnecessary or a waste of resources. Probably, all of the pre-suicidal persons are in need of some preventive therapy, and, of course, some need more intense treatment efforts than others. If we define "treatments" as human interac- tions in which there are some formal aspects or rules by which one set of persons (therapists) expend efforts to be helpful to other persons (the clients or subjects), it is apparent that treatment has a number of forms. For example, treatment may consist of an initial consultation or evaluation or a brief contact during a crisis giving immediate support. Depending on the needs of the sub- ject, appropriate treatment might involve the family, a peer group, a prolonged drug rehabilitation program, hospitalization--with or without various medical drugs--or long- term out-patient psychotherapy. I have sur- veyed the major psychological autopsy studies for what they reveal about youthful suicides and the treatment that was available, offered and/or accepted. Studies of adult suicides, using the psychological autopsy methodology, have clarified a number of suicide-related vari- ables. For example, intention, communica- tion of suicidal clues, stress factors, and the specific medical and psychiatric diagnoses. Studies of youth suicide based on reviews of records have been reported by Sanborn (1), Shaffer (2), Cosand, Bourque, and Kraus, (3) and others. All note that adolescent suicides are preceded by recognizable psychological maladjustment. In Shaffer’s 1974 sample of children’s suicide, (n=30), 30 percent were in treatment or were waiting to get into treat- 3-104 ment. Social withdrawal and friendlessness were common. Many children were recog- nized as having conduct disorders or emo- tional problems at school. Sanborn reviewed the lives of ten adolescent suicides in New Hampshire. Five had some difficulty in school adjustment. Four had threatened suicide previously. All the families were intact, but only two families described themselves as being "happy" families. Most of the youth suicides ap- peared to be impulsive rather than planned. Cosand and associates reviewed Sacramento coroner’s data. They concluded that impor- tant stresses on youth who commit suicide were loss of love, family conflicts, and psychiatric disorders which impeded adjust- ment to adult roles. They found multiple predictors of suicide and recommended im- proved training for physicians, police, families, employers, and school personnel in recognizing pre-suicidal symptoms and in im- proving communication with young persons. Only recently have investigators studied youth suicides in more adequate numbers using comparison or control groups. The largest research program has been under way for several years, directed by Dr. David Shaf- fer (4,5) in New York. In a project con- ducted by Dr. Mohammad Shafii (6,7) and his colleagues in Louisville, peers of the victim were used as controls. A third notable source of data are the reports beginning to come from the "San Diego Suicide Study" by Drs. Rich, Young, and Fowler (8). They com- pared suicides among persons under and over age 30. My own group in Los Angeles (9) published some pilot studies quite similar to Shafii’s, and we are now engaged in inves- tigating all youth suicides in California during a set study period. Finally, the Centers for Disease Control in Atlanta has been assem- bling information on youth suicide through psychological autopsies conducted in several different locations. All of the investigations have obtained a good deal of information about treatment and prevention, but these factors have not been consistently or careful- ly analyzed and interpreted, and at present R.E.Litman: Psychological Autopsies of Youth Suicide are considered to be quite obscure. Shafii and associates investigated 25 cases of youth suicide occurring between January 1980 and June 1983. Their subjects were 95 percent white and 90 percent males. The same standardized interview form was used to secure information about a matched con- trol, often the victim’s closest friend. No statistically significant difference was found between the victims and the control subjects regarding such variables as "broken home", separation from parents, or birth order. There were significant differences in ex- posure to suicide through suicidal siblings, friends, parents, or other relatives and dif- ferences in previous expressions of suicidal ideation, suicide threats, or suicide attempts. The frequent use of non-prescribed drugs or alcohol was associated with suicide, as was anti-social behavior, and "inhibited per- sonality." There had been previous psychiatric treatment for 9 of 20 suicide vic- tims and 4 of 17 control subjects. These in- vestigators agree with Cosand, et al. that a close relationship exists between suicidal wishes, threats, attempts, and completed suicide. Successful prevention involves reducing exposure to suicidal images and thoughts and replacing these with more posi- tive concepts. Three cases were presented by Shafii in con- siderable detail. A 17-year old white male shot himself in his bedroom. A week before his death, his mother called the pediatrician’s office and asked for help. "My son is depressed and moody", she said. "He has a hard time going to sleep. His personality has changed." A week later, the boy’s father called the pediatrician and expressed fear for his son who seems to be incoherent and bel- ligerent. "We are worried that he is taking drugs." There was a confrontation between son and father over this issue after which the son said, "You'll be sorry." The pediatrician suggested a psychiatric consultation, but the family felt it was not that serious, yet. A second case involved an 18-year old male who was having school problems and injuries which took him out of athletics. He began to make suicide threats to his girl friend and other people and to search for his biological father who had long since dropped out of sight. Said the boy, "I'm like my real dad. I'm just crazy." The clinical course was one of progressive dissatisfaction. He was having problems with his friends. A few days before his death, his mother tried to make an ap- pointment for him to see someone at a men- tal health clinic because of his withdrawal and his appearance of being spaced out with no plans for his life, but the victim expressed resentment and said he did not need help. The emphasis for Shafii is on the failure to get these suicide victims into treatment and keep them there. His group strongly stresses the grief and guilt reactions in surviving fami- ly persons and believes that the postvention efforts of their suicide research team may have been effective suicide prevention for the survivors that they interviewed. It would be instructive to review the cases of the non- suicides for factors which were associated with survival. By far the largest and most sophisticated study of youth suicide is being conducted by a group at the New York Psychiatric Institute led by David Shaffer. In a preliminary report, Shaffer stated that approximately half of the completed youth suicides had been in touch with the mental health system at some time. His findings indicate that slightly less than half of the victims were depressed and about the same number had a family history of suicide attempts. At least half of the male suicide victims had been in trouble because of impulsive behavior, learning difficulties and aggressive outbursts. Approximately half of the suicide victims had been using ex- cessive amounts of drugs or alcohol. About a third had made a serious suicide attempt. These investigators believe, that in all likelihood, other, better defined high risk groups will emerge from the study once the data have been fully analyzed. According to Shaffer, he has not considered the role of treatment, but in personal communications, indicated that a major problem has been 3-105 Report of the Secretary's Task Force on Youth Suicide keeping young people involved in a therapeutic contact. In his preliminary report, Shaffer indicates that it may be premature to attempt to evaluate prevention or treatment programs. However, I interpret these findings to indi- cate a need for effective anti-depressant treatment of these adolescents who are depressed. The prevention and treatment of early delinquency and early drug abuse remain as key unsolved issues. One program has stressed serial interviews with young women who had made suicide attempts. These women are often deprecated by their families, have rather poor personal relation- ships, and suffer from low self esteem. The program’s goal is to rebuild self esteem and self confidence through increased coping skills. On the basis of his experience to date, Dr. Shaffer recommends improved mental health courses such as seminars or workshops to help students and parents identify sig- nificant psychiatric problems. In particular, they should have information about the major psychiatric illnesses, eating disorders, and should be able to identify for themselves abnormal degrees of anxiety and depression. As Dr. Shaffer points out, suicide is only one possible bad outcome from adolescent psychiatric disorders; identification and treatment of other aspects would prevent a good deal of misery and disability in general. Dr. Shaffer believes that routine school- based screening and treatment referral for teenagers with depressive symptoms, espe- cially those who have fallen behind in school or who are getting into trouble, would be ef- fective prevention. He stresses teaching psychological strategies, for example, coping and problem solving skills to troubled teenagers. First courses would dramatize how to say "no" when offered drugs, or how to communicate and negotiate with family members and peers. Shaffer believes that better training for pediatric and psychiatric emergency room staffs is necessary to identify suicide at- tempters who are at especially high risk. It 3-106 would be good to have more well-publicized hotlines and drop-in clinics so teenagers would know about them and use them. Finally, Dr. Shaffer would like to see ex- panded insurance coverage for crisis situa- tions which may be life-threatening through self-destructive behaviors. Charles L. Rich, M.D. and his associates, Dr. Young and Dr. Fowler, have reported some of their investigations under the title "San Diego Suicide Study." They noted that when they compared completed suicides in people under 30 with completed suicides of people over age 30, there were many similarities and a few differences. In terms of psychiatric diagnoses, the younger group had significant- ly more drug abuse and anti-social per- sonalities, significantly less alcohol abuse, fewer affective disorders and fewer organic syndromes. Often, in the younger group, a drug or alcohol use disorder was combined with some other psychiatric diagnosis. Rich and his colleagues also noted that suicide is a particular problem of white males. They performed structured interviews on suicide cases in San Diego starting in 1981. They placed special emphasis on obtaining as complete a toxicology screening as possible. They tried to arrive at a consensus diagnosis based on DSM III criteria. They were surprised to discover that most of the people in the younger age group were not living alone. About half of their cases had some prior treatment and about a quarter ap- peared to be in treatment at the time of death. They remarked that these figures showed no change in rate in the past 25 years. "One might think that a quarter century of heightened awareness to the relationship be- tween psychiatric illness and suicide would have led to a higher treatment rate in such an obviously ill population." Significantly, more young people than older people hang them- selves, but as in previous U.S. studies, the use of firearms predominated. Rich emphasizes the frequency of alcohol and drug use disor- der particularly in the younger group, and he concludes that drug use may be the most im- portant single factor in the suicide rate in- R.E.Litman: Psychological Autopsies of Youth Suicide crease in youth in the United States. Considering stressors, they found the younger group had more separations and rejections compared to the older group, where the subjects had more mental illnesses. Overall, they found that there was an ex- tremely complex interplay of diagnostic categories. My colleagues and I are now (Spring, 1986) investigating youth suicides in California. We have noted in our early cases that al- though almost half of the subjects were known to mental health personnel, the deaths came as a surprise and a shock to al- most all of the involved counselor-therapists. The most conspicuous aspect of treatment failure revealed by psychological autopsies was that the adolescent was referred to therapy but did not make or keep the refer- ral appointment. Or, the family did not cooperate. The psychological autopsies underlined the warning of school problems and failures. We suggest exit counseling when students drop out. On a positive note, we feel that survival in our peer control group is related to the young person having at least one positive "role model", or stable older person who can be idealized. In twelve cases described by Litman and Diller (9), there were four therapy contacts, two in the suicide cases and two in the con- trols. Our tentative interpretation is that crisis interviews were insufficient for the chronic and multiple problems of the suicide cases. One of the controls benefited greatly from school counseling that eventually be- came long term. The counseling, originally for learning and behavior problems, helped him academically and also helped him cope better in his personal life. The cases dramatized the problems with con- fiding in a peer. While one person was led into effective counseling, another student confided to a friend that he was going to kill himself, but forbade the friend to tell anyone. After the suicide, the friend became suicidal himself, but was helped by the school coun- selor and a psychiatrist. In the older teenagers, alcohol and drug abuse contribute to the feeling of uselessness, failure, and con- fusion. "I just can’t get it together." Experience with suicidal alcoholics, especial- ly the failure of a one-on-one volunteer counseling program (10), convinced me of the importance of a team approach to suicidal substance abusers. To get the young person into therapy and hold him, we need to involve family, friends, and peers--a group process. The treatment experiences of most of the young persons who committed suicide could best be characterized as brief episodes of evaluative or supportive therapy. Both the families and the professional health workers tended to ignore and deny clues to suicide. Beyond that, the cases illustrate the diversity and multiplicity of the people and the problems. Case Illustration 1. J, age 19, was hospitalized at County Hospi- tal for short periods, once for a PCP psychosis and once for alcohol abuse. He was unemployed, from a broken family, had no goals, just existed. The final stressor event occurred when his girl friend left to join the army telling J, "You'll never amount to any- thing." His out-patient therapist was shock- ed when J shot himself. "J often told me he wished he were dead, but he said suicide was a sin, and he would never do it." In retrospect, the doctor felt that the problems were too many and too overwhelming and the therapy too little and too late. What might have made more of a difference would have been a placement off the streets into a structured environment, such as a work camp, an in-house drug rehabilitation program, or even possibly, the armed forces. Case lllustration 2. Institutions don’t always guarantee security. B was a 15-year old male who hanged himself 3-107 Report of the Secretary’s Task Force on Youth Suicide in his room at the juvenile detention facility. He had been in trouble in school, had stolen a bicycle, taken money from his mother’s purse, smoked marijuana and was sent to detention (rather than bailed out) in order to "teach him a lesson." Three weeks earlier, he overdosed on aspirin. Two days before, he cut his wrist, causing a noticeable lesion al- though no important structures were severed. The admitting social worker asked B if he was suicidal. B said no, not now. Later, staff persons, noting the cut wrist and negative attitude of B, asked about suicide precautions and were told by the social worker not to worry. In retrospect, the evaluator said that he now understands that suicidal teenagers do not necessarily present themselves as "depressed." Case lllustration 3. C, an age 16 1/2 female, was seeing a coun- selor once a week at an anti-drug abuse oriented community center. C was hard looking, dressed punk, acted tough. She was also a talented musician and poet, struggling with a chaotic home life and her own confus- ing bisexuality. When she hanged herself, she left a three page note beginning, "to let you know I didn’t want it to happen. Sorry. I just wanted to be accepted. Love you. So young, so brave, and yet so weak." The therapist was puzzled over the suicide. With some guilt, he admitted he had been seeing both C and her father separately and in- dividually in treatment, and maybe this arran- gement had been detrimental for C, since the therapist had considered the father to have the more important impairment. Case lllustration 4. The death of M, female, age 17, should have been prevented. She took an overdose of im- ipramine, a tricyclic anti-depressant, after being rejected by her boy friend. Her family took her to the hospital where she was ob- served briefly and discharged prematurely. At home, several hours later, she had a series of convulsive seizures and died. 3-108 M had been seen several times by a male so- cial worker who thought the therapeutic in- teraction was excellent. She came from an intact and supportive family, was a high achiever, and had been admitted to Harvard University. The psychiatric consultant who prescribed the imipramine was also surprised. In retrospect, they recalled that although M was talented and artistic, she had a poor self image and was overly dependent on her boy friend. There had been a previous overdose with tylenol. This case raises the problem of how much anti-depressant to prescribe as take-home medication for per- sons who have recently taken an overdose of other less toxic medicines. Other noted features include the following: Adolescent drug abuse seems to be closely associated with adolescent suicide, especial- ly in older (17-19) males. A suicide in the family is a major stress event, leaving sur- vivors at risk for suicide themselves; bereave- ment counseling is important in preventing further suicides. School problems and con- duct disorders are common in pre-suicidal adolescents. School counseling was impor- tant in helping some control cases avoid self- destructive acts. REFERENCES 1. Sanborn D, Sanborn C, Cimbolic P. A study of adolescent suicide in New Hampshire. Child Psychiatry. Hum. Dev. 3: 234-242, 1973. 2. Shaffer D. Suicide in childhood and early adoles- cence. J Child Psychol. Psychiat. 15: 275-291, 1974. 3. Cosand BJ, Bourque LB, Kraus JF. Suicide among adolescents in Sacramento County, California 1950-1979. Adolescence 17: 917-930, Winter, 1982. 4. Shaffer D, Fisher P. The epidemiology of suicide in children and adolescence. J Amer. Acad. Child Psychiatry 20: 545-565, 1981. 5. Shaffer D, et al. Governor's Youth Suicide Preven- tion Council Research and Evaluation Committee, Report for Albany Meeting, December 12, 1985. 6. Shafii M, et al. Psychological reconstruction of completed suicide in childhood and adolescence in Sudak etal. (eds) Suicide in the Young. pp 271-294, John Wright, Inc. Boston, 1984. 7. Shafii M, Carrigan S, Whittinghill JR, Derrick A. Psychological Suropsy of completed suicide in children and adolescents. er. J. of Psychiatry, 142 (9): 1061- 1064, 1985. 8. Rich CL, Young D, Fowler RC. The San Diego suicide study. Cohen-Sandler R (ed) Proceedings Eighteenth Annual Meeting American Association of Suicidology. pp 67-72. Toronto, Canada 1985. R.E.Litman: Psychological Autopsies of Youth Suicide 9. Litman RE, Diller J. Case studies in youth suicide in Peck ML, Litman RE, and Farberow NL (eds) Youth Suicide, Springer Co. New York, 1985. 10. Litman RE, Wold CI. Beyond crisis intervention in Shheiuman ES (ed) Suicidology. Grune and Stratton, 1976. 3-109 GAY MALE AND LESBIAN YOUTH SUICIDE Paul Gibson, L.C.S.W., Therapist and Program Consultant, San Francisco, California SUMMARY Gay and lesbian youth belong to two groups at high risk of suicide: youth and homosexuals. A majority of suicide attempts by homosexuals occur during their youth, and gay youth are 2 to 3 times more likely to at- tempt suicide than other young people. They may comprise up to 30 percent of completed youth suicides annually. The earlier youth are aware of their orientation and identify themselves as gay, the greater the conflicts they have. Gay youth face problems in ac- cepting themselves due to internalization of a negative self image and the lack of accurate information about homosexuality during adolescence. Gay youth face extreme physi- cal and verbal abuse, rejection and isolation from family and peers. They often feel total- ly alone and socially withdrawn out of fear of adverse consequences. As a result of these pressures, lesbian and gay youth are more vulnerable than other youth to psychosocial problems including substance abuse, chronic depression, school failure, early relationship conflicts, being forced to leave their families, and having to survive on their own prema- turely. Each of these problems presents a risk factor for suicidal feelings and behavior among gay, lesbian, bisexual and transsexual youth. The root of the problem of gay youth suicide is asociety that discriminates against and stig- matizes homosexuals while failing to recog- nize that a substantial number of its youth has a gay or lesbian orientation. Legislation should to guarantee homosexuals equal rights in our society. We need to make a con- 3-110 scious effort to promote a positive image of homosexuals at all levels of society that provides gay youth with a diversity of lesbian and gay male adult role models. We each need to take personal responsibility for revis- ing homophobic attitudes and conduct. Families should be educated about the development and positive nature of homosexuality. They must be able to accept their child as gay or lesbian. Schools need to include information about homosexuality in their curriculum and protect gay youth from abuse by peers to ensure they receive an equal education. Helping professionals need to accept and support a homosexual orienta- tion in youth. Social services need to be developed that are sensitive to and reflective of the needs of gay and lesbian youth. INTRODUCTION Suicide is the leading cause of death among gay male, lesbian, bisexual and transsexual youth.* They are part of two populations at serious risk of suicide: sexual minorities and the young. Agency statistics and coroner reports seldom reflect how suicidal behavior is related to sexual orientation or identity is- sues. The literature on youth suicide has vir- tually ignored the subject. Research in recent years, however, with homosexual young people and adults has revealed a serious problem with cause for alarm. *The terms "gay youth" and "gay and lesbian youth" will fre- quently be used to describe this population in the paper. Transsexual youth are included here because their problems are similar to those experienced by youth who have a minority sexual orientation. P.Gibson: Gay Male and Lesbian Youth Suicide Statistical Profile There is a high rate of suicidality among les- bians and gay men. Jay and Young found that 40 percent of gay males and 39 percent of lesbians surveyed had either attempted or seriously contemplated suicide (1). Bell and Weinberg similarly found that 35 percent of gay males and 38 percent of lesbians in their study had either seriously considered or at- tempted suicide (2). Homosexuals are far more likely to attempt suicide than are heterosexuals. A majority of these attempts take place in their youth. Bell and Weinberg found that 25 percent of lesbians and 20 per- cent of gay men had actually attempted suicide. Gay males were 6 times more likely to make an attempt then heterosexual males. Lesbians were more than twice as likely to try committing suicide than the heterosexual women in the study. A majority of the suicide attempts by homosexuals took place at age 20 or younger with nearly one-third occurring before age 17. Suicidal behavior by gay and lesbian youth, however, occurs today within the broader context of an epidemic increase in suicide among all young people in our society. Bet- ween 1950 and 1980, there was an increase of more than 170 percent in suicides by youth between the ages of 15 and 24 (3). The suicide rate for all age groups rose only 20 percent during that time. At least 5,000 youth now take their lives each year with the number believed to be significantly higher if deliberate auto accidents, victim precipitated homicides, and inconclusive coroner reports are taken into account. The rate of suicide attempts to completions is much higher among young people than any other age group with as many as 500,000 attempts an- nually. This leads us to believe that many times a suicide attempt by a young person is really a cry for help. Gay and lesbian youth have been a hidden population within the adolescent and young adult age group. Those programs and studies able to document suicidality in gay youth have found they have a high rate of suicidal feelings and behavior that places them at sub- stantially greater risk of taking their own lives compared to other youth (See Appendix A). Statistics from the Institute for the Protec- tion of Gay and Lesbian Youth in New York, the University of Minnesota Adolescent Health Program in Minneapolis, Roesler and Deisher in Seattle, and the Los Angeles Suicide Prevention Center consistently show that 20-35 percent of gay youth interviewed have made suicide attempts (4,5,6,7). Statis- tics from Minneapolis, Los Angeles and San Francisco find that more than 50 percent of gay youth experience suicidality including serious depression and suicidal feelings (5,7,8). The Larkin Street Youth Center in San Francisco found that among their client population of homeless youth, 65 percent of homosexual/bisexual youth compared to 19 percent of heterosexual youth reported ever being suicidal, and that gay youth had a rate of suicidality nearly 3.5 times greater than other youth (8). The Los Angeles Suicide Prevention Center in preliminary data from an unpublished study, found that the suicide attempt rate for gay youth is more than 3 times higher than that of heterosexual youth; their rate of suicidality is more than twice that of other youth (7). Why are feelings of self-destructiveness and suicidal behavior so prevalent among gay and lesbian youth? How can we learn to recog- nize these youth better and help them more effectively in coping with the problems that often lead them to want to take their own lives? The rest of this paper attempts to ad- dress these issues by providing an overview of the tasks and problems facing gay youth, an understanding of who they are, factors that place gay youth at risk of suicide, and an approach for society as a whole and the in- dividual helping professional in effectively helping these youth and preventing them from taking their lives. Tasks of the Gay Adolescent Gay youth face the double jeopardy of sur- viving adolescence and developing a positive identity as a lesbian, gay male, bisexual, or transsexual in what is frequently a hostile and 3-111 Report of the Secretary’s Task Force on Youth Suicide condemning environment. Contrary to popular belief, adolescence is not the time of our lives. It is a difficult and complex period of development filled with anxiety and few clear guidelines for helping youth resolve the problems they face, often for the first time, and making the transition to adulthood. Youth are going through physical changes, emotional changes, intellectual changes and sexual development all within the context of their particular culture, family, peer group, and capacity as individuals. They must ac- complish several formidable tasks including separating from their families while retaining a core sense of belonging (individuation), learning to form relationships with other people while fitting in with a social structure (socialization), establishing an integrated, positive, individual identity (identity forma- tion) and preparing themselves for the future in an increasingly complex and uncertain world (future orientation). Problems in accomplishing these tasks play a critical role in the suicidal feelings of any youth but present special hardships for those who are gay or lesbian. First they must come to understand and accept themselves in a society that provides them with little positive information about who they are and negative reactions to their inquiries. Second, they must find support among significant others who frequently reject them. Finally, they must make a social adaptation to their gay or lesbian identity. They must find where they belong and how they fit in with a social struc- ture that either offers no guidelines for doing so or tells them that they have no place. With the advent of the sexual revolution and gay liberation movement of the past two “decades, gay and lesbian youth have been in- creasingly aware of their feelings and coming to terms with their orientation at an earlier age than ever before. This has placed them into direct conflict with all of the traditional childrearing institutions and support systems of our society. Increasingly, this occurs while the youngsters are still living at home with their family, attending public school and developing a sense of their own self worth in 3-112 comparison with their peers and the expecta- tions of society as a whole. Problems Facing Gay Youth Lesbian and gay youth are the most invisible and outcast group of young people with whom you will come into contact. If open about who they are, they may feel some sense of security within themselves but face tremendous external conflicts with family and peers. If closed about who they are, they may be able to "pass" as "straight" in their communities while facing a tremendous in- ternal struggle to understand and accept themselves. Many gay youth choose to main- tain a facade and hide their true feelings and identity, leading a double life, rather than confront situations too painful for them. They live in constant fear of being found out and recognized as gay. The reasons for their silence are good ones. Gay youth are the only group of adolescents that face total rejection from their family unit with the prospect of no ongoing support. Many families are unable to reconcile their child’s sexual identity with moral and religious values. Huckleberry House in San Francisco, a runaway shelter for adolescents, found that gay and lesbian youth reported a higher incidence of verbal and physical abuse from parents and siblings than other youth (9). They were more often forced to leave their homes as "pushaways" or "throwaways" rather than running away on their own. In a study of young gay males, Remafedi found that half had experienced negative parental response to their sexual orientation with 26 percent forced to leave home because of con- flicts over their sexual identity (5). Openly gay and lesbian youth or those "suspected" of being so can expect harass- ment and abuse in junior high and high schools. The National Gay Task Force, in a nationwide survey, found that 45 percent of gay males and nearly 20 percent of lesbians had experienced verbal or physical assault in secondary schools (10). The shame of ridicule and fear of attack makes school a fearful place to go resulting in frequent ab- P.Gibson: Gay Male and Lesbian Youth Suicide sences and sometimes academic failure. Remafedi reports 28 percent of his subjects were forced to drop out because of conflicts about their sexual orientation (5). Gay youth are the only group of adolescents with no peer group to identify with or receive sup- port from. Many report extreme isolation and the loss of close friends. Gay youth also face discrimination in con- tacts with the juvenile justice system and foster and group home placements.* Many families and group homes refuse to accept or keep an adolescent if they know he or she is gay. A report by the San Francisco Juvenile Justice Commission found that gay youth stay in detention longer than other youth await- ing placement because of a lack of ap- propriate program resources (11). Many programs are unable to address the concerns or affirm the identity of a gay adolescent. They can be subjected to verbal, physical, and even sexual abuse with little recourse. Even sympathetic staff often don’t know how to re- late to a gay youth or support them in con- flicts with other residents. They frequently become isolated, ignored by youth and staff who feel uncomfortable with them. They are easy targets for being blamed and scapegoated as the "source" of the problem in efforts to force them to leave. The result of this rejection and abuse in all areas of their lives is devastating for lesbian and gay youth and perhaps the most serious problems they face are emotional ones. When you have been told that you are sick, bad, and wrong for being who you are, you begin to believe it. Gay youth have frequent- ly internalized a negative image of themsel- ves. Those who hide their identity are surrounded by homophobic attitudes and remarks, often by unknowing family mem- bers and peers, that have a profound impact on them. Hank Wilson, founder of the Gay and Lesbian Teachers Coalition in San Fran- *It is my observation that youth are experiencing more fre- quent contact with the juvenile court due to 1) increased conflicts in their home communities because of their sexual orientation which require intervention and removal from the home and 2) being open about their sexual identity at an earlier age than before. cisco, believes these youth constitute a large group who ares silently scapegoated, especial- ly vulnerable to being stigmatized, and who develop poor self esteem (12). Gay youth be- come fearful and withdrawn. More than other adolescents, they feel totally alone often suffering from chronic depression, despairing of life that will always be as pain- ful and hard as the present one. In response to these overwhelming pres- sures, gay youth will often use two coping mechanisms which only tend to make their situation worse: substance use and profes- sional help. Lesbian and gay male youth belong to two groups at high risk for sub- stance abuse: homosexuals and adolescents. Rofes found, in a review of the literature, that: Lesbians and gay men are at much higher risk than the heterosexual population for alcohol abuse. Ap- proximately 30 percent of both the lesbian and gay male populations have problems with alcoholism (13). Substance use often begins in early adoles- cence when youth first experience conflicts around their sexual orientation. It initially serves the functional purposes of (1) reduc- ing the pain and anxiety of external conflicts and (2) reducing the internal inhibitions of homosexual feelings and behavior. Prolonged substance abuse, however, only contributes to the youth’s problems and mag- nifies suicidal feelings. Several studies have found that a majority of gay youth received professional help for con- flicts usually related to their sexual identity (5,6). These interventions often worsen con- ditions for these youth because the therapist or social worker is unwilling to acknowledge or support an adolescent’s homosexual iden- tity. Many gay and lesbian youth are still en- couraged to "change" their identities while being forced into therapy and mental hospi- tals under the guise of "treatment." Those who seek help while hiding their iden- tity often find the source of their conflicts is never resolved because the therapist is un- 3-113 Report of the Secretary’s Task Force on Youth Suicide able to approach the subject. This silence is taken as further repudiation of an "illness" that dare not speak its name. A suicide attempt can be a final cry for help by gay youth in their home community. If the response is hostile or indifferent, they prepare to leave. Alone and frightened, they go to larger cities--hoping to find families and friends to replace the ones that did not want them or could not accept them. The English group "The Bronski Beat" describes the plight of the gay adolescent in their song "Smalltown Boy": Pushed around and kicked around, always the lonely boy You were the one they talked about Around town as they put you down But as hard as they would try just to make you cry You would never cry to them —just to your soul Runaway, turnaway, runaway, turnaway, runaway (14). Gay male, lesbian, bisexual, and transsexual youth comprise as many as 25 percent of all youth living on the streets in this country. Here, they enter a further outcast status that presents serious dangers and an even greater risk of suicide. Without an adequate educa- tion or vocational training, many are forced to become involved in prostitution in order to survive. They face physical and sexual as- saults on a daily basis and constant exposure to sexually transmitted diseases including AIDS. They often become involved with a small and unstable element of the gay com- munity that offers them little hope for a bet- ter life. Their relationships are transitory and untrustworthy. For many street youth, their struggle for survival becomes the fulfill- ment of a "suicidal script” which sees them en- gaging in increasingly self-destructive behaviors including unsafe sexual activity and intravenous drug use. Overwhelmed by the complexities of street life and feeling they have reached the "wrong end of the rainbow" a suicide attempt may result. While it has become easier in recent years to 3-114 be a gay male or lesbian adult it may be har- der than ever to be a gay youth. With all of the conflicts they face in accepting themsel- ves, coming out to families and peers, estab- lishing themselves prematurely in independent living and, for young gay males, confronting the haunting specter of AIDS, there is a growing danger that their lives are becoming a tragic nightmare with living only a small part of dying. UNDERSTANDING GAY AND LESBIAN YOUTH Lesbian and gay male youth are young people with a primary attraction to members of the same sex for sexual and intimate relation- ships. Bisexual youth have an attraction to members of both sexes for sexual and in- timate relationships. We use the term orien- tation rather than preference to describe this attraction because we still do not know how it originates. We are not certain to what ex- tent genetics, socialization factors or in- dividual choice determines either a homosexual or heterosexual orientation. Transsexual youth are young people who believe they have a gender identity that is dif- ferent from the sex they were born with. This includes young males who believe they are really females mistakenly born in a male body and young females who believe they are real- ly males mistakenly born in a female body. Sexual orientation and gender identity are separate issues for each individual. Transsexuals may have a heterosexual, homosexual, or bisexual orientation. Homosexuals are rarely confused about their gender identity with lesbians believing they are women and gay males believing they are men. There are indications that individuals may be predisposed to their sexual orientation from an early age. A gay or lesbian orientation in adolescence is not just a phase the youth is going through. Bell, Weinberg, and Ham- mersmith found that sexual orientation is likely to be formed by adolescence--even if the youth is not yet sexually active (15). P.Gibson: Gay Male and Lesbian Youth Suicide Childhood and adolescent homo- sexuality, especially pronounced homosexual feelings, can not be regarded as just a passing fancy...[it] seems to be relatively enduring and so deeply rooted that it is likely to continue as a lasting homosexual orientation in adult life. Huckleberry House found that, when given a choice, adolescents demonstrate a greater degree of conviction than confusion in iden- tifying their sexual orientation, with 75 per- cent self-reporting as heterosexual, 15 percent homosexual, 5 percent bisexual, and only 5 percent confused or undecided (9). Most youth who identify as heterosexuals and homosexuals will continue to do so as adults. Youth are more likely to underreport a homosexual orientation because of difficul- ties in accepting themselves and the fear of a hostile response. Jay and Young found that 56 percent of the lesbian respondents in their survey had previously identified as bisexual while only 16 percent currently did so (1). Forty-six percent of the gay males had pre- viously identified as bisexual while only 20 percent currently did so. Homosexuality is not a mental illness or dis- ease. It is a natural and healthy expression of human sexuality. In 1935, Sigmund Freud wrote that "Homosexuality...is nothing to be ashamed of, no vice, no degradation, it can not be classified as an illness" (16). In 1973, the American Psychiatric Association removed homosexuality from the list of psychiatric disorders and, in 1975, the American Psychological Association urged all mental health professionals to remove the stigma of mental illness long associated with a homosexual orientation. In 1983, the American Academy of Pediatrics en- couraged physicians to become involved in the care of homosexuals and other young people struggling with the problem of sexual expression (5). If homosexuality is not an ill- ness or a disorder, it can not be regarded as such to the extent that it occurs in the young. Gay and lesbian youth come from all ethnic backgrounds. The ethnicity of gay youth will reflect the ethnicity of youth in your com- munity or seen by your agency. The Institute for the Protection of Gay and Lesbian Youth reports the ethnic breakdown of youth it served, matched the population of New York’s public school system with 40 percent black, 35 percent white, 20 percent Hispanic, 2 percent Asian and 3 percent other (4). Huckleberry House in San Francisco found that more than half of their overall client population and gay youth seen by the program were ethnic minorities (9). There are far more gay youth than you are presently aware of. Kinsey found a sig- nificant amount of homosexual behavior among adolescents surveyed with 28 percent of the males and 17 percent of the females reporting at least one homosexual ex- perience (17,18). He also found that ap- proximately 13 percent of adult males and 7 percent of adult females had engaged in predominantly homosexual behavior for at least three years prior to his survey. This is where the figure that 10 percent of the population is homosexual comes from. It is difficult to assess the prevalence of a homosexual orientation given our knowledge that sexual behavior actually oc- curs along a continuum of feelings and ex- periences. Prevalence is even more difficult to estimate among adolescents because of the complex identity issues with which they are struggling and the scarcity of research on the subject. It is apparent, however, that a substantial minority of youth--perhaps "One in Ten" as one book suggests--have a primary gay male, lesbian, or bisexual orientation. Given the higher rates of suicidal feelings and behavior among gay youth in comparison with other young people, this means that 20- 30 percent of all youth suicides may involve gay youth. Parris believes that as many as 3000 gay and lesbian young people may be taking their lives each year (19). Coming Out: The Early Stages Coming out is the process through which a person comes to understand and accept his/her sexual identity and shares it with 3-115 Report of the Secretary's Task Force on Youth Suicide others. This is seldom a conscious undertak- ing for heterosexual youth who find that being "straight" is a given status in our society. It is as automatic as attending school or get- ting a driver’s license. However, identifying oneself as gay or lesbian is a long and painful process that only occurs gradually over an ex- tended period of time. Stages in the coming out process are identified in Appendix B with the ages reflecting those of gay and lesbian youth whom I worked with at Huckleberry House (20,21). - This population represents the bias of self-identified gay youth seeking services at a runaway program. It is impor- tant to recognize, however, that this process begins for many lesbian and gay youth at an early age with an awareness of their orienta- tion often developing by adolescence. It is then that they experience significant con- flicts involving understanding of whom they are, handling negative reactions from others and making a social adaptation which can lead to suicidal feelings and behavior. These conflicts must be resolved before the youth can develop a positive identity as a gay male or lesbian. The first stage in the development of a les- bian or gay identity is an awareness of being different. This often occurs several years prior to puberty with the youth seldom aware of what this feeling means or how it relates to their sexuality. Lewis, in describing this stage for young lesbians, notes that: Because our society and its process of socialization do not include a posi- tive vocabulary for same-sex attrac- tions (whether emotional or erotic), many girls experience only vague, un- definable feelings of "not fitting in" (22). Bell, Weinberg, and Hammersmith looked at numerous factors (i.e., family relationships) in attempting to determine how individuals develop a homosexual or heterosexual orien- tation (15). They provide evidence that this awareness of being different is related to the social roles of the child. During latency age years, the family often reinforces those roles, behaviors, attributes, and interests that are 3-116 stereotypically associated with being a male or a female in our society. For example, boys are expected to play outside more than girls and girls are expected to stay close to the house more than boys. Bell, et al. found that gay males and lesbians in their study tended to have atypical social roles in childhood that did not conform to gender expectations while heterosexuals tended to have typical social roles. Far fewer homosexual (11%) than heterosexual (70%) men reported having enjoyed boys’ activities (e.g., baseball, football) very much. Fewer of the homosexual (13%) than heterosexual (55%) women said they enjoyed typical girls’ activities (e.g., playing house, hopscotch) very much. This finding held true for a range of variables involving stereotyped male and female roles with gender nonconformity being the single most accurate indicator in childhood of a fu- ture homosexual orientation (15). However, they add a strong point of clarification for those who would force gender conformity on - a child in an effort to "prevent" homosexuality. Homosexuality is as deeply ingrained as heterosexuality, so that differen- ces in behaviors or social experiences of pre-homosexual boys and girls and their pre-heterosexual counterparts reflect or express, rather than cause, their eventual homosexual (orienta- tion). This finding does not account for the sub- stantial percentage of respondents giving answers that were atypical for their sexual orientation. Many children, however, who later identify as gay or lesbian begin to real- ize at this early age that they do not meet the social expectations of their families and other children. The second stage of the coming out process is an awareness of being attracted to mem- bers of the same sex. This also commonly oc- P.Gibson: Gay Male and Lesbian Youth Suicide curs prior to puberty with many gay and les- bian youth reporting childhood crushes on other children and adults. Bell, et al. found these sexual feelings typically occurred three years or so before any homosexual experien- ces and appear to be the most crucial stage in the development of adult homosexuality (15). Most children are unaware of the meaning and implications of these attrac- tions. However, for those who are able to make a connection between their "dif- ference"--homosexual feelings and gay or les- bian identity--depression and suicidal feelings may already be present. I always knew that I was gay. When Iwas 8 or 9 I would steal my mother’s Playgirl magazines and look at the pictures of men. I also remember seeing heterosexual couples and knowing I wasn’t like that. I would get very depressed about not being like other kids. Many times I would take a kitchen knife and press it against my chest, wondering if I should push it all the way in (23). Many adolescents experiencing conflicts re- lated to their sexual orientation report having their first homosexual experience around puberty. Some youth, however, first act on their feelings during adolescence. Young lesbians tend to have their first ex- perience at a later age than young gay males (1). Same-sex play and experimentation is relatively common prior to puberty with Kin- sey reporting that 60 percent of preadoles- cent boys and 33 percent of preadolescent girls described homosexual play at the time they contributed to the study (16,17). Pre- homosexual boys and girls often do not have a context in which to put their feelings and experiences. They have learned to hide sexual behavior from adults but have not developed an understanding of the stigma at- tached to homosexuality. Their initial ex- periences tend to confirm homosexual feelings. It is now, however, that a terrible thing happens to young people who will have a gay or lesbian orientation-- adolescence. Gay and lesbian youth will become distin- guished from other youth involved in preadolescent same-sex play by their progress through the developmental stages here identified and the persistence of homosexual feelings and experiences in spite of negative consequences. Adolescence With adolescence, many gay and lesbian youth have their first contact with homosexuality and it is all bad. They are told it is no longer acceptable to engage in sexual behavior with members of the same sex and that those who do are sick. The only images of homosexuals that society provides them with are derisive stereotypes of lesbians who are like men and gay men who are like women. Many experience their first per- vasive contact with the fear and hatred of homosexuality--homophobia. Nowhere are these harshly negative attitudes towards homosexuality more pronounced than in junior high and high school. These institutions are the brutal training grounds where traditional social roles are rigidly rein- forced. Boys are going to play sports and drink beer with the guys. Girls are going to start paying more attention to their physical appearance in the hopes of attracting boys. Adolescence will be the last stronghold of these stereotyped roles and behaviors be- cause young people are looking for identity. Homosexuality and gender nonconformity are threats to many youth and an easy target for their fears and anxieties about being "nor- mal." Youthwho have a growing awareness of a gay or lesbian orientation become painfully aware that they do not fit the "social script." They see the hostility directed towards homosexuals by others and hear taunts of "dyke" and "faggot" used indiscriminately by peers. They become alarmed and realize that they must make some social adaptation to the situation. Martin describes their predica- ment: In adolescence, young homosexually oriented persons are faced with the 3-117 Report of the Secretary’s Task Force on Youth Suicide growing awareness that they may be among the most despised...As this realization becomes more pressing, they are faced with three possible choices: they can hide, they can at- tempt to change the stigma, or they can accept it (24). These three adaptations are not mutually ex- clusive and are often present in the same in- dividual over time. Many youth initially try to deny a gay or lesbian orientation to both themselves and others. Those adolescents who understand and recognize they have a gay orientation will continue to hide their identity from family and peers for fear of ad- verse consequences. Finally, those who be- come open about their identity, confront those adverse consequences in an effort to win acceptance and support. Each adapta- tion contains specific problems which con- tribute to suicidal feelings and behavior. Self Denial All young people face tremendous pressures to desist from any homosexual behavior and develop a heterosexual orientation. It is easy to see why adolescents with predominantly homosexual feelings and experiences would try to deny a lesbian or gay identity. They have internalized an image of being a homosexual as wrong and dangerous to their physical and mental health. They have seen the stereotypes of lesbians and gay men and they don’t like them. These youth who don’t want to live like that decide they are going to conform to the social roles and start dating members of the opposite sex and become heterosexuals. Many youth engage in heterosexual sexual behavior in an effort to change their orienta- tion. This often turns out to be a losing bat- tle. Jay and Young found that 83 percent of the lesbians and 66 percent of the gay men in their survey had previously engaged in heterosexual sex (1). Bell and Weinberg similarly found that 87 percent of lesbians and 68 percent of gay males interviewed had prior heterosexual experiences (2). Two studies with gay male youth found that at 3-118 least 50 percent had prior heterosexual ex- periences (5,6). Jay and Young add that 55 percent of the lesbians and 46 percent of the gay males reported feeling negative about these experiences. Bell, et al. in their study on the development of sexual orientation conclude that: The homosexual men and women in our study were not particularly lack- ing in heterosexual experiences during their...adolescent years. They are distinguished from their heterosexual counterparts, however, in finding such experiences un- gratifying (15). The American Psychiatric Association notes in the 1980 edition of the Diagnostic and Statistical Manual of Mental disorders (DSM IIT) that "there is a general consensus that spontaneous development of a satisfactory heterosexual adjustment in individuals who previously had a sustained pattern of ex- clusively homosexual arousal is rare" (25). One potentially serious consequence of this heterosexual experimentation is pregnancy involving young lesbians or gay males that either occurs accidentally or in an effort to "prove" a heterosexual orientation. Youth who try to change a homosexual orientation and are unable to do so are at high risk of emotional and behavioral problems. They often develop feelings of hatred and rage that can be turned against themselves or others. They may engage in self-destructive behaviors such as substance abuse as an unconscious expression of feel- ings too painful to face. Others become in- volved in verbal and physical attacks against other homosexuals as a way of defending against their own fears. Finally, when the youth comes to recognize for the first time that he/she have a primary homosexual orientation, overt suicidal behavior may result. The DSM III includes a new disorder called "Ego-Dystonic Homosexuality" which describes many of the conflicts faced by youth engaged in denial of homosexual feelings P.Gibson: Gay Male and Lesbian Youth Suicide (25). It is characterized by "a desire to ac- quire or increase heterosexual arousal...and a sustained pattern of overt homosexual arousal...(that is) unwanted and a persistent source of distress." Associated features in- clude guilt, loneliness, shame, anxiety and depression. Age of onset occurs in "early adolescence when the individual becomes aware that he or she is homosexually aroused and has already internalized negative feel- ings about homosexuality." The course of the illness indicates that "in time, many in- dividuals...give up the yearning to become heterosexual and accept themselves as homosexuals...(with the help) of a supportive homosexual subculture." Remafedi notes that the usefulness of this term is still not known since distress is so prevalent among youth first recognizing a homosexual identity (5). However, it clearly identifies a phenomena in many young homosexuals that places them at a greater risk of taking their own lives. Those Who Hide Many youth are aware of their gay or lesbian identity but decide not to be open about it and try to pass as "straight" with their families and peers. They have seen the negative response to homosexuality from society and the brutal treatment of gays by their peers. Sometimes they have been the recipients of verbal or physical abuse as "suspected" homosexuals. Martin believes that hiding is the primary adaptation for gay and lesbian youth (24). He observes that many realize that their lives are based on a lie with "the socialization of the gay adolescent becoming a process of deception at all levels, with the ability to play a role." While remaining in- visible to others, the pain and loneliness of hiding often causes these youth serious harm to their mental health and social develop- ment. A serious consequence of this adaptation is that these youth suffer their fears and low self esteem in silence. They are unknown victims of scapegoating with every homophobic as- sault or remark they witness. They live in perpetual fear that their secret will be dis- covered. Gay youth become increasingly afraid to associate with others and withdraw socially in an effort to avoid what they per- ceive as a growing number of dangerous situations. They spend more and more time alone. Aaron Fricke relates the problems of hiding a gay identity in his book Reflections of a Rock Lobster: A Story About Growing Up Gay (26). He describes his response to being victimized by a homophobic assault as he was about to begin high school. I began to believe that everyone looked down on me and when anyone looked at me I thought I saw their seething hatred of me coming through. When I entered high school I was completely isolated from the world. I had lost all concept of humanity; I had given up all hopes of ever finding love, warmth or tender- ness. 1did not lie to myself, but I did keep others from thinking I was homosexual. I could refuse to ever mention my real feelings. That way, I'would never again suffer the conse- quences of being the individual I was. I retreated into my own world. The only goal left to me in life was to hide anything that could identify me as gay. I became neurotic about this. I once heard that gay people talked witha lisp. Iwas horrified when I dis- covered that I had a slight lisp, and it made me self-conscious about how I sounded every time I spoke. Self- doubt set in. I thought that anything I did might somehow reveal my homosexuality, and my morale sank even deeper. The more I tried to safeguard myself from the outside world, the more vulnerable I felt. I withdrew from everyone and slowly formed a shell around myself. Everyone could be a potential threat tome. I resembled a crustacean with no claws; I had my shell for protec- tion yet I would never do anything to hurt someone else. Sitting on a rock 3-119 Report of the Secretary’s Task Force on Youth Suicide under thousands of pounds of pres- sure, surrounded by my enemies, the most I could hope for was that no one would cause me more harm than my shell could endure. These youth suffer from chronic depression and are at high risk of attempting suicide when the pressure becomes too much to bear. They may run away from home with no one understanding why. A suicidal crisis may be precipitated by a minor event which ser- ves as a "last straw" to the youth. A low grade may confirm for the youth that life is a failure. An unwitting homophobic remark by parents may be taken to mean that the youth is no longer loved by them. Martin also believes this adaptation hinders the social development of gay and lesbian youth (24). There is an absence of social out- lets for gay youth that makes it very difficult for them to meet others like themselves. They shy away from attachments to friends for fear of getting too involved or experienc- ing rejection. Open relationships or displays of affection with others of the same sex is not tolerated in the gay youth’s home and social environment, making extreme secrecy a re- quirement in developing romantic attach- ments. (Indeed, these issues form the essence of discrimination against homosexuals in our society.) Consequently, lesbian and gay youth do not learn how to es- tablish and maintain intimate relationships in the way heterosexual youth do. Young gay males often experience their same-sex relationships as casual sexual con- tacts with strangers. Because of their age, many of these encounters occur in clandes- tine meeting places where gay males con- ‘gregate. Roesler and Deisher found that 76 percent of their subjects had met sexual partners in parks, 62 percent in theaters, and 32 percent in restrooms (6). Remafedi found that 63 percent of young gay males he sur- veyed had met other males in gay bars; only 28 percent said they had known their partner for a week prior to having sex (5). Martin ex- presses concern that these encounters condi- tion the young gay male to respond to other 3-120 gay males on a sexual level only. He often has not had the opportunity to develop courting behaviors other than direct sexual contact. Heterosexual adolescents learn to date and go through a series of socie- tally ordained procedures with sexual contact as a possible end result. The young gay male often learns to start with the end result, sexual behavior, and then attempts to develop the relationship (24). Young gay males face the risk of mistaking sexual feelings for deeper bonds of love. They may despair of the difficulties in form- ing lasting relationships on the basis of fleet- ing sexual encounters. Suicidal feelings may follow the failure of casual sex to meet the youth’s needs of intimacy and belonging. Young lesbians are even more isolated than young gay males in their efforts to form in- timate relationships. There are few meeting places for lesbians in our society and casual sexual contacts are a less frequent part of their development. Lewis writes that: Because women are socialized to have and maintain relationships, sexual exploration and experimenta- tion often takes place within the con- text of a relationship (22). With fewer social opportunities, however, young lesbians are often not able to form in- itial relationships with lovers until later adolescence or young adulthood. Suicidal feelings among young lesbians may be due to the extreme isolation they experience and the despair of being unable to meet others like themselves. Openly Gay and Lesbian Youth Those who accept their orientation and are open about it with others form a smaller but visible segment of the lesbian and gay male youth population. They learn that only part of developing an identity as a gay male or les- bian is coming to understand and accept your sexual orientation. Now they must find out P.Gibson: Gay Male and Lesbian Youth Suicide what their place is and where they belong within the confines of the traditional social structure available to them. There are few role models to emulate and society offers them little support in this process. Gay youth usually don’t begin to be open about their orientation until middle to late adolescence. Many of these youth will have an atypical so- cial role that includes gender nonconformity. Bell, et al. found that 62 percent of lesbians surveyed described themselves as "very mas- culine" while growing up (15). Remafedi found more than half of young gay males in- terviewed saw themselves as "less masculine” than their peers (5). Gender nonconformity may be more pronounced in youth first open- ly identifying themselves as gay. Sometimes it is a natural and permanent expression of who they are and sometimes it is a transition- al process youth go through in learning that they don’t have to behave in any particular way to be gay. Weinberg and Williams found that younger gay males identified themselves as effeminate three times more frequently than did older gay males (27). Gender nonconformity in gay youth may reflect natural qualities that do not fit cul- tural stereotypes (e.g., men who are gentle, women who are strong). Youth may have ex- pressed these attributes since childhood and will continue to do so as adults. Gender non- conformity may also fit the expectations that society sets for gay and lesbian youth. Gay youth are especially susceptible to cultural stereotypes while struggling to find an ap- propriate identity. One young gay male told me that he literally thought that he had to be "like a girl" because he was gay. There is not a diversity of gay male and lesbian adult role models for gay youth to pattern themselves after. For many young lesbians and gay men, the earliest images of adults they thought were homosexuals were people who fit the traditional stereotypes. One young lesbian recalled when she was a child there was a "tough look- ing" woman with a slight moustache who drove a pick-up truck and lived on the edge of town by herself. This woman was ostracized by the rest of the town and rumored to be a lesbian. The little girl both wondered and feared if she would grow up to be like her. Martin maintains that discrimination prevents adults from being more open about their homosexuality thus denying "suitable role models to gay adolescents who could demonstrate by example, sharing, and teach- ing that existing prejudices are false" (24). This is especially true for gay adults who work with children and adolescents. Gender nonconformity may finally be a con- scious effort to reject traditional roles and es- tablish a separate and viable identity. One young lesbian told me she threw away her dolls in disgust when she was a child. It is not unusual for individuals sharing a common identity to separate themselves from others by establishing particular behaviors, ap- pearances, terminology and interests. Ef- feminacy in young gay males and masculinity in young lesbians is often a way for them to affirm a homosexual identity and assist them in finding each other. According to Wolf, culturally defined masculine attire is "more strongly assumed by young women who are newly aware of their lesbianism and looking for a community" (22). Gay and lesbian youth take tremendous risks by being open about who they are. You have to respect their courage. They remain at high risk to suicidal feelings and behavior because of the pressures they face in conflicts with others about their homosexual orientation and the disappointments they experience at the initial hardships of an openly gay and les- bian lifestyle. Rofes warns that no myth is more dangerous to gay adolescents than the notion that "coming out" will insure them against feelings of self-destructiveness (19). The immediate conflicts that openly gay youth face are with their peers and family. Openly homosexual youth are an affront to a society that would like to believe they don’t exist. Our culture seems to have particular disdain for those gay youth who do not con- 3-121 Report of the Secretary’s Task Force on Youth Suicide form to gender expectations. Rejection or abuse can become so intense that suicidal feelings and behavior result. Openly gay youth are more likely to be forced to leave their schools and families and survive on their own. Those gay youth forced to become self suffi- cient prematurely find that they face the dis- crimination of society against both youth and homosexuals in trying to do so. Often these youth have not had vocational training and some have not completed their secondary education. They are discriminated against in finding housing and employment because of their sexual orientation. Perhaps most disap- pointing, gay youth find they often cannot depend on help from adult gay males and les- bians in getting established because of the fears adult homosexuals have of being seen as "recruiting" young people. Gay youth often become involved with a small and un- stable population of gay males and lesbians living on the streets. Here, they are at high risk at substance abuse, sexually transmitted diseases, and unstable relationships. The hardships of this lifestyle combined with the early rejection by family and peers may result in a suicide attempt. Oneyoung gay male involved in pros- titution attempted suicide after receiving a "hate" letter from his parents. Init his mother said she was sorry she had not gotten an abortion before he was born and his father said that he only had half of a son. The young man completed suicide two years later. A final area of difficulty for openly lesbian and gay male youth is in the forming and maintenance of intimate relationships with others. Having a lover is frequently a new ex- perience for gay youth. Lewis writes: The lesbian’s exploration of intimate experiences with other women is an emotionally turbulent process. It is, essentially, a second adolescence, complete with many of the symptoms common to the mainstream 3-122 heterosexual adolescent period (22). The lack of experience that youth bring to these relationships is compounded by the need for secrecy and lack of social supports for dealing with conflicts so common in homosexual relationships. These first romantic involvements often assume a dis- proportionate importance in the youth’s life. They serve to both affirm a lesbian or gay orientation and also fill unmet needs for love, caring, and friendship that have often been missing in the youth’s life. When the relationship ends, gay youth sometimes feel no one cares and nothing is left to live for. Ethnic Minority Gay Youth Ethnic minority youth (i.e., Black, Hispanic, Asian, and American Indian) comprise a sub- stantial number of youth who are gay, les- bian, bisexual, or transsexual. Ruth Hughes, Coordinator of Gay Youth Services at the Center for Special Problems in San Francis- co, reports that these youth face more severe social and cultural oppression than other gay youth and far more serious problems than other adolescents (29). Bell and Weinberg found that black gay males and lesbians at- tempted or seriously considered suicide at a rate less than white homosexuals but greater than black heterosexuals (2). However, they found that a higher percentage of suicide at- tempts by black homosexuals took place during their youth. 36 percent of black les- bians compared to 21 percent of white les- bians and 32 percent of black gay males compared to 27 percent of white gay males attempted suicide before age 18. This indi- cates that black gay youth may face particular hardships during adolescence. Ethnic minority gay youth face all of the problems that other gay and lesbian youth face growing up in a hostile and condemning society. They also face the same economic discrimination and prejudice confronted by other ethnic minority youth because of racism. Davis notes a dramatic increase in suicides among young blacks over the past two decades that has brought their suicide rate nearly equal to that of white youths (30). P.Gibson: Gay Male and Lesbian Youth Suicide Hendin, in his book, Black Suicide, offers an explanation: It does not seem surprising that suicide becomes a problem at such a relatively early age for the black per- son. A sense of despair, a feeling that life will never be satisfying, confronts many blacks at a far younger age than it does most whites (31). Ethnic minority gay youth additionally face racial discrimination from white homosexuals that is a reflection of their treat- ment by the majority culture. Dutton writes that the gay liberation movement has often failed to consider the needs of ethnic minorities while ignoring their issues and concerns (32). Jones adds that: Little has been written about Third World sexual minorities, and when generalities were made about our lifestyles, attitudes, and be- haviors, they were often made in reference to white cultures--white cultures being the basis for Third World cultures to deviate from or strive for (33). Finally, ethnic minority gay youth must con- tend with discrimination and special problems from their own ethnic group be- cause of their sexual orientation. Hughes believes that ostracism and separation from their own ethnic group is particularly painful and difficult for these youth to cope with: They expect acceptance by those like themselves who understand and have experienced oppression. Too often, blacks don’t want to face the issue and see homosexuality as a struggle for white gay males. Ethnic minority gay youth are seen as an "embarrass- ment" to their cultural group. There is more concern for daily survival is- sues than an increased under- standing of homosexuality (29). Jones adds that: Lesbians and gays growing up in Third World communities ex- perience just as much, if not more, oppression as heterosexual minority youth do in non-Third World com- munities. Unfortunately, most of the negative attitudes and oppressions bestowed upon lesbians and gays in Third World cultures are reactions to the influence that mainstream white culture has on it (33). Two issues that strongly effect ethnic minority gay youth are religion and the fami- ly. Ethnic minority cultures have historically believed that homosexuality is a sin accord- ing to the faiths to which they predominant- ly belong. Parents frequently use religion as the standard to evaluate homosexuality. A homosexual orientation in their son or daughter becomes incompatible with religious beliefs. Ethnic minority gay youth often internalize these religious values and feel guilty for having homosexual feelings and experiences, fearing they are condemned to hell. The family also plays a central role in the lives of these youth with strong expectations that they will fulfill social roles and perpetuate the extended family. A homosexual orientation is sometimes seen as a sign of disrespect to the family by the youth and a threat to the family’s survival. Ethnic minority gay youth have tremendous fears of losing their extended family and being alone in the world. This fear is made greater by the isolation they already face in our society as people of color. These ethnic minority gay youth who are rejected by families are at risk of suicide because of the tremendous pressures they face being gay and a person of color in a white homophobic society. Transsexual Youth Transsexual youth are perhaps the most out- cast of all young people and face a grave risk of suicidal feelings and behavior. Huxdly and Brandon found that 53 percent of 72 transsexuals surveyed had made suicide at- tempts (34). Harry feels that "transsexuals may be at higher risk than homosexuals and 3-123 Report of the Secretary's Task Force on Youth Suicide much higher risk than the general popula- tion" to suicidal behavior (35). Transsexual youth believe they have a gender identity dif- ferent from the sex they were born with. They often manifest this belief beginning in childhood through an expressed desire to be a person of the opposite gender, repudiation of their genitalia, gender nonconformity and cross dressing (25). These behaviors may subside by adolescence due to extreme pres- sures to conform to social expectations. Some transsexual youth, however, try to "pass" in junior high and high school as a per- son of the opposite sex or engage in increas- ingly pronounced behaviors that do not conform to gender expectations. These adaptations present serious internal and ex- ternal conflicts for these youth. All transsexuals are vulnerable to internaliz- ing an extremely negative image of themsel- ves. They experience tremendous internal conflict between this image and their persist- ent desire to become the person they believe they are. Heller notes that suicidal transsexuals tend to feel hopelessly trapped in their situation (36). These feelings may be particularly pronounced in young transsexuals who are forced to hide their identity. While wanting to change their sex, they are seldom able to do so and feel con- demned to a life they are convinced is a mis- take. The DSM III notes that transsexuals frequently experience "considerable anxiety and depression, which the individual may at- tribute to inability to live in the role of the desired sex" (25). This depression combined with a poor self esteem can easily result in suicidal feelings and behavior in transsexual youth. .Some transsexual youth, however, make in- creasingly braver attempts to live as a person of the opposite sex. They experience con- flicts in making a social adaptation to their believed identity. Many young transsexuals will adapt the most stereotyped roles and be- haviors traditionally associated with being a "male" and a "female" in our society. Like other youth, they are trying to define them- selves by rigid adherence to these roles. 3-124 Sometimes transsexual youth experience problems similar to this: A young transsexual male was ar- rested for soliciting an undercover police officer while in drag. He was taken to juvenile hall where he ex- perienced anxiety and confusion around his role in the unit with other boys. One time he reported it was his duty as the "only girl" to provide the other boys with sexual favors. Another time he broke down crying feeling as though he was being used and abused by the other males. A week later he made a suicide at- tempt. Transsexual youth who are open about their identity face extreme abuse and rejection from families and peers. Many are forced to leave their home communities and survive on the streets. Their prognosis in our society is poor and they are at high risk of suicide. Gender dysphoria is a disorder that we have little understanding of and a great deal of repulsion for. The only known course of treatment is to help transsexuals to adjust to their believed gender identity and obtain sex- reassignment surgery. Most transsexual youth, however, are unable to obtain or af- ford the help they need in resolving their identity conflicts. It is important to distinguish between transsexual youth and gay and lesbian youth who do not conform to gender expectations. Gender nonconformity is common among gay youth in both childhood and adolescence. Some gay and lesbian youth may experience gender identity confusion during adoles- cence in the coming out process because of the intense social pressures for gay males to be like women and lesbians to be like man. Gay youth may feel they actually have to be a person of the opposite gender to meet those expectations. Hughes, in her work with both homosexual and transsexual youth, emphasizes the importance of working with ayoung person over a period to time to deter- mine if they are truly a transsexual (29). Gay and lesbian youth come to recognize that P.Gibson: Gay Male and Lesbian Youth Suicide they neither want to change their sex nor live as a person of the opposite gender. RISK FACTORS IN GAY AND LESBIAN YOUTH SUICIDE Gay young people face the same risk factors for suicidal behavior that effect other youth. These include family problems, breaking up with a lover, social isolation, school failure, and identity conflicts. However, these fac- tors assume greater importance when the youth has a gay or lesbian orientation. Jay and Young found that 53 percent of gay males and 33 percent of lesbians surveyed believed their suicide attempts involved their homosexuality (1). Bell and Weinberg report that 58 percent of gay males and 39 percent of lesbians felt their first suicide at- tempts were related to the fact that they were homosexuals (2). Suicide attempts by gay and lesbian youth are even more likely to in- volve conflicts around their sexual orienta- tion because of the overwhelming pressures they face in coming out at an early age. General Bell and Weinberg found that initial suicide attempts related to homosexuality more fre- * quently involved acceptance of self and con- flicts with others for gay males, while lesbians tended to cite problems with lovers as the reason (2). Self acceptance may be especial- ly critical for young gay males who tend to have homosexual experiences and are aware of their orientation at a somewhat earlier age than lesbians (1,15). Conflicts with others may be more salient for young gay males "identified" as homosexuals. Gender non- conformity elicits a negative response from others for lesbian and gay male youth, but society seems to have particular disdain for effeminate young males. Young lesbians may experience more extreme social isola- tion, often reporting an absence of same-sex experiences or knowing others like them during adolescence. They also face stronger social pressures to fulfill the woman’s tradi- tional role of marrying and having children and may experience more depression related to not meeting social expectations. Problems with lovers may be especially critical for young lesbians because their sexuality is often explored within the context of their early intimate relationships. The earlier a youth is aware of a gay or les- bian orientation, the greater the problems they face and more likely the risk of suicidal feelings and behavior. Remafedi observes that: Younger gay adolescents may be at the highest risk for dysfunction be- cause of emotional and physical im- maturity, unfulfilled developmental needs for identification with a peer group, lack of experience, and de- pendence on parents unwilling or un- able to provide emotional support ()- He adds that younger gay adolescents are more likely to abuse substances, drop out of school, be in conflict with the law, undergo psychiatric hospitalization, run away from home, be involved in prostitution, and at- tempt suicide. The Los Angeles Suicide Prevention Center recently found that the strongest causative indicators of suicidal be- havior among gay youth were awareness of their sexual orientation, depression and suicidal feelings, and substance abuse--all before age 14 (7). A 14 year old gay male in- terviewed for this paper confirms that profile: When I was 11, I started smoking dope, drinking alcohol, and snorting speed every day to make me feel bet- ter and forget I was gay. I would party with friends but get more and more depressed as the night would go on. They would always make anti- gay remarks and harass gay men while I would just stand there. Late at night, after they went home, I would go down to the river and dive in--hoping I would hit my head on a rock and drown (23). 3-125 Report of the Secretary's Task Force on Youth Suicide Society It is a sobering fact to realize that we are the greatest risk factors in gay youth suicide. No group of people are more strongly affected by the attitudes and conduct of society than are the young. Gay and lesbian youth are strongly affected by the negative attitudes and hostile responses of society to homosexuality. The resulting poor self-es- teem, depression, and fear can be a fatal blow to a fragile identity. Two ways that society in- fluences suicidal behavior by gay and lesbian youth are: 1) the ongoing discrimination against and oppression of homosexuals, and 2) the portrayal of homosexuals as being self- destructive. It is the response of our society as a whole to homosexuality, and specifically those institu- tions and significant others responsible for their care, that pose the greatest risk to gay and lesbian youth. Gock believes that homophobia, the irrational fear and hatred of homosexuals, is the root of the problem (37). Gay males and lesbians are still routine- ly the victims of violence by others. In a recent survey of nearly 2,100 lesbians and gay men nationwide, the National Gay Task Force found that more than 90 percent had been victims of verbal and physical assault be- cause of their sexual orientation (10). Tacit and explicit discrimination against homosexuals is still pervasive in virtually all areas of life. Half of the States still prohibit homosexual relationships between consent- ing adults (37). Homosexuals are not al- lowed to legally marry and form "legitimate" long-term relationships. The vast majority of States and municipalities still discriminate against lesbians and gay men in housing, employment and other areas. Gay and les- bian youth see this and take it to heart. Rofes warns us against the myth that homosexuality, in and of itself, encourages suicide (13). There is nothing inherently self-destructive in homosexual feelings and relationships that could be a source of suicidal behavior. In his book, I Thought People Like That Killed Themselves Rofes maintains we have created a stereotyped 3-126 image of the unhappy homosexual in litera- ture and the media (e.g., Boys in the Band) for which suicide is the only appropriate resolution. This image is reinforced by the fact that homosexual characters in novels and films invariably kill themselves in the end. The myth is perpetuated by the absence of positive adult gay role models in our society where, historically, the only known homosexuals were those exposed by scandal and disgraced in their communities. Rofes maintains this creates a strong negative con- text for the early identity formation of young gay males and lesbians effectively socializing them into suicidal feelings and behavior. He sees a strong correlation between sexual orientation, social response to that sexual orientation, and subsequent suicidality in an individual. Self Esteem A predisposing factor in suicidal feelings among many adolescents is poor self esteem. This is especially true for gay adolescents who have internalized a harshly negative image of being bad and wrong from society, religion, family, and peers. For youth, a poor self-image contributes substantially to a lack of confidence in being able to cope with problems. The images of homosexuals as sick and self-destructive have impact on the coping skills of gay youth, rendering them helpless and unable to improve their situa- tion. Gay youth who have internalized a mes- sage throughout their lives of being worthless and unable to cope from abusive and chaotic families are at even greater risk. Youth with a poor self-esteem and poor coping skills are particularly vulnerable to suicidal feelings when confronting a problem for the first time. They really don’t know how to resolve it or even if they can. Gay youth are highly susceptible to suicidal feelings during the "coming out" process when first facing their own homosexuality and the hos- tile response it evokes in others. They may attempt suicide when they first realize they have homosexual feelings or a gay orienta- tion. Some youth deny their homosexual P.Gibson: Gay Male and Lesbian Youth Suicide feelings and engage in unconscious self- destructive behavior out of self hatred. Others try to "change" their orientation and make a suicide attempt when they recognize their homosexuality will not go away and is part of who they are. Many youth realize they are gay or lesbian but attempt to hide their orientation from others. They suffer from chronic loneliness and depression. They may attempt suicide because they feel trapped in their situation and believe they do not deserve to live. A suicidal gesture may be a cry for help from these youth for others to recognize and un- derstand their situation. Finally, those youth who are open about being gay, lesbian, or bisexual face continuous conflict with their environment. They remain vulnerable to suicide because they face these extreme pres- sures with a more fragile sense of self worth and ability to cope with life than other youth. Family Family problems are probably the most sig- nificant factor in youth suicide. Youth derive their core sense of being cared about and belonging from their families. Gay youth may make suicide attempts after being rejected by their families. For gay and les- bian youth forced to leave home, the loss of parental love and support remains a critical issue for them. Sometimes the youth’s sexual orientation becomes a convenient excuse for parents to reject a son or daughter they did not want. Youth from abusive and dysfunc- tional families are at even greater risk. Wandrei found, in comparing suicide at- tempts by lesbians and heterosexual women, that lesbians were more likely to come from broken homes (39). Gay and lesbian youth face more verbal and physical abuse from family members than do other youth. The National Gay Task Force found that more than 33 percent of gay males and lesbians reported verbal abuse from rela- tives because of their orientation and 7 per- cent reported physical abuse as well (10). These figures are substantially higher for youth open about their sexual orientation while still living at home. Sometimes this harassment becomes too much to bear for gay youth and a suicide attempt results. Gay and lesbian youth may feel suicidal be- cause of a failure to meet family expectations. All youth need approval from their parents. Some youth report only feeling loved by parents when they are fulfilling their parents image of who they should be. Gay youth often feel they can not meet their parents standards and may attempt suicide after real or anticipated disappointment by their families that they will not fit the social script of heterosexual marriage and grandchildren. This pressure is particularly strong for les- bians. Gay youth fear they will not have families of their own and be alone as adults with no one to care for them. Communication problems also play a serious role in family issues for gay youth. Many les- bian and gay youth hide their orientation from their parents out of fear of rejection. They have often seen a strong negative reac- tion to homosexuality by parents and siblings including homophobic remarks. The an- ticipated inevitable loss of love can precipitate a suicide attempt. Parris related a call to a suicide hotline in Washington, D.C: The youth said that he was gay and wanted to talk with his parents about it but was afraid because they were very religious. A week later, a man called...to say his son had committed suicide. They were calling an un- familiar number on their long dis- tance phone bill. By matching the man’s address...the tragic connection was made (19). Religion Religion presents another risk factor in gay youth suicide because of the depiction of homosexuality as a sin and the reliance of families on the church for understanding homosexuality. Many traditional (e.g., Catholicism) and fundamentalist (e.g., Bap- tist) faiths still portray homosexuality as 3-127 Report of the Secretary's Task Force on Youth Suicide morally wrong or evil. Family religious beliefs can be a primary reason for parents forcing youth to leave home if a homosexual orientation is seen as incompatible with church teachings. These beliefs can also create unresolvable internal conflicts for gay youth who adhere to their faith but believe they will not change their sexual orientation. They may feel wicked and condemned to hell and attempt suicide in despair of ever obtain- ing redemption. School Many gay and lesbian youth feel trapped in school settings because of a compulsory obligation to attend and the inability to defend themselves against verbal and physi- cal assaults. Schools do not adequately protect gay youth with teachers often reluc- tant to stop harassment or rebut homophobic remarks for fear of being seen as undesirable role models (19). Verbal and physical attacks against gay youth have increased in recent years as students become increasingly threatened by the presence and openness of peers with a lesbian or gay orientation. This abuse begins as early as late elementary school, becomes pronounced in junior high when youth are still immature, and continues into high school. The failure of schools to ad- dress this concern can be tragic: In Lebanon, Pennsylvania in 1977, a 16-year old boy fatally shot himself before entering the 10th grade. He left a suicide note explaining he could not return to school and sus- tain the abuse and ridicule about being gay from his classmates. A few friends at school supported (him) though they knew he was gay, but the majority ridiculed him without mercy. He skipped classes to avoid the torture and welcomed the sum- mer vacation as a respite. But he was already taking pills to escape the reality of the approach of another school term, when he would have to move from junior high to the even more sharply defined roles of senior 3-128 high. On September 3 he shared that anxiety with a friend and on Septem- ber 5 he shot and killed himself (40). The failure of schools to educate youth about homosexuality presents another risk factor to gay and lesbian adolescents. By ignoring the subject in all curricula, including family life classes, the schools deny access to positive in- formation about homosexuality that could improve the self esteem of gay youth. They also perpetuate myths and stereotypes that condemn homosexuality and deny youth ac- cess to positive adult lesbian and gay role models. This silence provides tacit support for homophobic attitudes and conduct by some students. Social Isolation Social isolation has been consistently iden- tified as one of the most critical factors in suicide attempts by youth. The isolation and alienation young people experience in all aspects of their lives can be overwhelming. Those youth hiding their identity often withdraw from family and friends out of fear of being discovered. They feel there is no one they can talk to and no one who will un- derstand. Tartagni, based on his experience teaching in public school, writes that "one of the loneliest people in any high school in America is the rejected and isolated gay adolescent" (41). This isolation may be more extreme for young lesbians who often report a total lack of contact with others like them- selves during high school. Joanne, in One Teenager in Ten, describes her feelings after realizing her lesbianism in adolescence: In October, I realized my lesbianism and I did not have someone gay to talk with. I recall the anguish I suf- fered looking back over my journal during that time period. "Please. Help me. Oh shit, I have to talk with someone...I have to tell someone, ask someone. WHO??!! Dammit all, would someone please help me? Someone, anyone. Help me. I'm going to kill myself if they don’t" (28). P.Gibson: Gay Male and Lesbian Youth Suicide Openly gay youth experience blatant rejec- tion and isolation from others. One young gay male related that his parents refused to eat at the dinner table with him after they learned he was gay. Male peers cruelly separate themselves from young gay males with jokes about not wanting to get AIDS. Gay youth frequently do not have contact with other gay adolescents or adults for sup- port. Parents often forbid them from as- sociating with people they "suspect" or know to be homosexuals. Youth service workers often feel uncomfortable talking with gay young people because of their prejudices and lack of understanding for who they are. The Los Angeles Suicide Prevention Center, in their recent study on gay youth suicide, ironi- cally found that gay young people rated so- cial support as being very important to them while simultaneously experiencing people as being more rejecting of them than did other youth (7). Substance Abuse Some gay and lesbian young people cope with the many problems they face by using al- cohol and drugs. The age of onset for sub- stance use among all youth has become lower in recent years and in 1985 is estimated to be 11.9 years for boys and 12.7 years for girls (42). This coincides with the age that many youth are becoming aware of a gay or lesbian orientation. Rofes found that lesbians and gay men have a higher rate of substance abuse than heterosexuals and found this to be correlated with increased suicidal feelings and behavior (13). Gay youth are especially susceptible to sub- stance abuse in trying to cope with the con- flicts of the coming out process. Remafedi believes there may be a higher rate of sub- stance abuse among gay youth than among gay adults (5). He found that 58 percent of young gay males he interviewed could be classified as having a substance abuse disor- der in the DSM III. Gay youth forced to live on the streets experience more severe drug problems. The Larkin Street Youth Center in San Francisco reported that more than 75 percent of their clients identified as gay had serious and chronic disorders (8). The Los Angeles Suicide Prevention Center found a strong correlation between substance abuse and suicide attempts among gay young people (7). Professional Help Perhaps no risk factor is as insidious or uni- que to the suicidal behavior of gay and les- bian youth than receiving professional help. The large number of gay youth who have had contact with mental health and social work services during their turbulent adolescent years would seem to be a positive indicator for improving their stability and future out- look. This is sadly not often the case. Many helping professionals still refuse to recognize or accept a homosexual orientation in youth despite growing evidence that sexual orien- tation is formed by adolescence (15). They refuse to support a homosexual orientation in youth despite the fact that homosexuality is no longer viewed as a mental disorder (25). They continue to insist that homosexual feel- ings are just a passing "phase", while making the goal of treatment arresting or changing those feelings and experiences. Martin poin- tedly describes this process: Pain and suffering are inflicted on the very young, whom society is sup- posedly protecting, under the guise of preventing the spread of homosexuality or of treating the in- dividual (24). He adds that some psychiatrists even advo- cate creating conflict, guilt and anxiety in adolescents concerned about homosexual feelings where none has previously existed. Youth who deny their feelings and ex- perience "ego-dystonic homosexuality" are especially vulnerable to this type of adverse treatment. Rather than helping these youth to accept and understand predominantly homosexual feelings and experiences, we see their denial as a "hopeful" sign that they can still develop a heterosexual orientation. When homosexual feelings persist after 3-129 Report of the Secretary’s Task Force on Youth Suicide treatment has attempted to change them, the youth despairs and is at potentially greater risk for suicide than if we tried to help him/her towards acceptance. Youth who are aware of their lesbian or gay orientation but hide it from others, may seek help without identifying their concerns about their sexuality. We often do not recognize these youth because we don’t acknowledge they exist. We are uncomfortable in discuss- ing or addressing the issue and consequently are unable to identify or resolve the source of the youth’s conflicts. A suicide attempt may be an effort by the youth to force the issue and bring it to our attention. It may also be an act of despair over a problem that they feel can not be addressed through profes- sional help. Even openly gay and lesbian youth are sub- jected to treatment with potentially adverse effects. Frequently, informing family and counselors that a youth is gay is the impetus for imposed treatment. We assume that the youth’s gay orientation is the source of the problem rather than the response of others to his/her being lesbian or gay. Encouraging these youth to change can cause regression in the development of a healthy gay identity and reinforce traditional stereotypes of homosexuals as sick and self destructive. This, in turn, further weakens the youth’s self-esteem and ability to cope with problems. Even those professionals who ac- cept the youth as gay or lesbian are often un- able to support the youngster in dealing with conflicts at home and in school. Youth Programs There is a critical lack of program resources for gay and lesbian youth. Many social and recreational programs for youth make no ef- fort to incorporate gay young people into their services. Few programs will accept or support a gay adolescent in their sexual orientation. Agency policies tacitly or ex- plicitly forbid the hiring of openly gay and les- bian staff, denying gay youth access to positive adult gay role models. Homophobic remarks and attitudes by youth and staff in 3-130 many of these programs go unrebutted. Con- sequently, gay youth do not use many of the youth service resources available to them or soon leave if they do. This increases their so- cialisolation and alienation from their peers. Other gay and lesbian youth who are wards of the juvenile court have little choice but to live in those placements to which they are referred. Here, they re-experience many of the problems they had in their home com- munities. Many foster families are rejecting of gay and lesbian youth, feeling less invest- ment than a youth’s natural family to keep- ing the youth in the home. Gay male and lesbian adults are prohibited in most States from being foster parents with gay youth again denied access to supportive adults who could serve as positive role models for them. Group home placements present special hardships for gay youth because abusive peers often live in the same home with them. Those programs without an on-site school re- quire gay youth to return to public school for their education. Program staff have seldom received training on issues and concerns re- lated to homosexuality. They are frequently unable to understand or work with gay youth effectively. Group homes become a living hell of harassment, isolation, and conflicts with other staff and residents offering gay youth little support and no resolution. A suicide attempt may be an effort to force removal from the placement and find a dif- ferent home. Many homes, however, will not accept gay youth and few offer specialized services to meet their needs. Relationships with Lovers We are all victims of the myth that our first love will be our one true love until death do us part (e.g, Romeo and Juliet). Young people are especially vulnerable to this mis- conception and breaking up with a lover is one of the most frequent reasons for their suicide attempts. The first romantic involve- ments of lesbian and gay male youth are a source of great joy to them in affirming their sexual identity, providing them with support, and assuring them that they too can ex- P.Gibson: Gay Male and Lesbian Youth Suicide perience love. However, society places ex- treme hardships on these relationships that make them difficult to establish and main- tain. Bell and Weinberg found that relation- ship problems were the single most frequently cited reason for the initial suicide attempts of lesbians (62%) and gay males (42%)(2). Intimate relationships are the primary focus of hostility and discrimination against homosexuals. Society severely restricts where homosexuals can meet, prevents public displays of affection between them, and does not allow legal marriages to be formed. Gay and lesbian youth suffer greater isolation than homosexual adults and far greater social deprivation than other adoles- cents. It is extremely difficult for them to meet other homosexuals and they frequently do not know anyone like themselves. Gay youth who hide their identity often form their first romantic attachments to unknowing friends, teachers, and peers. These are often cases of unrequited love with the youth never revealing their true feelings. Gay youth are fragile in these situations and may experience despair or suicidal feelings from never being able to fulfill their hopes for a relationship. Some gay youth bravely reveal their feelings and may attempt suicide after blatant rejec- tion by a teacher or the loss of a close friend. Young gay males often experience their first homosexual relationships as brief sexual en- counters in clandestine meeting places (e.g., parks). The extreme need for secrecy and anonymous nature of these contacts serious- ly hinder their further development. The in- tensity of sexual feelings that accompany these encounters can easily be mistaken for romantic attachment by young gay males. They may feel suicidal at the failure of these experiences to meet intimacy needs and the inability to fulfill the social expectation of sustaining the relationship. Young lesbians experience greater isolation than young males. They are less likely to explore their sexuality or have relationships during adoles- cence. They may feel suicidal at the despair of ever finding love in relationships with other women. Gay and lesbian youth develop intimate relationships at a later age than other youth and are unable to develop relationship skills in the manner of other adolescents. Their first romances are an emotionally turbulent trial and error process that resembles a second adolescence. Gay youth bring to these relationships extreme dependency needs resulting from the deprivation ex- perienced in their relationships with family and peers. They also are still in the process of forming their identity and have unresolved issues of guilt and poor self-esteem. When conflicts arise in homosexual relationships there are few social supports available to as- sist them. This is compounded for gay youth by their frequent need for secrecy and the fact that they may not be open about their identity with family and friends. Breaking up with a lover may confirm earlier negative experiences and concepts as- sociated with being a homosexual. Young lesbians often explore and define their sexuality within the context of their first relationships. A relationship failure for them may be synonymous with problems in developing a positive lesbian identity. For some gay youth, relationships become a way of filling needs for love and belonging miss- ing from family and peers. When the relationship ends, the youth feel as though they have lost everything. They fear that they will always be alone, that no one cares, and nothing is worth living for. Independent Living Gay and lesbian youth are more likely than other adolescents to be forced to leave home and become self sufficient prematurely. Some of these youth have been hiding their identities and can no longer stand the ex- treme isolation in their lives. Many others have been rejected by families and have dropped out of school, effectively forced out of their communities because of their sexual orientation. Gay youth come to large cities hoping to find others like themselves, legitimate employment, a lover and a new 3-131 Report of the Secretary’s Task Force on Youth Suicide "family." They soon become aware of the lack of opportunities available to them and become enmeshed in the problems of sur- vival. Suicidal feelings emerge as the hope for a new and better life begins to pale. Most gay youth are unprepared for the dif- ficulties they encounter. They are dis- criminated against in finding employment and housing by virtue of being both young and homosexual. Many have no vocational training and some were not able to finish high school. They often find limited support from the lesbian and gay male adult community who fear involvement with youth. Many are forced to turn to the streets for survival. A recent study on adolescent male prostitution found that nearly 75 percent identified them- selves as gay or bisexual, with family conflicts as the primary reason for leaving home (43). Many gay youth become homeless. Others depend on relationships with people they meet on the streets to obtain shelter and sur- vival needs. Gay youth living on the streets are at greater risk of suicide due to repeated exposure to chronic substance abuse, physical and sexual assault, and sexually transmitted diseases in- cluding AIDS. Their contact with the limited segment of gay adults involved in street life confirms a negative image of homosexuality and they remain unaware of the variety of positive adult gay lifestyles open to them. Their relationships are tenuous and compli- cated by dependence on their lovers for sup- port. Some gay and lesbian youth engage in increasingly reckless and self-destructive be- havior as an expression of the sadness and anger they feel because of the unresolved is- sues with their families and despair over their new life. A suicide attempt may result from a negative contact with their family, breaking up with a lover, or failure to make it on their own. AIDS (Acquired Immune Deficiency Syndrome) Gay and bisexual male youth again belong to two groups at high risk of contracting sexual- ly transmitted diseases: gay/bisexual males 3-132 and adolescents. Although the number of confirmed cases of AIDS and ARCs (AIDS Related Conditions) among adolescents is small, it is believed that cumulative exposure to the virus, beginning in adolescence, may result in a diagnosis or symptoms as a young adult. Gay and bisexual males have always been subject to a greater number of health problems through sexually transmitted dis- eases (STD). They comprise a substantial majority of confirmed cases of AIDS and more than 50 percent of adult gay males will contract Hepatitis Type B during their lifetime (44). Young people are taught in our society that sex is a secretive and spontaneous activity. Adolescent males are not encouraged to take responsibility for their sexual behavior; the vast majority do not take precautions in their sex practices. They engage in impulsive and unplanned sexual activity with grave conse- quences. Young people contract several mil- lion cases of STDs every year (45). Gay and bisexual male youth are particularly vul- nerable because of their need for secrecy in sexual contacts and the frequency with which they engage in unplanned sexual activity. Those gay and bisexual male youth living on ~ the streets face a substantially greater risk of exposure to STDs because of repeated sexual contacts in their relationships and prostitu- tion experiences. Street youth face addition- al exposure through intravenous drug use. Sexual experiences are important to gay male youth as a way of exploring and affirming their sexual orientation. Many do not take precautions and share a feeling of invul- nerability to future consequences that is common among all youth. Remafedi found, however, that 45 percent of young gay males interviewed had a history of STD's (5). The attitudes of young gay males towards ex- posure to AIDS ranges from denial to ex- treme fear to not caring. One young male said he was not concerned because "teenagers do not get AIDS." Another was convinced that a head cold he had developed was the first symptom of AIDS. Those who are at greatest risk may be those who simply P.Gibson: Gay Male and Lesbian Youth Suicide do not care whether they are exposed to the virus. Some gay youth have an uncaring ap- proach to life that reflects a "suicidal script.” They are more prone to self-destructive be- havior because of the severity of the problems they have experienced throughout their lives and specifically in relation to their sexual orientation. Contracting AIDS be- comes for them the fulfillment of alife of pain and suffering they no longer want to cope with. They feel that they deserve to die. Future Outlook A final risk factor for gay and lesbian youth suicide is a bleak outlook for the future. Young people have difficulty seeing a future life that is different from the present. Gay and lesbian youth fear their lives will always be as unhappy and hard as they presently are. They do not know that they will receive any more caring, acceptance, and support than they are getting now. The little information they have about homosexuality usually rein- forces these mistaken beliefs. Gay youth do not understand what life could be like as a gay male or lesbian adult. They do not have ac- curate information about homosexuality, positive role models to pattern themselves after, or knowledge of gay and lesbian adult lifestyles and communities. Lesbian and gay youth frequently don’t know that many les- bian and gay male adults lead stable, happy, and productive lives. They go through adolescence feeling lonely, afraid, and hope- less. Sometimes they take their own lives. ENDING GAY AND LESBIAN YOUTH SUICIDE We can substantially reduce the risk of suicide among gay and lesbian youth. The problem is clearly one of providing informa- tion, acceptance, and support to gay youth for coping with the pressures and conflicts they face growing up as homosexuals in our society. However, in addressing their con- cerns we confront two issues of greater mag- iety sa sh irre fi mn | i 7 pn. homophobia experienced by gay all parts of their lives is the primary reason for their suicidal feelings and behavior. iti i ). This is the issue we must address to save the lives of gay males and les- bians who are young. Society The first step in ending gay youth suicide is to end the discrimination against and stig- matization of homosexuals in our society. We have tenaciously clung to lies and prejudices about homosexuals for far too long. Too many lives have been brutalized and lost. A growing body of research con- tradicts our negative biases and assumptions about gay males and lesbians. We do not, as a society, want to continue to hold the un- tenable position of senselessly hurting others--especially the young. Gay males and lesbians need to be accepted as equal partners in our society. Laws should safeguard their individual rights and not per- mit discrimination against them in housing, employment, and other areas. Laws prohibiting homosexual relationships be- tween consenting adults should be repealed and marriages between homosexuals should be recognized. Special attention should be paid to the enforcement of laws that punish those who commit violence against homosexuals. Laws can help to establish the principle of equality for lesbians and gay men and define the conduct of others in their in- teractions with them. It is an even more comprehensive task to ad- dress the negative attitudes about homosexuality held by so many people. A conscious effort must be made to dispel the destructive myths about homosexuality at all levels of society. We must promote a posi- tive image of gay males and lesbians to reduce oppression against them and provide gay youth with role models to pattern themselves after. Massive education efforts need to take 3-133 Report of the Secretary's Task Force on Youth Suicide place that would provide people with ac- curate information about homosexuality. These efforts especially need to be directed to those who have responsibility for the care of the young including families, clergy, teachers, and helping professionals. The media needs to take responsibility for promoting a positive image of homosexuals that presents a variety of gay male and lesbian lifestyles. We must also take personal responsibility for revising our own homophobic attitudes and behavior as an ex- ample to others in the same way that we have worked towards revising discriminatory ra- cial attitudes and conduct. It is at the per- sonal level that we have the greatest impact on the lives of those around us. Third, we must directly address the issue of homosexuality in the young. Our society has historically denied the sexuality of young people. We must educate ourselves on the issues and problems related to sexual development in young people. Society needs to promote a positive image about sexuality and provide youth with accurate information on the subject. We need to recognize that youth are sexually active from an early age and that sexual orientation is frequently formed by adolescence. All youth need to be provided with positive information about homosexuality that presents it as a viable adaptation. We must accept a homosexual orientation in young people in the same man- ner we accept a heterosexual orientation. Finally, we need to assist gay and lesbian young people in the coming our process and support them in the many conflicts they presently face. Family Gay and lesbian youth need to receive accep- tance and understanding from their families if we are to reduce their risk of suicide. Parents need to be educated as to the nature and development of homosexuality in in- dividuals. They often feel guilty and ashamed upon first learning that their child is gay because they have been told that it is wrong and they are to blame. Parents should 3-134 know that homosexuality is a natural and healthy form of sexual expression. They do not need to feel bad about something that is good. Parents should also know that we still do not know the origins of a heterosexual or homosexual orientation. Research indicates a predisposition towards sexual orientation in children that limits the role of family in its development. Families have a great deal of influence on how their children feel about their sexual orientation. Parents should be made aware of the potential negative impact homophobic remarks and behavior have on their child. Homophobic conduct can be taken as rejec- tion by youth struggling with their sexual orientation or encouragement by other youth to victimize homosexuals as they grow older. Families need to take responsibility for presenting homosexuality in a positive context to their children. Parents need to ac- cept and understand a son or daughter with a homosexual orientation. Those parents who have difficulty accepting their lesbian daughter or gay son should get more infor- mation on the subject and not try to "change" them. They should let the child know they are still loved and cared about as individuals regardless of their sexual orientation. Ethnic minority families need to understand and accept their gay and lesbian children. Ethnic minority gay youth depend even more strongly on their extended family and culture for support because of the additional oppres- sion they face as a racial minority within society as a whole and the homosexual com- munity. Parents need to be educated as to the extent and diversity of lesbians and gay males within ethnic minority cultures. They need to understand that their child means no disrespect to the family and cannot be any dif- ferent from whom they are. Society needs to reinforce parental respon- sibilities for the care of their children, ir- respective of sexual orientation, until they become adults. Parents need to be held ac- countable for the abuse of their children re- lated to their homosexual orientation. We need to become more conscious of the extent P.Gibson: Gay Male and Lesbian Youth Suicide to which the abuse of gay adolescents occurs within their own families. Religion Religions need to reassess homosexuality in a positive context within their belief systems. They need to accept gay youth and make a place for them in the church and include them in the same activities as other youth. Religions should also take responsibility for providing their families and membership with positive information about homosexuality that discourages the oppression of lesbians and gay men. Faiths that condemn homosexuality should recognize how they contribute to the rejection of gay youth by their families and suicide among lesbian and gay male youth. Schools Public and private schools need to take responsibility for providing all students at the junior high and high school level with posi- tive information about homosexuality. Cur- riculum materials should include information relevant to gay males and lesbians as it per- tains to human sexuality, health, literature and social studies. Family life classes should present homosexuality as a natural and heal- thy form of sexual expression. Information on critical health issues such as AIDS should be presented to all students. Curricula should include values clarification around so- cial roles to increase the respect for in- dividual differences and reduce the stigma attached to gender nonconformity. A variety of gay male and lesbian adult lifestyles should be presented as positive and viable for youth. All youth should learn about prominent les- bians and gay males throughout history. So- cial studies courses should include issues relevant to gay male and lesbian concerns and provide youth with positive gay and les- bian adult role models in our society. Schools need to take responsibility for protecting gay and lesbian youth from abuse by peers and providing them with a safe en- vironment to receive an education. School staff need to receive training on how to work with gay youth and handle conflicts involving gay youth. Teachers should feel secure in being able to rebut homophobic remarks and defend gay youth against harassment. Strong disciplinary actions should be imposed on those who victimize gay and lesbian youth. It is important for schools to hire openly gay male and lesbian teachers to serve as role models and resource people for gay youth. Counseling services that are sensitive to the needs and concerns of gay youth should be available to them. Special educational programs may need to be developed for those youth who cannot be incorporated into exist- ing school settings to ensure that young gay males and lesbians receive an equal educa- tion. Social Support Gay and lesbian youth need access to the same social supports and recreational ac- tivities that other youth have. This would reduce their isolation and enhance their posi- tive social development. Communities need to develop social groups and activities (i.e., dances) specifically for gay and lesbian youth as a way of meeting others like themselves and developing relationship skills. Existing youth programs such as the Boy and Girl Scouts should incorporate gay youth into their activities. Youth programs such as Big Brothers and Sisters should enlist gay and les- bian adults to work with gay youth. It is very important for gay youth to see the potential of a happy and stable lifestyle as adults. Les- bians and gay men need to become more in- volved in supporting gay youth and being positive role models for them. This requires assurance for gay adults that they will not be harassed and accused of "recruiting" youth in doing so. Professional Help Lesbian and gay youth must have access to social services and professional counseling that is sensitive to their needs and able to ad- dress their concerns. This is critical to reduc- ing their risk of suicide. Sexuality is one of the most important issues facing all young 3-135 Report of the Secretary’s Task Force on Youth Suicide people. We need to be open about sexuality and accepting of homosexuality in young people. All social service agencies and men- tal health professionals working with youth need specialized training on homosexuality and issues relevant to gay and lesbian youth. We also need to address issues of suicide and depression in young people. Suicidality needs to be explored with youth who have a gay, lesbian, bisexual or transsexual identity. Problems related to a homosexual orienta- tion should be assessed as a possible reason for suicidal feelings. The goal of treatment should be to assist lesbian and gay youth in developing a positive identity and to support their sexual orientation in the conflicts they face with others. Additional counseling guidelines are provided in Appendix B. Youth agencies need to provide outreach to gay and lesbian youth to make them aware of services and assure them that they are wel- come. Gay youth are often afraid to seek help because of potential negative reactions from others. Programs should hire gay staff that reflect the population of gay youth under their care. Helping professionals should be prepared to offer referrals to gay- identified services and therapists if requested by the youth. It is an accepted premise in so- cial services that individuals have access to programs and staff that reflect their cultural background. This principle is no less true for gay young people who often would prefer to talk about their problems with a lesbian or gay man. Specialized services should be developed for gay and lesbian youth that reflect their par- ticular needs. Health care programs aimed at preventing AIDS and other sexually-trans- mitted diseases need to be directed towards young gay males. Alcohol and substance abuse programs need to target gay and les- bian youth as a population at risk. Pregnan- cy-related services should not assume a heterosexual orientation in young women and be prepared to discuss lesbian concerns. Vocational training and independent living skills programs may need to address special problems gay youth face in becoming self suf- 3-136 ficient and in being incorporated into an adult gay community. Residential Programs The juvenile justice system needs to take responsibility for ensuring that gay and les- bian youth receive fair treatment by the juvenile court and are placed in safe, nurtur- ing, and supportive environments. Special- ized training in working with and understanding gay youth should be provided to foster parents, group home personnel, treatment center staff, and juvenile hall counselors. Gay youth should be incor- porated into placements, whenever possible, where the staff has been taught how to sup- port gays in issues with other residents. It is critical for the juvenile court to show leader- ship in preventing discrimination against gay youth by prohibiting placements that refuse to accept them or that provide them with in- ferior care. The needs of some gay and les- bian youth might best be served in the immediate future by placement in a gay iden- tified foster or group homes. Extremely few such placements presently exist. The juvenile court should facilitate the licensing of gay male and lesbian foster parents along with the development of residential programs specifically for those gay youth who cannot be incorporated into existing place- ments. Research The lack of information about gay and les- bian youth suicide is a reflection of the op- pression of homosexuals by our society and the invisibility of large numbers of gay males and lesbians within the youth population. There is growing awareness that a serious problem exists but we have only started to break down the wall of silence surrounding the issue. Comprehensive research is needed to determine the extent and nature of suicide among young gay males, lesbians, bisexuals, and transsexuals. These studies need to ensure that the entire spectrum of gay youth is adequately represented includ- ing lesbians, homeless youth, and ethnic P.Gibson: Gay Male and Lesbian Youth Suicide minorities. This research can be the founda- tion for greater recognition of the problem and the allocation of resources designed to address it. Hopefully, the work done in recent years will serve as the beginning of the end of suicide among gay and lesbian youth. 3-137 Report of the Secretary’s Task Force on Youth Suicide APPENDIX A RISK FACTORS IN GAY AND LESBIAN YOUTH SUICIDE General Awareness/identification of homosexual orientation at an early age Self acceptance of homosexual orientation Conflicts with others related to homosexual orientation Problems in homosexual relationships Society Discrimination/oppression of homosexuals by society Portrayal of homosexuals as self destructive by society Poor Self Esteem Internalization of image of homosexuals as sick and bad Internalization of image of homosexuals as helpless and self destructive Identity Conflicts Denial of a homosexual orientation Despair in recognition of a homosexual orientation Family Rejection of child due to homosexual orientation Abuse/harassment of child due to homosexual orientation Failure of child to meet parental/social expectation Perceived rejection of child due to homosexual orientation Religion Child’s homosexual orientation seen as incompatible with family religious beliefs Youth feels sinful, condemned to hell due to homosexual orientation School Abuse/harassment of homosexual youth by peers Lack of accurate information about homosexuality Social Isolation Rejection of homosexual youth by friends and peers Social withdrawal of homosexual youth Loneliness and inability to meet others like themselves Substance Abuse Substance use to relieve pain of oppression Substance use to reduce inhibitions on homosexual feelings 3-138 P.Gibson: Gay Male and Lesbian Youth Suicide Professional Help Refusal to accept homosexual orientation of youth Refusal to support homosexual orientation of youth Involuntary treatment to change homosexual orientation of youth Inability to discuss issues related to homosexuality Residential Programs Refusal to accept/support homosexual orientation of youth Isolation of homosexual youth by staff and residents Inability to support homosexual youth in conflicts with residents Relationship Problems Inability to develop relationship skills like heterosexual youth Extreme dependency needs due to prior emotional deprivation Absence of social supports in resolving relationship conflicts Independent Living Lack of support from family Lack of support from adult gay community Involvement with street life AIDS (Acquired Immune Deficiency Syndrome) Unsafe sexual practices Secrecy/unplanned nature of early sexual experiences Future Outlook Despair of life as hard as the present Absence of positive adult gay/lesbian role models 3-139 Report of the Secretary's Task Force on Youth Suicide APPENDIX B COUNSELING GAY AND LESBIAN YOUTH Those of us who work with young people need to be able to identify gay and lesbian youth, accept them for whom they are and support them in resolving their problems. Many of these problems are directly related to their sexual orientation. If we can’t iden- tity these youth, we probably won’t be able to help them. The first step is being able to talk about sexuality concerns with any youth under your care. Sexuality Counseling Don’t be afraid to talk with youth about sexuality issues. You do not incur any liability for doing so. Initial interviews should include questions about the youth’s sexuality just as they include other issues that affect their life (i.e., family, school, substance use, suicide, and depression). It is appropriate to do further sexuality counseling with a young person if you have a good relationship with him/her and necessary if you feel that sexuality conflicts are an important part of the situation. It is good to examine your own attitudes and minimize prejudices so that youth can feel free to convey their feelings and experiences to you. The principle of nonjudgmental therapeutic intervention is especially important in working with gay and lesbian youth. Feel comfortable with your own sexuality in order to keep tensions be- tween you and your client to a minimum. Sexual Orientation Don’t be afraid to ask youth directly about their sexual orientation. Sexual orientation should be routinely included in questions and discussions related to sexuality concerns. Some youth will volunteer the information that they have a gay or lesbian orientation. If you strongly feel that a youth is gay, the only way to find out may be simply to ask. This does not reflect negatively on you, and your intuition is often correct. Even if you are 3-140 wrong, it rarely hurts your rapport with the youth if approached in a sensitive way. If you are unable to broach the subject with them, it is most likely a reflection of your own dis- comfort with the issue. Remember that one of the greatest risk factors in the problem gay youth face is the wall of silence surrounding the subject. The silence needs to be broken if you are to enter the lonely place where many gay and lesbian youth reside. It may be good to let youth know in some way that you accept young people regardless of their sexual orientation before asking them. Be prepared to give youth accurate and positive information about homosexuality. Assure them it is a healthy and positive form of human expression. Gay youth will be listen- ing closely. Acceptance Accept the youth’s sexual orientation as they report it to you. Their sexual identity should be based on the self reporting of their feel- ings and experiences. Do not label a youth as heterosexual or homosexual based on your own assumptions. Assure gay youth it is not sick, bad or wrong for them to be the way they are and that you are not going to try and change them. Let them know you care about them just as much after the disclosure as before. They are used to being rejected by others who find out they are gay. Respect them for being open and honest with you. It was probably hard for them to do and shows that they trust you. Sexual Orientation Confusion Do not assume a youth is confused about their sexual orientation if they identify as gay or bisexual. Many people both gay and straight have trouble accepting that an in- dividual is bisexual. It is important to validate bisexuality as a viable option for youth. However, some youth are genuinely con- fused about their sexual orientation. It is im- portant for them to know that it is alright to be confused. They should not feel pressured P.Gibson: Gay Male and Lesbian Youth Suicide to label themselves one way or another. A useful method in helping them to clarify a confused or undecided orientation is the Kinsey Percentage Scale. This technique al- lows the youth to be any combination of homosexual and heterosexual feelings and experiences that adds up to 100 percent. They can be 85 percent straight and 15 per- cent gay. Or they can be 40 percent straight and 60 percent gay. It is important to let them know you will accept them no matter where they fall on this scale. The purpose of this method is to give youth a context that al- lows them to identify their orientation along a continuum. It is easy to move from here to discussing specific feelings and experiences with them. Gender Identity Assure effeminate young gay males and mas- culine young lesbians that it is alright for them to be that way. Gender nonconformity is common among gay youth and may be a way for them to affirm their identity. Some gay youth, however, become confused by cul- tural stereotypes that insist gay men be like women and lesbians be like men. They feel they actually have to be a person of the op- posite gender in order to be gay. Be prepared to talk with them about their per- ceptions of what it is like being a young gay male or lesbian. Help them to separate so- cial adaptation issues from whether they real- ly believe they are a person of the opposite sex. Assure them they do not have to be any particular way in order to be gay. Transsexual youth will express a persistent desire to be a person of the opposite sex and live as that person over time. They will engage in frequent cross dressing and adapt the name of a person of the opposite sex. It is important for you to accept these youth for who they believe they are and call them by the name they want to be called. This is criti- cal to establishing basic rapport with these youth and effectively addressing their con- cerns. Self Esteem Gay and lesbian youth frequently suffer from low self esteem. They have often received a disproportionate amount of negative atten- tion because of their sexuality. Being gay has been the focus of problems and stigmatiza- tion for them. Assure them there is nothing wrong with being gay and that it is the response of others to homosexuals that is the source of the problem. Help them to develop pride in who they are and a positive identity as a gay male or lesbian. Sometimes they have had too much of their identity focused on their sexuality. It is easy for them to come to see themselves as sexual beings after be- coming known as homosexuals. Assure them that sexuality is only part of who they are. Ex- plore other areas of potential growth that give them a broader understanding of them- selves as individuals. Know the potential of gay youth under your care and work with them in a way that allows them to achieve more success than failure. Give positive feedback whenever possible. Be confident and optimistic of their ability to improve their situation and lead stable and happy lives as gay male and lesbian adults. Family Gay and lesbian youth sometimes mistake their parents inability to accept their sexual orientation as a rejection of them as in- dividuals. Frequently, parents still love their child but need time to come to understand and accept them as gay. Gay youth have trouble recognizing that an initial negative reaction by parents may change in the future. Help families to clarify their feelings for each other and encourage gay youth to be patient in gaining acceptance. Those gay and lesbian youth who have not come out to their parents should not be pressured to do so. Itis a per- sonal decision that they should make careful- ly. Finally, assure gay and lesbian youth that they too will have families as adults. While not the traditional family, their families will be comprised of those friends, lovers, and relatives who remain close with them over a long period of time. Their relationships can 3-141 Report of the Secretary’s Task Force on Youth Suicide be as rich and rewarding as those of other people. Being a gay male or lesbian does not mean that you are going to be alone. REFERENCES 1. Jay, K. and Young, A. The Gay Report: Lesbians and Gay Men Speak Out About Their Sexual Experiences and Lifestyles. New York: Summit, 1977. 2. Bell, A. and Weinberg, M. Homosexualities: A Study of Diversity Among Men and Women. New York: Simon and Schuster, 1978. 3. Centers for Disease Control. Center for Environ- mental Health. Youth Suicide Surveillance Report. Depart- ment of Health and Human Services. Atlanta, 1986. 4. Avicolli, T. Coming Out of the Dark Ages: Social Workers Told of Special Youth Needs. Philadelphia Gay News. May 9, 1986. 5. Remafedi, G. Male Homosexuality: The Adolescent’s Perspective. Adolescent Health Program, University of Minnesota: Unpublished, 1985. 6. Roesler, T. and Deisher, R. Youthful Man Homosexuality. Journal of the American Medical Associa- tion. Feb 21, 1972: 1018-1023. 7. Los Angeles Suicide Prevention Center. Problems of Suicide Among Lesbian and Gay Adolescents. Prelimi- nary Data. Los Angeles: Unpublished, 1986. 8. Larkin Street Youth Center. Client Statistics. San Francisco, CA. 1984. 9. Huckleberry House. Client Statistics. San Francis- co, CA. 1982, 10. National Gay Task Force. Anti-Gay/Lesbian Vic- timization. New York, 1984. 11. San Francisco Juvenile Justice Commission. Problems for Gay and Lesbian Youth Involved With the Juvenile Court. San Francisco, 1982. 12. Wilson, H. Personal Interview. Community Activist and Cofounder of the Gay and Lesbian Teachers Coalition. San Francisco, 1986. 13. Rofes, E. | Thought People Like That Killed Them- selves: Lesbians, Gay Men and Suicide. San Francisco: Grey Fox, 1983. 14. The Bronski Beat. The song "Smalltown Boy" from the album "Age of Consent." MCA Records, 1984. 15. Bell, A.,, Weinberg, M. and Hammersmith, S. Sexual Preference: Its Development in Men and Women. Bloomington, Indiana: Indiana University Press, 1981. 16. National Gay Task Force. Twenty Questions About Homosexuality. New York, 1978. 17. Kinsey, A., Pomeroy, W. and Martin, C. Sexual Be- havior in the Human Male. Philadelphia: Saunders, 1948. 18. Kinsey, A., Pomeroy, W., Martin, C. and Gebhard, P. Sexual Behavior in the Human Female. Philadelphia: Saunders, 1953. 19. Parris, F. Some Die Young. Washington Blade. Washington, D.C., May 17, 1985. 20. Morin, S. and Miller, J. On Fostering Postive Iden- tity in Gay Men: Some Developmental Issues. San Fran- cisco: Unpublished, 1977. 21. Gibson, P. Developing Services to Gay and Les- bian Youth. In Saunssing Lesbian and Gay Male Youth: Their Special Needs/Special Lives." Ed. by Bergstrom, S. and Cruz, L. National Network of Runaway and Youth Ser- vices, Inc., 1983. 22. Lewis, L. The Coming-Out Process for Lesbians: Sgreing a Stable Identity. Social Work, Sept-Oct:464- 69, 1 i 3-142 23. Fourteen Year Old Gay Male. Personal Interview. Berkeley, CA., 1986. 24. Martin, A. Learning to Hide: The Socialization of the Gay Adolescent. Adolescent Psychiatry, 10:52-65, 1982. 25. American Psychiatric Association. The Diagnostic and Statistical Manual of Mental Disorders. 3rd Ed. Washington, D.C.: American Psychiatric Association, 1980. 26. Fricke, A. Reflections of a Rock Lobster: A Story About Growing Up Gay. Boston: Alyson, 1981. 27. Weinberg, M. and Williams, C. Male Homosexuals: Their Problems and Adaptations. New York: Oxford, 1974. 28. Geron, A. ed. One Teenager in Ten. Writings by Gay and Lesbian Youth, Boston: Alyson, 1983 29. Hughes, R. Personal Interview. Program Coor- dinator for Gay and Lesbian Youth Services. Center for Special Problems. San Francisco, Ca. 1986. 30. Davis, R. Black Suicide in the Seventies: Current Trends. Suicide and Life Threatening Behavior 9:3, 1979. 31. Hendin, H. Black Suicide. Archives of General Psychiatry, 2:4, 1969. 32. Dutton, T. Nigger in the Woodpile. Fag Rag. Bos- ton, c. 1977. 33. Jones, A. The Need for Cultural Sensitivity in Work- ing With Third World Lesbian and Gay Youth. Counseling Lesbian and Gay Male Youth: their Special Lives/Special Needs. Ed. by Bergstrom, S. and Cruz, L. National Network of Runaway and Youth Services, 1983. 34. Huxdly, J. and Brandon, S. Partnership in Transsexualism, Part I: Paired and Non-paired Groups. Ar- chives of Sexual Behavior, 10:133-141, 1981. 35. Harry, J. Adolescent Suicide and Sexual Identi Issues. Submitted to the National Institute of Mental Healt for the Secretary's Conference on Adolescent Suicide. Washington, DC May 8-9, 1986. 36. Heller, J. Suicide and Sexual Issues. Suicide Prevention Center, Inc. (Place Unknown) 1983. 37. Gock, T. Suicidal Homosexual Theory as a Case of Anti-Gay/Lesbian Violence. Paper presented at American Public Health Association Meeting (112th). Anaheim, CA, 1984. 38. National Gay Task Force. Gay Rights in the United States and Canada. New York, 1982. 39. Wandrei, K. Sexual Orientation and Female Suicide Attempters. Oakland: Unpublished, 1985. 40. Suicide at Sixteen. Newswest, Jan. 20-Feb. 3, 1977. 41. Tartagni, D. Counseling Gays in a School Setting. School Counselor 26:26-32, 1978. 42. Morrison, M. Adolescence and Vulnerability to Chemical Dependence. Unpublished, 1985. 43. Adolescent Male Prostitution. Urban and Rural Sys- tems Associates. U.S. Department of Health and Human Services. San Francisco: 1982. 44. Kassler, J. Gay Men's Health. New York: Harper and Row, 1983. 45. Planned Parenthood. Teenage Sexuality Fact Sheet. San Francisco, 1985. ISSUES FOR SURVIVORS Curtis Mitchell, Delane, Florida SUMMARY The harvest of dead youth is only a small part of the total damage caused by the current crisis in teenage suicide. For each teenager who dies, a phalanx of survivors is suddenly burdened by devastating reactions. Their lifestyles are altered, their futures scuttled, their minds and bodies attacked by infir- mities. Many of them are so shaken that they require professional help. Unfortunately, the plight of survivors has been almost totally unrecognized. Only a few observers have studied their needs and formulated suggestions for their rehabilita- tion. Their books on the subject offer useful insights and protocols for regaining a semblance of normalcy. Successful therapy depends largely on what happens during what is now called the grief recovery process. A survivor himself, the author of this paper describes the steps taken by his family. Each step on the route to peace of mind, he warns, can also involve an emotional booby-trap that can defeat the entire process of libera- tion; it can also doom a griever to perpetual sorrow. Phases presenting unique difficul- ties include shock, denial, bargaining, anger, and guilt. Special danger exists, the author says, in the average survivor’s confrontation with stigma. Regrettably, he relates, the dis- covery that the rights of female grievers are frequently denied under the guise of chival- rous "protection." This well-meant denial can sometimes leave a life-long trauma. It must be corrected, and it can be corrected when women survivors, despite their grief, assert their rights and challenge their mas- culine care-givers. Under some circumstances, survivors require special help. Sources to be explored are named. So are procedures for self-help that may dissolve persistent grief fragments. Finally, aspects of community behavior are challenged. If these suggestions were fol- lowed, thousands of grievers who have been incapacitated might regain their lapsed roles as worthwhile and useful human beings. INTRODUCTION A strange new blight afflicts our earth. Its distemper infects thousands, perhaps mil- lions of our citizens. Neither virus nor isotope distributes its mischief. Its pathogens are man-made and are trans- mitted by bullets and drugs and broken hearts. Two examples. Ken and Mary Whitman were happy Floridians using their combined salaries to put their three sons through college. Jock, the oldest, came home during spring break. One night, the Whitmans returned to their home following a late meeting and found his body swinging from a backyard tree. Mary charged Ken, the father, with applying too much harsh discipline. He charged her with coddling the boy and countermanding his own orders. The two surviving sons took sides, one for each parent. The Whitmans 3-143 Report of the Secretary’s Task Force on Youth Suicide separated and fought bitterly over their divorce. Within months, that family had dis- solved. On the Pacific coast, Joan and Tom Miller ar- ranged for their son and daughter to help celebrate their 20th wedding anniversary. On the afternoon of the party, daughter Bar- bara arrived home late from her job. Passing the garage, she smelled gasoline fumes. Her mother’s lifeless body was sprawled over the steering wheel of her car. No suicide note could be found, nor any reason for the act. Family members became unhinged. Tom’s social drinking exploded into alcoholism. Barbara became convinced she could have saved her mother’s life if she had come home on time. Son Jim, 12, felt shut out and rejected. Acting out his rage, he stole a bike, destroyed school property, and was expelled. The State sent him to a foster home. Barbara attempted suicide, botched it, and was con- signed to a treatment center. The father refused help for his alcoholism and was fired. The doomed family lasted less than a year. Few observers will deny that an ancient and intractable malady seems to possess our spirits. Its infection spares neither nations nor continents. Its victims are a mob too large to count or comprehend. Their ail- ments fill a pharmacopoeia. Their depres- sion spins off a spray of shattered families, bankrupted careers, and poisoned dreams. The germ that spreads the plague is called suicide. The men, women, and children left behind are called survivors, and they are in trouble. So what else is new? Well, it’s new that we are finally aware that our young people are committing suicide at an unprecedented rate. And that every one of those deaths sends up a mushroom cloud that drops a fallout of pain, guilt, and depres- sion on the uncountable survivors left be- hind. Daily, the media trumpet a warning. Famous lecturer-on-love, Leo Buscaglia, says "In the United States 14 people between the ages of 3-144 15 and 24 kill themselves every day." That’s about 5,000 per year. The suicide rate for this age group has tripled over only three decades. Worse, for every death reported, there are an estimated 50 to 100 attempts. Dr. Art Ulene, the TV pundit, says, "Nearly 60 Americans take their lives every day and ten times that number attempt suicide." Paul Harvey, commentator and columnist, says that 13 school children are murdering themselves daily. Newspaper stories confirm their tallies. In Woods County, Wisconsin, five young people shot themselves to death last winter. Ten others in the same community made the attempt and failed. In the suburbs north of New York City, 17 suicides were recorded in a brief time span. Clusters of death have erupted among teenagers in and near Plano, Texas, on Chicago’s North Shore, in Lincoln, Nebraska, and in California. Neither sex is spared. More girls than boys attempt by a ratio of four to one. More boys than girls suc- ceed by a ratio of about four to one. One asks: are such figures accurate? The answer is no. No person on earth knows how many suicides take place in America. The figures are collected using an archaic system that is corrupted by indifference and cronyism. Assigned officials include not only coroners, physicians, and medical examiners, but in some places, deputy sheriffs, mor- ticians, and justices of the peace. Under- reporting is assured in many districts by the refusal of back-patting appointees to pin the stigma of suicide-in-the-family on a friend, a leading citizen, or an elected politician. The result is a shambles that has been called "the great cover up." Covered up also is the fact that our country has become the killing ground not only for any army of young people who are sick of life, but also of a much greater mass of survivors whose roll call mounts an- nually into the hundreds of thousands, per- haps into the millions. Who are they? We can tick them off on our fingers beginning at the family core and C.Mitchell: Issues for Survivors counting the husbands, wives, parents, children, brothers, sisters, grandparents, un- cles and aunts. Next, count their pastors, coaches, doctors, counselors, teachers, employers. Last, there are schoolmates, clubmates, and even filling stations atten- dants. What do those survivors want from life? Researchers have studied them and named their stations of the cross. Their journey is called the grief process. Its elements include shock, denial, bargaining, stigma, contamina- tion, isolation, fear of insanity, guilt, anger, rejection, ostracism, acceptance, and finally, deliverance. Their problem is that they don’t know what hit them. Nor has their life experience equipped them to cope well enough to stablilize their lives. In many ways, social workers report, they act and think like young babies. In our culture, society assumes a respon- sibility for babies. But where does one start? Our courts of law and the practice of jour- nalism offers one formula. Why not examine some witnesses? An investigator to whom all professionals are indebted is Edwin Shneidman, formerly co- founder and co-director of the Los Angeles Suicide Prevention Center, now located at the University of California. In one of his books, Voices of Death, he suggested four in- itiatives that can help grief-stricken sur- vivors. First, refuse to accept the stigma of suicide. Second, rid yourself of the notion that you, the survivor, might suffer a similar fate. Third, free your daytime thoughts and your night-time dreams of images of blood and violence. Fourth, restrain any obsession to learn the whys, the whats, and the if onlys. To assist the recovery process, Shneidman published an essay entitled "Care of the Bereaved." Here is his advice, paraphrased. A counselor should begin to work with a survivor-victim as soon as possible. Be aware that your survivor client will probably welcome a chance to talk with a pro. Expect to encounter powerful negative emotions such as irritation, anger, envy, shame, guilt, and the like. All of them must be ventilated. Obtain a medical examination by a physician. It will give you a baseline for checking developments. Reject the temptation to act as a voice of conscience and offer instead, the soft voice of reason. Avoid all banalities. Expect a slow recovery punctuated by setbacks. Expect working through the process to take a long time. Healing is rarely achieved in less than a year. It may extend to the grave. Insist on a program of health care includ- ing suicide prevention, intervention if needed, and postvention as a safeguard. Since Shneidman’s first publication, the world has adopted many of his recommenda- tions and the lot of survivors has improved. Hundreds of volumes now offer programs to prevent suicide. Only a handful are con- cerned about survivors. The latter are all worth reading, particularly a small volume titled Suicide by Jacques Charon, the American philosopher. His historical chap- ters provide valuable background, and his abstracts of what some of the world’s greatest minds have thought of suicide are inspiring. Or pick up Suicide and Grief by Howard W. Stone, a professional counselor. When he discusses pastoral care, his words are golden. His battle plan says: Keep active. Join group activities such as club, church, or charity. 3-145 Report of the Secretary’s Task Force on Youth Suicide Work hard at creating new friends. Bring into the open whatever hostility still festers. Rediscover hope. Find a meaning for your life. Living When a Loved One Has Died by Earl Grollman has earned its right to be called a classic. The Ultimate Loss by Joan Bordow reports the death of a child. A survivor herself, she calls it the most devastating of all deaths, and "an affront to our attempt at immortality as well as our sense of fair play." The Morning After Death by L.D. Johnson tells of a daughter’s death by accident on an icy highway. Dr. Johnson, professor of religion at Furman University, has given us the story of a man’s triumphant pilgrimage through tragedy. Don’t miss the chapter called "The Nature and Uses of Death." I liked what I found in Suicide Assessment and Intervention, a book of essays. Its fifth chapter by Barbara Bell Foglia directs atten- tion to bereaved children. They have never known death and cannot understand it, she says. So when it happens, include them in family conferences. Exclusion makes them feel unworthy. Above all, do not lie. Daddy has not gone on a long, long journey. Mommy has not been called to be with God. Dead is dead is dead. A motif that emerges from all these writings is a clear understanding of a survivor's suffer- ing. Iris Bolton writes of its many faces in her book, My Son...My Son... when she quotes her clients as insisting: "I am going crazy." "I cannot live without my man." "Nobody feels my anguish like I do, so how can anyone else understand my despair?” "A doctor told me that suicide can be in- herited and it frightens me that another person in my family may choose to die, 3-146 even me." "People think we're a bad family, that I'm a bad mother. How can I face anyone again?" "I can’t stop reliving the moment I found the body and seeing the blood." "If I let go, I'll explode.” But the record shows that unbearable pain can become bearable and that hopelessness can be converted into hope. I've seen it hap- pen. I'saw it happen when Curtis Mitchell Bolton, age 20, my grandson, and the son of Iris and Jack Bolton, shot and killed himself. It takes only a split second to create a survivor, a fami- ly of survivors, or a town full of survivors. I shall not recite the bad aftermath. I shall name some of the good things because they became the solid rock on which we built recovery. The first good thing was advice offered to the Boltons by a friend of the family who was also a psychiatrist. As Iris describes it, "That first visit, he took my husband and me and our three sons into a private room. His gaze lock- ed my eyes to his. "You will survive,” he declared firmly. You will survive, if you choose to do so.™ Next he advised that all important decisions be made by the whole family. Huddling together, conferring together during those first days, they discovered comfort in consen- sus. His third injunction was to look for some good to emerge from the horror of the mo- ment. "Seek it," he urged. I thought, "What an absurdity." Nevertheless, we began to search. Permit me to pause long enough to indulge in a personal reminiscence. When my grandson shot himself and when I first be- came the kind of strangulated survivor who usually emerges from this kind of experience, I endured a quality of pain previously un- known to me. It spread through the family, sparing no one. Because of it, | became a stu- C.Mitchell: Issues for Survivors dent of pain and of the unique agony that is so often the result of a suicide. What I am reporting now extends far beyond my own ex- perience. Indeed, I am indebted to many sur- vivors and owe special thanks to the members of support groups and to the expressions of members of the caring organization called Compassionate Friends. I must add this caveat. I can disclose no magic formula to anesthetize human suffering. No rule of thumb applies. The grief process is said to be a series of steps. These steps do not march in single file. Often, they behave like unruly children on a school bus. They change places. They roar and they whisper. In the midst of their grief, some survivors be- come so confused they literally cannot recall their ABCs. My own experience was a sort of free-fall into a bottomless black hole. My concern for survivors is that some do not plunge as I did. Instead, they soar, some of them into a booby trap. Think of grief as a minefield strewn with explosives. Survivors are not ordinarily equipped with mine detec- tors. Nor are many consultants. Some of the latter say, "Just take your pills. You'll be all right." Don’t believe it. Consider this situation. A family has lived for months or years in misery, enslaved by the al- coholic tyranny of a father. His suicide sud- denly liberates them. They think, "At last, I don’t have to submit to abuse and humilia- tion." But second thoughts occur. Our moral code asserts that nobody should feel good be- cause of another’s death. So guilt takes over. Without help, it may last for years, through additional decades of enslavement. Grief’s second step is usually called denial. Defending itself from shock, the human psyche rejects reality and tries to find arefuge in fantasy. The experience is like being tossed about in a cement mixer. Our orderly world is overturned. So are our most valued precepts. We have believed that a religious faith and a loving heart guaranteed a good life. The bitter fact is that life is no longer good. A dear one’s death has scuttled both our faith and future. So the ego tries to escape the pain by deny- ing the suicide. Circumstances are invented to "prove" the accident. Newspaper stories which repeat one’s fancies authenticate their veracity. This is another booby trap. Deep down, one knows that the story is a lie and that pretending otherwise involves a lifetime of lying. What such a survivor does not know is that keeping a secret--any secret--absorbs a prodigious amount of energy. Lack of energy depletes the body and mind for as long as one lives. Presently, fear invades one’s tis- sues, fear that the lie will be discovered. Close behind walks the monster called guilt. Ultimately, comes death. Next, make way for anger. Let’s face it. Anger elbows its way into grief in the majority of suicides. The deed is so surreptitious and so unfair. The brain protests, "Why wasn’t 1 told that his staying alive had become un- bearable? Now I'll never know." Or "Surely I could have been given an hour to argue against his decision to die." Or "It’s so unfair that I'm faced with a future alone, and without even a chance to plan." Yes, anger floods the mind, washing over all those near and dear. Nor does God escape. Which leads to another kind of ambush. Many survivors begin to play the game of "Who's to blame?" When an accident hap- pens, we Americans are the world’s fastest finger-pointers. "Who Done It" is a literary game. After a suicide, the question can be- come an obsession. A survivor usually learns the hard way that placing blame rarely solves a problem. To help overcome this hazard, a counselor requires steady thinking and affec- tionate guidance. In today’s world, both qualities are in short supply. Guilt provides another ambush. And guilt is almost inevitable. It comes bubbling out of one’s collection of If Onlys: if only I had been a better friend or parent, if only I had spent more time at home, if only I had known what I now know about depression. Teeming with a host of such questions, in some of us, the mind turns to What Ifs. 3-147 Report of the Secretary's Task Force on Youth Suicide What if I had shown how much love I could give? What if I had backed off instead of forcing a showdown? What if I had flushed those extra pills down the john, or buried that pistol, or burned that rope. William A. Miller wrote a useful book a few years ago called When Going to Pieces Holds You Together. He tells of a married couple who saved their money all their lives for a round-the-world trip. At age 65, he retires and they buy their tickets. They go to the air- port and check in. Their plane is announced, he suffers a heart attack, and dies. Miller tells us that the wife lapsed at once into the If Only ritual, mumbling over and over: "If only we’d never thought of this trip." "If only I'd got him to the hospital sooner." Another case. A father had twin sons of whom he was very proud. During a walk in the woods, the boys strayed into a pond and were drowned. The father sobbed, "If only I'd taught them to swim." He turned to al- cohol for 12 miserable years, repeating daily his self-accusation. Finally, he killed himself. Guilt spreads like a cancer, destroying self- esteem and sapping one’s energy. Thus weakened, many minds think, "If suicide was the only way out for my loved one, then it was because I failed him. So I'm responsible for his death and I must be a terrible person." The thought persists, nourished by apathy, and another survivor is ambushed. But there’s more. Psychologists tell me that guilt often turns to shame, and that shame is a consequence of a perceived failure. Failure demeans the ego. The ego, fighting back however it can, sometimes suppresses the idea of failure and stores it in the mind’s back alleys where it decays and stinks for a lifetime. Some of us turn belligerent and aggressive. Some of us seek revenge. Some of us hit the bottle. 3-148 Not a pretty picture, is it? Nor is it what the person who committed suicide wanted. His quest in most cases was simply for peace or surcease of pain. What I am saying is that each step of the grief process usually presents a potential pitfall. Survivors are dumped without warning into an emotional jungle. At first, they are like marionettes, reacting to the pull of forces they do not understand. They respond from the gut, as you counselors know. Those responses are really feelings. Let me repeat what some survivors said: "I feel utterly lost without the presence of the person who has left me." "I need the presence of someone to whom I can give my love." "My memory is missing. I've lost my brain." "I’m obsessed with that moment when I discovered the suicide and I relive it daily through all my five senses." "Driving my car, I must think carefully through each movement of my foot or hand. If I don’t, I may forget to apply the brake or to turn the steering wheel when I drive into my garage." "I can’t believe that I am alone. Once, I saw my wife standing at the foot of my bed." "When I catch myself laughing at a friend’s joke, I feel guilty." "People treat me as if I'm contagious." "I just want to cut and run." Can anyone doubt that many survivors need help?" But there’s more. In examining the grief process, let’s look at what is perhaps the stickiest wicket of them all. Its name is stig- ma. Stigma was once defined among the ancients as a sear left by a hot iron. I like that! Can you think of a hotter iron than suicide? Our modern definition says stigma is a stain or a C.Mitchell: Issues for Survivors mark of shame. And I'll buy that. For suicide surely leaves a stain, a judgment which is ap- plauded and endorsed by our society as an unalterable moral fact. Society’s attitude, I believe, is a sorry commentary on what we call Christian civilization. Trace it back a few thousand years. Mankind’s primeval fear of evil spirits and voodoo gods gave it a start. It became a so- cial monster about 400 years after the death of Christ when Saint Augustine, the Bishop of Hippo, proscribed suicide because of God’s sixth commandment which said, "Thou shalt not kill." Thus, suicide became a sin. What does its stigma do? To put it bluntly, it paints innocent bystand- ers with tar and feathers. Ifkilling is sin, then killing the self produces a sinner. For cen- turies now, the act has smirched, smeared, slurred and shamed millions of persons whose worst sin was loving another whose al- location of misery was too great to bear. What does it do? The telephone stops ringing. Invitations to parties, birthday celebrations, Christmas tree burnings, fetes which one has attended for years, no longer crowd one’s mailbox. Social gatherings that are attended become stressful because old friends steer small talk away from the mention of death or the name of the dear one who has died. Walking down the street, you see an old friend approaching. Your heart leaps with pleasure. The friend sees you and crosses to the other side. You know she is pretending. It happens. This ostracism leads to isolation. One sur- vivor reported, "I feel as if I'd been quaran- tined." Isolation turns thoughts inward, to apathy, to illness. Sometimes, it persuades one to build a shrine to one’s lost love. Have you seen them? Bedrooms where every col- lege pennant, every poster is in the exact spot where it was tacked a decade earlier. Or a dresser top bearing a half-smoked pipe, a golf ball, and enough change for bus fare. Or a cradle holding a doll. Often, the damage runs deeper, even into mental decline, a result of the absence of life- supporting relationships. Quickly, I must catalog a handful of addition- al hazards that threaten recovery. For in- stance, the "I'm Going Crazy" syndrome. It strikes early. Your world blurs as if you are submerged. Life wobbles along in slow mo- tion. Making a decision becomes difficult or impossible. Reaching the end of your tether, you conclude, "I'm going crazy." But you are not. Your are merely living through the pain- ful process of recovery. If you are a woman, almost certainly you will be afflicted with over-protection. Well- meaning and thick-headed males may take charge. Pastors and police are usually males. So are physicians, morticians, and cemetery lot salesmen. They want to be helpful but al- most without exception they impose their own code of chivalry on mothers, wives, and daughters. One issue is a women’s right to view the body of her loved one. Most males are biased to the contrary. Women are weak, unfitted for horrid sights. So they "protect" her. The men of the family are escorted to the mortuary. Not the woman. Recently, I read of a young mother whose need for a last look was denied by her menfolk. On the day of the funeral, she sat like a mad woman hug- ging the coffin in her arms until she was dragged away. Was her child really inside? We can be certain the question will haunt her all the days of her life. All this must be changed. But only feminine survivors can change it. To take a last loving look is their right. The change is coming and I hope soon. Another obstacle to deliverance from grief is what might be called Dangerous Days. Why dangerous? Because they bring to mind all the good things of one’s past life. In a nor- mal life, the best days are usually Mother’s day, Christmas, birthdays, and anniversaries. In one’s post-suicide life, one learns to avoid them. One of the worst days is the first an- 3-149 Report of the Secretary's Task Force on Youth Suicide niversary. The mind says, "Exactly one year ago, it happened. Right here in this room. I remember every detail." And the pain returns. But gradually one learns to dodge and to cope, and then you hate yourself for diminishing the goodness of the life you mourn. Occasionally, that self-hate thrusts you back into the abyss from which you have climbed and you feel doomed forever to eternal dam- nation. This too will pass. My daughter took the agony of her relapse to the same friend and guide who had counseled her in the beginning. Her attack was normal, he as- sured her. Almost certainly, it had originated in a remnant of her year-old grief. A rem- nant, he reminded her, was a fragment of something left over. She understood and her deliverance came quickly. We have not discussed how long deliverance from grief takes. Sometimes it lasts so long that it becomes an issue. Even old friends tend to magnify its length and to resent its persistence. A flesh wound usually heals quickly. A broken bone takes a bit longer. But a broken heart is bound to no time span. Impatient friends offer encouraging advice. "Let’s get it over, fella! We've got a job to do." Or "Come on kiddo! Stop your crying and pick up the pieces." Lucky, indeed, is the survivor who has a friend who understands that everyone recuperates in a different way and at a dif- ferent speed. Lucky, indeed, is the friend who can help a survivor to reach the peace- ful plateau of understanding that one never gets over such a loss but that presently one will learn to live with it. Along the way, many survivors wonder if they need help. For most, the answer is yes. Help may come from a professional or from a suicide survivor group. Hundreds of the lat- ter exist. Ask your doctor or your pastor. If you prefer a one-on-one relationship, ex- perts often recommend a trial run of three consecutive visits. Get to know each other, what goals he sets, what improvements he an- ticipates. If you prefer the caring com- 3-150 panionship of persons who have escaped from their own personal black hole, find a survivor group and listen to it. If you are on the same wavelength, there is no better therapy. The point I belabor is this: the casualties resulting from the suicide of a single teenager always extends far beyond his imagining. We know little about the path that he has chosen to follow but we know for certain that the ter- ritory he leaves behind is no Fun City. Earlier, I intimated that survivors have been neglected long enough. And I asked what services do they need most. My experience tells me that they need to know how others have felt in the same situation. Specifically, they need to be forewarned (as does the whole public) of the storms others have sur- vived and how able counselors have helped them to withstand the thunder and lightning. A perfect guidebook will never be written for every client. But our modern world which mobilizes religion, psychology, and medicine to make us more comfortable offers so many choices. Somewhere help is available. Often, it is within one’s own body. Patricia Sun, a California lecturer and healer, says this: "Every time you feel pain, every time you feel despair, every time you feel loss, every time you feel fear ... it is always your cutting edge." Some survivors who learn about the grief process feel cheated unless they are con- ducted through every emotional swamp and sinkhole. They must learn that no ladder will ever help them to escape. A grieving youth who lost his mother spent months bouncing around until his life settled down. He reported, "You build your own grief process and you build your own recovery. It’s not right or wrong or good or bad. It just is." Once one emerges, one becomes a member of a special brotherhood. Sascha Wagner, herself a survivor, describes them: "Some of us get desperate and we don’t make it through. Others manage to make it and there is a specialness to such persons. They have an inheritance and they kind of become C.Mitchell: Issues for Survivors a kind of living memorial to their dead children. Those who survive--that’s us-- make the difference that keeps the trees growing." So what do we want? What beyond mere sur- vival? Here is my wish list: That we not be castigated by the stigma of a suicide that we neither solicited nor ap- proved. That our loved ones, now dead, not be con- signed to oblivion through talk that avoids their existence. That our productivity as good citizens be ac- cepted socially politically, and industrially. That the patience of our peers will allow us, each in his own way, to gain the strength we need to balance and carry our life-long bur- dens. That in due time we may be surprised by joy. Finally, I would beg for survivor research that will help educate families in the grief process. ¢ Teach caregivers, educators, pastors and teenagers that liberation begins when isolation and depression and resentment are guided into productive channels. ¢ Teach that the swiftest healing comes to those who seek a blessing amid the debris of their messed up lives. * Teach that suicide is no disgrace when it is chosen to meet a positive need, or even to challenge an incredible fantasy. Surely, we live in a time of change for the bet- ter. Count the once-forbidden topics that have emerged from centuries of silence during our own generation. One can name women’s rights, minority rights, abused children’s rights, even Gay rights. It is time, at last, for survivor’s rights. REFERENCES Bordow, Joan. The Ultimate Loss. New York; Beaufort Books, 1944. Charon, Jacques. Suicide. New York; Charles Scribner's Sons. 1972. Grollman, Earl. Living When a Loved One Has Died. Bos- ton; Beacon Press, 1980. Hatton, Corrine Loing. Suicide Assessment and Interven- tion. New York; Appleton-Century-Crofts, 1977. Johnson, L.D. The Morning After Death. Nashville; Broad- man Press, 1978. Maris, Ronald W. Pathways to Suicide. Baltimore; Johns Hopkins University Press, 1981. Miller, William A. When Going to Pieces Holds You Together. Augsburg Press, 1976. Scarf, Maggie. Unfinished Business. New York; Doubleday and Company, 1980 Shneidman, Edwin S. Voices of Death. New York; Harper and Row, 1980. Stone, Howard W. Suicide and Grief. Philadelphia; Fortress Press, 1972. ACKNOWLEDGMENTS The author desires to acknowledge the assis- tance of The Compassionate Friends and of many suicide survivors’ support groups. Address correspondence or requests to the author at: 2679 Whitehurst Road, DeLand, Florida 32720. 3-151 PREVENTION OF ADOLESCENT SUICIDE AMONG AMERICAN INDIAN AND ALASKAN NATIVE PEOPLES James W. Thompson, M.D., M.P.H., Research Psychiatrist, Division of Biometry and Applied Sciences, National Institute of Mental Health, Rockville, Maryland INTRODUCTION In 1985 there was yet another round of media attention to the phenomenon of adolescent suicide in American Indian and Alaskan Na- tive (hereafter, "Indian") communities. This has been a long-standing problem, and a topic which has been periodically "viewed with alarm" by the media. It is also a subset + of a concern over similar recent trends in the majority population. Suicide, in combination with high rates for other self-destructive be- ' haviors (e.g., homicide, accidents, substance abuse, unwed motherhood), has meant an ongoing destruction of Indian communities. It is appropriate, then, to talk about preven- tion of these self-destructive behaviors. Indian suicide is a topic which has been peri- odically written about in the professional literature. I will not attempt to review that literature, or to repeat the work of such re- searchers as Shore (1) or Beiser (2) in epidemiology, or Berlin (3,4) Levy and Kunitz (5), and Ward (6) in suicide preven- tion. Rather, I will make some general com- ments about adolescent suicide among Indian people, its epidemiology, and ap- proaches to its prevention. Some of these points have been made previously by the above and by other authors, but is important to review them. Data Problems There is an a priori assumption, often made inthe press (and sometimes in the literature), 3-152 of an "epidemic" of suicide among adolescent Indians. Despite the importance of sound epidemiologic data to define whether indeed an "epidemic" exists, only a few articles have used epidemiologic methods to define the ex- tent and shape of the problem of suicide among Indian adolescents. Such work is necessary as a basis for appropriate clinical and community action. Unfortunately, however, studies in the area of Indian suicide present a multitude of difficulties. One problem is that suicide is a relatively rare event, especially in small populations such as Indian tribes. This is of concern because a very small change in raw numbers can look very large in terms of rates. These small changes (which result in large changes in rates) may be due to real changes in the prevalence in suicide, but can also result from problems in reporting suicide deaths. For ex- ample, different individuals from time to time may report and record suicide deaths in a community or county. They may or may not correctly identify Indians as Indians on death certificates. In addition, they bring to the process different attitudes about suicide, which affect their definition of suicide and their willingness to report it. Indian com- munities themselves may be reluctant to identify suicides as such for fear of adverse publicity, and this reluctance may wax and wane. With regard to "epidemics" among Indian AD \ ( \) \ ¢/ Jo 5 N J.W. Thompson: Prevention. . Among American Indian. . . people per se, it is not clear that suicide rates are higher in Indian communities than for the surrounding areas. To evaluate this, it is im- portant to use contemporaneous data for comparison from the same general area as the Indian data before deciding that a problem is specifically an Indian problem. ne frequent mistake is to compare the total | | U.S. suicide rate with the local Indian rate, or | | to compare both the total U.S. Indian and — non-Indian rates. Using data from one or a few tribes to generalize to all Indian people presents another difficulty. There are, depending on the definition used, more than 400 identifi- able tribal groups in the United States, which , are quite diverse with regard to size and cul- Hf . . ‘| ture. To assume that suicide rates, reasons x" | for suicide, and potential solutions to suicide among tribes are identical, or even similar, is | a serious mistake. For example, there are large differences in rates and patterns of suicide between Navajo and Hopi, the latter of whom is completely surrounded by the former. Such differences are easily "swamped" in reporting figures for large areas. Related to the problem of tribal diversity is the paucity of the data on non-reservation In- _dians. Less than half of Indians live on reser- vations; the remainder live in urban or rural communities, often far from reservations. It would be a serious mistake to assume that the nature and extent of the problem is the same for all of these communities. Studies that at- tempt to group these communities together may present an erroneous picture. Finally, there is a problem of what data are included by the various sources to determine suicide rates. For example, the Indian Health Service (IHS) areas have changed over time as have the definitions of an IHS "reservation state". Using IHS data to look at suicide trends, therefore, may present a problem. Two Studies Two examples of attempts to deal with the data problems follow. Pam Thurman sought data on suicide in Cherokee County, Oklahoma and also talked with people who could fill the gaps in the data. (She herself is a member of the com- munity, which allowed her access to such in- formal data sources.) She found that the suicide rate among Cherokees was not dif- ferent from the white suicide rate in the county (7). In the second study, Levy and Kunitz found that suicide rates varied widely within the . Hopi tribe, that suicides were often clustered within particular families, and that the Hopi rates rose and fell with that of other rural counties of Arizona (8). To understand the problem of adolescent In- dian suicide better, we clearly need further epidemiologic and services research. We need to look more closely at small, homogeneous areas, using appropriate com- parison groups (e.g., local rates for non-In- dians). We need better methods for collecting vital statistics including working closely with the tribes in a way that doesn’t penalize them for reporting suicides, e.g., providing the press with another sensational story. In addition, data need to be inter- preted in light of the culture of the com- munity and in light of its rural or urban character. With regard to services, research is needed on how the presence or absence of particular services, treatment personnel, or treatment facilities affects suicide rates. Finally, what "protects" some communities against suicide? This can be as revealing as learning about the communities with high rates. Here, we could and should study the similarities and differences between areas with high or low suicide rates. Strategies for Improving the Situation The first consideration in dealing with a problem is to be sure that it really is a problem. Suicide and other self-destructive 3-153 a = Report of the Secretary's Task Force on Youth Suicide behaviors are always found in a population, and even one case is cause for concern. There also may be peaks from time to time in the baseline rates of suicide in a community, just as there are in diseases and other social maladies, and these certainly constitute a reason for specific response in that com- munity. If an "epidemic" does occur, a more general response may be appropriate. Before responding, however, epidemiologic data should be obtained using methods which are congruent with the particular situation. To rely on news reports of a high suicide rate, lor on national aggregate statistics as a ration- ale to move into a particular community or | group of communities with large-scale plans to "fix" the problem may be grossly inap- I\ propriate, a waste of resources, and damag- | ing to the community. | When going into a community, whether to collect data or mount a program, it is ab- solutely necessary to work with community representatives initially and throughout the project and to conform with the culture and beliefs of the particular community. This is true in any community, but especially in In- dian communities, for Indian people have learned from long experience that when the white man wants to study them or help them with a new program, the outcome stands a very good chance of being negative. Finally, a solution in one community may be totally ineffective and inappropriate in another | community, even if the communities are ! nearby or seem similar in many ways. ( Itis also very important to understand suicide as one of several self-destructive behaviors which are "end-stage" behaviors. It is seldom that someone dies a self-inflicted death without previous events or conditions which led to the final behavior. The specifics of these previous events or conditions are key to the prevention and treatment of self- destructive behaviors. Although "end-stage" services (such as suicide crisis centers) are useful, prevention and treatment must begin much earlier. To rely on interventions at the last stage is analogous to trying to prevent renal disease by providing services to people 3-154 ready for dialysis. We must also find ways to look more definitively at several causes of death which may have a common etiology in- stead of looking separately at suicide, al- coholism, homicide, and accident statistics. Prevention How, then, can we prevent suicide in Indian youth? Some foci include: 1. Improving socioeconomic conditions. This is not a direct concern of psychiatric and other health care personnel and programs, but must be included as one of the keys to in- fluencing self-destructive behaviors. Specifi- cally with regard to Indians, until there is a solid economic base on reservations and in other Indian communities, it will be very dif- ficult to alter the rates of these negative be- haviors. The effective involvement of qualified Indian people in the process of economic and political development is a "must" (although simply terminating all help, in the name of "self-determination”, is very destructive). 2. Recognizing and treating underlying psychiatric disorders, providing services for Indian people which are adequate in number and quality, and coordinating with other health services. Clearly, many psychiatric disorders, mild and severe, are important risk factors for suicide and other self- destructive behaviors. Many of the disorders of youth which can lead to suicide are treatable. In spite of this, the mental health programs of the IHS have been chronically under-funded,; mental health, alcoholism, drug abuse, and general health services are often separated administratively and functionally; and there are far too few qualified treatment profes- sionals in the mental health, alcoholism, and drug abuse treatment programs. Also, be- cause mental health services are so decentralized in the IHS, there is little development of national or even regional strategies for effective service delivery. The majority of Indians are healthy, perhaps even healthier overall than at any time since the European invasion of the continent. But J.W. Thompson: Prevention. . Among American Indian. . . a great prevention challenge remains in help- ing adolescents and parents who suffer from alcoholism, drug abuse, and other psychiatric disorders. Case-finding, effective service delivery, and improved services per se are es- sential in terms of early identification and treatment of pathology which leads self- destructive behaviors. 3. Coordinating health (including mental health, alcoholism, and drug abuse) services with other human services. It is important to work with schools, employers, tribal leader- ship, and families in dealing with self-destruc- tive behaviors. A key actor is the primary care physician who is in an excellent position to recognize and deal with factors which may lead to violent behavior. This role may in- clude treating the patient, referring to qualified psychiatric specialty care, and making connections between schools, families, etc., thereby building a care-giving network or system for the troubled youth. This means dealing with basic problems early, not waiting for the last stage behaviors. 4. Addressing culture conflicts among young people. There is often an enormous conflict between the white and Indian cul- tures, which profoundly influences youths who are forming their personal identities. Having meaningful cultural experiences, as well as helping youth to deal with white cul- ture vis a vis their Indian culture, are impor- tant in the community, the family, and the schools. Some active treatment programs that address this conflict can be adopted as prevention models. These include the model boarding school on the Navajo reservation (9), and the Rainbow Lodge alcoholism treatment program in Canada (6). We have spoken about primary prevention (prevention of self- destructive behavior, per se), and of secondary prevention (early recognition and treatment of conditions which may lead to self-destructive behavior). Equally important is tertiary prevention: providing appropriate followup services to Indian youth who have received active psychiatric treatment or who have actually engaged in self-destructive behavior. This must be done with a recognition that the alienation which Indian youths feel toward the care-giving system can be profound. Therefore, active outreach must be per- formed in a culturally sensitive way. CONCLUSION Suicide and other self-destructive behaviors in Indian youth are not new problems. At the present time we do not have the information to determine the extent and shape of the problem in a given Indian community at a given point in time. From time to time, awareness of problems surfaces outside of the Indian community, but this may not be so much discovery of "epidemic," but rather highlighting a particular community at a par- ticular time. Prevention needs to be aimed not to the communities which have most recently made the press, but rather at com- munities which can be demonstrated by epidemiologic research to have a problem needing a specialized response. That response must be scaled to fit the nature and extent of the problem, not simply reacting with large scale initiatives when less expen- sive measures are called for. Both the gathering of data and the planning of respon- ses should be done in conjunction with the tribes themselves, with unobstrusive measures and with cultural sensitivity. We should remember that the challenge is not justsuicide, but all self-destructive behaviors, all of which are interwoven with one another. Finally, we should not be satisfied with prevention only at the "end stage", but work towards early recognition and treatment of conditions, social and psychiatric, which lead over time to self-destructive behaviors. REFERENCES 1. Shore JH: American Indian Suicide—fact and fan- tasy. Psychiatry, 38:86-91, 1975 2. Beiser M: Mental health of American Indian and Alaska Native children: Some epidemiologic perspectives. White Cloud Journal, 2(2):37-47, 1981. 3. Berlin IN: Prevention of emotional problems among Native American children: overview of develop- mental issues, Journal of Preventive Psychiatry, 1(3):319- 327, 1982. 4. Berlin IN: Prevention of adolescent suicide among 3-155 Report of the Secretary's Task Force on Youth Suicide some Native American tribes. Annals of the American Socity for Adolescent Psychiatry, 12(6):77-93, 1985. 5. Levy JE, Kunitz SJ: A suicide prevention program for Hopi youth. Presented at the annual meetings of the American Anthropological Association, Washington, D.C., December, 1985. 6. Ward JA: Preventive implications of a Native Indian mental health program: focus on suicide and violent deaths. Journal of Preventive Psychiatry, 2(3,4):371-385, 1984. 7. Thurman P: Personal communication, 1986. 8. Levy JE, Kunitz SJ, Henderson EB: Hopi deviance in historical and epidemiological perspective. In: J. Jor- ensen and L. Donald (Eds.), Essays in honor of David F. erle. Folklore Institute Press, Berkeley, in press. 9. Oetting E, Dinges N: Model Dorm Evaluation Sum- mary. Final Report on Indian Health Service Contract HSM 74-70-112, 1973. ACKNOWLEDGMENTS --oooeeeeenrennees - The author wishes to thank the following for their assistance: Committee of American In- dian and Alaskan Native Psychiatrists, American Psychiatric Association; Pam Thurman, graduate student in psychology, Oklahoma State University; Dr. James Jus- tice, Dr. Bill Douglas, and Maria Stetter, In- dian Health Service. 3-156 SUICIDE AMONG ASIAN AMERICAN YOUTH Elena Yu, Ph.D., Associate Professor, Community Health Sciences, School of Public Health, University of Illinois at Chicago, Chicago, Illinois, and Research Associate, Pacific/Asian American Mental Health Resource Center, Chicago, Illinois Ching-Fu Chang, Ph.D., Assistant Professor, Department of Sociology, Chung-Hsing University, Taipei, Taiwan William T. Liu, Ph.D., Professor, Department of Sociology, University of Illinois at Chicago, Chicago, Illinois, and Director, Pacific/Asian American Mental Health Resource Center, Chicago, Illinois Marilyn Fernandez, Ph.D., Research Associate, Pacific/Asian American Mental Health Resource Center, Chicago, Illinois Since the mid-1960s, the media have portrayed the positive stereotype of Asian American youth as a special population who work hard to pull themselves up by their bootstraps (New York Times Magazine, January 9, 1966; U.S. News and World Report, December 26, 1966, Time Magazine, March 28, 1983; Sunday Chicago Sun-Times, January 22, 1984; Chicago Tribune, January 15, 1986). Little attention, if any, was placed on the social problems faced by young Asian Americans in their struggle to excel and to es- tablish themselves in the U.S. society. Similarly, research on suicide among Asian American minorities has not focused on the age group 15 to 24 years old, but have instead tended to highlight the problems of the elder- ly (e.g., Bourne, 1973; McIntosh and Santos, 1981). OBJECTIVES This paper fills the knowledge gap concern- ing Asian and white American youth suicide by examining the national suicide data for two time periods in the age range 15 to 24 years. Aside from the apparent need for a descriptive database, such a comparison ser- ves four other objectives: (1) to determine the magnitude and direction of the ethnic dif- ferences in youth suicide between Asian and white Americans; (2) to examine the chan- ges in the suicide rate over time between the different groups; and (3) to explore plausible factors for the observed ethnic differences in Asian American youth suicide; and (4) to dis- cuss the implications of the research findings from a prevention perspective. Many of the findings in this paper are based on data extracted from death certificate records and submitted by each of the 50 States to the National Center for Health Statistics (NCHS). Created in 1960, NCHS is mandated to collect, analyze, and dissemi- nate statistical and epidemiologic data on the health of the nation. Because the size of the Asian American population has remained numerically insignificant until recently, na- tional mortality data for this special popula- tion are difficult to analyze and interpret even though they have existed for some time at NCHS. Furthermore, since analyses of 3-157 Report of the Secretary's Task Force on Youth Suicide such data require population denominators collected by the Bureau of the Census, the absence of intercensal estimates for Asian Americans in general and Chinese and Japanese in particular, has severely limited the use of these records for research pur- poses. For these reasons, meaningful cal- culation of suicide rates can be made only for the years 19701 and 1980. Of the 40-some Asian American subgroups enumerated in the last census, only the suicide statistics for Chinese and Japanese Americans will be presented in this paper for reasons of availability of data and confidence in the quality of the data. This decision should not be interpreted to mean that the statistics obtained for these two older Asian American groups in any way represent all Asian Americans. In the strictest sense, the term Asian Americans is a meaningful con- cept only insofar as it identifies the geographic origins of a group of people who are visibly and culturally different from the majority white population. However, the population itself is comprised of a number of diverse groups which, in many ways, are as different from one another as they are dif- ferent from other races. It should also be stated that mortality data in general, being collected primarily for legal purposes, have their inherent limitations. Reporting or classification errors are pos- sible. The magnitude of errors may vary by State as well as by specific information (e.g. sex may be more accurately recorded than ethnicity). Although many studies assess the 1. National mortality data prior to 1970 which contain infor- mation on Chinese and Japanese Americans have either been destroyed because they were not packed in tapes so as to be accessible for computer manipulation, or for those years in which they are available (e.g., 1968 and 1969), ap- propriate population denominators with detailed break- downs by age and sex for Chinese and Japanese are not available from the Bureau of the Census. 2. Data for other groups, such as the Filipinos, are available but are of dubious quality because they produce death rates which are improbable. Likewise, data for the Pacific Is- landers are extremely small, subject to severe fluctuations, and geographically confined to Hawaii and the West Coast. Consequently, a meaningful concept of national suicide statistics for the Pacific Islanders remains to be studied. 3-158 quality of medical recording in death certifi- cates, such studies have not been targeted to a specific population such as the Asian Americans. We do not know, therefore, the extent of underreporting or misreporting of race or of cause of death for Asian Americans other than what Yu (1982) reviewed in her earlier work on infant mortality. This ac- knowledgement of the limitations of data must also be balanced by an appreciation of the fact that the United States probably has one of the better maintained vital registra- tion systems of all modern nations, and mor- tality data extracted from death certificates are our only source of statistics on suicides. Thus, in the absence of alternative sources of statistics on suicide, we are forced to use the death certificate data. Caution, however, is clearly warranted in the interpretation of these statistics. SOCIODEMOGRAPHIC PROFILE OF ASIAN AMERICAN YOUTH Unpublished data based on 100 percent count of the U.S. Census show that there were only 89,342 Chinese Americans and 96,059 Japanese Americans between 15 and 24 years of age in 1970 (Liu and Yu, 1975). This age group represented 20.5 percent of the total Chinese American population and 16.2 percent of the total Japanese American population at the time (Table 1). (All tables appear at the end of this chapter.) By 1980, the number of 15 to 24 year olds increased by 63.5 percent (to 146,035) for Chinese Americans, and 25.4 percent (to 120,443) for Japanese Americans, compared to only 11.7 percent increase for white Americans. Table 2 shows the school enrollment pattern of 15 to 24 year olds by nativity and sex, based on the 1980 Census. In every age group ex- amined (15-24, 15-19, and 20-24 years), white Americans have the largest percentage not enrolled in school followed by Japanese Americans. Chinese Americans consistently have the lowest percentage not enrolled in school. Their rate, for the most part, is E.Yu: Suicide among Asian American Youth roughly half that found for white Americans. This is true regardless of nativity and sex>. In the late teens (15-19 years of age), the per- centage of Chinese not inschool is only about one-third of that for white Americans. Since this is true for both foreign- and native-born, cultural transmission of values concerning the importance of education in the Asian American communities is probably far more important than selective immigration per se in explaining the differential rates of school enrollment between these ethnic groups. The employment status of persons 15-24 years old who are not in school is shown in Table 3. Among foreign-born males not in school, the employment rate of white Americans is the highest of the three ethnic groups compared. About 82 percent of white Americans, 77 percent of Chinese Americans, and 73 percent of Japanese American youth who are not attending school are employed. Among foreign-born females, a larger percentage of Chinese and Japanese American youth (67 and 58 per- cent, respectively) are employed compared to white Americans (53 percent). Among native-born males, the employment rate of white and Chinese Americans are similar (about 79 percent), both being some- what lower than that found for Japanese Americans (85 percent). For native-born females, white American youth have the lowest employment rate (65 percent) com- pared to Chinese (84 percent) and Japanese Americans (81 percent). It appears that na- tive-born Asian American women are shouldering the responsibility of productive employment at a young age, probably to sup- port themselves as well as their parents and siblings. An examination of the employment status of those who are in school shows that among the foreign born 15-24 year olds, Japanese have 3. Tests of significance of differences in proportion were con- ducted for all the comparisons presented in Tables 2 to 4, inclusive. The differences between ethnic groups are statis- tically significant at the .05 level with one exception: the comparison of household type between white and Chinese Americans who immigrated before 1970 presented in Table 4. the lowest percentage of persons who are working while going to school (Table 3). This is true for males (26 percent, compared to 30 percent for Chinese and 32 percent for white Americans) as well as females (28 per- cent, compared to 34 percent for Chinese and 31 percent for white Americans). Among the U.S.-born youth, white Americans have the lowest percentage of persons who are working while going to school (36 percent for males and 35 percent for females), while Japanese Americans con- sistently have the highest percentage of per- sons who are both employed and in school (42 percent for males and 43 percent for females). Among foreign-born youth, the findings are just the opposite. Japanese American males and females have the lowest percentage of persons who are working and going to school, while the Chinese and white Americans have similar rates. Household type information on the 15 to 24 year olds are shown in Table 4. Among U.S.- born youth, the percentage of persons living in a married couple household varies from 64.3 percent for Chinese Americans, 65.4 percent for Japanese Americans, to 67.5 per- cent for white Americans. However, the per- centage of youth living alone, in group quarters, or in nonfamily households shows somewhat greater variability, with the Chinese having the largest percentage (25.6 percent) of the three groups living in these "other" type of households. No doubt this is due to the high percentage of Chinese na- tive-born youth who are enrolled in school, compared with the other two groups. For foreign-born youth, the data in Table 4 are broken down by year of immigration. A larger percentage of Asian Americans who were between the ages of 15 and 24 in 1980 and who immigrated to the United States in the 1970s are either living in group quarters or in nonfamily households. Again, this is most likely due to the large numbers of Asian Americans who are enrolled in school and living apart from their family. Japanese American youth who immigrated during the 1970s have a disproportionately large per- 3-159 Report of the Secretary’s Task Force on Youth Suicide centage (around 13 percent) living alone, compared with the other two groups (no more than 5 percent). Why this is so is far from clear. AGE-SPECIFIC SUICIDE RATES: 1970 and 1980 Tables 5 and 6 show the average annual age- specific and age-adjusted suicide death rates for white, Chinese, and Japanese Americans in 1980 and 1970, respectively. Following the convention of the National Center for Health Statistics where these data are managed, the U.S. population in 1940 was used as the standard population for age ad- justment. Across time and for all ethnic groups in 1980, male suicide death rates in the 15 to 24 age range have exceeded female rates. With a suicide rate of 13.79 per 100,000 population in 1970, which increased to 21.91 per 100,000 in 1980, white American male youth have the highest suicide rates among the three ethnic groups compared. Overtime, there was a 58.9 percent increase in suicide rates among white males 15 to 24 years old. The Chinese American male suicide rate increased even more (by 122.3 percent) from 3.63 per 100,000 population in 1970 to 8.07 in 1980. On the other hand, although Japanese American male suicide rates (11.97 per 100,000 in 1970 and 14.09 per 100,000 in 1980) have been higher than the Chinese rates, the rate of increase in suicide rates over time is not as dramatic as that found for Chinese Americans. Given the population increases that each of the three ethnic groups have experienced over time (see Table 1), the question arises as to whether the increase in suicide rates be- tween 1970 and 1980 may be a result of the population growth. A closer examination of the data show that the percent population in- crease is 13.7 percent for white male 15 to 24 year olds, 63.1 percent for Chinese American youth, and 30.6 percent for Japanese Americans. In short, the ratio of the rate of change in suicide rates to the rate of popula- 3-160 tion increase is highest for white American youth (4.3), lowest for Japanese Americans (1.7), and intermediate for Chinese Americans (1.9). This finding indicates that factors other than population growth ex- plains the increase in suicide rate over time. By comparison, suicide rates for white American females 15 to 24 years old have not changed significantly between the two census years. (4.21 per 100,000 in 1970 compared to 5.00 in 1980) even though the population within this age range has increased by about 10 percent. The suicide death rate for Chinese American females increased by 59.2 percent, which is less than the 63.8 percent increase in the population of females in the same age group, while the suicide death rates for Japanese American females dropped slightly from 5.51 per 100,000 in 1970 to 4.52 per 100,000 in 1980, during which time the population had actually increased by 20.3 percent. On the basis of the age-specific death rates and the above ratios, some readers may con- clude that suicide is not a major public health problem for Asian American youth, com- pared with white Americans. However, an examination of the proportional mortality statistics gives a different picture of the find- ings. PROPORTIONAL MORTALITY FOR SELECTED DEATHS While the age-specific rate for any cause of death is calculated using the population size of a given group as the denominator and the number of deaths from a specific cause for that particular group as the numerator; proportional mortality for any cause of death is obtained by using the total number of deaths for any given population as the denominator and the specific cause of death as the numerator. Because of the differen- ces in the denominator, it is possible for these two rates to give apparently contradictory in- formation. Table 7 presents the proportional mortality rates for suicide deaths by ethnicity in 1970 E.Yu: Suicide among Asian American Youth compared with 1980. One notes that suicide accounts for a much larger proportion of deaths among Asian American youth in 1980 than among white Americans. Among males, for instance, suicide represents 21.3 percent of all deaths for Japanese Americans, 15.1 percent for Chinese Americans and only 12.9 percent for white Americans. Among females, it constitutes 20.8 percent of all deaths for Chinese Americans, 14 percent for Japanese Americans, and 8.8 percent for white Americans”. There is another way of examining the proportional mortality data in Table 7, that is, making within-group comparisons over time. In this case, the percentage change in average annual proportional mortality for Chinese American 15-24 year olds between 1969-71 and 1979-81 is striking (200 percent for both sexes) compared to white (53 per- cent) or Japanese Americans (33 percent). However, it is important to stress that, to begin with, Japanese American proportional mortality rates had been very high in 1969-71, and they remained high in 1979-81, whereas the Chinese American rates were very low in 1969-71 and they increased dramatically in 1979-81. From a public health standpoint, this increase in proportional mortality rate over time for all three ethnic groups is cause for concern. It is therefore a serious miscon- ception to rely solely upon the age-specific suicide rate of 15-24 year olds and conclude that Chinese American youth do not have a suicide problem. What is a "high" or "low" suicide rate depends on what group or what year is used as the reference point for com- parison purposes. The high proportional mortality rates among Chinese Americans relative to their low age- specific death rates are most likely a conse- quence of competing causes of death. As a 4. Unfortunately, due to the exceedingly small denominators of Chinese and Japanese Americans in comparison with white Americans, application of statistical tests of sig- nificance fails to produce a p-value of .05 or less. Strictly speaking, then, no firm conclusions can be reached regard- ing the statistical significance of these different proportion- al mortality rates despite the fact that in some instances (e.g., Chinese female youth) the proportional death rate is more than two times that of the white female rate. rule, deaths due to accidents has always been one of, if not the, major competing cause of death for persons in the age range of 15 to 24 years. So long as the proportional mortality rates for accidents remain very high, if not the highest, of all causes of death, the proportion of deaths from suicide can be expected to remain relatively low. This becomes ap- parent when one calculates the potential years of life lost for different causes of death, as shown below. POTENTIAL YEARS OF LIFE LOST Using the 10 leading causes of death for the United States, Table 8 represents the dis- tribution of average-annual potential years of life lost before age 85 for Chinese and Japanese who die at age 15 or older. Poten- tial years of life lost before age 85 are calcu- lated by totaling the remaining years until age 85 for each person who committed suicide in his/her youth (i.e., in the age range between 15-24 years). For example, a person dying at age 20 would contribute 65 years to the total, while one who dies at age 70 contributes only 15 years. With this indicator it is possible to rank the different causes of death while in- cluding only deaths before age 85 and giving more weight to early deaths. Of the 10 leading causes of death presented in Table 8, potential years of life lost can be calculated for only 6 of them. This suggests that for Chinese Americans, the remaining four leading causes of death (diabetes mel- litus, Chronic Obstructive Pulmonary Dis- ease, Cirrhosis of the Liver, and Arteriosclerosis) occur at such older ages that the potential years of life lost ap- proaches, if not equals, zero. In terms of selecting health promotion and prevention priorities, the ranking of causes of death according to potential years of life lost is more useful than ranking causes of death according to the total number of deaths. A death at the age of 20 or older has a different impact, at least to the family and to society at large, than a death at the age of 3-161 Report of the Secretary’s Task Force on Youth Suicide 80 years. Furthermore, because calculation of the potential years of life lost is affected by the population size as well as by the age-specific death rates, it is more interesting to compare the potential years of life lost between groups for each cause of death rather than to look at the absolute figures themselves. Table 8 shows that all three ethnic groups have similar rankings of the potential years of life lost at age 15 or older for the 10 leading causes of death in the United States. Acci- dents head the list, followed by suicide. Thus, suicide is a serious problem in the Asian American population just as it is in the white American population. However, the groups differ greatly in the per- centage of potential years of life lost due to a particular cause of death. Accidents, for example, account for 73 percent or nearly three-quarters of the average annual poten- tial years of life lost for white Americans who die at the age of 15 or older during 1979-81, while they account for only 45 percent for Chinese Americans and 58 percent for Japanese Americans. On the other hand, suicide accounts for a much higher percent- age of potential years of life lost for Chinese (28 percent) and Japanese (25 percent) Americans, compared to only 15 percent among white Americans. Those concerned with the identification of major public health problems are well ad- vised to examine the distribution of potential years of life lost, as a first step to defining priorities. As a second step, rates of poten- tial years of life lost should be considered for identifying trends over time. Table 9 shows the average annual potential years of life lost by 10 leading causes of death for ages 15 to 24 years in 1969-71. The data indicate a higher percentage of potential years of life lost due to suicide for Japanese youth (18 percent) compared to Chinese (8.1 percent) and white Americans (10 percent). By 1979-81, it is the Chinese youth who had the highest percentage of potential years of life lost to suicide (28 percent) compared to 3-162 Japanese (25 percent) and white Americans (15 percent). THE NATIVITY FACTOR Nativity, or the decedent’s place of birth, is an important factor in Asian American mor- tality analysis. It is generally taken as a proxy measure for cultural upbringing and socioeconomic lifestyle, given the limited in- formation available from the death certifi- cate. In addition to age and sex, nativity is an important risk factor in the analysis of Asian American mortality data because among Chinese Americans 15-24 years of age, 60 percent are foreign-born, compared with 21.4 percent for Japanese Americans and 4.3 percent for white Americans. The variability in the proportion of foreign-born raises the question as to whether there are nativity dif- ferences in suicide mortality rates for Asian American youth compared with white Americans. Unfortunately, the nativity in- formation is not available in the national mortality data tapes for 1969-71. Analysis of the variable is thus confined only to the 1980 data set. Table 10 shows that at each age group, the suicide death rate for foreign-born youth is consistently higher than that found for the native-born. In the 15-24 age range, the rate for foreign-born Chinese is 7.1 per 100,000 population compared with 5.2 for native- born youth. For Japanese Americans, the nativity ratio is higher--the rate being 14.3 per 100,000 for the foreign-born youth com- pared with 8.1 for the native-born. Clearly, suicide is a serious problem in the Asian American population, and more foreign-born youth are at risk than the na- tive- born. Any prevention efforts should pay strong attention to the foreign born Asian American youth if we are to reduce ef- fectively the overall suicide rates. DISCUSSION The study of suicide as a socio-cultural phenomenon is a classic one, dating as far E.Yu: Suicide among Asian American Youth back as Durkheim's classic work (1897, trans- lated by Spaulding and Simpson, 1951). His innovative approach to the study of suicide involves inter-societal comparisons of suicide statistics over time and among dif- ferent segments of the population--an ap- proach which emerged as a result of his concern over societal integration and the na- ture of group cohesion. Insofar as research on Asian American suicide is concerned, the work has barely begun. This paper is perhaps one of the very first efforts at exploring the National Center for Health Statistics’ archival death files to analyze the inter-ethnic differences in suicide rates among Asian American youth. Previous efforts had been targeted to specific local areas, such as San Francisco (Bourne, 1973) or Hawaii, or to age groups other than age 15 to 24. It is obvious that we have bare- ly scratched the tip of the iceberg on Asian American suicide at the national level. What we have learned is that making inter- ethnic comparisons at one point in time has distinct disadvantages in that when the suicide rates for the reference population (in this case, white American youth) are high, Asian American rates always appear low by contrast. If one were to examine the Asian American suicide rates over time, however, one quickly discovers that these rates have in- creased dramatically. The reasons for their increase are not yet clearly understood, much less studied. We must admit that, theoretically, we have no adequate explanation, as yet for the lower age-specific suicide rates for Chinese and Japanese Americans as a group, compared to white Americans. One possibility is that overall rates for the 10 leading causes of death are lower for the two Asian American groups than for white Americans (Yu et al., 1985). Similarly, their age-specific suicide statistics are also lower than that found for white Americans. Thus, any Asian-white comparison of rates would always lead to the conclusion that the former appear to have few health or social problems. Nonetheless, as we have demonstrated in this paper, whenever proportional mortality rates are used for purposes of comparison, a different picture emerges. Proportional suicide rates are higher for Asian Americans than for white Americans, and these rates have increased dramatically over a ten-year period. An examination of the Census data indicates that Asian American youth are characterized by high enrollment in school, and State-level data indicate that they have low drop-out rates compared with white Americans (Yu, Doi, and Chang, 1985). The media have played an important role in highlighting the academic achievements of Asian American school-age children. What has not been em- phasized is the psychological pressure and emotional scars that the young have endured in order to sustain the expectations of parents and school teachers alike. Sociologically, it is important to realize that over the last twenty years, the United States has experienced an unprecedented influx of Asian immigrants whose educational levels and professional skills are at the highest levels compared to the earlier waves of Asian immigrants and European settlers. In the United States today, the proportion of Asian Americans with four years or more college is significantly higher than that of white Americans. Although the occupational return on education has not been as high for Asian Americans as one might expect had they been white Americans, large percent- ages of Chinese and Japanese Americans still hold high-prestige jobs compared with white Americans. This cohort of highly-educated professionals are concentrated in the 40-55 years age range, the age of parenthood with children in school. In interview after interview with Asian American high achievers, the public learned from the media that the children explained their drive to excel in terms of the shame that can befall their parents should they fail, and the glory they bring to their parents when they succeed. It comes as no surprise that we have a cohort of high-achieving Asian 3-163 Report of the Secretary's Task Force on Youth Suicide American parents who are putting tremen- dous pressure on their children to become even more successful then they are. The in- tensified pressure, and the sudden awareness of Asian American teenagers about their self-identity problems, are likely causes for the increased suicide rates among Asian Americans between 1970 and 1980. Native- born Chinese and Japanese in particular have a certain vulnerability in their self-con- cept in that most of them do not speak their parents’ language but are still perceived by the society at large to be non-native Americans. However, the foreign-born Asian American youth faces perhaps even more inner turmoil because of the inevitable clash of values held by their immigrant parents and the larger society, especially their American peers. The most recent U.S. Census data presented earlier (see Table 4) also show that, at least among the foreign- born Japanese American youth, a significant number (60 percent) of those who im- migrated in 1975-80 are living in households with no parents--that is, they either live alone, in group quarters, or in non-family households. The comparable figure for Chinese American youth who immigrated during that same period is 26 percent, which is also significantly higher than the rate for foreign-born white Americans (21 percent). Among those who immigrated in 1970-74, some 45 percent of the Japanese American youth and 16 percent of the Chinese youth are living in households without a parental figure, compared to only 6.5 percent of the white American youth. These findings sug- gest that a substantial percentage of Asian American youth are living without im- mediate familial support--a source of social support critical during the teenage years and early adulthood, especially for the uprooted (i.e., the foreign born). Much research remains to be conducted to examine the psychodynamics of the Asian American fami- ly, and the relationship between educational achievement and suicide among the young. 3-164 RECOMMENDATIONS To date, the only source of information on Asian American suicide comes from the death certificate data submitted by each of the 50 States to the National Center for Health Statistics. Death registration is generally regulated by State laws which con- form to a 1959 Model Act developed by the State registrars of vital records under the leadership of the National Office of Vital Statistics. Since disposing of a corpse without a permit is a serious violation, the registration of deaths in all but the most iso- lated areas is believed to be almost 100 per- cent. However, having a relative who commits suicide is a shameful experience for surviving relatives. Attempts to misreport suicide as an accident or as another type of death probably exist to an unknown degree among all ethnic groups. We do not know if such misreporting of the cause of death is greater for Asian Americans than for white Americans. If it is higher, such errors have not been accounted for in our statistical tables. We believe that a far more serious problem lies not in the registration of death itself, but in the recording of the personal particulars about the deceased. A report prepared by the National Center for Health Statistics (Woolsey, 1968) indicates that such par- ticulars are usually filled out by the funeral director, who obtains the information from a surviving relative. The funeral director also takes the certificate to the attending physician (or in the case of an unattended or violent death, to the medical examiner or coroner) for completion of what is known as the "medical certification of cause of death." This portion of the death certificate requires the signature of the physician or medical- legal officer. There is no built-in mechanism to check the accuracy of the funeral director’s identification or recording of the race item on the death certificate. Strictly speaking, questions remain as to the accuracy of the information even when such informa- tion is obtained from a surviving relative. To what extent, for example, do the survivors of E.Yu: Suicide among Asian American Youth a fourth-generation Japanese American identify the decedent as Japanese in response to the "race" item on the death cer- tificate? Although we have no empirical data to verify the accuracy of race recording, we suspect that, if there were gross errors, the unintentional misreporting of race is probab- ly greater among the native-born than among the foreign-born. The latter have a stronger ethnic identity than the former. Besides, the more monolingual native-born Asian Americans are, the more likely their "American-ness" will lead to their being clas- sified as white. This is of course pure speculation emboldened by the absence of data. Additionally, the National Center for Health Statistics issues to the 50 States model forms of the U.S. Standard Certificates each decade, and the States usually, but not al- ways, adhere to these in printing their own forms. Therefore, not all 50 States collect sufficiently specific race/ethnic information to allow for the systematic identification of Asian American subgroups such as Chinese, Japanese, Filipino, Korean, Vietnamese, Asian Indian, and others. Three recommendations can be made to as- certain the accuracy, if not to help minimize the misreporting of race or ethnic informa- tion on the death certificate. First, States which have not provided sufficiently detailed information on Asian American subgroups in their death certificates should be encouraged to do so. Increasingly, Asian Americans are no longer concentrated in just a few States. Perhaps the United States National Commit- tee on Vital and Health Statistics can be en- couraged to discuss this matter in their future meetings. Second, a special study can be made to do a follow-back mortality survey of a probability sample of suicides occurring to white, black, and Asian Americans, in order to ascertain the accuracy of the race or ethnic coding. For reasons of cost containment, such a study can be limited to States with the largest con- centration of Asian Americans: Hawaii, California, New York, Illinois, Texas and Washington. The inclusion of States which are recently experiencing the growing presence of Asian Americans--such as II- linois, Washington, and Texas--is important since the accuracy of the race/ethnic code may well vary with the density of Asian American population in a given geographic area. Methodological studies on the quality of the death certificates have been con- ducted by the National Center for Health Statistics for other purposes (Gittelsohn, 1982; Harris, 1980), but not for verifying the accuracy of the coding of Asian American deaths. Given the expertise available at the National Center for Health Statistics, it should not be difficult for the agency to col- laborate on such a methodological study with an advisory committee which includes Asian American researchers. Third, in the next decade when the U.S. Stan- dard Death Certificate will again be revised, serious thought should be given to allowing a separate identification of Asian Indians, Koreans, Vietnamese, Cambodians, and Laotians in the State vital registration forms. These immigrants have arrived in large num- bers as a result of the Immigration Act of 1965. Within two decades, they have grown in size to come close to the number of Chinese and Japanese who arrived over a 100-year span. The explicit identification of these ethnic groups in vital registrations will provide future investigators with a rich source of data on Asian American mortality patterns in general, and suicide in particular, at a relatively low cost. Additionally, a nation-wide suicide register can be established, with more detailed infor- mation about the decedent’s demographic and socioeconomic background, including family history of suicide and other types of mental illness, occupation, and type of work. The information obtained from the death register will greatly augment the quantity and quality of the data on suicide that are presently only available from the use of death certificates. It will also be a useful tool for monitoring deaths due to suicide, both for re- search and for prevention purposes. 3-165 Report of the Secretary's Task Force on Youth Suicide Special in-depth studies should be en- couraged through the grants and contracts mechanism, including requests-for- proposals and supplemental grants to exist- ing minority mental health research centers, by the Federal funding agencies such as the National Institute of Mental Health, the Na- tional Institute for Child Health and Development of the National Institutes for Health, and others. Such studies should specifically encourage the examination of ethnic differences in suicide, and within each group, the analysis of at least gender and nativity differences, if not other factors as well. From our work on the 1970 and 1980 data, it is clear that even among the Chinese and the Japanese youth, the increase in suicide varies by gender and by nativity. A thorough understanding of the cultural values and social structural factors which contribute to suicide would be most helpful in guiding the work of counselling and prevention programs. In short, we need to go beyond the limits of epidemiologic methods and venture into sociological, psychological, and anthropological studies of suicide. The subject deserves to be studied using an interdisciplinary approach, rather than just an epidemiologic one. By en- couraging young investigators to work close- ly with seasoned researchers in existing minority mental health research centers, the costs of research can be kept to a minimum and the collaboration of an interdisciplinary team of researchers assured. Proceedings of the Secretary's Task Force on Youth Suicide and other research findings on suicide that are forthcoming should be disseminated widely not only to research and prevention centers in the country, but also to school teachers and counselors alike. The latter come in close daily contact with the popula- tion most at risk of suicide. Their informed knowledge of the early signs of suicidal be- havior can go a long way towards the reduc- tion, if not prevention, of a major public health problem in the United States. 3-166 REFERENCES 1. Bourne, Peter G: Suicide among Chinese in San Francisco. American Journal of Public Health 63, 8 (August): 744-750, 1973. 2. Gittelsohn, Alan M: Annotated Bibliography of Cause-of-Death Validation Studies: 1958-1980. Vital and Health Statistics, Series 2. Data Evaluation and Methods Research; no. 89. DHHS Publication no. (PHS) 81-1368, 1982. 3. Green, Laura: Super Kids: Asian Americans. Sun- day Chicago Sun-Times, January 22, 1984. 4. Harris, Kenneth W: A Methodological Study of Quality Control Procedures for Mortality Medical Coding. Vital and Health Statistics: Series 2, Data Evaluation and Methods Research; no. 81. DHEW publication no. (PHS) 79-1355, 1979. 5. King, Haitung: Selected Epidemiologic Aspects of Major Diseases and Causes of Death among Chinese in the United States and Asia. Pp. 487-550 In: A. Kleinman, P. Kunstadter, E.R. Alexander, and J.L. Gale (Eds.), Medicine in Chinese Cultures. U.S. Department of Health, Education, and Welfare Public Health Service. National In- stitutes of Health DHEW Publication No. (NIH) 75-653. Washington, D. C.: Government Printing Office, 1975. 6. Liu, William T. and Elena Yu: Asian American Youth. Pp. 367-389 in Robert J. Havignurst (Ed.), Youth. The Seventhy-Fourth Yearbook of the National Society for the Study of Education. Chicago: University of Chicago Press, 1975. 7. McGrath, Ellie: Education, Confucian Work Ethic. Time Magazine: 52, March 28, 1983. 8. Mcintosh, J. L. and J. F. Santos: Suicide among minority elderly: A Preliminary Investigation. Suicide and Life-Threatening Behavior 11:151-166, 1981. 9. Petersen, William: Success Story, Japanese American Style. New York Times Magazine, January 9, 1966. 10. Spencer, Jim: Why Fu Lien Can Read: For Asian Americans, Learning is a Family Obligation. Chicago Tribune, January 15, 1986. 11. Woolsey, Theodore D: The 1968 Revision of the Standard Certificates. National Center for Health Statis- tics Vital and Health Statistics PHS Publication No. 1000- Series 4, No. 8. Washington, D. C.: Government Printing Office, 1968. 12. Yu, Elena S. H: The Low Mortality Rates of Chinese Infants: Some Plausible Explanatory Factors. Social Science and Medicine 16:253-265, 1982. 13. Yu, Elena S. H., Ching-Fu Chang, William T. Liu, and Stephen H. Kan: Asian-White Mortality Differences: Are There Excess Deaths? PR. 208-251 in Margaret M. Heckler (Ed.), Report of the Secretary's Task Force on Black and Minority Health. Washington, D. C.: U.S. Depart- ment of Health and Human Services, 1985. 14. Yu, Elena S. H., Mary Doi, Ching-Fu Chang: Asian American Education in Illinois. Chicago: Illinois State Board of Education, 1986. ACKNOWLEDGMENT The authors are grateful to Dr. Ching-Tung Lung for her technical assistance, and to Eugenia P. Broumas for her patience and en- couragement in the preparation of this paper. L91-€ Total White, Chinese, and Japanese American Population and 15-24 Year Olds 1970" and 1980° Census White Americans Chinese American Japanese American Age and Sex 1970 1980 % Change 1970 1980 % Change 1970 1980 % Change All Ages Both sexes 177,748975 188,371,622 +6.0 435,062 806,040 +85.3 591,290 700,974 +18.5 Male 86,720,987 91,685,333 +5.7 228,565 407,544 +78.3 271,300 320,941 +183 Female 91,027,988 96,686,289 +6.2 206,497 398,496 +93.0 319,990 380,033 +18.8 15-24 Year Olds Both Sexes 30,652,187 34,250,876 +11.7 89,342 146,035 +63.5 96,059 120,443 +25.4 Male 15,232,090 17,317,434 +13.7 45,572 74,332 +63.1 47,078 61,498 +30.6 Female 15,420,097 16,933,442 + 9.8 43,770 71,703 +63.8 48,981 58,945 +20.3 1. Data for 1970 are based on unpublished complete count (100 %) of the 1970 U.S. Census data prepared by the Bureau of the Census for the National Center for Health Statistics. They are more reliable than the figures reported in some published 1970 Census reports which are based on only 20% or 15% count. 2. Data for 1980 are based on 100% count of the 1980 U.S. Census data supplied by the Bureau of the Census to the Pacific/Asian American Mental Health Research Center. Table 1. YINOA UBDLIBWY UBISY Buowe apIoing :nA'3 891-€ School Enrollment of 15-24 Year Olds for White, Chinese, and Japanese Americans By Nativity and Sex: 1980 Census Foreign Born U.S. Born Male Female Male Female White Chinese Japanese White Chinese Japanese White Chinese Japanese White Chinese Japanese 15-24 Year Olds (733) (2224) (662) (742) (2159) (661) (16484) (1510) (2411) (1 6245) (1397) (2450) Total, in percent 100.0 100.0* 100.0 100.0* 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Not enrolled 52.8 24.8 36.6 54.3 28.9 40.1 50.0 23.0 32.3 51.8 26.0 33.6 Enrolled in: public school 39.3 62.3 46.4 38.1 59.2 46.6 43.3 61.7 58.7 41.2 59.4 58.9 church-related sch. 3.7 4.1 75 4.0 4.6 6.3 3.9 5.0 44 43 5.9 2.8 other private sch. 4.2 8.7 9.5 3.5 73 7.0 2.8 10.3 46 27 8.7 4.7 15-19 Year Olds (324) (1027) (203) (323) (885) (219) (8323) (721) (1155) (7902) (686) (1145) Total, in percent 100.0 100.0* 100.0 100.0 100.0 100.0* 100.0 100.0* 100.0* 100.0 100.0 100.0 Not enrolled 28.2 9.3 17.7 28.5 9.9 11.0 244 6.4 11.2 24.7 6.4 9.7 Enrolled in: public school 61.7 80.4 64.5 60.7 78.9 73.14 67.4 75.6 78.4 66.4 74.3 81.0 church-related sch. 4.9 55 9.4 6.2 5.1 9.6 5.4 8.0 6.6 6.1 8.5 4.1 other private sch. 53 4.9 8.4 4.6 6.1 6.4 2.8 9.0 3.9 2.8 10.8 5.2 20-24 Year Olds (409) (1197) (459) (419) (1274) (442) (8161) (789) (1256) (8343) (711) (1305) Total, in percent 100.0 100.0 100.0 100.0 100.0 100.0 100.0* 100.0 100.0 100.0 100.0 100.0 Not enrolled 72.4 38.2 449 74.2 42.0 54.5 76.2 38.1 51.8 77.5 449 54.5 Enrolled in: public school 215 46.8 38.3 20.8 45.6 33.5 18.7 48.2 40.7 17.4 45.0 39.6 church-related sch. 2.7 3.0 6.8 2.4 4.2 4.8 2.4 2.3 2.3 25 3.4 1.7 other private sch. 3.4 12.0 10.0 2.6 8.2 7.2 2.8 11.4 5.2 2.6 6.7 4.2 Source: Unpublished data from the 1980 Census tabulated by the authors. Data for Chinese and Japanese are based on the 5% white Americans are based on the .1% (B) sample tape. *Percent do not add up to 100 because of rounding errors. Sample Microdata (A) tape, while data for Table 2. 8pIoINS YINOA UO 82.104 XS. S,Aie}e108s ey} jo uoday 691-€ Schooling and Employment Status of White, Chinese, and Japanese Americans 15-24 Year Olds By Nativity and Sex: 1980 Census Foreign Born U.S. Born Male Female Male Female Not in school, Wiis number (387) in percent 100.0 Less than 16 years’ - Employed 81.9 Unemployed 7.2 Not in labor force 10.3 In school, number (346) in percent 100.0 Lessthan 16 years 16.8 Employed 31.5 Unemployed 2.3 Not in labor force 49.4 Chinese Japanese White (552) 100.0 76.5 6.3 16.5 (1672) 100.0 10.0 30.2 23 57.5 (242) (403) 1000 100.0 73.1 52.9 6.6 7.9 203 39.2 (420) (339) 100.0 100.0 6.2 18.3 25.7 31.0 1.9 2.6 66.2 48.1 Chinese Japanese White Chinese Japanese White Chinese Japanese (623) (265) (8249) (347) 65) (8418) (363) 1000 100.0 100.0 100.0 100.0 100.0 100.0 66.5 57.7 79.8 79.0 85.1 64.5 83.8 23 45 10.8 8.9 6.0 6.3 3.0 30.7 37.4 8.9 11.5 8.3 28.6 12.7 (1536) (396) (8235) (1163) (1631) (7827) (1034) 100.0 100.0 100.0 100.0 100.0 100.0 100.0 10.1 10.3 18.8 12.7 12.8 18.5 14.6 34.4 28.3 36.2 38.8 41.9 35.2 37.7 1.9 1.3 3.5 22 23 34 1.8 53.6 60.1 41.5 46.3 43.0 42.9 45.8 (124) 100.0 80.8 3.8 15.1 (1628) 100.0 12.5 43.4 2.4 41.7 Source: Unpublished data from the 1980 Census tabulated by the authors. Data for Chinese and Japanese are based on the 5% Sample Microdata (A) tape, while data for white Americans are based on the .1% (B) sample tape. 1. Since data on employment status are asked only of persons 16 years or older, data on the employment status of those between 15 to 16 years of age are not available. 2. Percent do not add up to 100 because of rounding errors. Table 3. YINOA URDIIaWY ueISY Buowe ep1oing :NA"3 0L1-€ Household Type of 15-24 Year Olds for White, Chinese, and Japanese Americans By Nativity and Year of Immigration: 1980 Census Foreign-Born by Year of Immigration Prior to Household Type Total U.S. -Born N.A. 1975-80 1970-74 1965-69 1969 White, (number) (34204) (32729) (357) (394) (218) (210) (296) in percent 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Married couple 67.3 67.5 61.6 53.6 68.8 73.8 66.6 One spouse absent 13.8 13.5 18.2 25.6 24.8 157 ns Other Living alone 44 44 48 4.8 14 1.9 6.1 Group quarters 7.0 7.0 8.7 7.9 28 3.3 7.1 Nonfamily 7.6 77 6.7 8.1 23 5.2 8.8 Chinese, (number) (7290) (2907) (72) (2300) (928) (752) (331) in percent 100.0 100.0 100.0 100.0 100.0 100.0* 100.0* Married couple 62.9 64.3 62.5 55.5 69.8 70.9 65.3 One spouse absent 13.7 10.1 12.5 18.8 13.8 12.4 124 Other Living alone 4.3 5.5 28 37 3.0 35 45 Group quarters 10.0 10.7 12.5 1.2 6.4 8.4 9.7 Nonfamily 9.0 9.4 9.7 10.8 7.0 49 8.2 Japanese, (number) (6184) (4861) (294) (585) (108) (81) (255) in percent 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Married couple 61.2 65.4 58.8 34.7 47.2 62.3 50.6 One spouse absent 11.7 12.1 16.0 5.3 8.3 8.6 16.1 Other Living alone 5.6 4.4 5.1 13.5 13.0 3.7 75 Group quarters 10.8 9.0 9.8 26.3 17.6 11.1 10.2 Nonfamily 10.7 9.1 12.2 20.2 13.9 13.6 15.7 Source: Unpublished data from the 1980 Census tabulated by the authors. Data for Chinese and Japanese are based on the 5% Sample Microdata (A) tape, while data for white Americans are based on the .1% (B) sample tape. 1. These are persons who cannot be said to have "immigrated" to the U.S. because they were born in U.S. Territories or possessions (and therefore not native-born). A small number of foreign-born persons for whom year of immigration information is missing may also be included in this category but the precise number cannot be ascertained. *Percent do not add up to exactly 100 because of rounding errors. Table 4. 8PIOINS YINOA UO 8210 Se] S,AIeje1088 ey) Jo uoday LLL-E Average Annual Age-specific’ and Age-adjusted (1940 U.S. Standard) Death Rates? for Suicide, Per 100,000 Population, by Specified Race: United States, 1980 White Chinese American Japanese American Age Group Total Male Female Total Male Female Total Male Female All ages, crude 13.31 20.57 6.43 8.27 8.26 8.28 9.08 12.57 6.14 Age-adjusted 12.54 19.41 6.20 7.97 7.93 8.02 7.84 11.08 5.00 5 - 14 years 0.52 0.75 0.28 030 0.61 0.86 169 15 - 24 years 13.55 21.91 5.00 6.39 8.07 4.65 9.41 14.09 4.52 25 - 34 years 17.48 26.99 7.98 7.13 8.59 5.72 12.18 16.72 7.82 35 - 44 years 17.03 24.27 9.93 9.01 8.94 9.09 3.10 12.68 6.39 45 - 54 years 17.69 24.55 11.18 12.28 10.77 13.89 8.75 9.81 8.22 55 - 64 years 17.54 26.52 9.59 12.34 9.37 15.52 9.93 12.38 7.78 65 - 74 years 18.28 32.41 7.45 24.35 25.85 22.61 6.61 11.17 217 75 - 84 years 20.91 46.18 6.03 33.51 21.82 44.32 25.01 39.56 15.75 85 years & over 19.45 53.28 492 56.13 64.10 49.93 62.59 139.76 19.50 Source: Division of Vital Statistics, National Center for Health Statistics, unpublished data calculated by the authors. 1. The numerator consists of 1979-81 cumulative number of deaths, the denominator is based on the total enumerated of the 1980 United States Census. 2. Excludes deaths of nonresidents of the United States. Table 5. YINOA UBDLIBWY UBISY Buowe apIoing nA"3 clL-€ Average Annual Age-specific’ and Age-adjusted (1940 U.S. Standard) Death Rates? for Suicide, Per 100,000 Population, by Specified Race: United States, 1970 White Chinese American Japanese American Age Group Total Male Female Total Male Female Total Male Female All ages, crude 12.29 17.75 7.09 10.89 12.06 9.60 9.24 10.68 7.99 Age-adjusted 12.28 17.90 7.22 11.60 12.48 10.45 8.55 10.01 7.22 5-14 years 0.35 0.52 0.17 -— -—- -— a — -— 15 - 24 years 8.97 13.79 4.21 2.98 3.63 2.92 8.70 11.97 5.51 25 - 34 years 13.94 19.33 8.68 10.61 11.91 9.26 14.88 16.91 13.29 35 - 44 years 17.79 22.86 12.90 14.62 13.94 15.37 10.42 11.23 9.95 45 - 54 years 21.18 29.00 13.83 19.47 20.00 18.80 11.52 11.61 11.44 55 - 64 years 22.89 34.80 12.21 20.74 25.15 15.03 6.40 725 5.54 65 - 74 years 22.09 38.03 9.81 39.00 49.54 24.88 9.62 13.89 6.36 75 - 84 years 23.03 46.67 7.42 56.17 54.19 58.75 30.75 26.66 33.61 85 years & over 19.42 46.70 4.38 175.09 84.71 271.74 86.11 162.82 -- Source: Division of Vital Statistics, National Center for Health Statistics, unpublished data calculated by the authors. 1. The numerator consists of 1969-71 cumulative number of deaths, the denominator is based on the total enumerated of the 1980 United States Census. 2. Excludes deaths of nonresidents of the United States. Table 6. 8pIoINS YINOA UO 82104 }SB] S,A1e}8108S 8y} JO Luoday ELLE Proportional Mortality Rate for Suicide Among White, Chinese, and Japanese American 15-24 Year Olds, by Sex: 1970 and 1980 Race and Sex 1970 1980 % Change White Americans Both Sexes 7.8 11.9 52.6 Male 8.1 12.9 59.3 Female 6.8 8.8 29.4 Chinese Americans Both Sexes 5.6 16.8 200.0 Male 49 15.1 208.2 Female 71 20.8 193.0 Japanese Americans Both Sexes 14.3 19.0 32.9 Male 14.2 21.3 50.0 Female 145 14.0 -3.4 Source: Unpublished data from the National Center for Health Statistics, calculated by the authors. Table 7. YINOA ueDIISWY UeISY Buowe epIoing :NA"3 V.iL-€ Average Annual Potential Years of Life Lost, in Percent and Rate Per 100,000 Population By 10 Leading Causes of Death at Ages 15-24 Years, 1979-81 White Chinese Japanese 10 Leading Causes Number Percent Rate Number Percent Rate Number Percent Rate 10 Leading Causes, total 2,024,922 100.0 2,189 100.0 2,948 100.0 Accidents 1,479,963 73.1 4321.0 975 445 667.6 1,712 58.1 1421.1 Suicide 301,708 14.9 880.9 607 27.7 415.4 737 250 611.6 Cancer 139,555 6.9 407.4 390 17.8 267.1 282 96 2339 Heart Disease 51,090 25 149.2 152 6.9 103.9 65 22 54.0 Cerebrovascular Disease 18,633 0.9 54.4 43 20 29.7 65 2.2 54.0 Pneumonia and Influenza 16,553 0.8 48.3 22 1.0 14.8 43 1.5 36.0 Diabetes Mellitus 7,215 0.4 21.1 -- - -- 22 0.7 18.0 C.O.P.D. 5,655 0.3 16.5 - -- - 22 0.7 18.0 Cirrhosis of the Liver 4,550 0.2 13.3 - - - -- - -- Arteriosclerosis 87 - 0.3 - -- - - - - Source: Unpublished data from the National Center for Health Statistics, calculated by the authors. Table 8. 8pIoINS YINOA UO 82104 YSe] S,AIe}e108S ay) Jo oday GLI-E Average Annual Potential Years of Life Lost, in Percent and Rate Per 100,000 Population By 10 Leading Causes of Death at Ages 15-24 Years, 1969-71 White Chinese Japanese 10 Leading Causes Number Percent Rate Number Percent Rate Number Percent Rate 10 Leading Causes, total 1,858,698 100.0 2,145 100.0 2,948 100.0 Accidents 1,372,497 73.8 44776 1,408 65.6 1576.3 1,625 548 1697.0 Suicide 178,642 9.6 582.8 173 8.1 194.0 542 183 565.7 Cancer 163,107 8.8 532.1 390 18.2 436.5 303 10.2 316.8 Heart Disease 48,252 2.6 157.4 65 3.0 72.8 130 44 135.8 Cerebrovascular Disease 29,055 1.6 94.8 65 3.0 72.8 173 5.8 181.0 Pneumonia and Influenza 41,557 2.2 135.6 22 1.0 24.3 108 3.8 113.1 Diabetes Mellitus 11,895 0.6 38.8 - -- - -- - - C.O.P.D. 6,782 0.4 22.1 - -- - 65 2.2 67.9 Cirrhosis of the Liver 6,803 0.4 22.2 22 1.0 24.3 22 0.7 22.6 Arteriosclerosis 108 0.4 -- - - - - - Source: Unpublished data from the National Center for Health Statistics, calculated by the authors. Table 9. YINoA UeDLIBWY UBISY Buowe apioing :nA"3g 9l1-¢ Average Annual Age-Specific’ and Age-Adjusted Deaths (1940 U.S. Standard Population) By Suicide, in Rate per 100,000, for Specified Asian American Groups, by Nativity: United States, 1980 Chinese Japanese Age in years Native-born ~~ Foreign-born Native-born Foreign-born All ages, crude 29 11.2 55 17.1 Age-adjusted 35 9.5 5.2 14.0 0-4 - - - - 5-14 0.5% - 1.1% - 15-24 52 71 8.1 14.3 25-34 5.9 7.5 7.9 18.7 35-44 4.1* 9.9 57 13.6 45-54 5.4* 14.2 55 13.6 55-64 2.4* 15.1 5.6 42.6 65-74 2.3* 37.3 5.4* 11.8* 75 and over - 45.2 6.8* 41.5 Source: Division of Vital Statistics, National Center for Health Statistics, unpublished data calculated by the authors. 1. The numerator consists of 1979-81 cumulative number of deaths, the denominator is based on the total enumerated of the 1980 United States Census. - The rates are obtained with numerators which consist of less than 5 persons. Table 10. 8pI2INS YINOA UO 82104 XS. S,Aie}e108s ay) Jo uoday BLACK YOUTH SUICIDE: LITERATURE REVIEW WITH A FOCUS ON PREVENTION F.M. Baker, M.D., M.P.H., Psychiatrist/Epidemiologist, National Institute of Neurological and Communicative Diseases and Stroke, National Institutes of Health, and Assistant Professor, Department of Psychiatry, The Johns Hopkins University School of Medicine ABSTRACT The national rates of completed suicide in the black population between 1950 and 1981 are presented including age-adjusted rates. Specific studies of suicide attempts and com- pleted suicides by blacks in several cities, e.g., New York, Philadelphia, Newark, Rochester (New York), and New Haven, are discussed. Methodological problems with existing studies and national suicide statistics are presented. Proposed theories of black suicide are reviewed. Based upon a summary of the characteristics of black suicide at- tempters reported in the literature, specific primary, secondary, and tertiary preventive strategies are suggested. INTRODUCTION Data from the National Center for Health Statistics (NCHS) have documented the in- crease in suicides by black Americans. Bet- ween 1950 and 1981 the suicide rates for black Americans increased: 114 percent for black males in the 15-24 age group and 33 percent for black females in the 15-24 age group (1). NCHS data also document the higher and increasing rates of suicide in white Americans (1). Other authors address this issue in another volume of the Report of the Secretary’s Task Force on Youth Suicide (2). This author will focus only on black suicide attempters and completers in this paper. By the 1980 census (3) blacks comprised 12 percent of the United States population with 41 percent of the black population between ages 16-39, the ages of highest risk for suicide among blacks (4,5,6). The increase in suicide by both black and white youth have resulted in the National Institute of Mental Health’s establishing suicide as one of its priority areas for 1986. But what specific information exists concern- ing the precipitating events, the psychosocial stressors, and interpersonal stressors that result in suicidal acts by black youth? Are there specific psychological theories which suggest etiology or mechanisms behind suicidal behavior in young Afro-Americans? To answer these questions and to identify specific primary, secondary, and tertiary preventive strategies, a literature review was conducted. The majority of the studies of black suicide attempters were completed in response to the work of Hendin (7) and the Black Revolution of the 1960s. The studies varied from theoretical papers which sum- marized mortality data and attempted to hypothesize an etiology for the observed in- creased rates of black suicide, through com- parative studies of black and white suicide attempters in several cities. Only one case- control study was identified and only two studies conducted a longitudinal assessment of a population during a period of four or 3-177 Report of the Secretary's Task Force on Youth Suicide more years. Although the majority of these studies are descriptive and I have some con- cerns with the methodological designs, I will review the known information in order to identify specific areas on which research should focus, as well as suggest various levels of preventive strategies. UNITED STATES BLACK POPULATION IN 1980 To understand the magnitude of suicidal be- havior within the black community, it is im- portant to review the demographic characteristics of Afro-Americans in the United States in 1980. The inaccuracies of the United States Census have been at- tributed to the failure of black persons to respond to the census as well as to the enumerators’ fear of urban neighborhoods. Acknowledging these imperfections, specific data are presented to provide a context for our discussion. In 1980 (2) 12 percent of the United States population was black--26.5 million persons. Fifty-three percent of blacks were females and 47 percent were males. Forty-four per- cent of black females were in the child-bear- ing years, between ages 15 and 44. The per- centage Afro-Americans in each age category by sex is shown in Table 1. Forty- one percent of black males and 41 percent of black females are between ages 16 to 39. Al- though black males outnumber black females through age 19, the loss of black males begins in the 20 to 24 age group with a significant difference noted in the 40 to 49 age group. Causes of premature death in black males ages 16 to 39 include homicide, suicide, acci- dents, and substance abuse (8,9,10). In fact, homicide is the leading cause of death for black males of ages 15 to 44 (8). This sex dif- ference is sustained throughout the latter years of the life cycle. Approximately 30 per- cent of the black population is below the poverty level compared to 9.4 percent of the total population. Table 2 shows the percent- age of individuals below the poverty level by race and age group. The average number of persons per black family is 3.7 and in white families, 3.9. In the black population 1,568,417 households were headed by a female without a husband present, with their own children under age 18. For whites, 3,166,397 households were headed by a female without a husband present, with their own children under age 18. DISTRIBUTION OF THE BLACK POPULATION BY AGE AND SEX % of TOTAL % of TOTAL % of TOTAL AGE BLACKMALES = BLACKFEMALES BLACK POPULATION 0-4 9.81 8.64 9.19 5-11 14.26 12.57 13.36 12-15 8.90 7.90 8.37 16-19 9.50 8.58 9.02 20-24 10.39 10.19 10.28 25-29 8.66 8.85 8.76 30-39 12.25 12.97 12.63 40-49 8.64 9.39 9.04 50-59 7.76 8.55 8.17 60-64 3.08 3.48 3.29 65-74 4.52 5.54 5.02 75-84 1.82 257 2.22 85 + 0.42 0.76 0.60 SOURCE: U.S. Census, 1980. Detailed Population Characteristics, U.S. Summary Section A. Table 1. 3-178 F.M.Baker: Black Youth Suicide NATIONAL STATISTICS ON COMPLETED SUICIDES In Health, United States, 1984 suicide rates were presented for blacks and whites by sex for the years 1950 through 1981 (11). Table 3 presents the rates per 100,000 population for ten year age groups. Figures 1-8 (Figures appear at end of chapter.) present graphic comparisons of white and black suicides by sex for each ten-year age group. In this 31- year period, suicides in white males increased 3.2 fold in the 15-24 age group and in the 25- 34 year age groups they nearly doubled. In all age ranges white males continued to have a steady increase in completed suicide. The increase in suicide by black males was also documented by national statistics. From 1950 to 1970, there was a 114 percent in- crease in the black male rate in the 15-24 age group and 106 percent increase in the 25-34 age group. By 1981, the suicide rate for black males showed a 134 percent increase in the 25-34 age group. During the thirty-one year period, 1950-1981, suicide declined sharply in black males beginning in the 35-44 age group and continued throughout the life cycle. The suicide rates for black females remained low in comparison to white females. Suicide rates in black females increased 111 percent for the 15-24 age group between 1950 and 1970. By 1981, this rate had declined to 33 percent greater than the 1950 rate. For black females, the 25-34 and 35-44 age groups had the highest suicide rates during the 1950 through 1981 period with peak rates occur- ring in 1970, a 119 percent and 85 percent in- crease respectively. For this 31 year period, white female suicide rates peaked in 1970 in the 25-54 age groups. By 1981, the suicide rate for white females in the 45-55 age group peaked, but only by 6 percent above the 1950 rate. The rate for the 15-24 age group had increased 81 percent and for the 25-34 age group, 50 percent. The Afro-American population of the United States is a younger population with a broader based population pyramid. The white population is older with a population pyramid that is more evenly distributed in all age groupings. Age-adjustment is a method that corrects for these age differences in the population of each race. Table 4 presents the age-adjusted suicide rates for the years 1950 to 1981. The age-adjusted rates (direct PERCENTAGE OF PERSONS BELOW THE U.S. POVERTY LEVEL BY RACE % Of Total % Of Total Age Group Black Population White Population under 16 38.6 11.6 16-21 33.1 12.0 22-24 26.9 11.1 25-34 22.0 7.5 35-44 20.5 6.4 45-54 20.5 55 55-59 23.1 6.6 60-64 27.4 8.4 65-69 31.0 9.5 70-74 35.4 12.0 75+ 39.8 16.6 SOURCE: U.S. Census, 1980. Detailed Population Characteristics, U.S. Summary Section A. Table 2. 3-179 08L-€ SUICIDE RATES BY SEX AND AGE PER 100,000 RESIDENT POPULATION 1950 1960 1970 1979 1980 1981 White Black White Black White Black White Black White Black White Black MALES 15-24 6.6 49 8.6 4.1 139 105 205 14.0 21.4 123 21.1 11.1 25-34 13.8 9.3 149 124 199 19.2 254 249 256 21.8 262 21.8 35-44 224 104 219 128 233 126 224 16.9 235 156 243 155 45-54 34.1 10.4 33.7 10.8 29.5 13.8 240 13.8 24.2 12.0 239 123 55-64 459 165 40.2 16.2 350 106 263 12.8 258 11.7 26.3 12.5 65-74 53.2 10.0 42.0 11.3 38.7 8.7 334 135 32.5 11.3 30.3 9.7 75-84 61.9 6.2 55.7 6.6 45.5 8.9 480 10.5 45.5 10.5 438 18.0 85+ 61.9 6.2 61.3 6.9 50.3 10.3 50.2 154 528 189 536 127 FEMALES 15-24 27 1.8 23 1.3 4.2 3.8 4.9 3.3 4.6 2.3 49 24 25-34 52 2.6 5.8 3.0 9.0 5.7 7.8 5.4 75 4.1 7.7 46 35-44 8.2 2.0 8.1 3.0 13.0 3.7 10.1 4.1 9.1 4.6 95 4.2 45-54 10.5 3.5 10.9 3.1 135 3.7 11.6 29 10.2 2.8 11.4 25 55-64 10.7 1.1 10.9 3.0 12.3 2.0 9.9 3.8 9.1 2.3 9.4 29 65-74 10.6 1.9 8.8 2.3 9.6 29 7.8 2.6 7.0 1.7 7.3 3.0 75-84 8.4 24 9.4 1.3 72 1.7 6.7 25 5.7 1.4 5.5 1.0 85+ 8.9 24 6.1 6.1 3.2 5.0 1.0 5.8 3.7 1.8 SOURCE: National Center for Health Statistics: Health, United States, 1984. Table 3. 8pIoINS YINOA UO 82104 XS. S,Aie}e108Ss ay} Jo uoday F.M.Baker: Black Youth Suicide method of adjustment using the 1940 United States population) show a 57 percent in- crease in completed suicide by black males, a 4 percent increase in white males, a 47 per- cent increase in black females, and a 13 per- cent increase in white females. LITERATURE ON BLACK SUICIDE In 1938, Prudhomme (12,13) addressed the issue of suicide among blacks in the United States population during the pre-World War II period at a time of black migration and legalized segregation. He emphasized restricted economic opportunities, rural living, and group solidarity facilitated by racism as factors contributing to the lower rates of suicide in blacks. The Civil Rights movement began in 1954 and evolved into the Black Revolution of the 1960s. Several major national leaders were assassinated in the late 1960s including Dr. Martin Luther King, Jr. and Malcolm X. In 1969, Hendin (14) reported on his sample of 25 black suicide attempters in New York who were identified through hospitalization. He emphasized the role of tenement living and concluded that black suicide was precipitated by the frustrations of ghetto life, discrimina- tion by whites, and aberrant black family pat- terns. In this psychoanalytic study Hendin discussed the attempters’ families which were characterized by absent, physically violent fathers and mothers who were brutal or left their children in the hands of others who were brutal to them. Self-hatred and in- tense rage characterized these suicide at- tempters, particularly the black males. In 1978, Hendin (15) described the mixture of despair and violence that characterized the struggle of ghetto-residing black suicide attempters who represented the poorest socioeconomic group among the black population. Hendin looked at a sample of black college students who were part of a sample of black suicide attempters that he studied for five years. He found that al- though they lived with their parents, they were involved in a link of "emotional dead- ness" which bound them to their parents. These students were absorbed and preoc- cupied with their own extinction as an ongo- ing part of their adaptation and used work (dull, demanding mental labor) "as a way of maintaining a distanced, uninvolved state to conceal that they had no right to live." (16) The high rates of homicide among black males and the increasing rates of suicide noted through the 1970s stimulated many comments and studies in the literature which attempted to explain these rates in the con- text of the Black revolution. Wolfgang (17) reviewed Philadelphia police records and concluded that there was a disproportionate- 1. Includes deaths of nonresidents of the United States. groups. AGE-ADJUSTED DEATH RATES FOR COMPLETED SUICIDE PER 100,000 U.S. RESIDENT POPULATION 1950’ White males, all ages, age-adjusted? 18.1 Black males, all ages, age-adjusted? 7.0 White females, all ages, age-adjusted? 5.3 Black females, all ages, age-adjusted? 1.7 2. Age-adjusted by the direct method of the total population of the United States as enumerated in 1940, using 11 age SOURCE: National Center for Health Statistics: Health, United States, 1984. 1960" 1970 1979 1980 1981 1756 182 186 189 189 7.8 99 125 11.1 11.0 5.3 7.2 6.3 5.7 6.0 1.9 29 2.9 2.4 25 Table 4. 3-181 Report of the Secretary's Task Force on Youth Suicide ly high number of "victim-precipitated" homicides in "which the victims have acted in such way as to bring about their deaths at the hands of others, often by being the first to use or threaten physical violence." In 1970, Seiden (18) reviewed the stresses on young, urban, black men: 1) excessive and consistent unemployment, 2) the resulting incapability to be useful to others and com- petent to make their own way, and 3) the potential for an increase in suicides as job op- portunities open up in a period of rapid, forced, and unequal change (resulting from the impact of the Black Revolution upon society). Seiden suggested that the expecta- tions and hopes created more rapidly than they could be fulfilled, resulted in intensified frustrations and despair. Bagley and Greer (19) criticized the 1969 work of Seiden for its small sample size, the absence of controls, and commented that if patterns of suicide in blacks reflected black alienation, then black suicide rates in New York City should be higher than white rates for all ages. Although these authors iden- tified 25 "black" suicide attempters from a medically treated, emergency room (ER) sample, their study was not comparable to any other studies completed in the United States. In addition to reporting a sample which mixed two cases of completed suicide with 23 cases of attempted suicide, these authors had a broad definition of "black" (economically disadvantaged and oppressed) and included Africans, Caribbean, Indians, Pakistanis, and persons from Cyprus in their "black" sample. All were recent immigrants to England. Although they used a control group of white suicide attempters matched for age, sex, and marital status, the ap- plicability of their data to a black United States sample is questionable. Bagley and Greer found that 48 percent of "black" suicide attempters compared to 12 percent of white suicide attempters were diagnosed as an acute situational reaction. Only 8 percent of "black" cases compared to 24 percent of white controls had brain damage or were psychotic. Although none of the "black" 3-182 suicide attempters were diagnosed as sociopaths or addicted to drugs, 12 percent of white controls were. "Black" suicide at- tempters in this sample were younger than the white controls. In 1973, Pederson, Awad, and Kindler (2) reported a sample of suicide attempters iden- tified from the Monroe County Psychiatric Case Register. The Case Register recorded all psychiatric contacts including ER visits, public and private psychiatric hospitaliza- tions, and visits to private psychiatrists. From 1964 through 1967, 1345 persons were seen as the result of a suicide attempt, an average of 336 suicide attempts per year representing 0.8 percent of all ER visits. Nonwhite (predominately black) and white attempters were compared. Nonwhite attempters were younger; 47 per- cent of nonwhite attempters were of ages 15- 24 compared to 36 percent of the white suicide attempters. In all age cohorts for this 1960s sample, the attempt rates for nonwhite females were higher than those of nonwhite males by an average of 4-5 percent. When nonwhite females were compared with white females, only in the 45 and older age group- ings were the percentage of nonwhite at- tempts less (5%) than the white attempts (15%). The majority (84%) of nonwhite suicide attempters were in the lowest socioeconomic group. In this sample of suicide attempters, 67 percent of white females and 78 percent of white males had some prior psychiatric contact (not specified). This contrasted with 49 percent of nonwhite females and 50 percent of non- white males having some prior psychiatric contact (not specified). When marital status was assessed, 28 percent of nonwhite females were separated compared to 9 percent of white females. Fifty-four percent of the non- white males were single, significantly greater than any other group. When the diagnoses of this Rochester, New York, sample of suicide attempters were reviewed, 25 percent of white male at- tempters and 13 percent of white female at- tempters were diagnosed as psychotic in F.M.Baker: Black Youth Suicide contrast to only 10 percent of nonwhite male and female attempters. Thirty-six percent of nonwhite male attempters in comparison to all other groups were diagnosed as having a neurosis. In this sample, the ratio of non- white male to nonwhite female attempters was 1:6 in contrast to a ratio of 1:3 among white suicide attempters. Only in this study were suicide attempters followed lon- gitudinally from 1964 through 1968. Thirty- five attempters completed suicide by 1968; 31 were white and 4 were nonwhite. Of the white suicide attempters who completed suicide, 16 were male and 15 were female. In the nonwhite sample, 1 male and 3 female suicide attempters completed suicide. The white suicide rate for Monroe County (Rochester, New York) was calculated as 10.51 per 100,000 and for nonwhite as 8.98 per 100,000 per year. In 1974, a case-control study of suicide at- tempters was published by Stein, Levy, and Glasberg (21). White and black suicide at- tempters were identified from psychiatric ad- missions to a large municipal hospital in New York City. White and black controls were matched for age (within 3 years), race, and the time of admission nearest to that of the suicide attempters. The controls were hospi- talized, psychiatric patients who denied suicidal ideation and had no recent or past suicide attempts. The authors focused on the role of a history of separation from close figures in the individual’s life as a risk factor for attempting suicide. They defined childhood separation as "a physical separa- tion from a parent, parent surrogate, or sib- ling, of six months duration occurring from birth to 17 years of age." Early childhood separation was defined as a separation occur- ring from birth to 7 years of age. The 48 white female suicide attempters and controls, the 49 white male suicide attempters and con- trols, the 48 black female suicide attempters and controls, and the 20 black male suicide attempters and controls comprised the total sample of 330. Black male and female suicide attempters in this study were younger than white suicide attempters; the mean age for males was 24.1 and for female, 25.6. Using education as an index of social class, all suicide attempters and controls in this study were of the same, lower social class. The authors found that only white male and white female suicide attempters had a greater num- ber of childhood and antecedent separations than controls. Black male and black female suicide attempters had significantly more (p<.100) early childhood separations com- pared to black controls. In this sample, black female suicide attempters had significantly more (p<.001) separations between 7-17 years of age than the controls. Further, a his- tory of antecedent separations was highest among white female suicide attempters fol- lowed by black female and black male suicide attempters, and, lastly, white male suicide at- tempters. When interaction between an- tecedent and childhood separation and suicide attempt was assessed, this study found no significant interaction. These authors suggested that the interaction between suicide attempts and antecedent and childhood separations may involve a variety of factors including type of separation ex- periences prior to and following separation, and the effect of threatened or psychological separations. Finally, the authors questioned whether childhood separation predisposed to maladaptive responses to separations in the adult. The role of the extended family (22) and the "adoption" mechanism of black families (23) were not addressed. Monk and Warshauer (24) compared com- pleted and attempted suicides in three ethnic groups in New York City for the years 1968 through 1970 for completed suicides and June 1971 through June 1972 for attempted suicides evaluated by the hospitals serving the target populations in East Harlem. Ninety-six suicides were completed between 1968 and 1970 and 359 suicide attempters were evaluated between 1971 and 1972. When age-adjusted suicide rates were com- piled for New York City for the 1960-1961 period, black males had a higher rate, 20.7 per 100,000 for persons age 15 and over, compared to a white rate of 17.1 per 100,000 for persons age 15 and over. Similar figures 3-183 Report of the Secretary's Task Force on Youth Suicide for the period of 1967-1968 found a black, age-adjusted rate of 16.8 and a white age-ad- justed rate of 18.5. The authors suggested that part of the difference in white and black suicide rates reported elsewhere could reflect differential reporting in the classifica- tion of deaths for the two groups. Specific problems with suicide statistics were discussed by Warshauer and Monk in 1978 (25) in a subsequent paper based on this study. Data on deaths from four New York City Health Districts with a significant ethnic minority population were compared with the records of the Office of the Chief Medical ex- aminer. Reports of suicidal deaths were received by the Health Department from the Office of the Chief Medical Examiner and were classified according to the Internation- al Classification of Disease (ICD) codes. The Office of the Chief Medical Examiner of New York City divided deaths into definite suicides and assigned suicides (deaths which could not be signed out as suicide) and deaths shown to be suicide upon investigation, but which were not signed out as such because no final determination was requested. As the unconfirmed black suicides in this sample used unusual methods twice as often as whites, blacks were classified as assigned suicides and did not appear in the Health Department statistics which were forwarded to the National Center for Health Statistics. Because of incomplete histories and less fre- quently used methods such as jumping, all black suicides were not classified in the definite suicide category. A further factor in- validating the New York City suicide figures was the impact of the change in the ICD coding which resulted in the failure to code “deaths which had been classified as suicide before the change in ICD coding. The change differentially affected black suicides. Thus, these authors demonstrated how the statistics on completed suicides by blacks could be underreported, locally and nation- ally. As assigned suicides may not be finally categorized until the toxicology report was returned, these cases would be delayed and the correct figures would not be reported to the Health Department. The extent to which 3-184 these problems exist in other municipalities across the nation is unclear. Lester and Beck (26) reported a sample of 124 white and 115 black suicide attempters admitted to a metropolitan Philadelphia hospital. Only subjects ages 40 and younger were included for comparison with the prior work of Hendin. Each suicide attempter was seen within 48 hours of admission by an ex- perienced clinician to obtain a history, clini- cal evaluation, and to review the patient’s state of mind. In a second interview a psychological technician obtained a detailed psychosocial history and administered the Beck Depression Inventory and the General- ized Expectancies Scale. When white and black suicide attempters were compared, five significant differences were identified from the 54 tests: black males were more likely to be Protestant, less likely to be living with others, more often had been separated from their fathers, scored lower on a test of vocabulary, and (if diagnosed as schizophrenic) were more likely to be diag- nosed as paranoid. When white and black female suicide attempters were compared, significant differences were identified in 12 tests: black females had not completed as many grades, had more unofficial marital ar- rangements (cohabitation and separation versus marriage and divorce), had worse physical health, used more alcohol, had made fewer previous suicide attempts, were more often Protestant, lived more often in a low rent district, had experienced more separa- tions from their fathers, and (if separated from their mothers) had experienced the separation at an earlier age. Black female at- tempters had poorer vocabulary scores, lower suicidal intent scores, and (if diagnosed as schizophrenic) were more likely to be diag- nosed as paranoid. When black male and black female attempters were compared, black females were more often living with others, more often unemployed, more likely to attempt suicide at home, and were less psychiatrically disturbed. The authors noted that the lower educational level of blacks, the differences in religious affiliation, and the high incidence of absent fathers among black F.M.Baker: Black Youth Suicide suicide attempters was probably reflective of socioeconomic status. Lester Black con- cluded that they did not find evidence of black self-hatred and rage in the psychologi- cal measures that were used in their Philadel- phia sample. The authors summarized their results as showing more similarities between black and white suicide attempters than dif- ferences. In 1976, Kiev and Anumonye compared black suicide attempters in Newark, New Jer- sey with a sample of white suicide attempters in New York City (27). They reported a male to female ratio of 1:1, higher than ratios reported by studies in other settings. Al- though these authors commented about sig- nificant alcohol abuse by black suicide attempters in their sample, this was not quan- tified. In 1977, Steele reported a sample of 275 suicide attempters who were identified from the ER of a general hospital in New Haven, Connecticut (28). Twenty-two percent (N=62) of these suicide attempters were black. Overall, this author found few dif- ferences between black and white attempters on 42 variables assessing mood, motivation, etc. Although white suicide attempters ap- peared to be more motivated to influence others by their suicide attempt, were more depressed, and tended to show more deliberation in their suicide attempts, the clinical significance of these statistical dif- ferences was questioned. Although white suicide attempters were found to be more deliberate in their attempts than black at- tempters, both groups were impulsive and the majority of both groups deliberated about their suicide attempt for only an hour or less. In view of these findings, Steele ques- tioned whether a separate psychology was needed for black suicide attempters and stressed the need to review the belief in men- tal health circles that blacks were less likely to engage in suicidal behavior. Baker (29) conducted a descriptive study of black suicide attempters evaluated in an ER setting in a New Haven general hospital which she contrasted with prior studies in this setting. Her 1980 sample of 56 black suicide attempters was compared with prior studies of suicide attempters in this setting by Steele, (28), Weissman, Pakal, and French (30), Weissman (31), and Fox and Weissman (32). Previous studies in this ER setting described a population of suicide attempters who were predominantly white, single females who im- pulsively took an overdose in the context of an argument with a significant other. Al- though diagnoses were not reported in Weissman’s three studies, the attempters studied had no prior psychiatric history. In 1980, blacks comprised 18 percent of the total suicide attempter population of 315. The 1980 sample of black suicide attempters was significantly different from those of pre- vious studies. Sixty-four percent of the black female attempters had a prior psychiatric his- tory; 54 percent had made a previous suicide attempt or gesture. Their primary diagnoses were 33 percent with affective illness and 31 percent with adjustment disorders with depressive features. Only 20 percent of black females had used alcohol prior to their attempt. In the 1980 New Haven sample, black male suicide attempters were marked- ly different from prior studies in this setting. Seventy-six percent had a previous psychiatric history, 35 percent had made a previous suicidal gesture or attempt, and 59 percent were diagnosed as psychotic (bipolar or schizophrenic). Twenty- nine percent of the back male suicide attempters had used al- cohol prior to their attempt. The primary method of attempt by the black suicide attempters in the 1980 study was drug overdose: 74 percent in black females and 71 percent in black males. The agents used had changed to include not only sedative-hyp- notic medications (methaqualone) and anti- anxiety agents (diazepam and chlordiazepoxide), but also over-the-counter medications (Sominex, Mydal, Nytol, Humphrey’s 11), and prescription medica- tion (insulin, penicillin G procaine, furosemide). All attempters were in Hol- lingshead-Redlich social class IV and V, the lowest socioeconomic groupings. The male 3-185 Report of the Secretary’s Task Force on Youth Suicide to female ratio in this sample was 1:2.3. In contrast to other studies of suicide at- tempters, Baker focused on the person who accompanied the attempter or came later to the ER. Sixty-four percent of black female and 35 percent of black male attempters were accompanied to the ER by a family member, usually the mother, a sibling or spouse, or children in declining order of frequency. In each case, the accompanying significant other was involved in or was aware of the psychosocial stressors that precipitated the suicidal act. Of the 31 patients referred to outpatient treatment, 22 percent entered treatment, 10 percent made one appoint- ment and dropped out, and 55 percent did not followup on their referral. In a 1968 sample of New Haven suicide attempters, Paykel, Hallowell, and Dressler (33) reported that of 38 percent of their sample of white and nonwhite attempters who were referred for outpatient treatment, only 16 percent showed up for their appointment. Baker suggested that greater focus on couple or family crisis intervention in the ER with the attempter and the accompanying per- son(s) could have two important benefits. First, it could identify outpatient psychiatric resources that could be used for future con- flict resolution so that the person would not have to attempt suicide to communicate dis- tress. Second, involvement of the significant other(s) at the time of initial ER or crisis cen- ter contact could facilitate the entry of the at- tempter into outpatient treatment. THEORIES OF BLACK SUICIDE Before turning to a discussion of preventive strategies, let us review the various theories developed to explain black suicide. In 1897, (34) Durkheim discussed the sociological dimensions of suicide. He related the rising suicide rate in the civilized world to a func- tional failure of State, church, and com- munity as the forces for social integration that they had been prior to the Industrial Revolution. Durkheim saw vulnerability to suicide as existing in people who were not in- tegrated into any religious, communal, or 3-186 family group. Even more vulnerable were in- dividuals who suffered a disturbance in the balance of their social integration: the single, widowed, and divorced having higher suicide rates than the married. Hendin (15) pointed out that Durkheim’s theory did not explain the high rates of suicide in Austria, a Catholic country. Nor did it explain the strikingly high suicide rates in Denmark and Sweden compared to the low suicide rate in Norway. He emphasized the need for a psychosocial approach to un- derstanding the differences in suicidal be- havior across cultures. Hendin stated that the Freudian construct (35) which sees suicide as a response to loss or abandonment of a loved object as insufficient in itself. Psychodynamically, rebirth, return, or reunion fantasies with the lost object may be seen as an attempt to undo or deny loss. Freud's instinctual frame of reference did not lead him to be concerned with the psychological impact of the social institutions of particular cultures or with psychosocial questions such as why suicide was very high in one country and low in another. Hendin suggested that more than an amalgam of Freud and Durkheim was needed to under- stand the varying motivations for suicide in different cultures and subcultures, the dif- ferences between genders and different age groups, and differences in ways of coping with love and loss, life, and death. More recent theories address the current context of black suicide attempters. Specific explanations of black suicide include: 1) urban stress, 2) the status-integration theory, 3) the black family deficit theory, and the 4) external restraint theory. The urban stress (frustration-aggression) hypothesis (Seiden, 16) proposes that compounded urban stres- ses associated with migration, poverty, un- employment, racism, poor housing, and poor education result in violence which often, though not always, takes the form of suicide. The status-integration theory suggests that as blacks work their way into the middle and upper-middle classes they inherit the economic, social, and psychological tensions F.M.Baker: Black Youth Suicide of their white counterparts. Davis (36) sug- gests that the more upwardly mobile blacks are, the more intense are the problems of ad- justment and assimilation into the American mainstream. These tensions produce a cor- roding sense of internal alienation which may result in self-destruction. The absence of higher suicide rates in black females provides some evidence against this theory. But, the greater probability of private care for this population may prevent its identification from ER statistics. A further bias involving completed suicide in middle-class black per- sons may be a collusion to prevent a diagnosis of suicide by the medical examiners office. These are methodological concerns in all studies of suicide attempters. The black family deficit theory presents the black family as being unable to meet the fun- damental needs of its members for survival, socialization, and the transmission of a viable cultural heritage. Pinderhughes (37), in dis- cussing the impact of poverty and racism on black families, described the result as the "vic- tim system." The work of Lewis and Looney (38) illustrated that working-class black families, just above the poverty level, were well-functioning units. These authors sug- gest that sufficient economic insecurity can destablize and then perpetuate dysfunction- al patterns of family function. The external restraint theory of Henry and Short (39) suggested that suicide varied in- versely with horizontal restraining factors (social relationships with others) and vertical restraining factors (social class and/or social status). These authors and Maris (40) sug- gested that the strength of the relational sys- tem of the individual defined by marital status, urban-rural residence, and ecological distribution serve as buffers; the stronger the relational system, the lower the number of suicides. Davis (36) viewed this external restraint theory as more relevant to black suicide. He questioned whether the decrease in overt racism and discrimination, which in the past had fostered group solidarity, as noted by Prudhomme (12,13), would decrease the strength of the relation- al system for young black persons in the 1980s. If this did occur, family ties would be left as the major insulation against the psychosocial stressors of daily living for the urban resident black youth. By this theory, if the family was disorganized, over-stressed, and/or dysfuncitonal, the black youth would have no ameliorating or buffering factors and would become a higher risk for suicidal be- havior. SUMMARY OF THE LITERATURE This review of the literature on black suicide has shown some divergent results across a variety of studies in several different cities. Several points of consensus as well as specific points of intervention follow. 1. Black suicide rates peak for both sexes be- tween ages 25 and 44. 2. Black suicide rates decline in both sexes after age 45. 3. Black males have a "double risk of death" due to their high rates of death from homicide and suicide. 4. Rates of completed suicide have remained consistently low in black females in com- parison to all other groups. 5. During a 31 year period the increase in the race-sex specific rates of completed suicide have been highest among black males (57% increase) and in black females (47% in- crease) in age-adjusted national statistics. 6. In specific States in specific years, suicide rates for black males exceed those for white males and white females in contrast to the overall national statistics. 7. Evidence of municipal and regional dif- ferences in the rates of black suicide attempts and completed suicides is provided by the Los Angeles data of Christian (41) and the data of Pederson et all. (20) from Rochester (NY), the only studies which reported higher rates of suicide attempts in black females compared with white females. 3-187 Report of the Secretary's Task Force on Youth Suicide 8. The accuracy of national suicide statistics may be flawed if black suicides are not coded as suicide due to administrative procedures, the use of alternative methods of suicide, and delayed reporting due to pending laboratory studies. 9. Only theories of suicide which have em- phasized interpersonal conflicts, familiar dis- cord, financial concerns, and the impact of poverty and racism upon the individual and the family, have stood the test of time and repeated observations in suggesting specific etiologies of suicide attempts and completed suicide by Afro-Americans. 10. Initiating crisis intervention techniques which include the person who accompanies the suicide attempter to the ER or crisis set- ting may: a) improve followup on referral to outpatient treatment and b) prevent future suicide attempts by making everyone in the attempter’s network more sensitive to the cues of distress in the system and aware of the resources to use to seek help. 11. Although the literature suggests that suicide attempters and persons who com- plete suicide are different populations, there have been no published studies to date that have contrasted black suicide attempters and blacks who completed suicide in the same geographically defined area in the same defined time period. 12. The population of black suicide at- tempters evaluated in some ERs in 1980 ap- pears to be changing: an increased proportion of attempters have a history of psychiatric treatment and a greater severity of psychiatric diagnoses. : PREVENTIVE STRATEGIES Three levels of prevention are discussed in textbooks (42). Primary prevention involves actions which prevent disease as exemplified by vaccination. Secondary prevention focuses on the treatment of illness, e.g. hospitalization of a psychotically disor- ganized individual. Tertiary prevention focuses on the rehabilitation of an individual 3-188 to facilitate return to productive function, e.g. postventive work with the surviving fami- ly members of a person who has completed suicide. Primary preventive strategies involving black suicidal youth should focus upon conflict resolution in the family and the clarification of expectations in various relationships. Helping black adolescents and youth to un- derstand the factors that they can control and the factors which are controlled by society may aid in clarifying the sources of frustration in the 1980s. As noted by Davis (43), high "in-group" stress may result from an individual’s family relationships, friendships, and personal relationships. The "extra group" stresses resulting from work relation- ships with other people and financial difficul- ties may be modified by support from the extended family and community groups such as churches or social clubs. Aiding black youth in sorting out the locus of stress and identifying effective action, should prevent a build-up of frustration to the point of impul- sive action. These educational activities could be centered in schools, churches, and adolescent drop-in centers and could be sponsored by black church groups, black businesses, and black fraternal organizations. Although New Haven represents only one study site, the population of suicide at- tempters has been studied over a 32-year period. With deinstitutionalization and the increased utilization of community mental health centers and general hospitals for in- patient psychiatric treatment, the population of psychiatric patients coming to the general hospital ER has included an increasing proportion of patients with prior psychiatric histories. Patients with affective illness (Tsuang, 44, Tsuang and Woolson, 45), schizophrenia (Roy, 46, Brier and Astrachan, 47), and substance abuse--particularly al- cohol (Motto, 48, Murphy, Armstrong, and Hermele, 49) are at increased risk for at- tempting suicide. Informational programs organized for the families of patients with psychiatric illness are important primary preventive strategies. Improving the family’s F.M.Baker: Black Youth Suicide knowledge of symptoms and behavioral changes characteristic of psychotic decom- pensation will enable the family to seek help before a frank psychotic episode occurs. The distressed and distraught youth can be brought to a crisis intervention setting for help before a suicide attempt signals that something is wrong. Another primary preventive strategy invol- ves the removal of all out-of-date and un- necessary medication from the home. As overdose remains a major method of suicide attempt for some black populations, remov- ing medications from the home would decrease the availability of medications to a distraught individual making an impulsive act. Prescribing non-lethal amounts of medication to patients in active psychiatric treatment who have diagnoses associated with an increased risk of suicide would be helpful, also. Secondary preventive strategies should focus upon the initial evaluation of the black suicide attempter, whether in the ER of a general hospital or the crisis unit of a com- munity mental health center. As noted by Baker (29), the person who accompanies the suicide attempter or comes later to the ER is usually aware of and involved in the events which precipitated the suicidal act. By in- itiating family and/or couple crisis interven- tion at the time of initial evaluation, conflict resolution can be facilitated and the pos- sibility of followup and entry into treatment can be improved. Further, the suicide at- tempter and his/her family can be provided with alternatives for help at a time of future crisis to prevent future suicide attempts through de-escalation of crises by the family seeking help and services at a crisis facility. Initially, the presistantly suicidal patient who presents repeatedly with suicidal gestures/at- tempts and does not follow through on refer- rals to treatment should be involuntarily hospitalized. Involuntary psychiatric hospitalization is recommended in order to break the cycle and initiate the treatment process. Tertiary preventive strategies have focused on the surviving significant others of the at- tempter. Fortunately the literature in this area is increasing (50,51,52). FUTURE RESEARCH DIRECTIONS As described in the above literature review, various studies have looked at suicide at- tempters and at those who complete suicide, sometimes in the same study. Only one case- control study compared black and white suicide attempters and childhood separation and only one study looked at the precipitat- ing events as well as the persons who accom- panied the attempter to the hospital. In order to detect the possibility of changing patterns in black suicide attempters which are indicated by increased psychiatric history, increased severity of psychiatric diagnoses, and the presence of psychosocial stressors such as those existing in the 1980s, more care- fully designed studies are needed. Although studies of black suicide attempters and completers have been conducted in New York City, Rochester (NY), Los Angeles, Detroit, Philadelphia, and New Haven, they were conducted in different years, they as- sessed different populations of suicide at- tempters, and emphasized the collection of different data. I suggest that the research ef- forts include a multi-site, prospective, descriptive study of black suicide attempters in at least five cities with large black popula- tions, to be completed in the same calendar year, with a sample selected by the same method in each site. The research protocol should collect data on: demographics, method of attempt, prior history of suicide at- tempt, history of prior ER contacts, history of prior psychiatric treatment, specific precipitating events, evidence of substance use, the person who accompanied the suicide attempter to the crisis setting, information about self-perception, family constellation, and the network of interpersonal and social relationships--particularly those changing or stressed in the period before the attempt. From this established data base, similarities and differences across the nation could be 3-189 Report of the Secretary’s Task Force on Youth Suicide identified. Subsequently, case-control studies using the identified sample of suicide attempters in the five cities could define specific risk factors and determine whether these risk factors varied in the five cities. Completed suicides in the same five cities could be reviewed for the same study period. Psychological autopsies as well as proxy in- terviews of the surviving next-of-kin to assess the precipitating events, current relation- ships, and self-image of the black persons who completed a suicide woud aid in clarify- ing the specific differences between the at- tempter and completer suicide populations. Where possible, an attempt should be made to identify suicide attempters and completers in all socioeconomic classes. CONCLUSION Specific strategies which can be implemented at the local/community level by societal in- stitutions and through federal research initia- tives have been suggested to address the national problem of black suicide. The im- plementation of these strategies has the potential to save lives and provide important data from which to develop future, more specific, preventive strategies. REFERENGCES -++-+--veeseesesersmssnnnsnnnnnnas - 1. National Center for Health Statistics. Health, United States, 1984. DHHS Publication (PHS) 85-1232. Washington, D.C.: U.S. Government Printing Office, 1984. 2. U.S. Department of Health and Human Services. Report of the Secretary's Task Force on Youth Suicide. Volume II: Risk Factors in Youth Suicide. Washington, D.C.: U.S. Government Printing Office, 1987. 3. United States Census. Detailed population characteristics - U.S. Summary Section A. 1980. 4. Davis R. Black suicide in the seventies: current trends. Suicide and Life-Threatening Behavior 1979; 9 (3):131-140. 5. Frederick G.J. Current trends in suicidal behavior in the United States. American Journal of Psychotherapy 1978; 32:172-200. 6. Seiden R.H. Mellowing with age: Factors influenc- ing the non-white suicide rate. International Journal of Aging and Human Development 1981; 13:265-284. 7. Hendin H. Black suicide. Archives of General Psychiatry 1969; 21: 407-422. 8. Poussaint A.L. Black suicide. In: Williams RA., (ed.) Textbook of Black-Related Diseases. New York: McGraw-Hill Book Company, 1975; 708-714. 9. Baker F.M. Black and white alcohol users in an emergency room setting: Implications for treatment. In: Brisbane F.L., Womble M., (eds.) Treatment of Black Al- 3-190 coholics. New York: The Haworth Press: 115-128. 10. U.S. Department of Health and Human Services. Report of the Secretary's Task Force on Black & Minority Health. Volume V: Homicide, Suicide, and Unintentional ijunes, Washington, D.C.: U.S. Government Printing Of- ice, 1986. 11. National Center for Health Statistics. Health, United States, 1984. DHHS Publication No. (PHS) 85- 1232. Washington, D.C.: U.S. Government Printing Of- fice, 1984. 12. Prudhomme C. The problem of suicide in the American Negro. Psychoanalytic Review 1938; 25:187- 204. 13. Prudhomme C. The problem of suicide in the American Negro. Psychoanalytic Review 1938; 25:327- 391. 14. Hendin H. Black suicide. Archives of General Psychiatry 1969; 21: 407-422. 15. Hendin H. Suicide: The psychosocial dimension. Suicide and Life-Threatening Behavior 1978; 8 (2):99-117. 16. Ibid 17. Wolfgang M. Victim-precipitated violence. Psychology Today; 1969, October. 18. Seiden R.H. We're driving young blacks to suicide. Psychology Today; 1970, August; 4 (3):24-28. 19. Bagley C., Greer S. "Black suicide:" Areport of 25 English cases and controls. Journal of Social Psychology 1972; 86 (2):175-179. 20. Pederson A.M., Awad G.A., Kindler A.R. Epidemiological differences between white and nonwhite suicide attempters. American Journal of Psychiatry 1973; 130 (10):1071-1076. 21. Stein M., Levy M.T., Glasberg H.M. Separations in black and white suicide attempters. Archives of General Psychiatry 1974; 31: 815-821. 22. Bass B.A, Wyatt G.E., Powell G.E., (eds.) The Afro- American Family: Assessment, Treatment, and Research Issues. New York: Grunet Stratton, 1982. 23. Martin E.P., Martin, J.M. The Black Extended Family. Chicago: The University of Chicago Press, 1978. 24. Monk M. Warshauer M.E. Completed and at- tempted suicide in three ethnic groups. American Journal of Epidemiology 1974; 100 (4): 333-345. 25. Warshauer M.E., Monk M. Problems in suicide statistics for whites and blacks. American Journal of Public Health 1978; 68 (4):383-388. 26. Lester D., Beck A.T. Racial background and suicidal behavior. Psychology 1975; 12 (2):3-5. 27. Kiev A. Anumonye A. Suicidal behavior in a black hetto. International Journal of Mental Health 1976; 5 2):50-59. 28. Steele R.E. Clinical comparison of black and white suicide attempters. Journal of Consulting and Clinical Psychology 1977; 45 (6):982-986. 29. Baker F. M. Black suicide attempters in 1980: A preventive focus. General Hospital Psychiatry 1984; 6:131-137. 30. Weissman M.M., Paykel E.S., French N., Mark H., Fox K., Prusoff B. Suicide attempts in an urban com- munity, 1955 and 1970. Social Psychiatry 1973; 8:82-91. 31. Weissman M.M. The Spi9enioiony of suicide at- tempts, 1960-1971. Archives of General Psychiatry 1974; 30:737-746. 32. Fox K., Weissman M. Suicide attempts and drugs; contradictions between method and intent. Social Psychiatry 1975; 10:31-38. 33. Paykel E.S., Hallowell C., Dressler D.M,, et al. Treatment of suicide attempters. Archives of General Psychiatry 1974; 31: 487-491. F.M.Baker: Black Youth Suicide 34. Durkeim E. Suicide: A study in sociology. New York: Free Press, 1951. 35. Freud S. Mourning and melancholia. In: Collected Papers. Vol IV. New York: Basic Books, 1917:152-170. 36. Davis R. Suicide among young blacks: Trends and perspectives. Phylon 1980; 41 (3):223-229. 37. Pinderhughes E. Afro-American families and the victim system. In: McGoldrick M., Pearce J.K., Giordano J., (eds.) Ethnicity and Family Therapy. New York: The Guildford Press, 1982:108-121. 38. Lewis J.M., Looney J.G. The long struggle: Well- functioning working-class black families. New York: Brunner/Mazel, 1983. 39. Henly AF. Short J.F. Suicide and Homicide. New York: Free Press, 1954. 40. Maris R.W. Social forces in urban suicide. Homewood, lllinois: Dorsey Press, 1969. 41. Christian E.R. Black suicide. In: Hatton C.L., Valente S.M., Bink A., (eds.) Suicide: Assessment and In- ferveniion. New York: Appleton-Century-Croft, 1877: 143-159. 42. Langgley D.G. Prevention in Pyshishy: Primary, secondary, and tertiary. In: Kaplan H.l., Sadock B.J., (eds.) Comprehensive Textbook of Psychiatry. IV Ed. Bal- timore, Maryland: Williams & Wilkins, 1985:1885-1888. 43. Davis R. Black suicide and social support systems: An overview and some implications for mental health prac- titioners. Phylon 1982; 43 (4):307-314. 44, Tsuang M.T. Suicide in schizophrenia, manics, depressives, and surgical controls - A comparison with eneral population suicide mortality. Archives of General ychiatry 1978; 35:153-155. 45. Tsuang M.Y., Woolson R.F. Excess mortality in schizophrenia and affective disorders - Do suicides and accidental deaths solely account for this excess? Archives of General Psychiatry 1978; 35: 1181-1185. 46. Roy A. Suicide in chronic schizophrenia. British Journal of Psychiatry 1982; 141:171-177. 47. Breier A., Astrachan B.M. Characterization of schizophrenic patients who commit suicide. American Journal of Psychiatry 1984; 141: 206-209. 48. Motto J.A. Suicide risk factors in alcohol abuse. Suicide and Life-Threatening Behavior 1980; 10:230-238. 49. Murphy G.E., Armstrong J.W., Harmele S.L. Suicide and alcoholism. Archives of General Psychiatry 1979; 36:65-69. 50. Shneidman E.S. Postvention and the suicide-vic- tim. In: Shneidman E.S., (ed.) Death: Current Perspec- tives. Palo Alto, California: Mayfield Publishing, 1984; 412-419. 51. Bolton |. Families coping with suicide. In: Hansen J.C., Frantz T.T., (eds.) Death and Grief in the Family. Rockville, Maryland: Aspen, 1984; 35-47. 52. Godney R.D. Survivor-victims and crisis care. Crisis 1985; Crisis 6:1-9. 3-191 Report of the Secretary's Task Force on Youth Suicide Rates of Completed Suicide by Sex and Age per 100,000 Resident Population 30 C ——— \\hite Ages 15-24 - Black 25 wm - White Males lL BM Black Males lL WF White Females L BF Black Females / 20 5 L s | j= - e 15k x L - f= << L oc L 10 - Sf 0 Figure 1. Rates of Completed Suicide by Sex and Age per 100,000 Resident Population Ages 25-34 30 — — — White Black = ite es 7” % = J ss wa WF = White Females BF = Black Females ~~ 20 RATE/100,000 a —_ o rrr Terr rrr rT Tr TT TTT TT Figure 2. 3-192 F.M.Baker: Black Youth Suicide Rates of Completed Suicide by Sex and Age per 100,000 Resident Population 30 Ages 35-44 B — ee \N Hite - WM = White Males + Black BM = Black Males [= WF = White Females 25 r BF = Black Females - WM ee —— Nt ——— 20 Ss i ™~ 15 oN C - l- < L nw = 10 ~ 5p 0 [ ] I ] | 1950 1960 1970 1980 YEAR Figure 3. Rates of Completed Suicide by Sex and Age _ per 100,000 U.S. Resident Population Go Ages 45-54 — me we \/ hhite 50 — Black [WM = White Males L BM = Black Males S - WF = White Females 40+ = Black Females 8 = i IW, ris hd ol TTS —— I 30 " ~~—w L SS — 20 L BM _— BE — ——— tof wo [ BF 0 C l 1 1 1 1950 1960 1970 1980 YEAR Figure 4. 3-193 Report of the Secretary’s Task Force on Youth Suicide } Rates of Completed Suicide by Sex and Age sol per 100,000 U.S. Resident Population i Ages 55-64 - — vn en \\/ Nit E 50 Black te WM i ~~ [WM = White Males ~~~ _ 40+ BM = Black Males ~~ L WF = White Females ~~ 8 - BF = Black Females - ~~ 3 - a 8 30 IN ju _ No << L oc L 20 = BM B WF — — — — 10 —_—— —-— i BF 0 — 1 I ] 1950 1960 1970 1980 YEAR Figure 5. } Rates of Completed Suicde by Sex and Age dor per 100,000 U.S. Resident Population - Ages 65-74 [ WM — — — White - ~N S50 WM = White Males Black - BM = Black Males NC - WF = White Females ~~ - BF = Black Females ~~ 40 ——— —_ I~ ~~ g8 | TS 2 wl ~~ = 30 IT be - << - [os 5 20 — : WF 0 Nm — mm i BF 0 1 1 1 1 1950 1960 1970 1980 YEAR Figure 6. 3-194 F.M.Baker: Black Youth Suicide Rates of Completed Suicide by Sex and Age per 100,000 U.S. Resident Population Ages 75-84 3 ~ WM 6p — = — White TS ——— Black “Sea ~ 50 wm - White Males Sa BM = Black Males RN eee eee ee WF = White Females BF Black Females 40 g L hd o e 30+ 5 L cd L E-¢ lL oc L 20+ 10} 0 + YEAR Figure 7. Rates of Completed Suicide by Sex and Age per 100,000 U.S. Resident Population Ages 85 and over WM Tm ~ 80 r ——— — White IN - ~ | Black NG —— 50 » WM = White Males Soe | BM = Black Males - WF = White Females - BF Black Females 40 — 8 L S oo - = 30 wo << - oc lL 20 = 10 0 BF 1 1 1 1950 1960 1970 1980 YEAR Figure 8. 3-195 HISPANIC SUICIDE IN THE SOUTHWEST, 1280-1982 Jack C. Smith, M.S., Senior Statistical Consultant, Division of Reproductive Health, Center for Health Promotion and Education, Centers for Disease Control, Atlanta, Georgia James A. Mercy, Ph.D., Chief, Intentional Injuries Section, Epidemiology Branch, Division of Injury Epidemiology and Control, Center for Environmental Health and Injury Control, Centers for Disease Control, Atlanta, Georgia Mark L. Rosenberg, M.D., M.P.P., Assistant Director for Science, Division of Injury Epidemiology and Control, Center for Environmental Health and Injury Control, Centers for Disease Control, Atlanta, Georgia SUMMARY Little is known about suicide among Hispanic Americans. We studied suicides among Hispanics of Mexican origin (Mexican Americans) in five southwestern States (Arizona, California, Colorado, New Mexico, and Texas) where more than 60 per- cent of all Hispanics in the United States reside (86% of whom are Mexican American). We obtained data on the num- ber of suicide deaths in the white population with Hispanics and Anglos (white, non- Hispanic) identified separately. Suicides of Anglos were used as a comparison group. The suicide rate for whites in the five south- western States was almost one-fourth higher than the rate for whites nationally. Suicide rates for the two ethnic groups, however, showed the rate for Hispanics to be less than the national rate for whites and one-half that of Anglos residing in the same area. The lower suicide rate for Hispanics relative to Anglos is seen for both males and females. The ratio of male to female suicides for Hispanics was greater than for Anglos (4.3 to 1 for Hispanics and 2.7 to 1 for Anglos). Proportionately more suicides occur among young Hispanics than among young Anglos. 3-196 More than one-third (34.6%) of Hispanic suicides occurred to persons less than 25 years of age compared to one-sixth (16.5%) of Anglo suicides. The 1990 health objective for the United States for suicide identifies young persons 15 to 24 years of age as the population on whom to focus national suicide prevention and intervention efforts. This objective is appropriate for the Hispanic population since, from our findings, the highest suicide rates for Hispanics are in the 20-24 year age group (17.1%). INTRODUCTION In 1978, suicide was the ninth leading cause of death for the white population in the United States with a rate of 13.4 deaths per 100,000 population (1). Although much suicide data for whites have been collected, analyzed, and reported in official publica- tions and scientific literature (2,3,4), very few studies have examined suicides specifically among Hispanics, the largest ethnic sub- group within the white population (5,6,7,8). Furthermore, no study has compiled a large data set to analyze and compare suicide J.C.Smith: Hispanic Suicide in the Southwest, 1980-1982 among Hispanics of Mexican origin with Anglos (white, non-Hispanics) living in the same geographic area. This paper updates data from a previously reported study (9) of suicide among Anglos and Hispanics in five southwestern States where more than 60 percent of the nation’s Hispanics reside. METHODS The Hispanic population of the United States is composed of three major and cul- turally diverse subgroups: Mexican Americans, Puerto Ricans, and Cubans. We studied the incidence of Hispanic suicide in the largest of these subgroups, Mexican Americans. Death certificates in each of the five south- western States (Arizona, California, Colorado, New Mexico, and Texas) allow for identification of Anglos and Hispanics separately; 86 percent of Hispanics in the five-State area are Mexican American (10). However, the States vary in the amount of mortality data they produce and publish on racial and ethnic groups in their annual vital statistics summaries. No State published suicide data for Anglos and Hispanics by the variables used in our study, namely, age, sex, and method of suicide. The Office of Vital Statistics of the respec- tive State health departments cooperated in this study by providing the Centers for Dis- ease Control (CDC) with either special tabulations or computer data tapes for all suicides of Anglos and Hispanics between 1977 through 1980. Classification of suicide as the cause of death was based on the Eighth Revision of the International Classification of Disease Adapted (ICDA-8) for 1977-1978 (11), and based on the Ninth Revision (ICD- 9) for 1979-1982 (12). The titles for the cause-of-death category for suicide are iden- tical under the Eighth and Ninth Revisions and the comparability ratio between the two revisions is near 1.0 (13). The classification of suicide by method was based on cause of death codes E950-E959 in the International Classification of Disease (11,12). We produced population data for calculating suicide rates for Anglos and Hispanics in the five southwestern States by using computer tapes from the Current Population Survey (CPS) (14). The population statistics from CPS were estimates based on a weighted na- tional sample. This sample was considered too small to provide reliable population es- timates at the State level by ethnicity, age, and sex. Therefore, suicide rates were centered for the selected time periods 1977- 1979 and 1980-1982 by using the mean num- ber of suicides for each 3-year period as the numerator and the mean population accord- ing to CPS data for each 3-year period as the denominator. Hispanic ethnicity from State health depart- ment suicide data in general is determined by the name of the decedent appearing on the death certificate. That is, if the decedent’s surname corresponds to a name on a list of Spanish surnames used by the State Office of Vital Statistics, then that person is categorized in the State mortality statistics as Hispanic. Hispanic in the Current Popula- tion Survey data is defined as a person of Hispanic origin who reported himself/herself as Mexican American, Chicano, Mexican, Puerto Rican, Cuban, Central or South American, or other Spanish origin (15). Itis not possible to determine from State mor- tality data using Spanish surnames whether Hispanic suicide victims were Mexican American, Cuban, Puerto Rican, or of other Spanish origin. However, because popula- tion statistics (10) show that 86 percent of Hispanics in the five southwestern States are Mexican American, we may reasonably as- sume that the preponderance of Hispanics in our study who committed suicide were Mexican American. Less than 1 percent of persons in the five southwestern States who were self-identified as Hispanic in the CPS surveys were nonwhite.* Our study assumes that the number of persons of nonwhite races who were identified as Hispanics by having a Spanish surname is equally small. Anglo is defined as white, non-Hispanic. *Special CDC tabulations of CPS data 3-197 Report of the Secretary's Task Force on Youth Suicide RESULTS The suicide rate for whites in the five south- western States was higher than the national suicide rate for whites for both the 1977-1979 period and 1980-1982 period (Table 1). However, when we separated the suicide rate for whites in the five southwestern States into rates for Anglos and Hispanics, we see that the higher rate for whites for both time periods was a result of a high rate for Anglos- a rate that was approximately twice the rate for Hispanics. However, the difference be- tween the rates for Anglos and Hispanics narrowed slightly between the two time periods. For the period 1977-1979 the rate for Anglos was 2.1 times higher than the rate for Hispanics compared with 2.0 times higher for the period 1980-1982. In both ethnic groups, more males than females committed suicide (Table 2). However, the ratio of male to female suicides was higher for Hispanics than Anglos. This higher sex ratio for Hispanics was consistent for all age groups. Suicides occurred at younger ages among Hispanics than among Anglos. Table 2 shows that 34.6 percent of all Hispanics who committed suicide were under age 25 com- pared with 16.5 percent of Anglos who com- mitted suicide under age 25. More than half (52.1%) of Anglos who committed suicide were over age 40, while about one-fourth (26.2%) of Hispanics who committed suicide were over age 40. The data for males show an even more pronounced difference be- tween the age of suicide for Anglos and Hispanics. More than half (53.5%) of the Hispanic men who committed suicide were under age 30. On the other hand, less than one-third (30.9%) of Anglo men who com- mitted suicide were under age 30. Three-year suicide rates for the period 1980- 1982 show that the age-adjusted overall rate for Anglos is 1.7 times the rate for Hispanics (16.9 and 10.0, respectively) (Table 3). The age-adjusted suicide rate for Anglo men was more than one and a half times the suicide rate for Hispanic men (25.9 and 16.8, respec- tively), while the rate for Anglo women was more than twice the rate for Hispanic women (8.9 and 3.6, respectively). Suicide rates for Anglos were higher than rates for Hispanics in every age group, both male and female, with one exception--young Hispanic males (15-19 years of age) had a suicide rate slight- Time Period United States’ White 1977-79 13.8 1980-82 13.2 *Rates per 100,000 population Population Survey data tapes. Population Survey data tapes. Three-Year Suicide Rates* for the Periods 1977-1979 and 1980-1982 by Race, Ethnicity and Geographic Area** **Five Southwestern States: Arizona, California, Colorado, New Mexico and Texas 1. Source: Number of suicides from National Center for Health Statistics mortality tapes. Population from Current 2. Source: Number of suicides from State health departments’ offices of vital statistics. Population from Current Five Southwestern States? White White Anglo Hispanic 17.4 19.4 9.2 15.9 17.8 8.8 Table 1. 3-198 J.C.Smith: Hispanic Suicide in the Southwest, 1980-1982 ly higher than Anglo males in the comparable age group. The pattern of suicide rates by age group is shown for Anglo men and Hispanic men in Figure 1. Clearly, Anglo males (with one ex- ception) have the highest rate in all age groups, with almost uniform rates from the twenties through the sixties, then a marked increase to the peak rate in the 70+ year age group. Hispanic males have a somewhat dif- ferent pattern: the rates peak in the 20-24 year age group and again, but lower in the oldest age group. The patterns for women in both ethnic groups have a somewhat similar unimodal curve with the lowest rates at the extremes of the age groups (Figure 2). Number, Percentage Distribution and Sex Ratio of Suicides by Race/Ethnicity, Age Group, and Sex, in Five Southwestern States* 1980-1982 Male Female Total Ethnicity & Cum. Cum. Cum. Sex Age Group No. % % No. % % No. % % Ratio Anglo <15 69 0.6 0.6 24 0.6 0.6 93 0.5 0.5 29 15-19 642 5.4 6.0 191 4.3 4.9 833 5.1 5.6 3.4 20-24 1436 12.0 18.0 338 77 126 1774 109 16.5 4.2 25-29 1542 129 309 459 10.4 23.0 2001 12.2 12.2 3.4 30-39 2257 19.0 494 883 20.0 43.0 3140 19.2 479 26 40-49 1499 125 62.4 738 16.8 58.8 2237 13.7 61.6 2.0 50-59 1544 129 75.3 776 176 77.4 2329 142 758 2.0 60-69 1342 11.2 86.5 518 11.8 89.2 1860 114 87.2 2.6 70 + 1617 13.5 100.0 473 10.8 100.0 2090 12.8 100.0 3.4 Total 11948 100.0 4400 100.0 16348 100.0 27 Hispanic <15 20 1.1 1.1 4 1.0 1.0 24 1.1 1.1 5.0 15-19 237 13.2 14.3 45 109 11.9 282 12.8 13.9 5.3 20-24 390 21.8 36.1 68 16.4 283 458 20.7 346 57 25-29 313 17.4 53.5 66 16.0 443 379 172 51.8 4.7 30-39 387 2168 75.1 99 239 682 486 220 73.8 3.9 40-49 166 93 844 61 148 83.0 227 103 84.1 2.7 50-59 130 72 91.6 43 10.4 934 173 7.8 919 3.0 60-69 76 42 95.8 18 44 97.8 94 43 96.2 4.2 70 + 75 4.2 100.0 9 2.2 100.0 84 3.8 100.0 8.3 Total 1794 100.0 413 100.0 2207 100.0 "a3 *Five Southwestern States: Arizona, California, Colorado, New Mexico, and Texas. Table 2. 3-199 Report of the Secretary’s Task Force on Youth Suicide Three-Year Suicide Rates* by Ethnicity, Age Group, and Sex Five Southwestern States**, 1980-1982 *Per 100,000 Population Anglo Hispanic Age Group Male Female Total Male Female Total <15 0.8 0.3 0.5 0.5 0.1 0.3 15-19 17.1 5.4 11.4 17.7 3.5 10.8 20-24 34.0 8.0 20.9 29.4 5.1 17.1 25-29 35.1 10.4 22.7 25.8 5.3 15.4 30-39 31.5 12.7 22.2 23.3 5.6 14.2 40-49 30.8 15.1 23.0 15.8 5.3 10.4 50-59 32.1 156.5 23.6 17.4 49 10.8 60-69 34.4 11.2 21.8 17.9 3.5 9.9 70 + 57.8 11.1 29.7 25.9 2.6 13.3 Total 26.5 9.5 17.8 14.4 3.2 8.8 Age-Adjusted Total 25.9 8.9 16.9 16.8 3.6 10.0 **Five Southwestern States: Arizona, California, Colorado, New Mexico, and Texas. Table 3. Suicide Rates* for Males by Ethnicity and Age Group, Five Southwestern States, 1980-1982 8 8 Rate Per 100,000 8 - ° 15-19 20-24 25-29 30-39 40-49 Age Groups [1 Anglo EXZ Hispanic * Per 100,000 Population Suicide Rates * for Females by Ethnicity and Age Group, Five Southwestern States, 1980-1982 Rate Per 100,000 Age Groups BX Hispanic [J Anglo * Per 100,000 Population Figure 1. 3-200 Figure 2. J.C.Smith: Hispanic Suicide in the Southwest, 1980-1982 Methods of suicide are shown in Table 4. The primary method of suicide for Anglos and Hispanics was the same. More than half the persons in both ethnic groups who com- mitted suicide used firearms or explosives (59.8% of Anglos and 59.2% of Hispanics). The second and third most frequently used methods of suicides were reversed for the two ethnic groups. Poisoning (22.5%) was the second most frequently used method for Anglos, and hanging, strangulation, and suf- focation (10.8%) were the third most fre- quently used methods. The reverse was true for Hispanics (20.7% of suicides by hanging, strangulation, and suffocation and 12.8% of suicides by poisoning). The patterns of suicide by method were similar for females in both ethnic groups, namely, both Anglo and Hispanic females had equally high percentages of suicide by firearms and explosives (42.8% and 44.4%, respectively). Poisoning was the second most frequently used method for both Anglo and Hispanic females (41.5% and 34.5%, respectively) and hanging, strangulation, and suffocation was the third most frequently used method (7.9% and 13.5%, respectively). While Anglo and Hispanic males had equal- ly high percentages of suicide by firearms and explosives (66.0% and 62.6%, respectively), Anglo males had almost twice the percentage of poisoning (15.5% and 7.8% for Anglo and Hispanic males, respectively) and almost half the percentage of hanging, strangulation, and suffocation (11.9% and 22.3% for Anglo and Hispanic males, respectively). DISCUSSION In interpreting the data, three possible data limitations should be kept in mind. First, suicide, to an unknown extent, is understated as a cause of death in vital statistics (16). This understatement is a result of difficulty in es- tablishing suicidal intent, practical considera- tions (such as a loss of insurance benefits), and the social stigma associated with suicide (which seems to be particularly important among youth). Although it has been suggested that suicide rates for Mexican Americans may be under- estimated, it is unlikely that underreporting alone is responsible for the low rate of suicide among Mexican Americans relative to Anglos (19). An investigation of the validity of reported suicide rates in eleven western Percentage Distribution of Suicides by Ethnicity, Sex, and Method Five Southwestern States*, 1980-1982 **UCDA-8 codes shown in parentheses.. Anglo Hispanic Method ** Male Female Total Male Female Total Firearms and explosives (E955) 66.0 42.8 59.8 62.6 44.4 59.2 Poisoning (E950-E952) 15.5 41.5 22.5 7.8 34.5 12.8 Hanging, strangulation, . suffocation (E953) 11.9 7.9 10.8 22.3 13.5 20.7 All other (E954, E956-E959) 6.6 7.8 6.9 7:2 75 73 Total 100.0 100.0 100.0 100.0 100.0 100.0 *Five Southwestern States: Arizona, California, Colorado, New Mexico, and Texas Table 4. Report of the Secretary’s Task Force on Youth Suicide States (four of which are in our study) did not reveal any evidence of systematic bias in reporting specific to any race or ethnic group (18). We investigated the possible effect that misclassification of suicide deaths as deaths due to undetermined causes might have had on the difference in the incidence rate of suicides for Anglos and Hispanics. We found that the difference could not be explained by a large number of Hispanic deaths being clas- sified as "cause undetermined". Even if we assumed the most extreme case, that is, no suicides by Anglos were misclassified as "cause undetermined", and all deaths of Hispanics classified as due to undetermined causes were actually suicides, the rate of suicide would still be two times greater for Anglos than for Hispanics. A second possible data limitation is the un- known extent to which undocumented aliens in the Hispanic population are under- represented in the CPS sample. The CPS is a sample of housing units and includes all persons who occupy a housing unit, without regard to residency status. No evidence indi- cates a differential enumeration of Anglos and Hispanics in the CPS data we used. Fur- thermore, if an undernumeration of Hispanics in the CPS data does occur, then the true Hispanic suicide rate is lower than reported in this paper, and hence, even more disparate from the Anglo suicide rates. A third possible data limitation is the un- known extent to which different methods of determining Hispanic ethnicity affect the rates, namely, Spanish surname (for vital statistics) and self-identification (for popula- tion statistics). Self-identification as used by the U.S. Census Bureau has been suggested as an important way to establish a consistent and uniform definition for Spanish ethnicity (19). Vital statistics, however, continue to rely on Spanish surname for ethnic iden- tification because of problems encountered with self-reporting of ethnicity on the death certificate**. **Personal communication with persons in the Office of Vital Statistics in the five southwestern States. 3-202 The statistical patterns of suicides by Anglos in the five southwestern States were similar to the statistical patterns of suicides by age, sex, and method for whites nationally (20). However, Hispanics in the same five south- western States had some important differen- ces in patterns of suicide when compared with Anglos, and hence, with whites nation- ally. Perhaps the most important difference in suicide patterns for Anglos and Hispanics was that the incidence of suicide for Hispanics was approximately half the in- cidence for Anglos. Very few studies cited in the literature have looked at the incidence of suicide for Hispanics (5,6,7) and these studies report only local data. Two of these studies (in El Paso (6) and Denver (7)) focused on Mexican Americans and can be compared with our findings because popula- tion statistics (10) indicate that almost 90 per- cent of Hispanics in the five southwestern States are Mexican American. As in our study, both of these studies found suicide rates for Mexican Americans to be ap- proximately half the rates for Anglos. A second major difference in the patterns of suicide for Anglos and Hispanics was the comparatively small proportion of suicides by Hispanic females compared with the propor- tion of suicides by Hispanic males. Our study found that the sex ratio for suicides by Anglos was similar to the sex ratio for suicides by whites nationally (21). We found, however, that among Hispanics, the ratio of male to female suicides was greater than that for Anglos in every age group. Overall, the suicide rate for Hispanic males was more than four times greater than the rate for Hispanic females. The higher Hispanic sex ratio was consistent with findings from the studies of suicides by Mexican Americans previously cited (6,7), and with a similar analysis of local area suicide data in Los An- geles (8). A study of age- and sex-specific suicide rates for Mexico City for the period 1975-1978 shows higher suicide rates for males than females in every age group, with an overall sex ratio of 3.7:1 (22). J.C.Smith: Hispanic Suicide in the Southwest, 1980-1982 A third and important difference in suicide patterns among Anglos and Hispanics was that suicide by Hispanics appears to be primarily a youthful phenomenon. Much at- tention has been focused on the increasing rate of suicide among persons under age 25 (23-32). During the period 1962-1972, the suicide rate for whites 15 to 25 years of age increased by almost 75 percent, while the suicide rate for whites of all ages increased by less than 10 percent (28). The finding that Hispanic victims were younger than Anglo victims in the five southwestern States is similar to the findings of other local area studies (6,7,8). The 1990 health objective for the United States for suicide identifies young persons 15 to 24 years of age as the population on whom to focus national suicide prevention and in- tervention efforts (33,34). It appears from our findings that the 1990 objective which focuses on youth suicide is quite appropriate for the Hispanic population, since the higher suicide rates for Hispanics are in the 20-24 year age group. Furthermore, our study shows that while the overall suicide rate for Hispanics is half the rate for Anglos, suicide rates are equally high for both Anglo and Hispanic teenage males (15-19 year age group) (17.1 and 17.7, respectively). There were some notable similarities in suicide patterns among Anglos and Hispanics. First, in both ethnic groups, more suicides occurred among males than females in every age group. Second, the distribution of age-specific suicide rates was bimodal for both Anglo and Hispanic males and unimodal for Anglo and Hispanic females which is similar to the national patterns for white males and females (35). Third, the sex- specific patterns of methods used for suicide were similar for the two ethnic groups and consistent with sex-specific patterns for whites nationally (35). Differences in the patterns of Anglo and Hispanicsuicide probably reflect aninterplay between the effects of the diminishing in- fluence of Mexican cultural traditions, the in- creasing influence of American culture, and the marginal socioeconomic status of Mexican Americans. Horowitz (36), in a study of culture and identity in a Chicano community in Chicago, identified two values heavily stressed in Mexican cultural tradition which are likely to protect against suicide: the concept of family honor and an emphasis on close family ties. A cultural incentive not to dishonor one’s family with a suicide and the ability of close family ties to decrease the risk of social isolation may work together to diminish an individual’s risk of suicide. Whether the incidence of suicide increases in the Mexican American population in the fu- ture may depend in part on the extent to which these cultural traditions continue to be held within Mexican American communities. The assimilation of Mexicans into the American culture has undoubtedly diminished the power of cultural traditions to influence behavior, particularly among Hispanic youth (36). This may be reflected in the younger ages of Hispanic suicide vic- tims uncovered in this study. Suicide rates among younger Hispanics were more similar to those for younger Anglos than rates for older Hispanics were when compared to those for older Anglos (Figures 1 and 2). Hispanic youths may be caught between the influence of traditional values and norms and their experiences in the larger social order which are likely to be heavily influenced by their marginal socioeconomic status. Psychiatric illness is most likely an important contributor to suicide among Anglos and Hispanics. In the general population, it has been estimated that 47 percent of all suicides occur in persons with serious depression or major affective disorders (37). The patterns and differences presented in this paper sug- gest that cultural and social variables may modify the expression and/or course of psychiatric illness in very important ways. Few data are available to analyze and explain the differences and similarities between suicide rates for Anglos and Hispanics. Regarding the availability of data for Hispanics, we agree with Braucht, et al: "Un- fortunately, the entire group of Hispanic 3-203 Report of the Secretary’s Task Force on Youth Suicide heritage Americans of Mexican, Puerto Rican and Cuban origin are buried within the ‘white’ group in nearly all available reports of data from the national vital statistics system and are not available for separate analysis. This state of affairs has resulted in an appall- ing lack of knowledge about the mortality of the Hispanic population in this country" (38). Further, Petzl pointed out that the meager amount of research done on suicide among minorities has largely focused on blacks and American Indians (28). This is because blacks and American Indians represent racial rather than ethnic categories, and can be studied through existing published vital statistics data. Although the lack of data on the health status of Hispanics is recognized (39) and better data is promised for the future (40), we are left for the present without sufficient data to examine why suicide patterns for Anglos and Hispanics differ in incidence and by age and sex. To explore reasons for differences and similarities between suicide rates of Anglos and Hispanics, it would be useful to have data on such variables as employment, education, family composition, and length of residence in the United States. With these data, one could begin to test hypotheses related to in- teractions between the cultural, socioeconomic, and experiential factors of the two ethnic groups and perhaps explain the differential incidence of suicide. REFERENCES: +++ srssrsmsmssmmssssismas . 1. Centers for Disease Control: Fifteen leading causes of death for metropolitan and non-metropolitan populations; United States, 1978. Atlanta, 1982. 2. Dublin LI: Suicide: a sociological and statistical study. New York: The Ronald Press Co., 1963. 3. Hirsch J: Part 1: Demography of suicide. Mental Hygiene 43(4): 516-525, Oct 1959. 4. Massey JT: Suicide in the United States 1950-1964. National Center for Health Statistics, Vital and Health Statistics, 20(5): Aug 1967. 5. Monk M, Warshauer EM: Completed and at- jempied suicides in three ethnic groups. Am. J. of Epidemiology 100 (4): 333-345, 1974. 6. Heather C, Hatcher D: Ethnic group suicide: an analysis of Mexican-American and Anglo suicide rates for El Paso, Texas. Crisis Intervention 5: 2-9, 1975. 7. Loya F: Increase in Chicano suicide, Denver, 3-204 Colorado: 1960-1975. What can be done? Proceedings of 1976 American Association of Suicidology Annual Meet- ing, Los Angeles, California, 1976 8. Loya F, Delyado O: Ethnic group rates of suicide and homicide in Los Angeles County: 1969-1979. Un- published study, 1981. 9. Smith JC, Warren CW, Mercy JA: Comparison of suicides fimiong ir los and Hispanics in five south- western States. Suicide and Life-Threatening Behavior 15: 14-26, 1985. 10. Bureau of the Census. Persons of Spanish Origin in the United States: March 1979, U.S. Government Print- ing Office, Yiastingion, D.C., Current Population Reports, Series P-20, 354, 1980. 11. National Center for Health Statistics. Eighth Revision International Classification of Disease, adopted in text for use in the United States. PHS Publication No. 1693, Washington, U.S. Government Printing Office, 1967. 12. World Health Organization. Manual of Internation- al Statistical Classification of Disease, Injuries, and Causes of Death (based on the recommendations of the Ninth Revision Conference , 1975). World Health Organization, Geneva, 1977. 13. Klebba AJ, Scott JH: Estimates of selected com- parability ratios based on a dual coding of 1976 death cer- tificates by the eighth and ninth revisions of the International Classification of Disease. Monthly Vital Statistics Report: 28(1 LL lement. National Center for Health Statistics. DHEW Publication No. (PHS) 80-1120. Washington, U.S. Government Printing Office, Feb. 29, 1980. 14. Bureau of the Census Current Population Survey, conducted March 1975-1980, U.S. Government Printing Office, Washington, D.C., (For information on the sam- pling procedure and sample size for each survey see Cur- rent Population Reports, Series P-20: 290, Feb. 1976; 310, July 1977; 329, Sept. 1978; 339, June 1979; 354, Oct. 1980; 361, May 1981. 15. Bureau of the Census: Persons of Spanish origin in the United States: March 1980 Puvancs Regort U.S. Government Prining Office, Washington, D.C., Current Population Reports, Series P-20: 361, May 1981. 16. Jobes DA, Berman, AL, and Josselson AR: The im- pact of psychological autopsies on medical examiners’ determination of manner of death. Journal of Forensic Sciences 31: 177-189, 1986. 17. Markides KS: Death-related attitudes and behavior among Mexican Americans: A review. Suicide and Life- Threatening Behavior 11(2): 75-85, Summer 1981. 18. Nelson FL, Farberow NL, MacKinnon DR: The cer- tification of suicide in eleven western States: An inquiry into the validity of reported suicide rates. Suicide and Life- Threatening Behavior 8(2): 75-88, Summer 1978. 19. Giachello AL, Bell R. Aday LA, Anderson RM: Uses of the 1980 census for Hispanic health services research. Am J Public Health 73: 266-274, 1983. 20. National Center for Health Statistics. Vital Statistics of the United States, 1977 Moralin. Part A, Vol Il, DHHS, Publication No. (PHS)81-1101, U.S. Government Printing Office, Washington, D.C., 1981. 21. National Center for Health Statistics. Advance Report, final mortality statistics, 1983. Hyattsville, Maryland, Sept 1985. Monthly Vital Statistics Report, 34 (6) Supplement (2). DHHS Publication No. (PHS) 82-1120. 22. Robertson MJ: The demographics of suicide in Mexico City; Preliminary analysis. Presented at the 15th Annual Meeting, American Association of Suicidology, New York, April 1982. 23. Hopper K and Guttenacher S: Rethinking suicide: Notes toward a critical epidemiology. International Jour- nal of Health Services 9: 417-438, 1979. 24. Bakwin H: Suicide in children and adolescents. Journal of Pediatrics 50:749-769, 1957. J.C.Smith: Hispanic Suicide in the Southwest, 1980-1982 25. Schrect A: Suicidal adolescents and children. Journal of the American Medical Association 188:1103- 1107, 1964. 26. Eisenberg L: Adolescent suicide: on taking arms against a sea of troubles. Pediatrics 66:315-320, 1980. 27. Shaffer D and Fisher P: The epidemiology of suicide in children and Young adolescents. J. of the Am. Academy of Child Psychiatry 20: 545-565, 1981. 28. Petzel SV and Cline DW: Adolescent Suicide. Chicago: The University of Chicago Press, pp 239-66, 1978. 29. Holinger PC: Violent deaths among the young: Recent trends in suicide, homicide and accidents. Am. J. of Psychiatry 136(9): 1144-1147, 1979. 30. Peck M: Youth: special suicide risk group. Presented at the National Institute of Mental Health Suicidology Institute, Los Angeles, California, Jan 1971. 31. Mercy JA, Tolsma DD, Smith JC, Conn JM: Pat- terns of youth suicide in the United States. Educational Horizons: 124-127, Summer 1984. 32. Centers for Disease Control: Violent deaths among persons 15-24 years of age - United States, 1970-1978. Mos sicity and Mortality Weekly Report 32(35):453-457, 1983. 33. Department of Health and Human Services: Remoting Health/Preventing Disease, Objectives for the Nation. U.S. Government Printing Office, Washington, DC, Fall 1980. 34. Silver BJ. Goldston SE, and Silver LB: The 1990 objectives for the nation for control of stress and violent behavior: Progress report. Public Health Reports 99(4): 374, July-Aug 1984. 35. Centers for Disease Control: Suicide Surveillance, 1970-1980, April 1985. 36. Horowitz R: Honor and the American dream. New Brunswick, New Jersey: Rutgers University Press, 1983. 37. Miles CP: Conditions predisposing to suicide: A review. Journal of Nervous and Mental lliness 164(2): 231- 246, 1977. 38. Braucht GN, Loya F, and Jamieson KJ: Victims of violent death: A critical review. Psychological Bulletin 87(2): 309-333, 1980. 39. Brandt EN: Prevention as policy. Public Health Reports 97(5):401, Sept-Oct, 1982. 40. National Center for Health Services Research. Hispanic health services research: Dorothy Rice, closing remarks. Research Proceedings Series, DHHS Publica- tion No. (PHS) 80-3288, Sept 1980. ACKNOWLEDGMENT ----ccoooeeeemennnnnns - The authors wish to acknowledge the cooperation and data provided by the Office of Vital Statistics in Arizona, California, Colorado, New Mexico and Texas. Partial support for this research was provided by the National Institute of Mental Health. 3-205 THE ROLE OF VOLUNTEER WORKERS IN SUICIDE PREVENTION CENTERS Barbara P. Wyatt, Arlington, Virginia Lay volunteers have been called the most im- portant single discovery in the history of suicide prevention. Eighty percent of the suicide prevention centers in the United States operate with non-professional volun- teers as their primary staff (3). These centers represent the only wide scale use of trained lay volunteers for the delivery of clinical ser- vices which have traditionally been provided by professionals (4). Volunteers in the 1980s Volunteerism is a uniquely American phenomenon. Throughout our history, visitors to this country have observed and remarked upon the resourcefulnes of Americans in organizing themselves and giving of their time, their money, and their talents to address their common problems. It has been estimated that today, more than 37 million volunteers actively participate in community programs and services throughout the United States (2). The stereotypical view of volunteers as non- professional suburban housewives--which perhaps was accurate twenty years ago--is well out of date in 1986. Since 1975, the num- ber of women employed outside the home has exceeded the number of full-time homemakers. Organizations such as the Junior League, Girl Scouts, PTA, and the League of Women Voters, whose work had for years depended upon an apparently inex- haustable supply of suburban homemakers, contracted severely during the past decade. 3-206 At the same time, organizations directed at solving specific community problems such as drug abuse, mental retardation, physical handicaps, and suicide have generally been able to recruit enough volunteers to staff their programs. The gap left by homemakers entering the workforce has been filled large- ly by single working professionals, retired professionals, and disabled people (6). Given the demands upon their time as well as their professional experience, today’s volun- teers tend to be highly selective about the programs with which they become involved. They actively pursue their own interests and often look for opportunities that are directly relevant to their professional careers or which will reflect favorably upon them in a career context. This kind of pragmatic altruism--as opposed to altruism for its own sake--has produced a new breed of volun- teers who are educated, skilled and highly motivated. In the past decade, many corporations began to support their employees’ volunteer pur- suits both by allowing employees reasonable amounts of time away from work to engage in approved volunteer activities and by en- couraging and even underwriting the forma- tion of in-house volunteer organizations which participate in various community ser- vice programs. While these corporations are certainly not unmindful of the public rela- tions value of their efforts, they are also aware that many, if not most, of these ac- tivities deal with issues and problems that B.Wyatt: The Role of Volunteer Workers. . . directly effect employees’ lives and health. When the cost of absenteeism, impaired productivity, and health insurance is taken into account, corporate support of volun- teerism may reasonably be viewed as good business practice as well as good public rela- tions. It should also be noted that departments and agencies of the United States government, as well as the United States military and many State and local government organizations, endorse many employee volunteer activities and support participatory volunteer programs. Development of Suicide Prevention Services In 1960, there were fewer than half a dozen suicide prevention centers in the United States. By the end of the 1960's, there were more than 100 centers; by the end of the 1970’s, approximately 200. Approximately 200 centers are still in operation. During the three decades, 1950-1980, the overall suicide rate in the United States remained roughly static. Therefore, it may be inferred that the growth of suicide prevention centers corre- lates to increased awareness of the problem, the technology (telephone) to respond to it, and the mobilization of concerned citizens who were willing to undertake action to deal with it (3). Suicide prevention centers are sometimes controversial undertakings within a com- munity. Such controversy may have a nega- tive impact on a center’s ability to win community endorsement, recruit volunteer workers, and raise funds which are vital to its existence since the services it provides are free. Part of this controversy is rooted in the stig- ma still associated with suicide in western society. Many people are uncomfortable dis- cussing the subject and seek to avoid it al- together. Others may even deny that the problem exists in their community. Still others believe that discussion of suicide and the advertisement of prevention services will actually cause or inspire suicidal acts. While empirical data do not support the latter view, facts do not necessarily reverse emotionally derived opinions. Because there had been no consensus of ex- pert professional opinion on the operation and ethical conduct of suicide prevention centers until recently, some centers have found themselves inadvertently in conflict with local medical societies, social service agencies, law enforcement authorities, courts, churches, or other community or- ganizations and institutions. In the early 1980s, the American Association of Suicidol- ogy (AAS) promulgated operating standards for crisis intervention centers as well as train- ing and performance evaluation criteria for volunteer crisis workers. An accreditation program, under the aegis of AAS, has made significant progress in alleviating public dis- trust and professional skepticism about crisis intervention techniques and volunteer crisis workers. Organization and Structure of Suicide Prevention Centers The organization and structure of suicide prevention centers is determined by the ser- vices offered. The minimum service is usual- ly a telephone hotline whose basic mission is to respond to people in crisis. This response might involve sympathetic listening and "talk- ing things through," referral to professional counseling services, or initiating emergency intervention in the case of immediately life- threatening circumstances. Depending on their financial and staff resources, some centers have expanded their services to in- clude public awareness and education programs, extensive contact and followup programs with callers to the center, and grief counseling services for families of victims. A center’s organizational structure typically includes an advisory board and/or board of directors, executive director, consulting staff, administrative and clerical staff, and lay volunteer staff. The center could also be ex- pected to have ongoing working relation- ships with community emergency services, Report of the Secretary’s Task Force on Youth Suicide social service agencies, mental health ser- vices, hospitals, and private health care providers. Distinction should be made among three categories of personnel and levels of involve- ment with the center’s operations: 1. Paid staff: Usually the executive director and at least a portion of the full-time administrative and clerical staff. 2. Professionally-trained volun- teers (e.g., mental health profes- sionals): Usually members of the advisory board and the consulting staff who donate their professional expertise and services in support of the center. 3.Lay volunteers: Non-profes- sionals who have been trained to provide those crisis intervention ser- vices which have traditionally been provided only by professionals. While crisis intervention workers are typical- ly lay volunteers, it is not unusual for profes- sionally trained people to volunteer to answer hotlines and serve on crisis interven- tion teams. Both Motto and McGee em- phasize that all crisis intervention staff, regardless of their professional training, as well as all consulting staff, should be required to go through the same training programs as lay volunteers and should be subject to the same monitoring and evaluation procedures. Further, Motto and McGee both concluded that people with professional training are not demonstrably more effective than lay volun- teers as crisis intervention workers (3,4). McGee goes on to say that conflicts between lay volunteers and professionals who become directly involved in delivery of crisis interven- tion services can present a significant chal- lenge to center management. He urges that all crisis workers be measured exclusively against the center’s overall performance standards, irrespective of any individuals professional credentials or lack thereof. 3-208 Recruiting and Screening of Volunteers Volunteer recruiting is probably a full-time job at most suicide prevention centers. Sour- ces of volunteers include the general public, college students (especially graduate stu- dents), faculty members, corporations, government entities with active volunteer programs, and community volunteer bureaus. Unpaid advertising is the most usual method of volunteer recruiting. This may take the form of notices in newspaper columns, public service announcements, and flyers and posters distributed to target groups and or- ganizations. A public relations campaign in- volving media interviews and speeches by a center spokesperson before local civic groups and professional organizations may also be incorporated into the overall recruit- ing effort. Paid advertising, such as classified newspaper advertisements, is another alternative. However, since suicide prevention centers often operate on tight budgets, paid advertis- ing may be viewed as less effective than a resourceful volunteer recruiter. The initial screening process begins with an interview at the center. Upon arrival, the volunteer may be asked to complete a per- sonal data sheet which includes general back- ground information. The respondent is then interviewed by two or three designated inter- viewers. These interviewers might be some combination of consulting staff, mental health professionals, center administrators, and active volunteer staff. Psychological testing may reasonably be incorporated into the screening process. Initial screening may involve two or three visits to the center, which is in itself a method of screening. The objective of the screening process is to elicit information regarding past and present life patterns and to provide evidence of emo- tional stability, integrity, receptivity to learn- ing, perceptiveness, and responsiveness to human needs. Incidence of prior suicide at- tempts would almost certainly disqualify the B.Wyatt: The Role of Volunteer Workers. . . applicant for direct crisis intervention work although the person might be assigned to other tasks within the center at the interviewers’ discretion (4). Applicants who pass screening interviews and testing are accepted into the center’s training program. Training may be viewed as an extension of the screening process since some aplicants will drop out or show them- selves to be unqualified for crisis intervention work as they are exposed to the real world conditions of the center’s operations. Training Training programs in suicide prevention centers ideally combine theoretical informa- tion with practical experience in crisis inter- vention. A thorough training program requires 18 to 25 hours of concentrated ses- sions extending over a three to six week period plus concurrent parallel reading. Elements of a good training program include: e orientation to the center’s methods of operation, record keeping procedures, performance criteria, and ethical stand- ards. e lectures by mental health professionals and suicidologists. e films, video tapes, selected reading. * listening to audio tapes of actual calls to hotlines (with all personal identification information removed). e role playing. e observing actual phone calls to the cen- ter. e working on the phones under close su- pervision. By the end of the training period, volunteers have been gradually introduced into their roles as crisis intervention workers and there has been ample opportunity to discover in- dividuals who are obviously unsuited for the job. McGee reports some instances in which lay volunteers were given no formal training but were put directly to work on the telephones and trained on the job, as it were, after they had been interviewed. In other cases, volun- teers were put through marathon two and three day training sessions and began work immediately thereafter. While the results of the short but intensive training sessions have not been evaluated in comparison with the longer sessions, it is thought that one object of the marathon sessions was to produce a large number of volunteer workers very quickly while the longer, more deliberate ses- sions were aimed at producing genuinely dedicated volunteers who were likely to remain with the center over a long period of time (3). Some centers do not use lay volunteers at all but depend upon psychiatrists, psychologists, nurses, clergymen, and other professional counselors to serve as crisis workers. Train- ing may be bypassed altogether in the belief that these people are already sufficiently trained to counsel people in crisis. Ex- perience indicates, however, that crisis inter- vention and suicide prevention work require unique and special skills. Specialized train- ing for all who are engaged in suicide preven- tion should be considered mandatory (4). Regular and systematic in-service training is also highly recommended. Extended train- ing may take the form of lectures, seminars, or free-for-all discussion sessions. Oppor- tunities for the exchange of information, ex- periences, ideas, and even complaints are vital to the success of the center’s work and to maintaining staff morale in a highly stress- ful environment. It should be remembered that volunteers-- particularly those who work late night and weekend shifts--may have lit- tle or no face-to-face contact with center per- sonnel, consulting staff, and other volunteers unless such sessions are deliberately initiated and scheduled. Equally necessary is regular evaluation of the crisis worker’s performance. Observation of the worker taking calls and review of written case records prepared by the volunteer are an essential part of center management and can constitute the basis for performance 3-209 Report of the Secretary’s Task Force on Youth Suicide evaluations as well. When these procedures are combined with regular personal discus- sions with consulting staff and center ad- ministrators, performance evaluation can become a part of the volunteer’s continuing in-service training (4). Program Implementation The programs of a suicide crisis center can be divided into three broad categories: 1. prevention, which includes education and awareness programs; 2. intervention, which includes telephone hotlines and emergency services; and 3. postvention, which includes fol- lowup, outreach, and counseling for the families of victims. As a practical matter, chronology does not reflect the evolution and focus of the crisis center. Intervention is the center’s initial and principal "reason for being," while prevention and postvention programs, and services are later outgrowths and extensions of a successful center’s basic service. Intervention The telephone is the basic tool of crisis inter- vention. Some centers limit their service ex- clusively to hotlines, while others have walk-in facilities and/or crisis intervention teams which go to the caller’s location in ex- treme emergencies. In 80 percent of all suicide intervention centers, the people who answer the phones and respond to emergen- cy situations are volunteers. As back-up resources, volunteer workers may have access to the center’s consulting staff, private practitioners who take referrals from the center, law enforcement and emer- gency medical personnel, and community so- cial service and mental health agencies. The basic job of the crisis worker is first, to establish communications with the caller and second, to make an assessment of the caller’s condition (lethality assessment). In the case 3-210 of an immediately life-threatening situation- which is a relatively infrequent occurrence- the first priority is to determine the caller’s location and dispatch assistance. More usually, the crisis worker’s role is to es- tablish a rapport with the caller, to listen to the person’s description of his problems, and to work with him in setting a course of action. Shneidman has suggested that the role of a therapist is that of ombudsman for the patient; the same can be said of the crisis worker vis-a-vis his client. The worker’s ob- jective is to increase the options available to the caller, to reduce the caller’s sense of pain and isolation, to listen, to offer hope and help, to suggest alternatives, to play for time, and with the caller’s permission, to involve others (7). Some centers actively discourage any face- to-face involvement between crisis workers and clients. On the other hand, many centers allow workers, at their own discretion, to meet callers away from the center’s premises while cautioning against over-involvement in any particular case. Other centers train selected workers to follow the progress of clients through individual personal visits. Crisis teams are another alternative. Usual- ly comprised of two people, teams are care- fully selected and trained to go out into the community and intervene directly in life- threatening situations. There may be some risks to the worker’s personal safety in these circumstances, and there is certainly a greater likelihood of a worker’s becoming over-involved with a particular case (4). Nevertheless, crisis teams have proven to be highly effective in many instances. Once a crisis center has established its reputation in the community, it is not unusual for police, fire-rescue services, and hospitals to request that a crisis intervention team be sent to a location where a suicide emergency is in progress. This represents acknow- ledgement by law enforcement and emergen- cy services personnel that crisis intervention is indeed highly specialized work and best un- dertaken by those who are specifically B.Wyatt: The Role of Volunteer Workers. . . trained for it. The variety of approaches and techniques used in crisis intervention are limited only by the center’s financial resources and the time constraints of its volunteer crisis workers. Prevention Many suicide prevention centers assume the responsibility for community awareness and education. While the responsibility for these initiatives and programs is largely within the purview of the professionals affiliated with the center, volunteer crisis workers may be included in a particular program designed for a particular group. For example, college-age crisis workers have been shown to be espe- cially effective in working with teenagers, presumably because their proximity in age engender a natural rapport (1). Again, methods and approaches vary widely and are limited only by the center’s resources and the ingenuity of its staff. Postvention Postvention refers to followup and outreach activities directed to individuals following a suicide attempt and to counseling services for families of suicide victims. In addition to following up at regular inter- vals the progress of callers to the center, out- reach may also involve responding to referrals from hospital emergency rooms, law enforcement authorities, social service agen- cies, clergy, teachers, and other concerned individuals. In these cases, the center in- itiates contact, either by telephone or by per- sonal visit, with people who have attempted suicide and extends an offer of help. Ross describes another approach called "Befrienders" in which volunteers who were able to give large amounts of time were selected to befriend and work with particular individuals who had attempted suicide. Befrienders were carefully matched to clients on the basis of personality, attitudes, sex and age, and were thoroughly briefed in advance on the details of the case to which they were assigned. They were required to be available for personal visits several hours each week and to be available by telephone as much of the remaining time as possible. Obviously, this involves a substantial time commitment on the part of the volunteer befriender; however, results of the program as an alter- native means of post-episode care were con- sidered encouraging (5). Outreach programs directed toward provid- ing support to families of suicide victims during the bereavement process may be in- itiated in response to a family’s request for help in coping with their grief. In other in- stances, the center might be contacted by coroners, hospital emergency room staff, funeral directors, or concerned friends who are aware of the circumstances and the family’s need for assistance. In some grief counseling programs, the cen- ter serves as facilitator in establishing contact and networks among families who have suf- fered the same tragedy. The center may also provide referrals to professional counselors and offer seminars and discussion programs dealing with the particular problems of sur- vivors. Special attention is paid to siblings of adolescent victims who are known to be a particularly high risk. Volunteer crisis workers are, of course, involved in the delivery of all of these services. As a corollary to postvention programs, some crisis workers are trained to conduct psychological autopsies. This involves gathering information from those closest to the victim which may shed light on the exter- nal events and family dynamics leading up to the suicide. Because of their training, crisis workers often recognize the significance of information that may have been ignored or overlooked by family, friends, and others in- volved in the case. Such information is often valuable to researchers and therapists who are trying to understand and deal with self- destructive behavior. CONCLUSION Volunteerism is a time-honored custom, in- 3-211 Report of the Secretary’s Task Force on Youth Suicide deed almost an obligation of American life; but to equate volunteerism with amateurism is to do an injustice to volunteers everywhere. In no case is this more evident than in volun- teer crisis workers. Crisis intervention workers are skilled and talented people who take their work very seriously. Most of them have undergone extensive training and devote a minimum of six hours per week to their "volunteer" work. Their contributions to suicide prevention have been recognized by the American Association of Suicidology which, since 1972, has admitted qualified volunteers to full AAS membership. To call crisis intervention workers lay volun- teers and non-professionals detracts from their level of training, skill and commitment. They have earned the respect of their profes- sional colleagues and they have earned the right to be called para-professionals. REFERENCES --ooeereereerrennuenennenes 1. Farberow NL: Suicide in adolescence: Prevention and treatment. In: Golombek, H. and Garfinkel, B.D. (Eds.) The Adolescent and Mood Disturbance. New York, Inter- national Universities Press, 1983. 2. Flanagan J: The Successful Volunteer Organiza- tion. Chicago, Contemporary Books, Inc., 1981. 3. McGee RK: Crisis Intervention in the Community. Baltimore, University Park Press, 1974. 4. Motto JA, Brooks RM, Ross CP, and Allen NH: Standards for Suicide Prevention and Crisis Centers. New York: Human Sciences Press, 1978. 5. Ross CP and Motto JA: Group counseling for suicidal adolescents. In: Sudak, H.S., Ford, A.B. and Rush- ks, N.B. (Eds.) Suicide in the Young. Littleton, PSG Inc., 6. Rubin N: The New Suburban Woman. New York: Coward, McCann and Geoghegan, 1982. 7. Shneidman E: Definition of Suicide. New York: John Wiley & Sons, Inc., 1985. 3-212 PREVENTING SUICIDE BY IMPROVING THE COMPETENCY OF CAREGIVERS Bryan L. Tanney, M.D., F.R.C.P., Clinical Director, Psychiatric Emergency Services, Calgary General Hospital, Calgary, Alberta, Canada SUMMARY All caregivers should be competent in the tasks needed to make direct, personal, and immediate interventions with persons-at-risk for suicidal behaviors, especially when a per- son is actively considering suicide. The core tasks of recognition, risk rating and resource referral define the role of a "gatekeeper" in the intervention process. A review of caregivers’ competency in this role emphasizes the need for better learning and training experiences in suicide. Available learning resources are summarized. The cur- riculum of a program that aims to prevent suicide by improving the gatekeeping skills of all caregivers is described, along with the in- frastructure for delivering a two-day presen- tation to large numbers of caregivers. The Foundation Workshop described in this paper has successfully involved more than 4,000 participants. Prospects and pitfalls for its expansion to a larger, perhaps national, level are briefly discussed. INTRODUCTION Even in the most ideal of situations, not all suicides are preventable. The belief that some, and probably the majority, are preventable sustains the efforts of helpers to persons-at-risk for suicide. But the continu- ing and apparently increasing rate of suicide implies that helping efforts of all kinds are not as successful as might be hoped. Before discarding the accumulated experien- ces of efforts in suicide prevention as un- workable and demanding a new direction with innovative prevention strategies, it is im- portant to realize that the impact of helping efforts has always been difficult to evaluate accurately. Only the outcomes of unsuccess- ful interventions, which result in suicide at- tempts or completed suicides, are recorded and the difficulties in obtaining accurate and reliable statistics are well known. In addi- tion, few studies claim to have sufficiently controlled all of the many environmental, personal, and historical variables which lead to a decision for suicide. In defining the directions for suicide prevention activities, we are still in an age of empiricism. The choice of effective prevention strategies must derive from the rationale and the rationality of the proposals themselves. THE RATIONALE FOR "PREVENTION BY INTERVENTION" An episode of suicidal behavior is initiated when a person actively considers suicide as a problem-solving behavior, usually as a result of some precipitant stressor. An episode is usually time-limited and persons may ex- perience one or more during their lifetime. Every episode demands therapeutic atten- tion. Self-destructive behavior is not a pre- requisite in the definition of an episode, 3-213 Report of the Secretary’s Task Force on Youth Suicide although it offers the best index or marker of the experience. We suggest that the time im- mediately proximal to the suicidal behavior, a prelude phase, is a focal point for rational, effective, and high impact suicide prevention activities by any caregiver (Figure 1). A number of similarities have been recog- nized in persons-at-risk during this prelude to suicidal behavior. Ringel (48) described it as the pre-suicidal state. Shneidman charac- terized the thought process during this time as one of constriction, and more recently he has described a number of other com- monalities (53). If the thoughts, feelings, and actions of those contemplating suicide share some common features, then caregivers can devise intervention responses and activities applicable to any episode and any person-at- risk, whatever the origins, predisposing fac- tors, or immediate precipitants. The interactions between a person at risk and the caregiver’s planned activities together make up the process of suicide intervention. The process can be described and modelled, and roles, tasks (and accompanying skills), as- signed to both partners in the interaction (34). The duration of this pre-suicidal or prelude phase is variable, but it is estimated that 80 percent of persons-at-risk warn of their movement towards self-destruction during this time (14,24,33,49,62). These cries for help communicate distress and ambivalence as well as statements about suicidal intent. Because the outcome too often is suicidal be- havior, we must conclude that many of these "care-eliciting behaviors" (28) are unsuccess- ful. Why is this so? It must not be assumed that the directed communications of the suicidal person are ambiguous, unclear, or directed to the wrong person. They are often repeated and directed towards more than one poten- tial helper or resource (12,49). Some 23 dif- ferent studies confirm that medical caregivers have significant amounts of direct contact with persons-at-risk in the time preceding suicidal behaviors (references available). Though less exact, similar data in- dicates that other helpers are also ap- proached (13,24,33). To understand why these directed communications apparently fail to generate caring and helping from others, we must look to the helpers--those A PRAAAAAMAAAMAAAAASAIAAMAARANAAA Ae Pre diSPOSING me Suicidal Behavior--The Course of an Episode eee 1° Prevention gp | <—————— 2° Prevention m——| Precipitant Prelude H— Postvention ei Figure 1. 3-214 Tanney: . . .Improving the Competency of Caregivers who receive the messages of distress. Who are these helpers? A medical doctor is the first choice of those who seek formal help for suicidal or mental health crises (13,26,51,56) although many and diverse other caregivers are also approached. The caregivers avail- able for help and support during a suicidal be- havior episode can be defined as emergent or designated (45). Designated caregivers have received training to act in a specific helping role to suicidal persons. Emergent caregivers do not have this background, but may have the opportunity to prevent suicidal behavior as the person to whom a person-at- risk turns for help. What is the helper’s role? Some caregiving persons offer support to the person-at-risk while others work directly to resolve the suicidal situation. The aim for all caregivers is to prevent suicide as the outcome of the episode. Terminating the episode by closing the door to suicide and opening or re-open- ing alternative actions and solutions is an ac- tive process of intervention and involvement. The caregiver in this process has been called a "helper of the first instance" (6) or a frontline worker, but the role itself is well described as that of "gatekeeper" (56). The gatekeeper’s role includes specific com- petencies in suicide prevention activities as well as effective use of interpersonal and communication skills. Not all of the neces- Exposure to Learning Experiences About Suicide in Some Community Caregivers Caregiver Group (n) No Exposure Education (193) 56% Clergy (47) 51% Child Care (87) 49% Medicine 7) 43% Social Work (280) 36% Nursing (248) 36% Psychology (140) 27% - Table 1. sary skills have been defined, but the role-re- lated tasks include: 1) recognizing the risk of suicide (Schapira’s "spotter" 52); 2) rating the degree of suicidal risk; 3) referring to ap- propriate resources. We call these the 5 R’s: Recognition, Risk Rating, and Resource Referral. If competence in these tasks were expected of all caregivers, we believe that many episodes ending in suicidal behaviors would be effectively prevented. THE PRESENT STATUS OF CAREGIVERS AS GATEKEEPERS A. Prior Education and Training Experiences Professional schools and clinical disciplines offer few organized courses in the study of suicidal behavior (44). Presentations during clinical practicum experiences review suicide risk factors, but virtually ignore therapeutic, clinical, personal, and professional issues. In two studies, 90 percent (62) and 48 percent (50) of physicians surveyed recalled no in- struction in suicidal behaviors during their professional medical training. Boldt (6) found that more than 90 percent of 143 reporting agencies had no workers with any significant training in this area. More that 50 percent of caregivers participating in our program recalled no prior learning experien- ces in the area of suicide. (See Table 1.) B. Self-Perceptions Forty-four (58) and 58 percent (50) of physicians surveyed indicated a desire or will- ingness to learn more about the area of suicide. In a membership survey of the Canadian Association for Suicide Prevention (8) these caregivers already involved in the field clearly expressed a need for more infor- mation and training in dealing with persons- at-risk. C. Reports and Recommendations Throughout this century, the need for better education and preparation of caregivers has 3-215 Report of the Secretary’s Task Force on Youth Suicide been emphasized by both educators and suicide researchers. The first recommenda- tion of a World Health Organization Work- ing Group (10) stated that training in the detection and management of suicidal per- sons should be a requirement in the educa- tion of all professional helpers in the field of public mental health. D. Knowledge and Skills Three of four studies (7,27,30,47) assessing family physicians’ responses to a multiple choice questionnaire about suicide produced disappointing scores. Three studies involv- ing different groups of designated caregivers (29,30,47) found that their medical prac- titioners achieved the highest score of any of the groups tested. Other studies reported a better response to major clinical issues in- volved in the recognition and management of suicidal persons than to strictly factual or statistical items (30,35,40,50). In comment- ing on the knowledge and skill competence of caregivers, there was repeated concern with the apparent failure of these caregivers to apply the awareness they do possess to their work with suicidal persons (38). Twen- ty percent of Steele’s (57) medical students and the same percentage of Whittemore’s 78 physicians (62) specifically avoided the issue of suicide in clinical situations. E. Caregiver Attitudes Towards Suicide and Persons-at-Risk In working with persons at risk for suicide, medical caregivers may be not only un- responsive (55) or passive (4), but also over- tly hostile. In retrospective chart reviews of completed suicides, (5,19,31,61) the caregiver’s rejection of the patient (at- tributed to feelings of hate, hostility, anger, and anxiety) is a regular finding. Surveys of caregivers utilizing questionnaires (21,39), semantic differentials (2,22), objective checklists (15), or the rating of clinical vignettes (2,25,43) consistently reflect this negative attitude (Table 2). Two studies of caregiver’s responses to 3-216 suicidal persons support the general finding that a therapist’s initial attitude and emotion- al reactions are of crucial significance to his clinical decision-making (15,25). Although positive and supportive attitudes appear to develop with increased contact with suicidal persons (22,23,39,62), the majority of caregivers do not regularly encounter suicidal persons in the course of their clinical practices. There is also disagreement about the congruency between personal and professional attitudes toward suicidal per- sons (22,42). If caregivers’ attitudes, both personal and professional, can be the source of significant resistance to their becoming ef- fective agents for suicide prevention, it is im- portant that training experiences ensure the opportunity for open discussion of these is- sues (17). These observations about caregiver com- petence in preventing suicide are disquieting. There is a definite and recognized deficiency in the abilities and attitudes of caregivers to undertake this helping role. If caregiver in- terventions are to be a successful strategy for suicide prevention, additional training is needed to upgrade the competencies of caregivers in the tasks that ensure successful intervention. AVAILABLE LEARNING MATERIALS Training programs, curricula, and learning Caregiver Attitudes/Feelings About Suicidal Persons Negative Anxiety Unfavorable Uncomfortable Hostile Contempt, not satisfying to treat Unsympathetic Denial, avoidance Aggressive-avoidant Passivity Rejection Betrayal Guilt Failure Table 2. Tanney: . . .Improving the Competency of Caregivers materials are part of the ephemeral litera- ture. The support of the Suicide Information and Education Center enabled a review of these materials, which included the oppor- tunity to view more than 150 audio-visual productions. Additional information was gained in an informal survey of trainer/educators active in delivering suicide prevention learning experiences. A. Content A recognizable body of "core knowledge" (36, p.23) needed for the gatekeeping role and a scientific literature supportive of this content is readily available and well docu- mented. For example, cues and clues to recognizing the person-at-risk have changed little since the suggestions made by Sym (59) and Fairbank (18). In the curriculum con- tent of virtually all training/education programs reviewed, mastery of the tasks of the gatekeeping role (5 R’s) is included in the participant objectives. The content of audio- visual productions intended for caregiver audiences is more restricted with major em- phasis on the task of rating suicidal risk. Only one audiovisual attended to caregiver at- titudes (11) and fewer than a half dozen specifically deal with the intervention process (3,9,32,41). A similar emphasis on the risk rating task was noted in the cur- riculum and program outlines. B. Formats for Learning Materials Written text (books, papers, or monographs) continue to be the major format for present- ing materials on suicidal behavior to caregiver audiences. Some variant of the "Facts and Myths" questionnaire (54) is the most standardized and widely used learning aid available. Although often not developed for caregiver audiences, the number of audio-visual productions related to suicidal behaviors has doubled in the past decade. The most modern incorporates video disc and computer technology in a self-directed learning experience (20). C. Learning Experiences A comprehensive review of issues concern- ing education and training in suicidology stressed the need to develop "imaginative dis- semination programs” (36, p.23). A learning experience in suicide prevention should in- clude: 1) a curriculum outlining content, ob- jectives, and the schedule of learning experiences; 2) available learning materials in different formats that are supportive of and integrated with the content and process of learning and; 3) available background materials which adequately prepare the trainer or educator presenting the program. There are many learning experiences in suicide prevention that meet these standards. Although their curriculum content may be remarkably similar, most programs were uni- quely developed for a particular and homogeneous audience of caregivers usually defined by the group to whom their helping services are directed; for example, crisis or distress center workers, correctional institu- tion personnel, school board employees. Curriculum developers and trainers both agree that one day (7 hours) of caregiver par- ticipation is a minimum requirement and the inclusion of practicum and/or simulation ac- tivities often lengthens participant committ- ment to two or more days. Few experienced "disseminators" (36) support the develop- ment of presentations that are individualized for a particular caregiver group. Ex- perienced trainers/educators virtually all present a core or foundation of similar material to all caregivers with some tailoring of situations and examples to the experiences of the participants. A program in suicide prevention is more than an organized learning experience. In addi- tion to meeting all of the criteria for an effec- tive learning experience described above, it must have: 1) an organizing infrastructure that monitors, evaluates and modifies both the program and the presentation itself, and 2) a mechanism for ensuring optimal delivery of the learning experience to its intended audiences. Two programs that offer stand- ardized learning materials intended for 3-217 Report of the Secretary’s Task Force on Youth Suicide caregivers of different disciplines and varying levels of expertise are available. Although not fully meeting the standard defined above, the Suicide Prevention Training Manual (1) has been used widely for almost a decade. The other program (9) intended for distribu- tion to a broad audience is technically superb in content and curriculum materials. Unfor- tunately, it was never aggressively marketed or promoted, and the materials are no longer available from the copyright holder. Deciding whether an educational/training experience will succeed in improving caregiver competency requires consideration of learning materials, learning activities, and curriculum organization. Rating of suicidal risk is the major content focus of most avail- able experiences, with almost no attention to ensuring caregiver effectiveness in the over- all intervention process. There is no program presently available and appropriate for use by all caregivers that offers integrated learning materials and activities in its cur- riculum. THE FOUNDATION WORKSHOP Over the past five years, we have developed, piloted and successfully implemented a program which offers a standardized learning experience in the foundations of suicide prevention (46). The workshop presenta- tion assumes that increasing the competen- cies of caregivers in dealing with persons-at-risk during an episode that might end in suicide is an effective suicide preven- tion strategy. The intended audience of the program includes caregivers of any discipli- nary and theoretical orientation or level of expertise. To our knowledge, it has been delivered to more caregiver participants than any integrated standardized program on suicide prevention ever available in North America. A. Descriptive Summary Development team (disciplines): Psychiatry, social work, psychology; with input from clergy, education, counseling, nursing, and family medicine. Intended audience: Any group of emergent or designated caregivers. Aim: To provide all caregivers with the com- petencies to intervene in an episode of suicidal behavior until either the immediate danger is alleviated or further assis- Content and Learning Activities of the Foundation Workshop Module / Hours Learning Activity Audio- "Discovery" visual Discuss Survey Worksheet Learning Simulations Lecture Introduction / 1 X X Attitudes / 3 X X x Knowledge / 3 x x x x Skills:Model / 2 x X Skills: b ¢ Simulations / 4.5 Resourcing: X Networking / 0.5 14* 0.75 3.25 1.75 1.25 1.25 3.25 1.5 (* includes 4 breaks of 0.25 hours) Figure 3. 3-218 Tanney: . . .Improving the Competency of Caregivers tance/resources can be obtained. Curriculum: Model: adult learning as inservice or continuing professional education. Learning Experience: structured skills (16,37), workshop. Formats: large and small groups (maxi- mum 15 participants/trainer) with team teaching (two trainers minimum). Learning Activities: audiovisuals (2); surveys, questionnaires (3); simulation; worksheet exercises; discussion group; discovery learning (30+ visuals); lec- tures (maximum 15%) with participant handouts. Content: modular. B. Developmental History In the pilot phase, the workshop was presented eight times to a total of 434 par- ticipants. The audiences for these presenta- tions were caregiver groups, both homogeneous and heterogeneous in com- position, and included helpers in urban and rural communities, childcare workers, staff of a psychiatric institution, and native coun- selors. Evaluation at the end of the presen- tation by these diverse groups of caregivers supported both the curriculum organization and the content focus. The development phase required the preparation of trainers who could present the Foundation Workshop. This prepara- tion, to be discussed later, included the presentation of the workshop by a co-leader or apprentice with a more experienced trainer. These apprenticeship presentations, involving 27 workshops and 1250 par- ticipants, were also used to continue the development of the workshop learning ex- perience. C. Program Evaluation At the conclusion of the development phase, three experienced trainers in different centers across North America were con- tracted to review and critique the program materials and curriculum. Later, three other caregivers reviewed the entire program by considering its potential use as a training vehicle for a State-wide mental health delivery system (60). Participant evaluations are completed at the end of the workshop. Fewer than 1 percent of participants drop out during the two day experience. There are more requests for an extra day or added time than there are dropouts. Fewer than 1 percent of par- ticipants fail to recommend this workshop to other colleagues and caregivers. All modules of the workshop receive positive comments, with the practical experiences involving both large and small group simulations garnering the most positive support. It is notable that the evaluations of the workshop with respect to its value, strengths, and deficiencies are remarkably uniform across the different groups of participant caregivers. Evaluation of the Foundation Workshop’s impact on caregivers over an extended time span has been proposed. At present, end of workshop evaluations from all participants are centrally reviewed by a group of experienced workshop presenters. Feedback from this ongoing audit is used continuously to adapt and up- date the program materials and learning ac- tivities. Delivery of the Foundation Workshop to All Caregivers The Foundation Workshop is suitable for presentation to any group of caregivers inter- ested in mastering a lifesaving task. To sus- tain this amount of program activity and to ensure that the quality of the workshop ex- perience is maintained, requires an in- frastructure that must be sponsored and supported. At the most practical level in this structure, the essential component is the trainer or educator who delivers the workshop presentation. A training program which offers certification to trainers, who can then conduct the Foundation Workshop, has been developed. 3-219 Report of the Secretary's Task Force on Youth Suicide The Training for Trainers program involves experience as a workshop participant, fol- lowed by a two or three day training ex- perience in effective use and understanding of the curriculum. Because the stand- ardization of both content and process of the Foundation Workshop is quite rigorous, a 300 page curriculum manual (45) emphasiz- ing theory, content, and practical process is provided. An apprenticeship program in which more experienced trainers assist the newly qualified in the presentation of their first full workshop is also encouraged. Seven presentations of the Training for Trainers program have, to this time, involved more than 150 potential trainers. The model for program delivery requires considerable effort and expense before any caregivers actually participate in the workshop itself. With this large front-end ex- penditure, cost-effectiveness increases with the involvement of more candidates at a Training for Trainers session, and with every workshop that is delivered by the certified trainers. It is vitally important to minimize trainer drop-out. This is accomplished by an initial screening procedure, by the evaluation of readiness for apprenticing at the con- clusion of Training for Trainers, and by the apprenticing process itself. For certified trainers, an active network keeps them up- to-date with developments and modifications to the workshop information and materials. We have been gratified by the entrepreneurial spirit that has developed within the network of trainers and believe that it reflects their ownership, investment, and identification with the quality of the workshop experience. Despite the enthusiasm, it is apparent that active sponsorship of this program is re- quired because of both the heavy initial in- vestment and the objective of delivering the workshop to a large number of caregivers. Several avenues for development are avail- able. The development of competency as a basic "Rescuer" in cardiopulmonary resus- citation (CPR) is a similar task, and this program has attracted many organizing spon- 3-220 sors. Its success affords a useful model for the delivery of a basic learning experience to large and diverse groups of caregivers. In the area of suicide prevention, the American As- sociation of Suicidology could adopt this program as part of its objective to certify in- dividual caregivers. Professional organiza- tions of designated caregivers (physicians, nurses, clergy, social workers, psychologists) or the administrators of human resources sys- tems could make a commitment to offer the Foundation Workshop to all of their caregivers. Although such support would resolve many organizational hurdles, their endorsement should only be offered to a program of proven excellence. While awaiting the support for large-scale implementation, it is useful to summarize the successes of the Foundation Workshop program. Following its original development for caregivers in Alberta, Canada, presenta- tion of the Foundation Workshop using the Training for Trainers delivery model has been extended into two other caregiving sys- tems, one geographical and the other system- based. In addition, pilot presentations to special audiences have been offered. These included selected inmates of a medium security correctional facility who were developing a peer support model, and lay persons in a number of rural communities where the formal network of caregiving sup- port resources is diffuse and less important then the support provided by family and friends. Although unplanned, materials and activities from the Foundation Workshop ap- pear in courses offered in professional schools at three Canadian universities. As of April 1986, more than 4000 persons have par- ticipated in the Foundation Workshop, and two workshops per week are being con- ducted by a network of more than 80 certified trainers. DISCUSSION As a model program for upgrading the com- petencies of caregivers in effective suicide prevention, the experiences of the Founda- tion Workshop point to two major issues Tanney: . . .Improving the Competency of Caregivers which require further consideration. One addresses program content, and the other concerns the process of delivering the learn- ing experience to large numbers of caregivers. Both issues have been a focus for some resistance to the widespread use of this program. The content issue questions the effectiveness of a general and basic experience in suicide prevention/intervention for audiences of caregivers who work almost exclusively with a specific at-risk population (adolescents, in- mates, crisis center clients). This concern is surprising because similar content and objec- tives can be identified across many of the programs available for specialized target audiences. Besides discouraging costly duplication, we believe that it is very impor- tant to encourage a common learning ex- perience for all caregivers. A shared baseline of competencies across disciplines, caregivers, and agencies facilitates com- munication between caregivers. Confidence in other professionals at this personal level strengthens the network of available resour- ces and improves the continuity of care for persons-at-risk. In several communities and institutions, workshop participants who took part in simulations dealing with referral to appropriate resources within their systems, later reported making a similar referral in a real-life situation, with the very same caregivers in the referring and resource roles. Finally, in presenting the Foundation Workshop to participants with widely diverse formal qualifications and experiences, we were struck by a remarkable similarity in both the process and the content of the group dis- cussions and learning activities across the dif- ferent types of caregivers. The support for a basic or Foundation program does not imply that learning ac- tivities for specialized or targeted caregiver groups are not valuable. Parallelling the CPR model, it is most appropriate to present these materials at a secondary or advanced level which can be added after the basic Foundation curriculum has been delivered. The availability of advanced or specialized learning experiences which further develop caregiver competencies only increases the likelihood of effective suicide prevention. Before suggesting any further extension of this strategy beyond the basic learning ex- perience, it is important to emphasize again that improving the competencies in suicide intervention of all caregivers is a large and daunting task. The program requires a large front-end loading in terms of dollars, time, and human resources involvement. This re- quires the continuing support of organiza- tions which can maintain and nourish an active network of trainers. In our efforts to encourage the large-scale delivery of the Foundation Workshop, two possible sources of resistance by potential sponsors have been identified. The first con- cerns the scope and size of the program itself. It is possible that the organization required to deliver and maintain this program is seen as a task too large to handle. There is, admit- tedly, some reality to this concern. The alter- native explanation is a much more difficult and onerous one. Different disciplinary or professional groups may have a sense of in- vestment or ownership in being the desig- nated agents of suicide prevention. They would have much difficulty accepting or sup- porting a program which aims to develop the same baseline of competencies in all caregivers. This limitation broaches the larger question of the relationship between the many different professions, institutions, and agencies which affirm some interest and investment in the issues of suicide preven- tion. SUMMARY 1. Persons-at-risk for suicidal behaviors con- tinue to ask for help. Their caregivers of all disciplines and levels of expertise need and want to improve their abilities as suc- cessful agents of suicide prevention. 2. Intervening in an episode during the prelude to suicidal behavior deploys caregiver resources at a focused point of potentially great impact. The process of 3-221 Report of the Secretary’s Task Force on Youth Suicide intervention can be modelled and under- stood by assuming that a helping caregiver and person-at-risk are identifiable roles. Effective intervention is possible for any caregiver competent in the tasks of the gatekeeping role: recognition, risk-rating, and resource referral. 3. Training opportunities to develop these competencies are available, but many learning experiences are too limited in their content or their intended audience. The Foundation Workshop offers a model program for the development of basic in- tervention competencies that has been successfully presented to more than 4000 caregivers. 4. The successful experience of the car- diopulmonary resuscitation program can be adapted to deliver a training experience that enhances skills in suicide intervention to large numbers of different caregivers. The basic requirement for large-scale delivery involves the availability of an ade- quate number of certified trainers. Al- though a rigorous Training for Trainers experience produces trainers who are committed to and invested in the workshop experience, the need for active sponsors to develop and maintain the net- work that organizes and monitors program delivery remains an unresolved problem. CONCLUSION The need for innovative training and educa- tion programs directed to caregivers active in suicide prevention has repeatedly been recognized and recommended. Our ex- periences in the development and delivery of the Foundation Workshop afford a success- ful model of one such program. With ap- propriate support, a large-scale training program for caregivers that ensures their basic competencies in suicide intervention is a rational and workable suicide prevention strategy. REFERENCES 1. American Association of Suicidology (1977). 3-222 Suicide Prevention Training Manual (1st ed.). Merck Sharp & Dohme, West Point, PA. 2. Ansell, E., McGee, R.K. (1971). Attitudes Toward Suicide Attempters. Bulletin of Suicidology, Fall, 22-28. 3. Bay State Film Productions (Producer) (1978). Handling Suicide Threats. Springfield, Mass. 4. Birtchnell, J. (1983). Psychotherapeutic Con- siderations in the Management of the Suicidal Patient. American Journal of Psychotherapy, 37/1, 24-36. 5. Bloom, V. (1967). An Analysis of Suicide at a Train- ing Centre, American Journal of Psychiatry, 123/8, 918- 925. 6. Boldt, M. (Chairman). (1976). Report of the Task Force on Suicides to the Minister of Social Services and Community Health (The Honorable Helen Hunley). 7. Burdick, B.M., Holmes, C.B., Wain, R.F. (1983). Recognition of Suicide Signs by Physicians in Different Areas of Specialization. Journal of Medical Education, 58, 716-721. 8. Canadian Association for Suicide Prevention (1981). Membership Survey Data. Unpublished manuscript. 9. Center for Studies of Suicide Prevention, National Institute of Mental Health. (1973). Suicide Prevention and Crisis Intervention. Charles Press, Philadelphia, PA. 10. Changing Patterns in Suicide Behavior (1982). Report of a W.H.O. Working Group (Athens, Sept. 29 to Oct. 2, 1981). Euro Reports and Studies, 74, (E,F). 11. Communications Media, University of Calgary (Producer), Ramsay, R., Tanney, B.L., Simkin, R. (Direc- tors). (1982). Cause of Death. (Film). University of Calgary, Calgary, AB. 12. Delong, W.B. & Robins, E. (1961). The Com- munication of Suicidal Intent Prior to Psychiatric Hospitalization: A Study of 86 Patients. American Journal of Psychiatry, 117/8, 695-705. 13. Diekstra, R.F.W., de Graaf, A.C., van Egmond, M. 1984). On the Epidemiology of Attempted Suicide: A mple-Survey Among General Practitioners. Crisis, 5/2, 108-118. 14. Dorpat, T.L., Ripley, H.S. (1960). A Study of Suicide in the Seattle Area. Comprehensive Psychiatry, i, 349-359. 15. Dressler, D., Prusoff, B., Mark, H., Shapiro, D. (1975). Clinical Attitudes Toward the Suicide Attempter. Journal of Nervous and Mental Disease, 150/2, 146-155. 16. Drum, D., Knott, E. (1977). Structured Groups for Facilitating Development. New York: Human Sciences Press. 17. Eisman, E.J. (1977). Training Paraprofessionals in Suicide Intervention. Proceedings of Tenth Annual Meet- ing, American Association of Suicidology, (pp. 89-92). Bos- ton, Mass. 18. Fairbank, R.E. (1932). Suicide: Possibilities of Prevention by Early Recognition of Some Danger Signals. Journal of the American Medical Association, 98, 1711- 1714. 19. Flood, R.A., Seager, C.P. (1968). A Retrospective Examination of Psychiatric Case Records of Patients Who Subsequently Committed Suicide. British Journal of Psychiatry, 114, 443-450. 20. Gentry, J.H. & Woods, J. (1986). Suicide Interven- tion: Assessing Teenagers at Risk. Department of Health and Human Services, Public Services, Rockville, Maryland. 21. Ghodse, AH. (1978). The Attitudes of Casualty Staff and Ambulance Personnel Towards Patients Who Take Drug Overdoses. Social Science and Medicine, 12, 341-346. 22. Goldney, R.D., Bottrill, A. (1980). Attitudes Toward Patients Who Attempt Suicide. Medical Journal of Australia, 2, 717-720. Tanney: . . .Improving the Competency of Caregivers 23. Gurrister, L., Kane, R.A. (1978). How Therapists Perceive and Treat Suicidal Patients. Community Mental Health Journal, 14/1, 1-13. 24. Hawton, K., Kane, R.A. (1976). General Practice Aspects of Self-Poisoning and Self-Injury. Psychological Medicine, 6, 571-575. 25. Hawton, K., Marsden, P., Fagg, J. (1981). The At- titudes of Psychiatrists to Deliberate Self-Poisoning: Com- parison with Physicians and Nurses. British Journal of Medical Psychology, 54/4, 341-348. 26. Hawton, K., Grady, O., Osborn, M., Cole, D. (1982). Adolescents Who Take Overdoses: Characteristics, Problems and Contacts With Helping Agencies. British Journal of Psychiatry, 140, 118-123. 27. Heimburger, E.M., McCallum, R.N., Pratt, M. 1980). Facts About Suicide: How Knowledgeable is the rimary Care Physician? Missouri Medicine, 77/6, 295- 298. 28. Henderson, A.S. (1974). Care Eliciting Behavior in Man. Journal of Nervous & Medical Disease, 159, 172. 29. Hipple, J. (1981). Differences Between Profes- sional Helpers in the Ability to Assess Suicidal Signs. Proceedings of the 14th Annual Meeting American As- sociation of Suicidology. (pp. 17-18). 30. Holmes, C.B., Howard, M.E. (1980). Recognition of Suicide Lethality Factors by Physicians, Mental Health Professionals, Ministers, and College Students. Journal of Consulting and Clinical Psychology, 43/3, 383-387. 31. Ironside, W. (1969). latrogenic Contributions to Suicide and a Report on 37 Suicide Attempts. New Zealand Medical Journal, 69/443, 207-211. 32. Krajic, K.E. (1985). Four Approaches to Interview- ing About Suicide. (Videotape). Albert Einstein College of Medicine, Bronx, New York. 33. Kreitman, N., Chowdhury, N. (1973). Distress Be- havior: A Study of Selected Samaritan Clients and Parasuicides ("Atternpted Suicide" Patients) Part 2: At- titudes and Choice of Action. British Journal of Psychiatry, 123/572, 9-14. 34. Lang, W.A,, Ramsay, D.R., Tanney, B.L., Tierney, R. (1986). A Suicide Intervention Model. (Submitted for publication). 35. Macintosh, J.L., Hubbard, R.W., Santos, J.F. (in press), Suicide Facts and Myths: A Study of Prevalence. ath Studies. 36. Maris, R. (Chairman) (1972). Education and Train- ing in Suicidology for the Seventies. IN: H. Resnick and C. Hawthorne (Eds.), Suicide Prevention in the Seventies. Washington: U.S. Government Printing Office. 37. Middleman, R. go . The Pursuit of Competence in Structured Groups. A. Maluccio (Ed.). Promoting Com- petence in Clients. New York: The Free Press. 38. Murphy, G.E. (1975). The Physician's Respon- sibility for Suicide, II: Errors of Omission. Annals Internal Medicine, 82, 305-309. 39. Patel, A.R. (1975). Attitudes Towards Self-Poison- ing. British Medical Journal, 2, 426-494. 40. Porkorny, A.D. (1960). Characteristics of Forty-Four Patients Who Subsequently Committed Suicide. Archives of General Psychiatry, 2, 311-323. 41. Polymorph Films (Producer). (1982). Psychiatric Emergency Care, Part |: Pre-Hospital Care of the Suicidal Patient. Distributor: N/A. Source: Media Review Digest, 1982. 42. Ramon, S. (1980). Attitudes of Doctors and Nurses to Self-Poisoning Patients. Social Sciences and Medicine, 14A, 317-324. 43. Ramon, S., Bancroft, J.H.J., Skirmshire, A.M. J1873), Attitudes Towards Self-Poisoning Among icians and Nurses in a General Hospital. British Jour- nal of Psychiatry, 217, 257-264. 44. Ramsay, R., Ost, A, Tanney, B.L. (1986). Suicidol- ogy Courses: A Survey of Canadian Universities. Un- published manuscript. 45. Ramsay, R., Tanney, B., Tierney, R., Lang, W. (1983). A Suicide Prevention Training Program: Trainers’ Handbook. Calgary: Canadian Mental Health Association (Alberta Division). 46. Ramsay, R.F., Tanney, B.L., Tierney, R.J., Lang, W. (1986). A Curriculum and Training Program for Suicide Prevention: Help for the Helpers. Manuscript submitted for publication. 47. Reid, P., Smith, H, (1990); Knowledge About Suicide peng Members of Helping Agencies in Ireland. Journal of the Irish Medical Association, 73/3, 117-119. 48. Ringel, E. (1973). The Pre-Suicidal Syndrome. Psychiatrica Fennica, 209-211. 49. Robins, E., Gassner, S., Kayes, J., Wilkinson, R.H., Murphy, G.E. (1959). The Communication of Suicidal In- tent: A Study of 134 Consecutive Cases of Successful (Completed) Suicide. American Journal of Psychiatry, 115, 724-733. 50. Rockwell, D.A., O'Brien, W. (1973). Physicians’ Knowledge and Attitudes About Suicide. Journal of the American Medical Association, 225/11, 1347-1349. 51. Royal, P. (Chairman) (1979). Report of the Com- mittee on the Nature of, and Response to, Personal and Family Crisis in the Province of Alberta. Volume | - The Report. Minister of Social Services and Community Health, Edmonton, Alberta. 52. Schapira, K., Davison, K., Hug, Z. (1978). The Role ofthe Family Doctor in the Prevention of Suicidal Attempts. Aspects of Suicide in Modern Civilization. In: H.Z. Winnik, and L. Miller feds) Proceedings of the 8th International Congress on Suicide Prevention and Crisis Intervention. Jerusalem, Jerusalem Academic Press, 183-189. 53. Shneidman, E.S. (1984). Aphorisms of Suicide and some Implications for Psychotherapy. American Journal of Psychotherapy, 38, 319-328. 54. Shneidman, E.S., Farberow, N.L. & Leonard, C.V. (oes Some Facts About Suicide: Causes and Preven- tion. (Publication No. 101). Washington, D.C. U.S.Govern- ment Printing Office. 55. Shneidman, E.S., Farberow, N.L., Litman, R.E. (1970). The Psychology of Suicide. Science House, New York. 56. Snyder, JA. (1971). The Use of Gatekeepers in Crisis Management. Bulletin of Suicidology, 7, 39-44. 57. Steele, T.E. (1975). Evaluation of First-Year Medi- cal Student's Ability to Recognize Suicidal Potential. Jour- nal of Medical Education, 50, 203-205. 58. Stoudemire, A.,, Thompson, T.L., Mitchell, W., Grant, R.L. (1982-83). Family Physicians Perceptions of Psychosocial Disorders: Survey Report and Educational Implications. International Journal Psychiatry in Medicine, 12(4), 281-287. } 59. Sym, J. (1637) Life's Preservative Against Self-Kill- ing. Dawling & Fawne; London, England. 60. Tanney, B.L., Ramsay, D.R. (1985). Certifying All Caregivers in the Foundations of Suicide Prevention: A Proposal Whose Time Has Come. Proceedings of the An- nual Meeting of the American Association of Suicidology. Toronto, Ontario. 61. Wheat, W.D. (1960). Motivational Aspects of Suicide in Patients During and After Psychiatric Treatment. Southern Medical Journal, 53/3, 273-278. 62. Whittemore, K.R., Nugent, J., Boom, P. (1972). Suicide and the Physician: Experience and Attitudes in the Community. Journal of the Medical Association of Geor- gia, 61, 307-311. 3-223 THE SAMARITANS AND THE PREVENTION OF YOUTH SUICIDE Richard G. Katzoff, M.S., Treasurer, Samaritans USA, W. Alton Jones Campus, The University of Rhode Island, West Greenwich, Rhode Island SUMMARY Samaritans U.S.A. is a network of 14 branch hotlines in the eastern United States bound together by a common set of principles and practices. The primary purpose of the branches is to provide a 24-hour crisis line for the lonely, despairing, and suicidal, employ- ing a method called "befriending". Seven branches have outreach programs for intervention and prevention of youth suicide. Two exciting new models have been developed by Samaritans branches in Boston, Massachusetts and Providence, Rhode Is- land. In Boston, the Samaritans are implementing a special hotline staffed by trained teenagers called "Samariteens." In Rhode Island, the Samaritans have developed a suicide awareness curriculum to be taught by high school teachers that has recently been made part of the required high school curriculum Statewide. INTRODUCTION To understand the Samaritans and their role in the prevention of youth suicide, one must first understand the Samaritans themselves. Founded in 1953 in Great Britain by Chad Varah, the Samaritans now have 275 branches worldwide--180 branches and more than 22,000 volunteers in Great Britain alone. In the United States there are 14 3-224 branches with more than 700 volunteers. The Samaritans represent a unique approach to the prevention of suicide. Samaritans branches are non-professional services which offer "befriending" rather than coun- seling. While the Samaritans have the back- up services of professional counselors and consultants, the telephones are always answered by volunteers who have been care- fully chosen for their ability to make an equal-level relationship, to listen, and to befriend the troubled person without offer- ing judgments, unwanted advice, or un- solicited intervention. The Samaritans do not trace calls or take any other unrequested action, thus insuring that the service is com- pletely confidential. HISTORY This model differs in some ways from most other suicide prevention agencies in the United States. To appreciate the Samaritans commitment to confidentiality and befriend- ing, it is useful to understand the history of the Samaritans movement and the develop- ment of befriending as an approach to the prevention of suicide. In 1935, Chad Varah’s first duty after ordina- tion as an Anglican priest was to officiate at the burial of a fourteen year old girl who killed herself when her menstruation started, not understanding menstruation and having R.Katzoff: The Samaritans and Prevention of Youth Suicide no one to ask. Moved by this experience, Varah began to counsel his parishioners about human sexuality. Soon he developed a reputation as a sex therapist and clients sought him out in great numbers--many of them suicidal. Over the years, he was moved to expand his ministry and his counseling ac- tivities to include outreach to the suicidal. On the first of November 1953, Varah opened both a telephone and walk-in coun- seling service for the suicidal at St. Stephen Walbrook. Within a short period of time, both the number of callers and the number of people coming in for counseling far ex- ceeded Varah’s ability to respond. Publicity for Varah’s services had attracted not only people in need of help but also a large num- ber of people who felt they could give help. While some of these people offering assis- tance were professionals, most were ordinary people and instinctively Varah began to in- volve these volunteers in his work. These people did not see themselves as counseling the suicidal but merely as helpers who could pour tea or lend an ear to others as they waited for Varah. In just a few weeks, Varah began to notice that the number of people he was seeing personally began to diminish; those who did come to him for counseling were easier to help because of the time they had spent with the volunteers. He also noted that the people he was seeing were rightly judged to be in need of the kind of profes- sional attention the volunteers could not give. This fascinated him and he began to try to discover what the volunteers were doing and why was it so successful. By both interviewing and observing the volunteers he realized that they were provid- ing a new kind of listening therapy. They provided an ear, but not advice. They provided sympathy and love, but no judg- ments. They avoided discussions of religion and God and just listened to what a caller or visitor had to say. They were "befriending." Befriending, as developed by Chad Varah’s experience with these volunteers became his new work. He met with the volunteers regularly, recruited and screened new volun- teers, and dismissed volunteers whom he thought were not providing "befriending". The name Samaritans came out of a headline describing Varah as the "Samaritan Priest"-- a compliment he passed on and ascribed to the volunteers. For the next five years, Varah experimented with and perfected the Samaritans’ particular approach to the suicidal and in 1959 he began to support and encourage the development of other branches in Great Britain. In 1960, the ap- proach went worldwide as Samaritans opened new branches in Hong Kong and Bombay. Samaritans are "...themselves and what they have to give--namely, their personal concern, their time, attention and friendship. The Samaritan listens, accepts, cares; and this can make all the difference for those who feel that no one has time for them, that they are rejected, and that nobody cares" (1). This simple statement by Chad Varah describes the heart of the Samaritans movement and why the approach of befriending spread so quickly and so far. In 1974, the Samaritans movement was brought to the United States by Monica Dickens who founded the first U.S. branch, the Samaritans of Boston. The second U.S. branch, the Samaritans on Cape Cod, initial- lyshared a board with the Boston Samaritans. In 1980, Samaritans USA was incorporated to set standards for new branches, to support the development of new branches in the United States and to fundraise for the Samaritans on a national scale. PRINCIPLES AND PRACTICES The Seven Principles and Practices, as agreed upon by the Council of Management at its meetings in June and November 1981, are the standards by which Samaritans branches operate in the United States and worldwide. They are: Seven Principles 1. The primary aim of the Samaritans is to be available at any hour of the day or night to 3-225 Report of the Secretary’s Task Force on Youth Suicide befriend those passing through personal crises and in imminent danger of taking their own lives. 2. The Samaritans seek to alleviate human misery, loneliness, despair, and depression by listening to and befriending those who feel that they have no one else to turn to who would understand and accept them. 3. Acaller does not lose the freedom to make his/her own decisions, including the decision to take his/her own life, and is free to break contact at any time. 4. When a person asks help of the Samaritans, the person’s identity and everything he/she has said is completely confidential within the organization un- less permission is freely given by the caller for all or part of such information to be communicated to someone outside the or- ganization. A Samaritan volunteer is not permitted to accept confidences if a con- dition is made that not even the Director should be informed of them. 5. Samaritan volunteers, in befriending callers, will be guided and actively sup- ported by experienced leaders who will have the advice, when required, of profes- sional consultants. 6. In appropriate cases the caller will be in- vited to consider seeking professional help in such fields as medicine and social work, and material help from other agencies. 7. Samaritan volunteers are forbidden to im- pose their own convictions or to influence callers in regard to politics, philosophy, or religion. Seven Practices 1. Samaritan volunteers are carefully selected and prepared by the local branch in which they are to serve. 2. The Samaritans are available at all hours to callers, and may be contacted (anonymously if desired) by telephone or personal visit, or by letter. 3. When a caller is believed to be in danger 3-226 of suicidal action, the Samaritan is par- ticularly encouraged to ask the caller’s permission to maintain contact during the crisis. 4. Samaritans offer longer-term befriending of callers where appropriate, while recog- nizing that the Branch may, from time to time, have to set limits. 5. Samaritans listen to those concerned about the welfare of another person, and, if satisfied that the third person is despair- ing, depressed, or suicidal, may discreetly offer befriending. 6. Samaritans are normally known to callers only by a forename and contacts by callers maybe made only through the branch cen- ter. 7. Samaritan branches are banded together in a legally constituted association whose Council of Management represents all the branches and reserves to itself the ap- pointment of the person in charge of each branch (2). While the primary purpose of all Samaritans branches is to provide a 24-hour crisis line for the lonely, despairing, and suicidal, almost all the branches have outreach programs per- taining to suicide and its prevention. Like many other suicide prevention agencies, the Samaritans have been involved in the schools as their primary outreach--speaking to clas- ses and assemblies, holding workshops for parents and educators, and befriending sur- vivors after a suicide has taken place. While the principles and practices remain constant throughout the Samaritans branches in the United States, the outreach programs of the various branches differ greatly. SAMARITANS OF BOSTON The oldest of the Samaritans branches in the United States, in operation since 1974, the Samaritans of Boston receive almost 250 calls or visits daily. Noticeably more and more of these callers are young people who are seriously depressed or suicidal. Since 1982, the Samaritans of Boston have focused spe- R.Katzoff: The Samaritans and Prevention of Youth Suicide cial attention on the problem of teenage suicide through their Youth Outreach Project. Instead of waiting for young people to call or visit, the Samaritans focused their efforts on prevention through education. They thoroughly researched the extent of the problem in Massachusetts and designed educational materials based on the results of that research. They began their outreach by distributing their pamphlets to all schools and colleges in the State of Massachusetts. They followed this with an offer to provide a speaker for any class or staff group that was interested. Both the oral and the written information was divided into two sections one for the stu- dents themselves and one for teachers and counselors working with those students. For the professionals, the booklet, "Teens: Depression and Suicide", delineated the problem, identified warning signs of depres- sion and suicide risk in children, offered sug- gestions on how to help, and provided information on where to go for help. The in- formation is presented in a clear and direct manner, Viz.: Parents, teachers, and friends of depressed or suicidal young people often ask the Samaritans what they can do to help. The im- portant thing is to pay attention. Encourage them to talk. Listen. Be on their side. Reas- sure without dismissing. Don’t panic. Remember that no one is suicidal all the time. Thoughts of self destruction arise at times of crises, but lives can be saved by understanding and support. Learn to recognize the signs of serious depression and suicide risk. Eight out of ten suicides give definite warnings, ver- bal or behavioral, of their inten- tions... These signs of depression do not invariably mean that young people are contemplating suicide, but they alert you to the need to explore more carefully their state of mind.... Don’t be afraid to ask, "Do you some- times feel so bad you think of suicide?" ....Discussing suicide openly is one of the most helpful things you can do. It shows that you are taking the person seriously, and that you care. If the answer is "yes," follow through by asking, "Have you thought about how you might do it?"....It is vital not to under- estimate the danger by not asking for details. If you think there is immediate danger, DO NOT LEAVE THE PERSON ALONE. Stay with him until the crisis passes or help arrives... The Samaritans are always available to help you or the person in danger.... If the person is hallucinating, affected by drugs or alcohol, if an attempt has begun or is imminent, do not try to go it alone. Stay with the person, and contact any of the following: an ambulance service in your town, your local police, emergency room of a local hospital, a trusted adult, the Samaritans (3). The Samaritans also designed pamphlets geared to young people themselves, offering much of the same information in an easily un- derstandable style. These pamphlets include the warning signs of depression and suicide risk but are presented in question form. Myths and facts about suicide are also in- cluded, places to turn for help, and guidelines for handling an emergency. The Youth Outreach Project of the Samaritans of Boston has been a tremendous success. Now coordinated by a full-time per- son, the materials described above have been mailed annually since 1982 to 1500 schools and colleges. In addition, Samaritans have given more than 500 talks on the subject of teenage suicide and have followed-up on more than 2500 requests for additional infor- mation. The number of teenagers contacting the Samaritans of Boston has increased over the life of the program from 10.8 percent of the total client volume in 1982 to 16.6 per- cent in 1985 (4). Report of the Secretary’s Task Force on Youth Suicide In 1985, the program was expanded to in- clude information on the relationship of al- cohol to youth suicide. Two pamphlets, one oriented to teenagers and the other to their parents, were published in 1985, both en- titled, "Drinking and the Teenager." In 1986, the Youth Outreach Project of the Samaritans of Boston made an additional commitment to the prevention of adolescent suicide through the creation of a special and separate telephone hotline for teenagers, staffed by teenagers themselves who are called SAMARITEENS. SAMARITANS ON CAPE COD The Samaritans have developed multiple relationships with the schools in Cape Cod, Massachusetts. They have provided workshops for teachers, both for credit and not for credit, on befriending and suicide awareness. They also make presentations to student groups and church youth groups, usually about once a month that often in- clude a screening of the film, "Urgent Mes- sages." They are often called into a school after a suicide to provide postvention ser- vices. In addition, this group consults with an active chapter of Students Against Drunk Driving (S.A.D.D.), a relationship that not only sup- ports the activities of that organization, but provides access and legitimacy for the Samaritans among a large population of stu- dents. SAMARITANS OF RHODE ISLAND The unique small size of Rhode Island is con- ducive to Statewide activities and interven- tions. The Providence branch of the Samaritans serves the whole State of Rhode Island, both in its telephone service as well as its outreach programs. In 1985S, the Rhode Island Samaritans conducted almost 200 suicide information seminars in the school- rooms of Rhode Island. Realizing that this approach only began to meet the need that 3-228 existed in the schools, the Samaritans of Rhode Island began work on a model that would institutionalize the teaching of suicide awareness Statewide in the schools. Merely believing that the education of stu- dents in the classroom is an effective tool in suicide awareness and prevention is not enough to garner the kind of support neces- sary to implement a new Statewide program. With an issue as emotional as adolescent suicide, documentation and evaluation are necessities. The Samaritans of Rhode Island first developed a manual to be used in the teach- ing of suicide awareness in the classroom. The author of the manual, George J. Fincik, is both a Samaritan and a secondary school English teacher. The manual consists of five lessons: 1. Developing a compassionate attitude toward suicide and its victims. 2. Acquiring knowledge about suicide. 3. Developing an awareness of the signs sig- nalling suicide. 4. Developing befriending skills. 5. Building a support system. Through these five lessons the manual in- tends to focus the attention of the teaching community on adolescent suicide, to involve teachers in a realistic way in suicide preven- tion, and to introduce befriending as a skill and approach. Fincik, in his introduction to the manual, states this intent most eloquent- ly: We can no longer perceive our roles only as dispensers of knowledge. The stu- dents sitting in front of us are not mere receptacles for ideas, facts, values, trends, data, or events. Those sitting in front of us are caught up in the pressure- cooker of modern life. These human beings are sometimes seething with feel- ings that are incomprehensible, with a sense of helplessness, hopelessness and futurelessness that is emotionally debilitating, with ideas that confuse R.Katzoff: The Samaritans and Prevention of Youth Suicide rather than clarify, and with demands that would defy a Hercules. Such human beings are hardly in a condition to take our classroom activities seriously. And who but teachers are in the best position to observe students individually or in groups, to sense the emotions seeth- ing under the surface, to monitor subtle changes in behavior, and to realize that the student in reality might be a hurting human being? But if the teacher lacks knowledge about suicide--its causes and its warning signs-- the teacher very well could be unaware that a life and death drama might be developing. It is too late to rewrite the lesson plans of life once the student rips up the original copy (5). With the help of a grant from the National Conference of State Legislatures, the Samaritans of Rhode Island embarked on a pilot program to assess the effectiveness of the manual and its awareness program in four Rhode Island schools. Four corresponding, non-participating schools were used as con- trols. Both the experimental schools as well as the control schools represented a sampling of teenagers in the State: urban, suburban, rural and suburban/rural mixed. In October 1985, baseline data were col- lected at all eight schools. Following that, the Samaritans held a two day workshop for the participating teachers from the four ex- perimental pilot schools. Four sessions were developed by the Samaritans to help the teachers confront their feelings and attitudes about suicide and to inform them further “about youth suicide. The sessions were: 1. Suicide information, statistics, studies, and programs. 2. Attitudes toward death and suicide. 3. Befriending: the art of active listening. 4. The manual, crisis intervention protocol, and school protocols. Samaritans staff reported that teachers par- ticipating in these sessions were extremely positive in their evaluations, and felt more skilled and knowledgeable. Following this training, the teachers implemented the cur- riculum model from the manual. Followup assessments were done at all eight schools with four different measures employed as assessment tools: knowledge about suicide, student attitudes toward suicide, personal knowledge of suicide, and feelings of hopelessness. The pilot program based on both participant evaluation and external evaluation was con- sidered a success. The Samaritans, in collaboration with the Rhode Island Department of Education, the Rhode Island Department of Health, and the Rhode Island Task Force on Adolescent Suicide Prevention developed a plan to en- sure that this model was implemented Statewide. The Rhode Island Department of Education has incorporated suicide prevention into their mandatory health cur- riculum. In conjunction with Rhode Island College, the Samaritans will offer a two credit graduate course in suicide education so that teachers can learn the new mandated cur- riculum. The Rhode Island Department of Health has set an internal objective for 1990 that states "...that greater than 60 percent of young people, ages 15 through 24, should identify a suicide prevention hotline." And the State legislature through a bill introduced on behalf of the Task Force on Adolescent Suicide Prevention, has made an appropria- tion of $35,000 to the Samaritans of Rhode Island to coordinate the implementation of the Statewide suicide curriculum through the continued refinement of the manual and the training of health teachers Statewide in its use. The preliminary results of the Samaritans of Rhode Island study were presented at the American Association of Suicidology meet- ing in April 1986 and will appear in the Rhode Island Journal of Medicine in Sep- tember 1986. A followup study is scheduled for June 1986. 3-229 Report of the Secretary's Task Force on Youth Suicide SAMARITANS OF THE MERRIMACK VALLEY Located in Lawrence, Massachusetts, the Samaritans of the Merrimack Valley in- tegrate their outreach efforts with the efforts of other agencies in the Merrimack Valley. They consult with a high school peer coun- seling project, "The Connection," that produces an improvisational theater program on adolescent suicide. They par- ticipate in the Town of Andover’s Assess- ment-Support-Knowledge (A.S.K.) Program, providing wallet cards with hotline numbers to all the school children in the area aswell as integrated and coordinated preven- tion and intervention programs. Their outreach efforts also include visits to 20 to 25 schools in the Lawrence/Haver- hill/Lowell area, direct mailings of suicide re- lated articles, showings of the video "Teen Suicide--What Can We Do?", and postven- tion services to communities that have ex- perienced a suicide. SAMARITANS IN KEENE Like all Samaritans branches, the Samaritans in Keene, New Hampshire, reach out to the schools, youth groups, and colleges in their area with brochures and offer to present on- site programs. Last year the presented al- most 35 programs to groups of young people. The Keene Samaritans are also taking ad- vantage of the experiences of other Samaritans branches in the United States, both in their current efforts as well as in their plans for new outreach activities and programs. They currently distribute a brochure entitled "Suicide Prevention--A Guide for Students" developed by the Samaritans of Rhode Island. In addition they are in the process of writing a grant proposal that would fund both a Samariteens line such as that developed by the Boston branch as well as an improvisational youth theatre group similar to the one supported by the Merrimack Valley branch. 3-230 SAMARITANS OF SOUTH MIDDLESEX Serving the Framingham, Massachusetts area, the Samaritans of South Middlesex direct-mail a brochure to all schools in their locale. Similar to the other Samaritan branches, they offer a followup visit includ- ing a showing of the film "Urgent Message." One unique aspect of the outreach of the Samaritans of South Middlesex is that they emphasize to their young audiences that they can call collect. This, the Samaritans believe, encourages more young people to use the service who might otherwise fear the Samaritans number appearing on the parents’ telephone bill. SAMARITANS OF FALL RIVER/NEW BEDFORD The Samaritans of Fall River/New Bedford is the newest full branch in the Samaritans USA network, having opened in the spring of 1984. Similar to the other full branches of the Samaritans, the Fall River/New Bedford group has reached out beyond the hotline and walk-in service to the schools and col- leges in its area. Two schools have videotaped the Samaritans’ presentation and copies are available in their school libraries whenever anyone wants to see them. In addition, the Samaritans have been invited to produce and plan a workshop on "befriending" for all guidance counselors in the Fall River, Massachusetts system. The Fall River/New Bedford Samaritans are par- ticularly interested in sharing the model of befriending-acceptance without judgment— with key professional people in the schools. OTHER BRANCHES The other seven branches that currently make up Samaritans USA are considered either probationary or preparatory branches, still working at meeting all the guidelines and standards for full status. Located in Chicago, New York City, Albany, Hartford, Washington, D.C., the South Shore of Mas- R.Katzoff: The Samaritans and Prevention of Youth Suicide sachusetts, and South Central New Hampshire, these branches are focusing on providing 24 hour befriending services to their areas. Major outreach activities by Samaritan branches are prohibited prior to attaining full status. Samaritans USA is still a relative newcomer to the field of youth suicide prevention in the United States. Until recently located primarily in New England, Samaritan branches are now opening in a wider geographical area as indicated in the list of probationary and preparatory branches. As the number of branches spreads throughout the United States, sodoes the im- pact of the Samaritans approach to youth suicide. The Samariteens program in Boston and the suicide awareness curriculum in Rhode Island are quite possibly models for the entire country. And the heart of the Samaritans, befriending, is an approach that has worked and is working worldwide. REFERENCES +i sossissessomersemsessransan ‘ 1. Varah, Chad. The Samaritans: Befriending the Suicidal. Revised edition, 1985. Great Britain: St. Ed- mundsbury Press, 0 09 466110 3. 2. Ibid. 3. Teens: Depression and Suicide. The Samaritans of Boston. 4. The Samaritans Youth Outreach Project: 1982 - 1985. The Samaritans of Boston. 5. Fincik, George J., A Teachers Manual for the Prevention of Suicide Among Adolescents. Providence, Rhode Island: The Samaritans, Inc., 1985. 3-231 EVALUATION AND MANAGEMENT OF SUICIDAL RISK IN CHEMICALLY DEPENDENT ADOLESCENTS John E. Meeks, M.D., Medical Director, Psychiatric Institute of Montgomery County, Rockville, Maryland SUMMARY The chemically dependent youngster who becomes suicidal appears to be characterized by a strong family history of chemical abuse, obvious signs of depression and hopeless- ness, a stated desire to die, and a recent loss or separation, most commonly, parental separation or divorce. Recognizing the crisis and providing care and protection is the most crucial aspect of the prevention of suicide in these cases. However, the underlying illness is chronic and requires long term treatment with con- tinued alertness to the possibility of a recur- rence of suicidal risk. INTRODUCTION The correlation between heavy drug use and suicidal behavior is evident to a wide range of clinicians (1,6,7,9). According to Frances (4), half of all suicides are associated with al- cohol use. In recent years, large scale studies of mortality in psychiatric patients show that a history of alcoholism or drug dependency greatly increases the likelihood of untimely death among both psychiatric outpatients and inpatients (1,6). Much of the increased mortality in these patients can be attributed to suicide. Recent studies of adolescent populations have confirmed these trends in young people, a finding that does not surprise therapists who treat adolescents and are familiar with this phenomenon in their clini- 3-232 cal practices. The strong relationship between severe depressive illness and suicidal behavior has also been widely recognized (7,8,9). A high rate of depressive symptomatology is also ob- vious in chemically dependent patients. In- deed, Robins and Alessi (9) have theorized that drug abusing depressives are a special subpopulation with an exquisite sensitivity to dysphoric affects and a very limited capacity to tolerate them. Thus they would be more likely to use chemicals in an effort to ameliorate depression and more likely to be tempted to turn to suicide as a permanent solution to their painful state. In spite of considerable interest and research, however, the relationship between the syndromes of depression and the syndromes of addiction is not completely clear. For example, recent literature suggests that addictive behavior and the depressive syndromes are two separate illnesses with some degree of over- lap. Itis also clear that a great deal of depres- sion encountered in chemically dependent individuals is a result of addiction rather than its cause (2,5,10). Many elements of the addictive experience increase the likelihood of depression with at- tendant feelings of helplessness, despair, and suicidality. These factors are enhanced in adolescents because of the young person’s relative lack of a time perspective, the ten- J.Meeks: Evaluation & Management of Suicidal Risk... dency to be action oriented, and the heightened impulsiveness characteristic of immaturity. Even if drug-using, depressed adolescents do not represent a specific high risk population as suggested by Robins and Alessi, many, al- most inevitable effects of habitual drug use increase the likelihood of suicidal behavior. As mentioned earlier, many of the drugs themselves produce central nervous system depression which is often accompanied by subjective dysphoria. Alcohol, for example, is notorious for producing a morose and ir- ritable mental state when the blood alcohol level begins to diminish. Other drugs produce a depressive frame of mind on withdrawal, particularly cocaine (35). Most of the depression produced by chemi- cal dependency, however, results from its im- pact on overall life style. The inability to control one’s actions creates a sense of help- lessness and demoralization. In addition, the adolescent heavily involved with drugs amas- ses a constantly increasing load of guilt. This guilt is related to alienation from family and other responsible adults, accumulating failures and disloyalties, and personally unac- ceptable behaviors which the adolescent had to perform in order to get drugs, or did them because his/her judgment and self control were impaired by intoxication. This increas- ingly heavy mental baggage of regret and remorse is held just out of consciousness by denial, projection, minimalization, and heavier use of drugs. It looms behind the adolescent as a monstrous shadow, however, ever growing and ever closer to overwhelm- ing the youngster’s self-esteem and pleasure in life. The problems produced by a lifestyle of heavy drug use are accentuated during times of withdrawal because of the physiological depression and irritability which accompany the process of detoxification from many of the psychoactive drugs. The withdrawal from cocaine is frequently accompanied by suicidal ruminations (5), although actual suicidal behavior has not been as frequently reported during withdrawal as it has been during the state of intoxication itself. For ex- ample, many alcoholics who commit suicide are discovered to have high blood alcohol readings at death. There is considerable agreement that suicidal behavior in drug-involved individuals is frequently triggered by experiences of loss. Very often these losses are the direct result of drug use. For example, the patient may get into serious legal trouble, face major disrup- tions of family life, be dropped from an ath- letic team or experience some other loss or life failure clearly resulting from drug use. On the other hand, these individuals seem to be vulnerable to losses that are unrelated to the drug problem, such as the death of a parent or geographic move which disrupts support networks. It is as though the chemi- cally dependent individual is more likely to experience loss than the average person, but may also be less able to deal with the psychological consequences of loss experien- ces. CLINICAL PROFILE OF HIGH RISK ADOLESCENTS Evidence of Depression According to Robins and Alessi (9) the most accurate indicator of the likelihood of a serious suicide attempt is the adolescents’ stated sense of hopelessness and a definite statement that they "wish to die." - The author’s clinical experience supports the ob- servation that this information is best ob- tained in an individual interview with the adolescent which should include empathic listening and very direct and persistent ques- tioning regarding the adolescent’s true feel- ings. Diane was a 17-year old girl with a history of periodic rage outbursts, periods of depression, marijuana, alcohol and cocaine abuse, and difficulty in making and keeping friends. She was hospital- ized following a suicide threat which was judged to be serious although it was not accompanied by any attempt. The patient had a stormy early hospital ex- 3-233 Report of the Secretary's Task Force on Youth Suicide perience but was gradually able to discuss her feelings about her mother’s death by suicide and her own difficulty in main- taining a positive mood. She improved and was discharged to her home. Con- tinuing outpatient individual and group psychotherapy aftercare helped Diane maintain an adequate adjustment in her adoptive family until the adoptive father’s work required a geographic move. Although Diane was placed in psychotherapy in the new location and seemed to be making an adequate adjust- ment, she called a friend in her old neigh- borhood and complained that she was not able to make close friends in the new location and that she had begun to use cocaine again. A week later she called another friend stating enigmatically that she "just wanted to hear her voice". A week later she overdosed on her an- tidepressant medication and died. This tragic case reminds us that adolescents often offer meager clues which they expect us to perceive and actively pursue. Unfor- tunately, as in the case of Diane, these clues are often very confusing. Diane’s friends were touched by her telephone calls and since they saw mainly the somewhat abrasive, behaviorally disordered facade which Diane used to protect herself from being hurt in de- pendency relationships, they simply did not recognize the warning. It is unknown whether Diane provided any warning to her new therapist. It is very likely that no clues were offered although Diane had lived in the new location for more than six months. Diane’s depression, substance abuse, and death also illustrate other important risk fac- tors. As previously mentioned, her mother was a suicide victim. The mother committed suicide on a holiday evening during Diane’s seventh year. Diane herself committed suicide within a week after that time ten years later. At the time of her hospitalization the psychologist who tested Diane noted, "thoughts of death are often on her mind." Diane also showed a pervasive difficulty in 3-234 forming and maintaining comfortable de- + pendent relationships. She had a mild learn- ing disorder which interfered with her effective functioning in school in spite of a tested 1.Q. in the superior range. All of these factors contributed to her reputation as a troublemaker and smart-aleck. They also disguised to some extent her hunger for ac- ceptance and affection. Finally, the geographic move with its disruption of Diane’s therapy and friendship relationships probably contributed to her sense of loss and triggered the overwhelming sense of hope- lessness and depression that led to her suicide. A second brief case vignette offers addition- al examples of the same risk factors. Dawn, a 14-year old Caucasian female, was raped by a girlfriend’s father at a slumber party one year prior to admis- sion. Because of the complex set of cir- cumstances which probably included some poor legal advice, no action was taken against the rapist. Dawn responded to the event with severe psychological problems. She became very rebellious at home, slept excessive- ly, lost about 20 pounds and appeared to lose interest in all of her previous ac- tivities, including her school work. She began to act out sexually, caught gonor- rhea, became pregnant, and had an abor- tion. She also began to use drugs heavily, particularly marijuana and cocaine, and started running away from home for prolonged periods. Her depression only became clearly evident, however, when her parents separated and began plan- ning for a divorce two months prior to her admission. When her mother told her of their plans, Dawn became profoundly depressed and made a very serious suicide attempt, taking all the medica- tions she could find in the family medicine cabinet. She was hospitalized in a coma and transferred for psychiatric treatment when medically cleared. At the time of hospitalization, she appeared less depressed and was demonstrating J.Meeks: Evaluation & Management of Suicidal Risk... some of her previous defensive measures. Psychological testing reported defenses of "denial, avoidance, and externalization". Psychological test- ing also revealed visual perceptual motor problems for which Dawn seemed to be attempting to compensate. CASE SURVEYS To provide some numerical support to anec- dotal clinical impressions, the last nine youngsters admitted for serious suicide at- tempts who also reported drug abuse, were reviewed. Certain features are interesting, although is is obvious that we cannot draw definite conclusions from such a small group or even state which hypotheses might best ex- plain the case characteristics. Two findings are extremely striking. A clear history of alcoholism within a family member (no more distant than uncle or grandparent) existed in eight of the nine youngsters (five girls and four boys). Three fathers, one mother, and four other close family members (two grandfathers and two uncles) had docu- mented alcoholism. In some cases, several family members were known to be alcoholic. The second striking finding was that asepara- tion or divorce had occurred within the pre- vious two years in six of the nine cases. In two other cases, the probability of separation was being openly discussed by the parents at the time of hospitalization. A history of a suicide attempt or completed suicide was noted in a family member in only two of the nine cases and only three cases had recently experienced a geographical move. Four of the youngsters had evidence of primary learning disability through both his- tory and psychological testing. It is difficult to say exactly what these obser- vations mean, but they suggest that the population is a highly vulnerable one. There is a major hereditary susceptibility for sub- stance abuse. Central nervous system dys- function or instability demonstrated through the presence of learning disabilities was also evident in four of the youngsters and a fifth had a seizure disorder. Perhaps these vul- nerabilities, coupled with problems created by drug abuse help to explain the youngster’s difficulty in maintaining self-esteem and the ease with which they could develop a sense of helplessness and hopelessness that might make suicide attractive to them. Often, de- pendency relationships which might have supported the vulnerable youngsters were tenuous, particularly within their families and, as we have seen in several cases, these tenuous support systems were further dis- rupted by marital disharmony or separation. Perhaps the subjective sense of their lives is better conveyed by Marti, a 15-year old girl who wrote in an English essay, in the hospital’s school, "Why can’t I love anyone? Why can’t I care about myself? Why can’t I keep friends? Why do I hurt people all the time? Why can’t I be straight? I can’t think of any solu- tions. All (I) can think about is the questions and the problems that face me." The psychologist described Marti as "an ex- ceedingly discouraged person who sees her- self as undergoing a great deal of suffering and torment.... The depressed feelings that Marti has are compounded by family problems.... This combination leads Marti to feel quite hopeless about things and there are signs that acting on suicidal feelings presents a continuing threat in her case.... Marti feels quite abandoned by other people.” Sometimes this sense of abandonment simp- ly reflects reality. Sean, a 15-year old white male with a history of bedwetting until age 11, played with matches as a child, and was generally destructive. He was completely rejected by both of his divorced parents. The mother refused to take custody. The father hospitalized Sean after a serious suicide at- tempt but rejected the recommendation for long-term residential treatment or for out- patient treatment since he felt Sean did not deserve "to have another penny spent on his 3-235 Report of the Secretary's Task Force on Youth Suicide meanness." He stated instead, that it was his intention to take him home, let him decom- pensate and then "turn him over to the authorities." On psychological testing Sean "painted a picture of a sordid, morally barren world, in which even loved ones cannot be trusted." MANAGEMENT OF THE SUICIDAL CHEMICALLY DEPENDENT ADOLESCENT The first priority in the treatment of the suicidally active, chemically dependent adolescent is risk identification. As men- tioned previously, the recognition of suicidal risk is not difficult if the index of suspicion is high. The initial evaluation of all adolescents should include careful review of all pertinent suicide risk factors including the degree of hopelessness and stated desire for death, his- tory of suicidal behavior in a relative or friend, history of previous attempts, and the history of recent losses or separations. It ap- pears from the limited data mentioned in the study of the nine cases above that parental separation or divorce seems to be a par- ticularly dangerous loss experience for this population of youngsters. Once a youngster with a high risk of suicide is identified, it is important to remember that continuing caution is indicated since most of the conditions which predispose the youngster to depression and suicidal be- havior are chronic. These basic vul- nerabilities and pathological defenses do not yield easily to treatment and are difficult to alter on a permanent basis. One can easily be overly optimistic about patients who are in treatment. The process of psychotherapy itself temporarily satisfies emotional and de- pendency needs. The experience of being cared for may produce behavioral improve- ment which is not, however, readily internal- ized in the youngster’s psychic structure. The fact is that these youngsters remain highly vulnerable to loss experiences. Any changes in living conditions perceived by youngsters as traumatic can produce new suicidal peril for them and their apparent gains may rapid- 3-236 ly vanish. At the time of acute suicidal risk, aggressive treatment is necessary to protect the youngster. One cannot expect complete cooperation from either the adolescent or the adolescent’s parents in all cases. For the adolescent, suicide often appears as a con- sciously desired solution to his/her chronic unhappiness while parents may consciously or unconsciously accept this outcome also. In some instances, such as Sean’s father, the parent even seems to be actively and con- sciously encouraging the adolescent’s despair and sense of abandonment. Adolescents may resist help because they are ambivalent about desiring death and they fear dependency. Often, they force helpful adults to prove their determination and genuine desire to be of assistance by making the therapeutic process as difficult and un- rewarding as possible. However, when provided firm, clear support of sustaining life and firm disavowal of suicide as an acceptable action, the adolescent will usually abandon suicidal efforts. At times, supervision of the adolescent is necessary on an almost constant basis to insure the youngster’s safety, but as a rule, this kind of caution does not need to be extended over time. When adolescents are prepared to give unequivocal assurances that they will not harm themselves and they will report any suicidal urges to a responsible adult, the promise can usually be trusted. In the same way that adolescents are usually truthful in reporting their plan to kill them- selves when asked directly within a framework of care and concern, their promises that they will not harm themselves- -if delivered with appropriate affect to a trusted adult--are a reliable indicator of their ability and willingness to protect themselves. The active treatment of the concurrent chemical dependency is also very important in these cases. Continuing use of drugs un- dermines self-esteem and produces a sense of alienation from caretakers, both of which increase the danger of suicide. As a rule, traditional psychotherapy alone is J.Meeks: Evaluation & Management of Suicidal Risk... not sufficient to gain and maintain abstinence and control of addictive behavior in chemically dependent adolescents. Long- term traditional psychotherapy is usually necessary to ameliorate the psychological vulnerabilities which may have initiated the excessive dependency on drugs and also to help the adolescent tolerate the task of mastering those developmental achieve- ments which were disrupted by chronic in- toxication and the preoccupation with the drug life. In addition, special remediation and ego building approaches may be needed to help the adolescent master the environment more effectively and internalize a sense of com- petence. These ego building, cognitive aspects of treatment may include special education as well as social skills training as in- dicated for specific adolescents. Practical, sensitive assistance in gaining school and/or vocational successes as well as help in developing and maintaining a network of friends is also important. The youngster’s environment should be evaluated also. If a neighborhood or school is particularly competitive, fragmented, cold, or unfriendly, the therapist may need to be- come involved in efforts to create community atmospheres more supportive of the fragile adolescent. Assistance in locating ap- propriate Alcoholics or Narcotics Anonymous support groups may be very im- portant in maintaining abstinence and in providing needed emotional support derived from group membership and group recogni- tion of achievement. Family therapy is always a necessary part of the treatment process. Some families are dif- ficult to treat. Often these youngsters have been a longstanding source of stress and un- happiness to the parents. The disturbed adolescent may even have been a significant element in the marital discord or separation commonly seen in these families. At the same time, many of the parents are chemical- ly dependent themselves or suffer from ego weaknesses similar to those observed in the adolescents. The result of all these negative factors is that these parents often present in a very unappealing manner. They may be overtly or subtly rejecting the adolescent patient, contemptuous of therapy (to some extent because they have had many treat- ment failures), and uncooperative. At times, the parents appear very child-like and self- centered and do not hold the attitudes toward child-rearing which mental health workers would find acceptable. In spite of all these comments, many parents can be involved in therapy if they are ap- proached with understanding and with an ac- curate perception of the special needs of their disturbed adolescent children. It is usually necessary to be very firm in involving the parents in the treatment process. This may go as far as insisting that they assume their parental responsibilities, if necessary, by involving protective services and other legal approaches to force them to behave responsibly. Firm limits on acting out be- havior are often necessary. Interestingly, al- though this leadership is often initially greeted with anger, the anger does not seem to interfere with long-term efforts to involve the families in treatment. At some level, these adults realize that they have a parental responsibility and, like an acting out adoles- cent, to some extent they are inviting us to take a firm stand to demand their best pos- sible performance on behalf of their children. A complete discussion of family therapy ap- proaches is obviously not the purpose of this paper. It should be mentioned briefly, however, that multi-family therapy provides a model in which families who have achieved progress in developing or regaining a suppor- tive parental role in their child’s life can help new families. The "experienced" families can calm and support the shakier new parents and guide them through the treatment process with greater tact and with less narcis- sistic injury than professionals alone can offer. Didactic techniques, which include substan- tial education regarding the addictive process, its depressive component, and the effectiveness of therapy, also can be very im- 3-237 Report of the Secretary’s Task Force on Youth Suicide portant in orienting the family to the nature of the treatment process and encouraging them regarding the possibilities of a positive outcome. The value of psychopharmacology in the treatment of these adolescents is difficult to evaluate generally. Some adolescents, par- ticularly those with a clear family history of major depressive illness, do seem to benefit by antidepressants. These medications are a reasonable part of the treatment plan, par- ticularly in cases where treatment is begun in a hospital setting. It is crucial, however, not to expect too much of the antidepressant drugs, particularly in youngsters with his- tories that include clear cut ego defects and external traumas such as most of those we have described in this paper. It is important not to forget that the most common drug used in suicide attempts and successful suicides is the antidepressant. REFERENCES --ooooeverriininenieenns . 1. Black DS, Warrick G and Sinokur G: Excess mor- tality among psychiatric patients. The lowa record-linkage study. JAMA 253:5961, 1985. 2. Cummings CP, Prokup CK, and Cosgrove R: Dys- horia: The cause or the result of addiction? The sychiatric Hospital 16:131-134, 1985. 3. Famularo R, Stone K, and Popper C: Preadoles- cent alcohol abuse and dependence. Am. J. Psychiatry 142:1187-1189, 1985. 4. Frances RJ: Quoted in: Biochemical abnor- malities can be linked to suicide. Am. Med. News, January 17, 1986. 5. Gawin FH, and Kleber HD: Abstinence Symprmatsiogy and psychiatric diagnosis in cocaine abusers. Arch. Gen. Psychiatry 43:107-113, 1986. 6. Martin RL, Cloninger CR, Guze SB, and Clayton PJ: Mortality in a follow-up of 500 psychiatric outpatients: II. Cause-specific mortality. Arch. Gen. Psychiatry 42:58-66, 1985. 7. Murphy GE: On suicide prediction and prevention. Arch. Gen. Psychiatry 40:343-344, 1983. 8. Pokony AD: Prediction of suicide in psychiatric patients. Arch. Gen. Psychiatry 40:249-257, 1983. 9. Robins DR, and Alessi NE: Depressive symptoms and suicidal behavior in adolescents. Am. J. Psychiatry 142:589-592, 1985. 10. Schuckit M: Genetic and clinical implications of al- coholism and affective disorders. Am. J. Psychiatry 143:140-147, 1986. 3-238 OVERVIEW OF EARLY DETECTION AND TREATMENT STRATEGIES FOR SUICIDAL BEHAVIOR IN YOUNG PEOPLE Susan J. Blumenthal, M.D., M.P.A., Chief, Behavioral Medicine Program, Health and Behavior Branch, National Institute of Mental Health, Rockville, Maryland David J. Kupfer, M.D., Professor and Chairman, Department of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania INTRODUCTION AND OVERVIEW During the past five years, it has become ap- parent that suicidal behavior among young people represents an important public health problem requiring the development and im- plementation of detection and information strategies at the national policy level. Suicide is now the second leading cause of death in young people. For example, between 1970 and 1980, 49,496 of the nation’s youth, 15 to 24 years of age committed suicide. Within this one decade, the suicide rate for this age group increased 40 percent (from 8.8 deaths per 100,000 population in 1970 to 12.3 per 100,000 in 1980), while the rate for the remainder of the population remained stable. Young adults 20 to 24 years of age had approximately twice the number and rate of suicides as teenagers 15 to 19 years old. This increase in youth suicide is due primari- ly to an increasing rate of suicide among young men. Rates for males increased by 50 percent (from 13.5 to 20.2 per 100,000) com- pared to a 2 percent increase in females (from 4.2 to 4.3 per 100,000) between 1970 and 1980 so that by 1980 the ratio of suicides committed by males to those committed by females in this age group was almost five to one. A first step in our review of this area is to define early detection. Itis important to real- ize that completed suicide is a low-base-rate phenomenon in that, fortunately, it does not occur that frequently (1). However, this rare event status does not compromise our major objective, which is to increase our ability to determine who is at greatest risk for com- pleted suicide. We would argue that target- ing one’s efforts to prevent successful suicide--that is, to detect all behavior that leads to a final common pathway, suicide completion--is best done if we understand the various domains through which suicidal behavior emerges. These domains are not just risk factors but spheres of vulnerability. Because they can be detected and manipu- lated, they also represent opportunities for intervention. In this review, we will attempt to describe several levels of detection and to integrate our proposed model of risk factors with a multi-threshold level model of detection (Table 1). Unfortunately, up to the present, reports on detection of suicide have often been flawed by poor methodology. Moreover, available detection and interven- 3-239 Report of the Secretary’s Task Force on Youth Suicide tion studies have not been adequately evaluated, nor has sufficient attention been given to the development of a conceptual model of detection and intervention. We believe that there are three major levels of detection. ¢ The first level represents the need to "red flag" high risk groups. It is basically a detection awareness strategy to follow individuals who may possess certain genetic and biological risk factors that in- teract with environmental factors in such away that these individuals may be at risk for developing behavioral problems and/or psychiatric disorders that are present in Levels II and III. Level I also includes environmental stressors that are linked with suicidal behavior. e Level II represents the detection of major behavioral/environmental problems. Here we are dealing with symptomatic children and youth in whom assessment and intervention may be re- quired. Problems such as emotional dif- ficulties, running away from home, and poor self-esteem may be identified at this particular level. e Level III represents the detection of a psychiatric disorder of sufficient severity to require assessment and intervention by mental health professionals. These three levels will be described in greater detail later in this review. Level | - Detection Awareness: Level Il -Detection of Major Problem: Level lll-Detection of Psychiatric Disorder Levels of Detection Red flagging high-risk groups for awareness and educational purposes. May require assessment-intervention (academic problems, self-esteem, being the victim of child abuse). Requires assessment and treatment. Table 1. 3-240 Model of suicidal behavior Our next task is to describe the five domains that comprise our theoretical model of suicidal behavior. We believe that five domains organized as a matrix provide a simple but appropriate model for considering these five risk factors for clinical investiga- tion as well as for education and clinician in- tervention (Figure 1). We believe this overlapping model of risk, shown graphically as a series of interlocking Venn diagrams, represents a compelling alternative to no- tions of final common pathways or parallel schemas (2). A major clinical research strategy using this new model will be to develop weightings for each of its major com- ponents. For example, in applying this model, the breakup of a relationship might be a final humiliating experience that triggers a depressive episode in a young person with a family history of affective disorder. Such an individual may also have poor social supports which interact with the other identified risk factors to increase the individuals vul- nerability for suicide. The question is, at what level and in what degree do each of these factors contribute to suicide potential? Is the degree of overlap of all factors the most significant criterion? Or, we may wish to pose such questions as, what makes 15 per- cent of the people who suffer from an affec- tive disorder end their lives by suicide while the other 85 percent donot? Using this over- lapping model, we may learn that the sub- group of affective disorder patients who commit suicide have a greater overlap of other risk domains such as increased hope- lessness, impulsiveness, fewer social sup- ports, a recent humiliating life experience, and/or stronger family history of affective dis- order or suicidal behavior. Domains of risk matrix What are the five domains that comprise our risk matrix? The first is a careful description according to psychiatric diagnosis. Second, personality traits relating to suicide, such as aggression, impulsiveness, and hopelessness, are important in and of themselves in charac- S.Blumenthal: Overview of Early Detection and Treatment.. terizing suicide since they may represent per- sonality styles that cut across diagnostic groupings. In addition, this domain includes certain personality disorders, such as border- line personality disorder and antisocial per- sonality disorder, which are more highly correlated with suicidal behavior and repre- sent risk factors. The third domain is con- cerned with psychosocial factors, social supports, life events, and chronic medical illness. For example, early loss, greater num- ber of negative life events, the presence of chronic medical illness, and fewer social sup- ports increase the risk for suicide. The fourth area is the identification of genetic and fami- ly factors that predispose an individual to suicide. Previously, investigators have sug- gested that the genetics of suicide may be in- dependent of the genetics in a family history relating to specific psychiatric disorders such as affective disorder or alcoholism. The final factor in the matrix is the neurochemical and biochemical variables which are currently under active investigation in an attempt to identify either a biologic abnormality or a vul- nerability state for suicide. With respect to children and youth, each of these domains is at least mentioned in the available literature. A number of theoretical issues, however, still need to be considered. These center around such questions as, what are the commonalities across psychiatric diagnoses which increase suicidal risk; does the mere presence of such a disorder with the overlapping of the domains create the in- creased risk; or, is it both? Psychiatric diagnosis The diagnostic picture for youth suicide is not clear-cut. Only a few studies on completed suicide have been conducted in this age group. However, in the adult literature, we know from four major studies with sufficient sample size (three retrospective and one prospective (3-6)), that if one sums the find- Biology Psychosocial Life Events, and Chronic Medical Overlap Model of Suicide Behavior (Five Domains) lliness Psychiatric Disorder Family History and Genetics Personality Traits Figure 1. 3-241 Report of the Secretary's Task Force on Youth Suicide ings about the association of suicide with psychiatric disorder from these studies, more than 90 percent of the victims had a psychiatric disorder and less than 10 percent had no mental disorder. The findings in the adolescent literature are quite similar. A recent study by Shafii (7) found that 95 per- cent of the adolescent suicide victims had an associated psychiatric disorder by DSM-III criteria. A high percentage of these young people had an affective disorder--76 percent had major affective disorder or dysthymia as compared to 28 percent in the control group. His work (8) also suggests that 70 percent of youngsters who end their lives by suicide have associated substance abuse, 70 percent have a history of antisocial behaviors, 65 per- cent have "inhibited" personality traits, and 50 percent had made a previous suicide at- tempt. Suicidal behavior of parents, rela- tives, and friends, along with a parental history of emotional problems and abuse, were also significant variables. An earlier study by Shaffer (9) suggests that several per- sonality traits are characteristics of youngsters who end their lives by suicide, in- cluding tendencies to be withdrawn, perfec- tionistic, impulsive, or aloof. Preliminary data from a large ongoing psychological autopsy study of adolescents (10) suggest that at least a third of the young people in the study who ended their lives by suicide had an associated conduct disorder and that one quarter of the sample population were suf- fering from a depressive disorder. In addi- tion, a high percentage of these youth abused alcohol or drugs. Approximately 50 percent of these young people had a family history of suicidal behavior (11). Suicide attempts in this age group have likewise been linked to depressive symptoms. The co-morbidity of antisocial and depressive symptoms appears to be a particularly lethal combination in youth (10,12). In sum, it is likely that the symptom triad of aggressiveness, impulsiveness, and depres- sive symptoms represents a major contribu- tion to risk for suicide across the life cycle. In addition, substance abuse represents a major risk factor for youth suicide especially when 3-242 linked with affective symptoms and impulsive personality traits. Personality factors Even though conduct disorders and border- line personality disorders are highly as- sociated with adolescent suicide (7-10,13,14), assessment of personality fac- tors has been impeded by lack of stand- ardized measures for these characteristics in young people. In addition, assessment of personality at the time of a suicide attempt is confounded by the distress experienced by the individual concerning the event. Finally, there is a continuum of traits and disorders associated with suicidal behavior in youths with such behavior in adults. Therefore, characteristics appear to be stable over the life cycle (15). It is proposed that certain diagnostic categories from the DSM-III In- fancy, Childhood, and Adolescence section correspond to, and in effect eventually develop into, certain personality disorders (e.g., schizoid disorder of childhood and adolescent lead to schizoid personality disor- der; avoidant disorder of childhood and adolescence leads to avoidant personality disorder; conduct disorder leads to antisocial personality disorder; oppositional disorder leads to passive aggressive personality disor- der; and identity disorder leads to borderline personality disorder). The presumption is that the childhood or adolescent condition is diagnosed if the in- dividual is under age eighteen, and the adult personality diagnosis is used after age eighteen whenever the personality psychopathology has persisted at an intensity sufficient to meet disorder criteria (15). In addition, these personality variables may also have biological correlates (i.e., serotonin deficiency related to increased impulsiveness and aggressiveness) which interact with en- vironmental factors. It should also be noted that the co-existence of Axis I depression and Axis I conduct disorder or borderline identity personality disorder may represent an ex- tremely risky combination of factors. Recently, the relationship of personality S.Blumenthal: Overview of Early Detection and Treatment. variables to cognitive styles has been studied in suicidal behavior. In one study (16) cogni- tive characteristics of rigidity, impulsiveness, and field dependence were contrasted in a group of suicide attempters and a group of nonsuicidal psychiatric controls. The suicide attempt group was characterized by greater rigidity in a divergent thinking task; using multivariate analysis while controlling for age and diagnosis, field dependence was also more characteristic of the suicide attempters, but only in the 19-34 age group. Impulsive- ness did not differentiate the two groups. The results were interpreted as supporting a hypothesis of a cognitive predisposition to at- tempting suicide. Inasimilar manner, Beck and colleagues (17) intensively studied 207 patients hospitalized with suicidal ideation, but not recent suicide attempts, at the time of admission. During a followup period of five to ten years, fourteen of these patients committed suicide. Of all the data collected at the time of hospitaliza- tion, only the Hopelessness Scale and the pessimism item of the Beck Depression In- ventory correlated with the eventual suicides. A score of 10 or more on the Hope- lessness Scale correctly identified 91 percent of the eventual suicides. Taken in conjunc- tion with previous studies showing the relationship between hopelessness and suicidal intent, these findings indicate the im- portance of degree of hopelessness as an in- dicator of long-term suicidal risk in hospitalized depressed patients. Beck (per- sonal communication) has now been examin- ing the congruence of cognitively rigid individuals and the level of hopelessness in the development of suicidal ideation and be- havior. Neuringer (18) has suggested that cognitively rigid individuals faced with naturally occurring life stress are unable to generate alternative solutions to their problems; as a result, they are inclined to develop ideas of helplessness and hopeless- ness, which, in turn, heighten the risk of suicidal ideation and/or behavior. In support of this model, Schotte and Cum (19) found that college students under high life stress who performed poorly on an interpersonal problem solving task, the Means-End Problem solving procedure (20), reported greater suicidal ideation as measured by the Scale for Suicide Ideation (21). While other issues of impulsiveness and aggressiveness in children need to be examined systemically, these data represent the current level of knowledge regarding the association of specific personality factors and suicide in young people. Finally, in an extensive review of psychoso- cial and cognitive aspects of adolescent suicide, Petzel and Riddle (22) concluded that adolescent suicide completers are even more isolated, less visible, and more dis- turbed than suicide attempters. Social isola- tion and impulsiveness were reported in a number of studies, and suicidal behavior within the family was associated with in- creased adolescent suicide attempts. They described a host of familial, social, school, and emotional problems, as well as physical illness, as interacting to increase the suicide risk. Petzel and Riddle concluded with a recommendation for clinical research ap- proach using an interrelationship of multiple factors. Psychosocial factors Although the data base is limited, there is considerable convergence of findings in the area of family and environmental factors in relation to youth suicidal behavior (23). Adolescents who make suicide attempts are characterized by considerably increased life stress and have had many losses (particularly early loss) and significant changes within the nuclear family as compared with other psychiatrically disturbed youngsters, depressed adolescents, and the general population. They have also had both physi- cal and psychiatric illnesses. Precipitating events are often humiliating and are almost invariably interpersonal problems between the adolescent and his parents or peers. The social and familial background of these adolescents is marked by parental death, divorce, and separation. The general relationship with parents is often troubled, 3-243 Report of the Secretary’s Task Force on Youth Suicide and discord is a frequent characteristic of the marital relationship. Adolescents who at- tempt suicide have a greater number of nega- tive life events, fewer social supports, and fewer personal resources than adolescents who do not. In addition, increased contact with suicidal behavior in the environment has been noted as putting certain vulnerable youth at greater risk. Family history and genetics A family history of suicide is a significant risk factor for suicide. Explanations for this as- sociation include identification with and im- itation of a family member who has committed suicide, transmission of genetic factors for suicide, and transmission of genetic factors for psychiatric disorders such as affective disorders (24-28). A study of psychiatric inpatients revealed that (a) half of the persons with a family history of suicide had attempted suicide themselves, and (b) more than half of all patients with a family history of suicide had a primary diagnosis of affective disorder (29). Astudy of the Amish, a religious group with a 100-year history of non-violence, no alcohol or drug abuse, a high degree of social cohesion, no divorce or family dissolution, and a philosophy of suicide as the ultimate sin, has demonstrated, quite unexpectedly, that suicides do occur among this group. Twenty-six suicides have been documented among the Amish of southeastern Pennsylvania between 1880 and 1980. Twenty-four of the 26 individuals who committed suicide were diagnosed with a major affective disorder, and the suicides occurred in four primary pedigrees. This re- search suggests possible genetic factors in both the transmission of affective disorders + and suicide (30). Another study of suicides in the general population found that six of 100 suicide completers also had a parent who committed suicide. This rate was eighty- eight times higher than predicted (31). Investigations have suggested a high concor- dance rate for suicide in identical twins (32,33). While ten sets of identical twin pairs who both committed suicide have been 3-244 reported in the literature, there has been no report in which both fraternal twins have committed suicide (34). In another study, a greater incidence of suicide was found in the relatives of the control group (28). In the well-known Copenhagen adoption study, a greater incidence of suicide was found in the biological relatives of adoptees who com- mitted suicide than in their adoptive relatives (as compared to adoptee controls) (24). The fifty-seven adoptees who committed suicide had 269 biological relatives, of whom twelve committed suicide (4.5 percent) and had no adopting relatives who committed suicide. By comparison, only two of the 269 biologi- cal relatives of fifty-seven matched control adoptees (0.7 percent) and none of 150 adopting relatives committed suicide. In another adoption study comparing suicide in persons with known depressive illness and matched controls, these same investigators again found a greater incidence of suicide among the biological relatives of the probands (3.7 percent). Of 407 biological relatives, fifteen (0.5 percent) committed suicide; only one of 187 adopting relatives committed suicide (S. Kety, personal com- munication). These studies suggest that we may be able to separate the contribution of a family history of suicide and a family history of affective dis- order to isolate high-risk groups for both re- search and clinical purposes. Issues of family history and genetic factors are complicated not only by concordance for psychiatric diag- noses in families but also by the environment in terms of identification and imitation of suicidal behavior by family members over long periods of time. Biological factors Recent biochemical investigations of suicidal behavior have shown that suicide victims and violent suicide attempters have alterations in the function of a brain neurotransmitter, serotonin, which has been measured by ex- amining a major metabolite of serotonin, 5- hydroxy- indoleacidic acid (5-HIAA), in the cerebrospinal fluid (CSF). Other studies S.Blumenthal: Overview of Early Detection and Treatment.. have measured serotonin and imipramine binding in the brains of suicide victims. Fur- thermore, reduced central serotonergic ac- tivity is associated with suicidal behavior, not only when there is a diagnosis of unipolar depressive disorder but also in association with a range of other psychiatric disorders. This research has found a common biochemi- cal association with aggression, impulsive- ness, and reduced serotonergic function. Some studies suggest that the findings of decreased serotonin and violent suicide at- tempts may increase the risk of completed suicide ten-fold at one-year followup (35). Arsonists, for example, show a very high in- cidence of violent suicide attempts (36). But even with the promising 5-HIAA data, we must urge caution. While low 5-HIAA levels are associated with violent suicide attempts, low 5-HIAA levels are found in patients with diverse psychiatric illnesses and also in groups of normal controls (37). An in- creased incidence of depressive illness has been found in the relatives of both patients and normals with decreased CSF 5-HIAA (38,39). While the serotonergic data repre- sent the most compelling current evidence for a biological correlate of suicidal behavior, other biological factors (neuroen- docrinological, neuro- physiological) are also being investigated actively. It is expected that information derived from such studies will strengthen the relative weight of biologi- cal factors in our overlap model of suicide. EARLY DETECTION AND TREATMENT Several suggestions have been made about how to prevent suicide attempts in children and adolescents, but none of them have been evaluated. It is not known whether voluntary agencies providing help at times of crisis have had a major preventive effect. It has been suggested that more care in the prescribing of psychotropic drugs for young people may prevent overdoses (40-42). However, this is unlikely to have any impact on very young adolescents because they visit their general practitioners before overdoses less often then their older counterparts (43), and usual- ly attempt suicide with non-prescribed anal- gesics (44). In school children, educational measures, including use of the media, aimed at modifying attitudes to self-poisoning have also been proposed (41). Finally, some data demonstrate that States that have strict gun control laws have lower suicide rates (45). Other public health measures have also been found to be effective. In Great Britain, for example, the rate of suicide decreased when the type of domestic gas was changed from a toxic to a non-toxic form (24). Evidence now available from two school- based programs demonstrates that teachers, counselors and other students were increas- ingly able to deal with suicidal students fol- lowing crisis training for counselors, inservice training for teachers, and curriculum addi- tions for students (46). In one program, stu- dents were described as becoming more willing to ask friends directly about their suicidality and were less likely to view a suicidal statement as "nothing to worry about." Students and teachers both reported increased knowledge about the mental health referral process. While there is no dearth of literature on in- tervention techniques for youth suicidal be- havior, the results of studies to date are compromised by poor methodology, lack of control groups, lack of evaluation and fol- lowup, and by the fact that most of these in- terventions are not based on a conceptual model of detection and intervention. These points are easily demonstrated by examining the published studies on whether volunteer agencies providing help at times of crisis have had a major preventive effect. A second form of "intervention" is represented by the many school-based programs that have provided in-service training for teachers and crisis training for counselors. However, no followup data are available to evaluate their impact. In sum, the available data base points to both lack of proven efficacy of these approaches as well as insufficient methodol- ogy to provide the tools to evaluate them. At this point, we have been unable to 3-245 Report of the Secretary’s Task Force on Youth Suicide demonstrate, beyond the importance of crisis support, that these interventions decrease the actual rate of suicide, although it is ac- knowledged that these efforts may play a sig- nificant role in providing needed support and education about suicide. Risk detection levels It is appropriate now to return to the three levels of risk detection to provide greater detail. We will give a brief overview of each level and then present several examples. Individuals whom we would place in Level I, which can best be labelled detection aware- ness, are not actively suicidal or in immediate danger of suicide (Table 2). However, in- dividuals at this level have certain risk factors of which we ought to remain aware. For ex- ample, the offspring of affectively ill or sub- stance abusing parents, the offspring of a person who has died by suicide, close con- tacts with suicides and suicidal people, and abused and neglected children would com- prise the Level I group. Level I would also include children who have recently been under extreme stress, such as divorce of parents, moves, the presence of a chronic ill- ness either in the children or in the family, or thc cent death of a parent or a close rela- tive. 1. “hould be pointed out that one can think of ti. * individual as having relatively lit- tle control o. =r Level I problems. And, as Level | - Detection Awareness a. Offspring of affectively-ill or substance abusing parents b. Offspring of suicides and suicide attempters c. Close contacts with suicides and suicidal people (prevention of contagion) d. Abused and neglected children e. Children who have recently been under severe stress --Divorce of parents --Move --Death of parent/relative Table 2. 3-246 mentioned earlier, Level I contains in- dividuals with a high threshold of genetic loading for psychiatric disorders or suicidal disorders as well as those individuals who have experienced major environmental stressors. Recently, Salk et al. (47), sug- gested that several "early" risk factors, name- ly respiratory distress for more than one hour at birth, no antenatal care for the mother before twenty weeks of pregnancy, and chronic disease of the mother during preg- nancy differentiated adolescent suicide vic- tims from matched controls. Level II, shown in Table 3, is characterized by major problems that do not meet criteria for a psychiatric disorder. Young individuals who fit into Level II may require assessment, intervention, and perhaps even treatment; but the treatment is not for a DSM-III psychiatric diagnosis. Individuals at this level generally show some amount of distress, presence of symptoms, and/or decrease in function. Examples of symptoms at this level may include pronounced academic problems, the presence of learning disabilities, increas- ing interpersonal relationship difficulties, a major loss, or severe self-esteem problems. These individuals may be exemplified by youth who run away, adolescents who have an unwanted pregnancy, or children who are undergoing major stresses and become symptomatic. Extreme aggressiveness or feelings of hopelessness also characterize Level II in young people. Individuals at Level II may indeed become suicidal and are at risk for suicide attempts and suicide com- pletion. They may be individuals who have demonstrated difficulties relating to Level I and have moved from Level I to Level IL However, we must remember that in- dividuals can also move from Level II to Level III or I to III, or appear de novo at Level IT or IIL Level III represents the detection of suicidal youth who have major psychiatric disorders (Table 4). When any individual is identified at this level, assessment and an intervention component are required with the interven- tion representing active treatment aimed at S.Blumenthal: Overview of Early Detection and Treatment.. a specific psychiatric diagnosis. Such diag- noses in children and youth may include af- fective disorders, conduct disorders, schizophrenia, eating disorders, substance abuse, and adjustment reactions. Even though personality disorders are not diag- nosed before the ages of eighteen, they can be diagnosed in those youth between the ages of eighteen to twenty-four. In younger individuals, such disorders as conduct disor- ders, identity disorders, oppositional be- havior, avoidant disorders, and overanxious Level Il - Major Problem Awareness a. Academic problems Learning disability Runaways ~ 0 ao Qo Having an unwanted pregnancy TQ Requires assessment and intervention/treatment (not related to psychiatric diagnosis) Anything that is not a major psychiatric disorder Symptomatic, but does not meet criteria for a DSM-III, Axis | disorder Interpersonal relationship difficulties Self-esteem and sexual identity problems Children who are undergoing major stress or loss and are symptomatic Aggressivity, hopelessness, personality traits and styles Table 3. Level lll - Major Psychiatric Disorder a. Affective disorder Conduct disorder Schizophrenia Eating disorder Substance abuse ~~ 0 ao oT Alcoholism Adjustment reactions J @ Identity disorder Oppositional disorder Separation anxiety disorder x Avoidant disorder Overanxious disorder Necessitates appropriate assessment and evaluation Has a treatment component aimed at a specific psychiatric diagnosis Table 4. 3-247 Report of the Secretary's Task Force on Youth Suicide disorders would fall into Level III. In addi- tion, as was previously discussed, certain dis- orders of childhood and adolescence may be predictive of adult disorders (i.e., conduct disorders lead to antisocial personality disor- ders). It should be pointed out that medical illness also can be treated at any of the three levels of suicidal potential. The interactions of the overlapping matrix model (2) of risk with the three-level detec- tion intervention schema should be readily grasped. For example, at Level 1, the so- called awareness level, one may have a genetic loading for a psychiatric disorder, particularly an affective disorder, or have a family history of suicide which increases the individual’s risk of suicide. In addition, there may be stressors in that young person’s life that impact directly on the adolescent. At Level II, we include the genetic, biological, and environmental domains, but add to them personality style and traits, which may also play a major contributory role to risk so that a young person begins to have more of the domains or risk factors interacting by the overlap. Therefore, the risk for suicidal be- havior increases as the individual meets Level II criteria. Future studies will deter- mine whether the appropriate risk model is an additive or a multiplicative model. Final- ly, at Level III, which requires a psychiatric diagnosis, all of the domains and levels of the overlap model are apparent and interacting. Thus, genetic and biological loading, per- sonality traits and style, the environment, and psychiatric diagnoses clearly interact at the Level III stage. Obviously, individuals can move among these levels; and it is hoped that individuals at Levels II and III, through ap- propriate intervention, will return to Level I where intervention is not necessary. Persons who are in clinical remission from a psychiatric disorder, however, should remain "red-flagged." Intervention and treatment The next issue to be considered is interven- tion and treatment, which should be reviewed at each of the three levels. Level I, 3-248 detection awareness, requires a strong com- ponent of education. This means education of teachers, parents, and health care profes- sionals in detection awareness. For example, although controversial, we consider the education of young people themselves about their risk and about what they can do to prevent further development an appropriate arena for active discussion. We believe that direct information concerning suicide risk in relation to family history of suicide, the presence of alcoholic parents, or parents with affective disorders should be conveyed by health professionals in an age-appropriate style to young people. We will also have to assess whether a heightened awareness among physicians of suicidal behavior will reduce the scope of the problem. The issue of physician recognition of mental illness is extremely complex, and both patient and physician factors have been studied to explain the lack of recognition. Nevertheless, it should be possible to design an intervention program in which the educa- tional component for physicians focuses on suicide and related symptoms, diagnoses, and behaviors with specific emphasis on adoles- cents and young adults. Awareness education It is critically important to educate health care practitioners outside the mental health mainstream about the various levels of awareness that comprise our model. These individuals include family practitioners, in- ternists, pediatricians, obstetricians, and gynecologists, who, while trained to deal with stressful health issues such as chronic illness or unwanted pregnancy in the young popula- tion, may not be aware of the additional risk imposed by the factors we have identified as part of our matrix. In addition, many young people may present to their doctors with physical complaints that are somatic repre- sentations of their psychic distress. For the same reasons, it is important to ex- tend our educational activities to health and social service personnel who work within the juvenile justice system. Health care profes- S.Blumenthal: Overview of Early Detection and Treatment.. sionals need to pay particular attention to life cycle issues that emerge during puberty and adolescence. We cannot overemphasize the need for education of health professionals. It is apparent from the adult literature that more than 80 percent of the people who com- plete suicide have seen a physician within weeks to months before the attempt and may have accumulated many months of prescribed medication which they can use to end their lives (3,5,6). Thus, physicians must be educated to diagnose psychiatric syndromes and suicidal behavior and to inter- vene and refer when appropriate. Guidelines for treatment, particularly when pharmacologic intervention is involved, need to be learned. The second task in detection awareness is best described as information gathering and documentation by health care professionals about the various factors mentioned earlier under detection awareness. Documentation in patient charts by pediatricians or internists concerning Level I factors is extremely im- portant. In essence, we are arguing for a type of red flagging of these individuals, even as a lifetime red flag. The major objective here is to keep children and youth at Level I through education, recognition of individual risk, and instruction about what to do if more factors develop so that we can prevent them from moving on the Level II or IIL The third task is "environmental detoxifica- tion." It is important to point out to the fami- ly of anyone at Level I the need to "detoxify" the home from fire arms, medication, and other potential means of suicide. These kinds of public health measures (gun control, non-lethal domestic gas, and removing toxic substances from the home) are important in- terventions in reducing suicide. The detoxification should certainly take place at Levels II and III, but we would argue that it is equally appropriate for Level I. Level II, major problem awareness, requires a level of active intervention and treatment to deal with behavioral problems, personality issues, or specific life events. It is important to develop age-appropriate assessment scales to determine degree of hopelessness, aggressiveness, personality, and other relevant characteristics that are associated with increased suicide risk. Interventions should include primarily behavioral interven- tions (for example, cognitive behavior train- ing, psychotherapy, educational interventions for learning disabilities, self-es- teem training, stress management, and group activities). We believe that these types of in- terventions should deal appropriately with various academic problems, self-esteem is- sues, stressful life events, personality issues, and runaway problems seen in adolescents at Level II. This brings us to Level III, which deals not exclusively, but certainly with psychiatric dis- orders. Here, the use of age-appropriate as- sessment and diagnostic scales by health care practitioners is important, as is the issue of referral. When does the health care profes- sional refer? What are the specific cir- cumstances that require immediate intervention by the psychiatric system? Ob- viously, there are specific interventions for diagnosed psychiatric disorders including psychotherapeutic and psychophamacologic treatments. In addition, we feel very strong- ly that the early development of bipolar af- fective illness or schizophrenia in young people requires immediate intervention for both the patients and their families. Strategies for treating specific psychiatric syndromes may play a major role in prevent- ing suicidal behavior. Examples include the use of lithium carbonate early in the course of bipolar affective disorder to prevent fu- ture manic episodes or the use of neurolep- tics in the treatment of schizophrenia. Cognitive or interpersonal psychotherapies aimed at cognitive distortions and problems in relationships that occur in affective ill- nesses can help minimize symptoms and in- teractional styles that may occur with a chronic, untreated illness. It has been demonstrated that suicidal risk is probably high in the early years of bipolar disorder and schizophrenia. Therefore, aggressive treat- ment during the period of initial illness might 3-249 Report of the Secretary’s Task Force on Youth Suicide indeed reduce the risk of completed suicide. Treatment strategies for Level III using the overlap model approach would be to treat the associated psychiatric condition but at the same time to "red flag" the high risk patient, paying particular attention to en- vironmental stresses and psychosocial sup- ports (Table 5). Psychotherapeutic and psychosocial treatment modalities used in conjunction with pharmacotherapy may in- crease the compliance rate of high-risk in- dividuals who are most prone to commit suicide. In addition, psychotherapeutic treatment may improve interpersonal relationships and reduce the cognitive distor- tions that frequently occur with depression and suicidal thinking. Since suicidal patients are difficult to sustain in treatment and fre- quently drop out, the use of clinic facilities, clinic support, or network systems to ensure that such individuals will stay in treatment is an important strategy. In addition, such programs provide a type of social support through which patients, families, and clinicians can form an alliance that provides education, treatment, and family support over long periods of time. Reports from clinic facilities for the treatment of affective disorders in New York, Tennessee, and California confirm this phenomenon (48). They suggest that the rates of suicide in these patient groups are much lower than would be expected and that such system approaches have a "protective role." Multidimensional intervention components This type of treatment strategy illustrates the interaction of the overlap model of suicide risk and the model of detection awareness levels and intervention. Affective disorder clinics such as those described above provide interventions aimed at all five domains of suicide risk: education to patients and their families (family history and genetics); development of network and support sys- tems (psychosocial and environment); iden- tification and treatment of associated psychiatric disorder (psychiatric disorder, 3-250 biological factors), and psychotherapeutic in- terventions (personality factors, psychiatric diagnosis, family problems). Interventions such as these clinics provide, which encapsu- late as many domains as possible, increase the possibility of preventing suicide in high-risk persons. These programs follow persons in remission as well as during illness; therefore, individuals remain "red flagged." Other psychiatric illnesses that require similarly aggressive interventions include al- coholism and drug abuse, particularly among youth, and disorders of childhood and adoles- cence such as conduct disorders and antiso- cial behavior. Interventions that involve as many domains as possible, including family treatment, environmental modification, and treatment of the behavioral disorder, will maximize prevention of suicidal behavior. Again, the need to detoxify the home should be apparent. In each case, one should follow crisis management principles, use psychotherapeutic and/or psychophar- macologic interventions when appropriate, but also include environmental interven- tions. It is important to educate the family and, when necessary, to treat the family. Several final points with respect to interven- tion and treatment issues include the need to develop followup strategies and a schema to ensure that children at risk will not engage in recurrent suicidal behavior. Difficult issues often arise for the clinician in treating such cases. In general, young suicidal patients are Issues in the Treatment of Suicidal Behavior Therapeutic style - rapport, directness Reassurance Therapeutic interventions a. Altering cognitive rigidity b. Modification of hopelessness Medical aspects - pharmacotherapy Education - patient and family Countertransference Table 5. S.Blumenthal: Overview of Early Detection and Treatment.. difficult to manage and may seem at times un- rewarding to treat, and the child and family may be excessively demanding. These fac- tors may make the physician feel increasing- ly helpless and inadequate. It is important for health care professionals to keep these is- sues in mind and not to communicate nega- tive attitudes and messages to the patients or their families. CONCLUSION To summarize, our recommendations can be grouped into three areas: educational, clini- cal, and research. We have placed a great deal of reliance on education interventions, for example, at Level I, "red flagging" high- risk children, detoxifying the home, and developing rapid and economical screening batteries for general practitioners and pediatricians to use as early detection tools. It is important to highlight the need for good record keeping of suicidal behavior, psychosocial stresses, and family history data. Regarding detoxifying the home, even though national gun control efforts have not succeeded at this point, it is appropriate to argue that stricter gun control is a method for decreasing suicide among youth. Studies have shown that States with strict gun control have lower suicide rates (45). While most detection efforts have focused on Level III (which are still inadequate because psychiatric illness is underestimated in youth), relatively little has been done at Level II and almost nothing at Level L Recent studies show increasing rates of af- fective disorder and conduct disorder in young people (49). Even so, very few people really understand the relationship of any of these detection levels to suicidal behavior and where to go for help for the problems identified in each of the levels. With respect to the clinical arena, it is important to edu- cate clinicians about the diagnostic criteria for psychiatric illnesses in young people and the most effective treatments for specific psychiatric syndromes in youth. From our discussion, it should be apparent that future directions for research operate at every one of the three levels that we have described and that all three clearly need considerable atten- tion. However, research should not be confined to one level because there is considerable over- lap across levels. It is hoped that the outline of these strategies can significantly enhance our approaches to the early detection and in- tervention of suicidal behavior in young people and, thereby, prevent this tragic loss of human life in our country. REFERENCES 1. Porkorny AD: Prediction of suicide in psychiatric atients: Report of a prospective study. Arch Gen sychiatry 40:249-57, 1983. 2. Blumenthal SJ, Kupfer DJ: Generalizable treat- ment strategies for suicidal behavior. Ann NY Acad Sci, in press. 3. Barraclough B, Bunch J, Nelson B, Sainsbury P: A hundred causes of suicide: Clinical aspects. Br J Psychiatry 125:355-73, 1974. 4, Dorpat TL, Ripley HS: A study of suicide in the Seattle area. Comp Psychiatry 1:349-59, 1960. 5. Robins E, Murphy GE, Wilkinson RH, Gassner S, Kayes J: Some clinical considerations in the prevention of suicide based on a study of 134 successful suicides. Am J Public Health 49:889-99, 1959. 6. Hagnell O, Rorsman B: Suicide in the Lundby Study: A controlled prospective investigation of stressful life events. Neuropsychob 6:319-32, 1980. 7. Shafii M: Presented at the National Conference on Risk Factors for Youth Suicide, Bethesda, Maryland, May 8-9, 1986. 8. Shafii M, Carrigan S, Whittinghill JR, Derrick A: Psychological autopsy of completed suicide in children and adolescents. Am J Psychiatry 142:1061-64, 1985. 9. Shaffer D: Suicide in childhood and early adoles- cence. J Child Psychiatry 15:275-91, 1974. 10. Shaffer D: Adolescent suicide. Ann NY Acad Sci, in press. 11. Crimley FE: The adolescent suicide attempt: Acar- dinal symptom of a serious psychiatric disorder. Am J Psychother 36:158-65, 1982. 12. Chiles JA, Miller LM, Cox GB: Depression in an adolescent delinquent population. Arch Gen Psychiatry 37:1179-84, 1980. 13. Crumley F: Adolescent suicide attempts and bor- derline personality disorder: Clinical features. Southwest Med J 74:546-49, 1981. 14. Shaffer D, Fisher P: The epidemiology of suicide in children and adolescents. J Am Acad Child Psychiatry 20:545-65, 1981. 15. Frances A, Blumenthal SJ: Personality disorders and characteristics, presented at the National Conference on Risk Factors for Youth Suicide, Bethesda, Maryland, May 8-9, 1986. 16. Patsiokas AT, Clum GA, Luscomb RL: Cognitive characteristics of suicide attempters. J Consult Clin Psychol 47:478-84, 1979. 17. Beck AT, Steer RA, Kovacs M, Garrison B: Hope- lessness and eventual suicide : A 10-year prospective study of patients hospitalized with suicidal ideation. Am J Psychiatry 142:559-63, 1985. 3-251 Report of the Secretary’s Task Force on Youth Suicide 18. Neuringer C: (ed.) Psychological assessment of suicidal risk. Spingfield: Charles C. Thomas, 1974. 19. Schotte DE, Cum GA: Suicide ideation in a college population: A test of a model. J Consult Clin Psychol 50:690-6, 1982. 20. Platt J, Spivack G: (eds.) Manual for the Means End Problem Solving Pr ure (MEPS): Ameasure of in- terpersonai problem solving skill. Philadelphia: Hah- nemann Medical College and Hospital, Department of Mental Health Services, Hahnemann Community MH/MR Center, 1985. 21. Beck AT, Kovacs M, Weissman A: Assessment of suicidal intention: The scale for suicidal ideation. J Con- sult Clin Psychol 47:343-52, 1979. 22. Petzel SV, Riddle M: Adolescent suicide: Psychosocial and cognitive aspects. Adolescent Psychiatry 9:343-98, 1981. 23. Hirschfeld R, Blumenthal S: Personality, life events and other psychosocial factors in adolescent depression and suicide: A review. In: Klerman G, ed. Suicide among adolescents and young adults. American Psychiatric Press, in press, 1986. 24. Blumenthal SJ: An overview of suicide risk factor research, presented at the Annual Meeting of the American Psychiatric Association, Los Angeles, May 1984. 25. Roy A: Family history of suicide. Arch Gen Psychiatry 40:971-74, 1983. 26. Sohujsinger F, Kety SS, Rosenthal D, Wender PH: A family study of suicide. In: Schou M, Stromgren E, eds. Origins, prevention and treatment of affective disorders. New York: Academic Press Inc., 277-87, 1979. 27. Tsuang M: Genetic factors in suicide. Dis Nerv Sys- tem 38:498-501, 1977. 28. Tsuang M: Risk of suicide in the relatives of schizophrenics, manics, depressives and controls. J Clin Psychiatry 44:396-400, 1983 29. Roy A: Risk factors for suicide in psychiatric patients. Arch Gen Psychiatry 39:1089-95, 1982. 30. Egeland JA, Sussex JN: Suicide and family load- ing for affective disorders. JAMA 254:915-18, 1985. 31. Farberow N, Simon M: Suicide in Los Angeles and Vienna: An intercultural study of two cities. Public Health Rep 84:389-403, 1969. 32. Haberlandt W: Aportacion a la genetica del suicidio. Folio Clin Int 17:319-22, 1967. 33. Haberlandt W: Der suizid als genetisches problem (zwillings-and familier analyse). Anthrop Anz 29:65-89, 1965. Mig Zaw K: A suicidal family. Br J Psychiatry 189:68-9, 1981. 35. Asberg ML, Traskman L, Thoren P: 5-HIAA in the cerebrospinal fluid: A biochemical suicide prediction. Arch Gen Psychiatry 33:1193-97, 1976. 36. Linnoila M, Virkkunen M, Scheinin, Nuutila A, Rimon R, Goodwin FK: Low Sershidspinal fluid 5- hydroxyindoleacetic acid concentration differentiates im- pulsive from nonimpulsive violent behavior. Life Sci 33:2609-14, 1983. 37. Brown GL, Goodwin FK, Bunney WE: Human ag- gression and suicide: Their relationship to neurop- sychiatric diagnosis and serotonin metabolism. In: Ho BT, Schooler JC, Usdin E, eds. Serotonin in biological psychiatry. New York: Raven Press, 287-307, 1982. 38. Sedvall G, fy B, Gullberg B, Nyback H, Wiesel FA, Wode-Helgodt B: Relationships in healthy volunteers between concentrations of monoamine metabolites in cerebrospinal fluid and family history of psychiatric mor- bidity. Br J Psychiatry 136:366-74, 1980. 39. van Praag HM, de Haan S: Depression vul- nerability and 5HT prophylaxis. Psychiatry Res 3:75-83, 1980. 3-252 40. White HC: Self-poisoning in adolescence. Br J Psychiatry 124:24-35, 1974. 41. Morgan HG: (ed.) Death wishes? The under- standing and management of deliberate self-harm. Chichester: Wiley, 1979. 42. British Medical Journal. Annotation: children and parasuicide. 283:337-8, 1981. 43. Hawton K, O'Grady J, Osborn M, Cole D: Adoles- cents who take overdoses: Their characteristics, problems and contacts with helping agencies. Br J Psychiatry 140:118-23, 1982. 44. Hawton K, Goldacre M: Hospital admissions for adverse effects of medicinal agents (mainly self-poison- ing) among adolescents in the Oxford region. Br J Psychiat 141:166-70, 1982. 45. Boyd J: Increase in rate of suicide by firearms. N Engl J Med 308:872-4, 1983. 46. Horsfall JS: Final evaluation report. Project #615. Intervention/ prevention seeking solutions to self destruc- tive behavior in children. Barrett T (Project Director), Sopris West, Inc., 1982. 47. Salk L, Lipsitt LP, Sturner WQ, Reilly BRM, Levat RH: Relationship of maternal and perinatal conditions to eventual adolescent suicide. Lancet 1:624-627, 1985. 48. Jamison KR: Bipolar disorders and suicide. Ann NY Acad Sci, in press. 49. Robins LN: Changes in conduct disorder over time. In: Farran DC, McKinney JD, eds. Risk in intellec- tual and psychosocial development. New York: Academic Press, Inc., 227-59, 1986. SPECIFIC TREATMENT MODALITIES FOR ADOLESCENT SUICIDE ATTEMPTERS Paul D. Trautman, M.D., Assistant Professor of Clinical Psychiatry, Columbia University College of Physicians and Surgeons, and New York State Psychiatric Institute, New York, New York INTRODUCTION The treatment of adolescent suicide at- tempters is a particular interest of mine. I have been working with this population for about five years, along with David Shaffer, M.D., to develop interviews which identify the symptoms and problems of suicide at- tempters. More recently, I worked with Mary Jane Rotheram, Ph.D, to test treat- ment strategies targeted as specifically as possible to those problems. I approach the topic of treatment with a bias towards brief psychotherapy and outpatient management. As director of the Child and Adolescent Depression and Suicidal Disorder Clinic at the Presbyterian Hospital in New York City, 1 am also interested in the problems of train- ing staff to work effectively with this difficult group of adolescents. STUDIES OF TREATMENT FOR SUICIDE ATTEMPTERS I can state quite simply that there are no specific treatment modalities for adolescent suicide attempters. That is, there are no treatment studies--psychotherapeutic, be- havioral, or psychopharmacologic --which show that a clearly defined treatment ap- proach is superior to no treatment or to some other treatment. There are many descrip- tions of treatment--individual, family, group, insight-oriented, behavioral, cognitive, and so forth--but no evidence that suicide at- tempters who are so treated might not have done just as well without that treatment. Only a few studies of treatment in adults provide limited support for the idea that con- tact with a helping professional is better than no contact for the prevention of suicidal be- havior. e Greer & Bagley (1971) showed that sub- jects who had two or more visits with a professional were less likely to make another suicide attempt than untreated subjects; however, the treated subjects were self-selected and may simply have been healthier and better motivated to change. e Motto (1976) made regular telephone contact with one-half of 853 people who dropped out of outpatient therapy after hospitalization. At the four-year follow- up, 5 percent of "contact" and 8 percent of "no contact" subjects had committed suicide; this difference approached, but did not achieve, significance at the 0.05 percent level. e Welu (1972) telephoned and visited a random sample of subjects at home. These subjects were more likely to attend outpatient visits and were less likely to make another suicide attempt than non- contacted controls. 3-253 Report of the Secretary's Task Force on Youth Suicide * Studies by Ettlinger (1978), Chowdhury, et al. (1973), and Gibbons, et al. (1978) failed to demonstrated any impact of out- reach or time-limited case work on reat- tempt rates. © Liberman and Eckman (1981) showed that subjects receiving 32 hours of in- patient behavioral therapy did better on a variety of measures of mood than those receiving insight-oriented psycho- therapy, but the groups did not differ in their reattempt rates. The last observation speaks to a problem which has already been raised by others, namely, that suicide attempters have a num- ber of problems such as mood disturbance, drug and alcohol dependency, aggression, etc., but it is difficult to know which problems will improve in therapy. Mood may change in the short run, for example, but suicidal be- havior may not change in the long run. DEVELOPING A TREATMENT PROGRAM The second part of this paper addresses a series of questions which I keep asking myself in trying to develop an effective, practical treatment program: What conditions need treatment? Who wants treatment? 1 2 3. Who gets treated? 4 What general treatment approaches are useful with adolescents? 5. What can we learn from psychotherapy studies of other adult and child popula- tions? 6. Are any medications of use? This paper will try to provide some, not definitive, answers to these questions. What conditions need treatment? A review of the literature reveals relatively few characteristics which distinguish suicide attempters from other adolescent psychiatric 3-254 patients. Ideally, treatment should be designed to change a problem, symptom or constellation of symptoms (diagnosis). A valid diagnosis carries information about etiology, natural history, and sometimes, treatment. A suicide attempt is not a diag- nosis since it is associated with many different causative factors and diagnoses. SPECIFIC FOCI FOR TREATMENT ® Major Depressive Disorder (MDD) ® Aggression, Conduct Disorder ® Associated Physical Illness * Drug and Alcohol Abuse Parental Psychiatric Illness ® Marital Conflict Parent-Child Conflict MDD. Depression powerfully increases the risk of suicide in adults. While only 25-30 percent of suicide attempters can be said to be depressed, depression is a treatable disor- der, at least in adults, and therefore, should not be overlooked. Aggression and conduct disorder. Aggres- sion and suicidal behavior often go hand-in- hand. Shaffer (1974) noted that a majority of young adolescents who committed suicide manifested antisocial behavior before their deaths; this was also found to be true in the on-going New York Study of Adolescent Suicide (Shaffer & Gould, 1985). Fifty per- cent of black suicide attempters, but only 10 percent of whites, had recently been in trouble with the law (Breed, 1970). Dr. Meeks pointed out that hopelessness and guilt are often associated with antisocial ac- ting out. Associated physical illness. Studies have shown higher rates of current medical illness among older adolescents who have at- tempted suicide than among age-matched peers (Garfinkel et al. 1982; Hawton et al. 1982). I include pregnancy among physical conditions associated with suicidal behavior. P.Trautman: Specific Treatment Modalities. . . Appropriate medical management and education can be expected to produce better physical and psychiatric functioning and im- prove self-esteem. Drug & alcohol abuse. This problem was al- ready discussed by Dr. Meeks. Thirty to 40 percent of adolescent suicide attempters have parents with high rates of alcoholism (Cohen-Sandler et al., 1982; Garfinkel et al. 1982). Depressive spectrum disease may af- fect the families of some suicide attempters, so that some members become depressed, some are alcoholics, some are both alcoholics and depressed, and some remain disease free (Van Valkenburg et al. 1977). Parental psychiatric illness. Psychiatric ill- ness is very common and serious among the parents of suicide attempters. For example, maternal depression can cause disturbance in children; depressed mothers show decreased emotional involvement, disaffec- tion and increased hostility towards their children (Weissman et al. 1972). Maternal depression is also a barrier to compliance; this will be discussed later. Suicidal preoc- cupation in a parent and conscious or uncon- scious wishes to be rid of a child may push some children towards suicide as a solution to their parents’ problems (Margolin & Teicher, 1968; Sabbath 1969, 1971). Marital conflict. Adolescent suicide at- tempters come from homes with high rates of marital conflicts and are more likely to have heard recent talk of separation and divorce than psychiatric controls (Stanley & Barter, 1970). Parent-child conflict. This is the most im- portant external factor in adolescents’ suicide attempts; 70 percent of our adoles- cent girls report suicide attempts precipitated by arguments with parents. Their parents often exhibit extremes of ex- pectation and control, alternating between over-protectiveness and indifference, withdrawal, and an inability to respond to adolescent crisis (Trautman & Shaffer, 1984). Who wants treatment? A suicide attempt is a life threatening event and one would think that parents would be eager to avail themselves of professional ser- vices to make sure it did not happen again. Yet this is not so. NON-COMPLIANCE WITH AFTERCARE Analysis of Suicide Attempters Seen in Emergency Rooms (ER) 23% were evaluated and completed 15 sessions of brief psychotherapy or were still in treatment. 10% were referred to other clinics or hospitalized. but: 20% did not keep any outpatient ap- pointments. 19% dropped out during the initial as- sessment period (first or second ap- pointments). 27% completed two diagnostic visits but refused treatment or dropped out during treatment. Source: Trautman & Rotheram, 1986, unpublished In a consecutive series of 77 adolescents treated in the emergency room for self- poisioning or other self-injury, 23 percent failed to keep a followup appointment in the child psychiatry clinic and 19 percent kept only one followup appointment (Trautman & Rotheram, 1986, unpublished). This oc- curred in spite of vigorous efforts by telephone and letter to reschedule missed visits. 3-255 Report of the Secretary’s Task Force on Youth Suicide Our experience is similar to that of others: ATTEMPTER NON-COMPLIANCE WITH AFTERCARE 44% of 50 8-17 year olds (82% female) did not keep an appointment within one week of ER discharge (Taylor & Stansfeld, 1984). 61% of 27 10-17 year olds (50% hospi- talized) did not keep followup recom- mendations (Litt, Cuskey, & Rudd, 1983). 88% of 138 children and adults (34 10- 19 year olds) did not follow outpatient care recommendations. 75% did not follow recommendations for voluntary admission (Bogard, 1970). 55% of 29 adults did not keep out- patient appointments (Paykel, et al., 1974). 73% of 296 adults ("moderate or high risk for suicide") did not keep out- patient visits (Knesper, 1982). Taylor & Stansfeld (1984) examined 50 8-17 year olds (82% female) who had been ad- mitted to a pediatric ward following a deliberate self- poisoning. All were given a followup appointment in the psychiatric out- patient clinic within one week of discharge but 44 percent failed to attend. Only 39 per- cent of 27 10-17 year olds (half of whom were briefly hospitalized) complied with recom- mended followup within one year after their attempt (Litt, Cuskey & Rudd, 1983). In a study of 138 child and adult attempters, of whom 34 were 10-19 years old, only 12 per- cent of those recommended for outpatient care attended, and only 25 percent of those recommended for voluntary admission to an inpatient or day treatment facility actually turned up (Bogard, 1970). Compliance for the adolescent subjects was not reported separately. Paykel et al. (1974) reported 45 percent compliance among 29 adult patients given an outpatient clinical referral following 3-256 an attempt. In a study of 296 adult patients who were judged to be at moderate to high risk for suicide at an emergency room visit, only 27 percent kept an outpatient appoint- ment (Knesper, 1982). Compliance with pediatric referrals from emergency to other clinics or from the pediatrician to child psychiatrist is generally in the range of 50-70 percent (Litt & Cusky, 1980; Hildebrandt & Davis, 1975; Lefebre et al., 1983; Bergman, Corbin & Haber, 1982; Bacon, 1985). What contributes to compliance after emer- gency treatment for a suicide attempt? In the Taylor & Stansfeld (1984) study, attenders had more psychiatric symptoms, particularly depressive symptoms (depression, in- sominia, and loss of appetite), were judged to have greater intent to die (as opposed to goals such as escape, help-seeking, or manipulation), and were more likely to have received a psychiatric diagnosis than non-at- tenders. This finding is similar to findings among adult patients that low levels of anxiety and/or depression and high levels of paranoid and sociopathic symptoms con- tribute to drop-out from psychotherapy (Baekeland & Lundwall, 1975). Who gets treated? This paper has already shown that 40 percent of attempters do not get evaluated and only 20 percent of the subjects (at best) complete a three-month brief therapy program. Studies of drop-out rates from adult psychiatric clinics show that the median num- ber of sessions attended is about six, and that 30 to 65 percent of patients drop out unilaterally, i.e., before their therapists think they should. In a study of 102 adolescent out- patients, Viale-Val et al. (1984) found: ® 23% did not turn up for the first visit. ® 25% dropped out after one, two, or three sessions (assessment). ® 26% dropped out unilaterally after four Or more sessions. * 10% were referred away. P.Trautman: Specific Treatment Modalities. . . e Only 14.7% stayed in treatment for a median of eight sessions. Suicide attempters, children referred for school problems and externalizing disorders, and minority and low-income patients were less likely to stay in treatment. One can conclude two things: 1.Adolescents are not more likely to drop out of therapy than adults, as is often stated. 2.People want brief treatment and fast results: either they get it and leave or they do not get it and leave. Therefore, brief, crisis-oriented treatment makes sense for most patients. Many suicide attempters have problems of a brief nature; studies by Henderson et al. (1977) and Facy et al. (1979) illustrate this point. SUICIDE ATTEMPT SUBTYPES A. Formal Psychiatric Illness Multiple Adverse Social Factors Methods of High Endangerment B. No Formal Diagnosis Acute Familial or Interpersonal Crisis Methods of Low Endangerment Suicide attempters in Group A are older adolescents and a greater proportion are boys. Suicide attempters in Group B are younger, mostly girls, and have problems which are often quickly resolved, with or without treatment. What general treatment principles are useful? We know from studies by Ricks (1974) and Kolvin et al. (1981) (see reviews by Shaffer, 1984 and Dulcan, 1984), that an effective therapist for children is active, assertive, ex- planatory, and responsive, not passive. S/he uses community resources and meets with parents. S/he uses longer treatment for more severely disturbed patients and shorter treat- ment for less disturbed patients. The therapist’s age, sex, race, religion, ex- perience, and theoretical orientation make little difference (Parloff, et al. 1979), al- though Viale-Val et al. (1984) found that sex- matching of patient and therapist was associated with better treatment compliance. Effective treatment for depression is charac- terized by: a. High treatment structure. b. A clear, well-planned rationale. c. An emphasis on skills training. a The independent use of skills outside the treatment context (i.e., homework). e. An emphasis on self-attribution for in- creased skillfulness (i.e., not only be- havioral change but also the ability to say, "I did this myself and did a good job of it"). What can we learn from other studies? Cognitive behavior therapies hold promise for the treatment of depression and suicidal behavior in adolescents. Beck et al. (1979) described the cognitive triad of depression: a negative attitude about oneself, the world. and the future. They argue that dysfunctional beliefs (e.g., "my friends don’t really care about me") cause sad moods and lead the subjects into maladaptive behaviors (e.g., avoiding others). Several studies of adults have shown that cognitive therapy is as effective as tricyclic antidepres- sant medication for the continuing treatment of depression, with better treatment com- pliance. Beck’s cognitive therapy seems well-suited for adolescents: It is systematic, highly struc- tured, and didactic. The patient and therapist work together to identify and solve problems, and the patient is instructed to carry out homework assignments, to gather information about himself, monitor mood and behavior, and try out new behaviors. Meichenbaum (1977) focuses on the think- ing processes involved in performing a task. He believes that a patient’s behavior is in- fluenced not primarily by environmental 3-257 Report of the Secretary's Task Force on Youth Suicide events, but what the patient says to himself about these events. He trains children and adults to use coping self-instructional thought to deal with problem situations. Spivack & Shure (1974) noted that children with behavior problems were poor interper- sonal problem-solvers. They train children in two types of social reasoning--first, to think of alternative solutions to conflict situations, and second, to predict the likely consequen- ces of the various solutions. Lewinsohn (1974) proposes a behavioral theory of depression which has three major assumptions: 1) a lack of pleasant events (reinforcement) stimulates depressive "be- haviors" such as dysphoria and fatigue; 2) the lack of reinforcement is a sufficient explana- tion for symptoms of depression, and 3) the amount of reinforcement is a function of the number of potentially reinforcing events for the individual, the number of potential rein- forcers the environment can provide, and the skill of the individual in eliciting these rein- forcements. Treatment involves the use of activity schedules, identification of reinfor- cers, and training in social and assertiveness skills and desensitization. Family therapy interventions move the focus of attention from the attempter to pathologi- cal family interactions which promote suicidal behavior or which the suicidal be- havior is meant to solve. Disturbances in family structure including role conflicts, blur- ring of role boundaries (e.g., the child who is given a parental role, or the mother who un- dermines her parental authority by saying, "Johnny, stop yelling! Okay?"), dysfunctional alliances across boundaries (e.g., a child who joins one parent in discrediting the other), failures of communication, secretiveness, and rigidity with inability to accept change or tolerate crisis, may promote suicidal acting out (Minuchin, 1974; Richman, 1979, 1981; Fishman & Rosman, 1981). These systems-theory approaches are essen- tially descriptive rather than etiologic, but are useful in that they readily lead to defin- ing the tasks of treatment, for example, get- 3-258 ting the parents to unite on rules for the child’s behavior, or removing parental responsibilities from the adolescent. The clinician must take care not to let family sys- tems issues blind him to the immediate psychopathology of the adolescent. Depres- sive delusions, for example, are reason to admit the patient to a hospital, no matter how solid the family relationship. On the other hand, removal from the home might also be indicated in the absence of major psychopathology when open marital conflict has pushed the adolescent to suicidal acting out. Parent-child conflict is the most common im- mediate precipitant of suicidal behavior and family sessions are an essential, if not the only, component of successful management. (A useful discussion of combined individual and family treatment is provided by Stein- hauer, 1985). The goals of the initial family sessions are to decrease destructive family in- teractions, increase communication among family members (including discussion of the adolescent’s suicidal ideation and parents’ destructive wishes), and to help the family identify solutions to the current crisis (Rich- man, 1979; Perlmutter & Jones, 1985). These solutions may be readily apparent once the family is able to state explicitly the nature of their crisis, and are already within their problem-solving repertoire. Specific family training programs in com- munication skills and problem solving have been shown to be effective (Robin, 1979; Guerney, Coufal & Vogelsong, 1981) but whether this kind of explicit skill training is necessary is unclear. In a study comparing problem-solving communication training (PSCT) with an alternate family therapy ("family systems", "psychodynamic", or "eclectic"), only PSCT families objectively used problem-solving techniques, but both groups reported significant subjective decreases in family conflicts and disputes, as compared to controls (Robin, 1981). In a study comparing communication skills train- ing to unstructured group therapy for mother-daughter pairs, the skills training was P.Trautman: Specific Treatment Modalities. . . superior in enhancing expressive and em- pathic skills as well as the general quality of the relationship (Guerney et al. 1981). A word of caution should be added about compliance with family therapy approaches. Drop-out rates from behavioral family ap- proaches are high among lower socio- economic and high-risk families (Wahler et al. 1977, cited in Werry, 1979). Shapiro & Budman (1973) reported significantly higher drop-out rates from family therapy than from individual therapy and emphasized that the father’s enthusiasm for treatment is very im- portant to the continuance of that treatment. Group therapy. Very little is written about group therapy for adolescent suicide at- tempters. It is easy to understand why: groups for adolescents are not easy to start or maintain under the best of circumstances unless you have a captive population in a hospital. One needs 10 referrals to be suc- cessful in starting a group of five. Only a large medical center would have enough patients at any one time to start up a group exclusively for suicide attempters; and the group would have to be continuous, not time- limited. An indication of the difficulty of running an outpatient group for adult attempters is provided by Comstock & McDermott (1975) who conducted open groups (that is, the patients were free to attend the group as long as they wished). One hundred five patients were so treated, and the median number of sessions attended was six. Only 20 percent attended the group for three months or longer. The number of patients who were of- fered but refused group treatment is not stated. This supports the earlier contention that, given a choice, the majority of at- tempters want very brief, supportive treat- ment. Costock & McDermott (1975) provide some useful goals for short-term groups: a. Identify situational differences that lead to suicidal preoccupation; b. Point out that action without reflection accounts for many suicides; label inci- dents of impulsive acting out as such; teach group members how to alter their tendency to act impulsively, especially under stress. c. Emphasize that alternative behavior is possible for individuals contemplating suicide; d. Foster psyological mindedness, par- ticularly taking responsibility for one’s behavior, self-observation, questioning motivation, identifying mood correctly, and examining differences between what the patient said and what she/he wanted to convey. Other non-spcific beneficial effects of group therapy include, learning that others in the world share one’s problems, support by peers, role-modeling, ventilation, and ac- quisition of social skills such as conversation- al skills and the use of eye contact (Yalom, 1970). A skillful leader must ensure that role-modeling does not work negatively, that is with hopelessness, suicidal ideation, and suicidal behavior spreading contagiously to all group members. With its greater oppor- tunities for frustration and provocation, a group experience may be more likely than in- dividual treatment to bring out aggression in a suicidal person (Mullan & Rosenbaum, 1975) which may be meted out on other group members (DeRosis, 1975). Glaser (1978) and Ross and Motto (1984) described group therapy techniques and ex- periences with hospitalized and non-hospi- talized adolescent suicide attempters. The latter used group therapy with suicidal adolescents. After a two-year followup of 17 subjects, they found no reattempts or com- pleted suicides. Glaser suggested that group therapy may be a useful alternative to family therapy for the adolescent who is in florid rebellion against his parents. What drugs are useful? Many reports demonstrate the superiority of tricyclic anti-depressants (TCAs) and ECT over placebo for the treatment of major depressive disorders in adults. Endogenous 3-259 Report of the Secretary's Task Force on Youth Suicide or melancholic symptoms (e.g., early morn- ing awakening and weight loss) respond well to TCAs in the short- and long-term (Ander- son, 1982). Depressive delusions are generally resistant to TCAs but respond to ECT. There were no well-controlled drug studies of children or adolescents before 1977; three or four have appeared recently, only one of which is a study of adolescents. In a double blind study of pre-pubertal children with MDD, Puig-Antich et al. (1979, 1985a) found no differences between impramine (IMI) and placebo; the response rate was high, about 60 percent in both groups. This is a much higher placebo response rate than that found in adult studies, which is typically about 30 percent. A small study by Kashani et al. (1984) shows a trend (p .09) for the su- periority of IMI in a pre-pubertal sample of nine subjects. Puig-Antich et al. did find that high plasma levels of IMI were associated with significantly greater improvement as compared to placebo or low plasma IMI. This finding was also reported by Geller et al. (1985) in a study using nortriptyline. In an open study using IMI (5 mg/kg) in 34 adolescents, Ryan et al. (in press) found that 44 percent improved; there was no relation- ship between plasma level of IMI and im- provement. In a study comparing amitriptyline and placebo, Kramer and Feiguine (1981) found nossignificant drug su- perority. Drugs for mania and bipolar disorders. Mania is extremely rare in children but in- creasingly common in adolescents. About one-fifth of adult bipolar patients report that their symptoms began before age 19, and 10 percent report onset before age 12! (Perris, 1966; Winokur et al. 1969; Carlson et al. 1977; Loranger and Levine, 1978). Many adolescents with bipolar disorder are mis- diagnosed as schizophrenic; a patient of mine, who was doing headstands on his hospi- tal bed, was diagnosed as "borderline." Mania and depression appear to be equally common first manifestations of bipolar disor- der, but after onset, manic episodes outnum- 3-260 ber depressive episodes by about 3:1 (Carlson and Strober, 1978). About one- fifth of adolescents admitted for a depressive episode eventually develop bipolar disorder. Early onset has a worse prognosis both for frequency of episodes and suicide. Family risk for depressive disorder is greater for bipolar than unipolar (depression only) patients. Lithium is effective for controlling the symptoms of mania in adolescents and for preventing recurrence, as is true in adults (Delong, 1978; Youngerman & Canino, 1978). Possible complications of lithium treatment include hypothyroidism, proteinuria (renal damage), adverse effects on learning, concentration and memory (Judd et al. 1977) and inhibition of bone growth. Carbamazipine, a drug related both to the TCAs and promazine (a neuroleptic) is effec- tive in adults and in lithium-resistant subjects (Nolen, 1983). In summary, this paper has tried to make the following points about treatment: 1. We are a long way from developing a specific treatment strategy for adoles- cent suicide attempters, and on one treatment that will be effective for this diverse group. 2. There is great resistance to treatment on the part of adolescents and their families. 3. People want brief treatment and quick results, especially low-income and minority patients (Acosta, Yamamoto & Evans, 1982). 4. Good therapy with adolescents is active, teaches skills, uses outside resources, engages the patient in problem-solving, and involves the family. 5. Cognitive-behavioral approaches meet the needs for brevity and activity. 6. We need innovative approaches to: ® Educate families about the therapy process; P.Trautman: Specific Treatment Modalities. . . e Structure the therapies which focus quickly on specific problems, e Promote home visits to understand problems families face and to reach out to those who will not or cannot come to the office, and ¢ Develop strategies which will appeal to minority and low-income families. REFERENCES 1. Acosta FX, Yamamoto J, Evans, LA (1982): Effec- tive Psychotherapy for Low Income and Minority Patients, New York, Plenum Press. 2. Andreason NC (1982): Concepts, diagnosis and classification. In Paykel E. (ed): Handbook of Affective Disorders. New York, Guilford Press, pp 24-44. 3. Bacon K (1985): Mothers compliance with pediatric mental health referrals. npublished manuscript, Adelphi University. 4. Barter J, Swaback D, Todd D (1968): Adolescent suicide attempts: A followup study of hospitalized patients. Arch Gen Psych., 19:523-527. 5. Baekeland F, Lundwall L (1975): Dropping out of treatment: A critical review. Psychological Bull 82:738- 783. 6. Beck AT, Rush AJ, Shaw BF, et al. (1979): Cogni- tive Therapy of Depression. New York, Guilford Press. 7. Bergman D, Corbin S, Haber J. (1982): Analysis of a program for mental health referrals from a pediatric clinic. J Dev Beh Pediatrics , 3:232-235. 8. Bogard HM (1970): Followup study of suicidal atients seen in emergency room consultation. Amer. J. sychiat., 126: 141-144. 9. Breed W (1970): The negro and fatalistic suicide. Pacific Soc Rev. 13:156-162. 10. Carlson GA, Strober M. fora) Manic-depressive iliness in early adolescence. J Acad Child Psychiatry 17:138-153. 11. Carlson GA, Davenport YB, Jamison K (1977): A comparison of outcome in adolescent and late-onset bipolar manic-depressive illness. Am J Psychiatry 134:919-922. 12. Choquet M, Facy F, Davidson F (1980): Suicide and attempted suicide among adolescents in France. In: Farmer R, Hirsch S (Eds): The Suicide Syndrome. Lon- don, Croom Helm. 13. Chowdhury N, Hicks RC, Kreitman N (1973): Evaluation of an aftercare service for parasuicide (at- tempted suicide) patients. Social Psychiatry, 8:67-81. 14. Clum G, Pastsiokas T, Luscomb RL (1979): Em- pirically based comprehensive treatment program for parasuicide. J Consult Clin Psychology. 47:937-945. 15. Cohen-Sandler R, Berman R, King R (1982): Life stresses and symptomatology: Determinants of suicidal behavior in children. J Am Acad of Child Psychiatry., 21:398-403. 16. Comstock BS, McDermott, M (1975): Group therapy for patients who attempt suicide. Int J Group Psychotherapy., 25:44-49. 17. ryn L, McKnew DH, Zahn-Waxler C, Gershon ES (1986): velopmental issues in risk research: The off- spring of affectively ill parents. In: Rutter M, Izard CE, Read PB (Eds): Depression in Young People, Develop- mental and Clinical Perspectives. New York: The Guilford Press. 18. Delong GR (1978): Lithium carbonate treatment of select behavior disorders in children suggesting manic- depressive illness. J Pediat., 98:689-694. 19. DeRosis L (1975): Karen Horney's theory applied to psychoanalysis in groups. In: Rosenbaum M, Berger MM (eds): Group Psychotherapy and Group Process. New York, Basic Books. 20. Dulcan MK (1984): Brief psychotherapy with children and their families: The state of the art. J Am Acad Child Psychiatry. 23:544-551. 21. Durkheim E. (1951): Suicide: A Study in Sociol- ogy. Glencoe, ll, The Free Press. 22. Elkins R, Rapoport JL (1983): Psychopharmacol- ogy of adult and childhood depression: An overview. In: Cantwell DP, Carlson GA (Eds): Affective Disorders in Childhood and Adolescence, An Update. New York, SP Medical & Scientific Books. 23. Ettinger R. (1975): Evaluation of suicide preven- tion after attempted suicide. Acta Psychiatr Scand. 135 (suppl 260):1-135. 24. Facy F, Choquet M, Lechevallier Y (1979): Research d'une typologie des adolescents suicidant. So- cial Psychiatry. 14:75-84. 25. Fishman HC, Rosman BL (1981): Atherapeutic ap- roach to self-destructive behavior in adolescence: The amily as the patient. In: Stuart IR, Wells CF (Eds): Self- Destructive Behavior in Children and Adolescents. New York, Van Nostrand Reinhold Co. 26. Garfinkel BD, Froese A, Hood J (1982): Suicide at- tempts in children and adolescents. Am J Psychiatry 139:1257-1261. 27. Geller B, Perel JM, Knitter EF et al (1983): Nortrip- tyline in major depressive disorder in children: Response, steady state plasma levels, predicitive kinetics and phar- macokinetics. Psychopharmacol Bull 19:62-65. 28. Gibbons JS, Butler J, Urwin P etal (1978): Evalua- tion of a social work service for self-poisoning patients. Br J Psychiatry. 133:111-118. 29. Gibbons JS (1980): Management of self-poison- ing: Social work intervention. In: Farmer R, Hirsch S. (Eds): The Suicide Syndrome. London, Croom Helm. 30. Glaser K (1978): The treatment of depressed and suicidal adolescents. Am J Psychother. 32:252-269. 31. Goldacre M, Hawton K (1985): Repetition of self- poisoning and subsequent death in adolescents who take overdoses. Brit J Psychiatry, 146, 395-398. 32. Greer S, Bagley C (1971): Effect of psychiatric in- tervention in attempted suicide: A controlled study. Brit J Med, 1:310-312. 33. Guerney B, Coufal J, Vogelsong E (1981): Relationship enhancement versus a traditional approach to therapeutic/preventative/enrichment parent-adolescent programs. J Consult Clin Psychol, 49:927-939. 34. Hawton K, O'Grady J, Osbourne M, Cole D (1982): Adolescents who take overdoses: their characteristics, problems and contacts with helping agencies. Brit J. sychiatry, 140:118-123. 35. Henderson AS, Hartigan J, Davidson J, et al. (1977): A typology of parasuicide. Brit J Psychiatry, 131:631-641. 36. Hildebrand T, Davis M (1975): Home visits: A method of reducing the pre-intake dropout rate. J Psychiatr Nursing Mental Health Services, 13:43-44. 37. Judd LI, Hubbard B, Janowsky DS, et al. (1977): The effect of lithium on the cognitive functions of normal subjects. Arch Gen Psych, 34:352-357. 38 Kashani J, Shekion WO et al (1984): Anitriptyliane in children with major depressive disorder: A double blind crossover pilot study. J Am Acad Child Psychiatry, 23:348- 351. 3-261 Report of the Secretary’s Task Force on Youth Suicide 39. Kiev A (1975): Psychotherapeutic strategies in the management of depressed and suicidal patients. Am J Psychother 29:345-354. 40. Knesper D (1982): A study of referral failures for potentially suicidal patients: A method of medical care Sslustion. Hospital and Community Psychiatry, 33:49- 52. 41. Kolvin E, Garside RF, Nicol AR, et al f1987} Help Starts Here: The Maladjusted Child in the Ordinary School. London Tavistock. 42. Kramer AD, Feiguine RF (1981): Clinical effects of amitriptyline in adolescent depression. J Am Acad Child Psychiatry, 20:634-644. 43. Lefebre A, Sommeraver J, Cohen N et al (1983): Where did all the "no-shows" go? Canadian J Psychiatry, 28:387-390. 44. Lewinsohn PM (1974): Clinical and theoretical aspects of yeplession. In: Calhoun KS, Adams HE, Mitchell KM (Eds): Innovative Treatment Methods in Psychopathology. New York, John Wiley & Sons. 45. Liberman R, Eckman T (1981): Behavior therapy vs. insight-oriented therapy for repeated suicide at- tempters. Arch Gen Psychiatry 38:1126-1130. 46. Litt], Cuskey W. (1980): Compliance with medical regimes during adolescence. Ped Clin NA 27:3-15. 47. Litt, Buskey W. Rudd S. (1983): Emergency room evaluation of the adolescent who attempts suicide: Com- pliance with follow-up. J Adolescent Health Care: 4:106- 108. 48. Loranger AW, Levine PM (1978): Age at onset of bipolar affective illness. Arch Gen Psychiatry 35:1345- 1348. 49. Margolin NL, Teicher JD (1968): Thirteen adoles- cent male suicide attempts—-Dynamic considerations. J Am Acad Child Psychiatry 7:296-315. 50. Meichenbaum DH (1977): Cognitive-Behavior Modification, An Integrative Approach. New York: Plenum Press. 51. Minuchin S (1974): Families and Family Therapy. Cambridge, Harvard University Press. 52. Modestin J (1985): Antidepressive therapy in depressed clinical suicides. Acta Psychiatr Scand. 71:111-116. 53. Motto J (1976): Suicide prevention for high-risk persons who refuse treatment. Suicide and Life Threaten- ing Behavior. 6:223-230. 54. Mullan H, Rosenbaum M (1975): The suitability for the group experience, In: Rosenbaum M, Berger MM (Eds): Group Peyenictherapy and Group Function. New York, Basic ks. 55. Olig RM, Staton RD, Beatty WW, et al. (1985): An- tidepressant treatment of children: Clinical relapse is un- related to tricyclic plasma concentrations. Perceptual and Motor Skills. 60:879-899. 56. Nolan WA (1983): Carbamazepine, a possible ad- junct or alternative to lithium in bipolar disorder. Acta Psychiat Scand. 67:218-225. 57. Parloff MB, Waskow IE, Wolfe BE (1979): Research on therapist variables in relation to alo ress and outcome, In: Barfield SL, Bergin AE Dray Handbook of Psychotherapy and Behavior Change: An Empirical Analysis. New York, Wiley. 58. Paykel E, Hallowell C, Dressler D, et al. (1974): Treatment of suicide attempters - A descriptive study. Arch Gen Psychiatr, 31:487-491. 59. Perlmutter RA, Jones JE (1985): Problem solving with families in psychiatric emergencies. Psychiatr O. 57:23-32. 60. Perris C (1966): A study of bipolar (manic-depres- sive) and unipolar recurrent depressive psychoses. Acta Psychiatr Scand. 42 (Suppl 194):9-189. 3-262 61. Petti TA, Law W 41982) Imipramine treatment of depressed children: A double-blind pilot study. J Clin Psychopharmacol 2:107-110. 62. Preskorn S, Weller E, Weller RA, et al. (1982): Depression in children: Pelstienship between plasma im- ipramine levels and response. J Clin Psychiatr, 43:450- 453. 63. Puig-Antich J, Perel JM, Lupatkin W et al (1979): Plasma levels of imipramine (IMI) and des- methylimipramine (DMI) and clinical response in pre- pubertal major Gepressive disorder: A preliminary report. J Am Acad ¢hild ychiatry. 18:616-627. 64. Puig-Antich J, Perel JM, Lupatkin W, et al. (1985a): Imipramine effectiveness in pre-pubertal major depressive disorders: |. Relationship of plasma levels to clinical response of the depressive syndrome. Arch Gen Psychiat. 65. Puig-Antich J, Lukens E, Davies M, et al. (1985b): Psychosocial functioning in pre-pubertal major depres- sion disorders: |. Interpersonal relationships during the depressive episode. Arch Gen Psychiat 42:500-507. 66. Puig-Antich J, Lukens E, Davies M, et al. (1985c¢): Psychosocial functioning in major depression disorders: Il. Interpersonal relationships often sustained recovery from affective episode. Arch Gen Psychiat 42:511-517. 67. Richman J (1979): The family therapy of at- tempted suicide. Family Process 18:131-142. 68. Richman J (1981): Family treatment of suicidal children and adolescents. In: Stuart IR, Wells CF (Eds): Self-Destructive Behavior in Children and Adolescents. New York, Van Nostrand Reinhold Co. 69. Ricks DF (1974): Supershrink: Methods of a therapist judged successful on the basis of adult out- comes of adolescent patients. In: Ricks DF, Thomas A, Roff M (Eds): Life History Research in Psychopathology (Vol 3). Minneapolis: University of Minnesota Press. 70. Robin AL (1979): Problem-solving communication training: Abehavioral approach to the treatment of parent- adolescent conflict. Am J Fam There. 7:69-82. 71. Robin AL (1981): A controlled evaluation of problem-solving communication training with parent- adolescent conflict. Behav Ther 12:593-609. 72. Ross CP, Motto JA (1984): Group counseling for suicidal adolescents. In: Sudak HS, Ford AB, Rushforth NB (Eds): Suicide in the Young. Littleton, Mass, John Wright PSG. 73. Ryan ND, Puig-Antich J, et al. (in press) Imipramine in adolescent major depression: Plasma level and clinical response. Acta Psych Scand. 74. Sabbath JC (1969): The suicidal adolescent-The expendable child. J Am Acad Child Psychiatry. 8:272-289. 75. Sabbath JC (1971): The role of the parents in adolescent suicidal behavior. Acta Paedopsychiatr (Basel) 38:211-220. 76. Shaffer D. (1974): Suicide in childhood and early adolescence. J Child Psychol Psychiatry 15:275-291. 77. Shaffer D. (1984): Notes on peycrotierany re- search among children and adolescents. J Am Acad Child Psychiatry 23:552-561. 78. Shapiro RJ, Budman SH (1973): Defection, ter- mination and continuation in family and individual therapy. Family Process 12:55-67. 79. Spivack G, Shure M (1974): Social Adjustment of Young Children: A Cognitive Approach to Solving Real- Life Problems. San Francisco: Jossey-Bass. 80. Stanley EJ, Barter JT (1970): Adolescent suicidal behavior. Am J Ortho 40:87-96. 81. Steinhauer PD (1985): Beyond family therapy. Toward a systemic and integrated view. Psychiat Clin NA 8:923-945. 82. Taylor E, Stansfeld A (184): Children who poison themselves. |. A clinical comparison with psychiatric con- P.Trautman: Specific Treatment Modalities. . . trols. Il. Prediction of attendance for treatment. Brit J Psychiatry 122:1248-1257. 83. Toolan, JM (1975): Suicide in children and adoles- cents. Am J Psychother 29:339. 84. Trautman P, Shaffer D. (1984): Treatment of child and adolescent suicide attempters. In: Sudak H, Ford A, Rushforth N {Bus Suicide in the Young, Littleton, Mass: John Wright PSG 85. Van Valkenburg C, Lowry M, Winokur G, et al. (1977) Depression spectrum disease versus pure depres- sive disease. J Nerv Ment Dis. 165:341-347. 86. Viale-Val G, Rosenthal RH, Curtiss G, Marohn RC (1984): Dropouts from adolescent psychiatry: A prelimi- nary study. J Am Acad Child Psychiatry. 23:562-568. 87. Weissman MM (1979): The psychological treat- ment of depression. Arch Gen Psychiatry 36:47-56. 88. Weissman MM, Paykel ES, Klerman GL (1972): The Serressse woman as a mother. Social Psychiatry 7:89-108. 89. Welu T (1977): A followup program for suicide at- tempters: Evaluation of effectiveness. Suicide and Life Threat Behavior 71:17-29. 90. Winokur G, Clayton PJ, Reich T (1969): Manic Depressive lliness. St. Louis, CV Mosby Co. 91. Wolkind S, Rutter M (1985): Sociocultural factors. In: Rutter M, Hersov I: Child and Adolescent Psychiatry. London, Blackwood. 92. Yalom ID (1970): The Theory and Practice of Group Psychotherapy. New York, Basic Books. 93. Youngermen J, Canino | (1978): Lithium car- bonate use in children and adolescents. Arch Gen Psych. 35:216-224. 94. Wahler R, Leskye G, Rogers E (1977): The insular family. Read at the Banff International Conference on Be- havior Modification, Banff, Alberta. 95. Werry JS (1979): Family therapy, behavioral ap- proaches. J Am Acad Child Psychiatry 18:91-102. 3-263 PERSPECTIVES OF YOUTH ON PREVENTIVE INTERVENTION STRATEGIES Iris M. Bolton, M.A., Executive Director, The Link Counseling Center, Atlanta, Georgia SUMMARY Because a youthful suicide has a powerful ef- fect on thousands of young people, it is im- portant to involve them in developing and carrying out preventive strategies which they might view as helpful. Their depth and maturity of thinking, their insights, honesty, and frankness can provide us with additional solutions to the agonizing tragedy of suicide. Their opinions regarding the statistics on the increasing rate of teen suicide, why some young people choose suicide while others choose life, and what we as professionals might do to help prevent this tragedy can offer us guidance in our examination of this phenomenon. INTRODUCTION It is estimated that every 90 minutes a young person completes the act of suicide. Nation- ally, suicide increased in youthful popula- tions by 136 percent between 1960 and 1980 (5.2 to 12.3/100,000). According to the Na- tional Center for Health Statistics, suicide is the second leading cause of death for 15-24 year olds. For every young person who com- pletes suicide, another 100 attempt suicide unsuccessfully, some becoming paralyzed or disabled for life. These numbers are evidence of a major health problem in the United States. Young people have displayed remarkable in- sight and depth of understanding when ad- dressing the reasons of why some of their 3-264 peers choose suicide while others choose life. They make cogent suggestions for effective and helpful intervention approaches. As helping professionals, we must continue to solicit their viable and important contribu- tions, adding these to the already-existing data, so that we might obtain additional resources in our fight against this national tragedy. Today’s youth have clearly spoken to us of the stress and anxiety that they experience in today’s complex world. Many of them have been profoundly touched by the suicide of a friend or loved one. It is, therefore, impera- tive that they be given full consideration in addressing this problem. It is appropriate that they become aware that they are not powerless, that they can take charge of their lives, and that they can assist their friends and peers by becoming part of the solution to suicide prevention. We have solicited and received comments and opinions from young people throughout the country about why so many of their peers have chosen to give up on life while others did not choose this drastic action. We have asked them to report to us anonymously their suggestions for effective prevention techni- ques aimed specifically at young people be- tween the ages of 14-24. Questionnaires (Appendix A) were distributed with the help of youth counselors, youth workers, and suicide prevention coordinators around the I.Bolton: Perspectives of Youth... country. Two specific age populations were targeted; high-school age youth, and college students up to the age of 24. Participants were provided with self-addressed, stamped envelopes for direct return of the informa- tion. Age and gender were indicated on the return questionnaire and, in many cases, the respondent chose to identify him/herself by name and address (the latter was optional). Questionnaires were returned by 82 stu- dents, including 54 from high school youth and 28 from college youth. Among the total number of respondents, 35 percent were male, 45 percent were female; 20 percent replied anonymously. All of the responses were written with the exception of one high school group from the State of Georgia which submitted a five-minute video discus- sion. Students from twelve States were in- vited to participate including a sampling from the Northeast, Southeast, North Central, South Central, Northwest and Southwest. States included were New York, Maryland, Pennsylvania, Georgia, Minnesota, Kansas, Texas, Washington, Montana, Illinois, North Dakota, and California. Responses were received from Washington, California, Min- nesota, Kansas, Maryland, and Georgia. DOCUMENTATION OF OTHER RESOURCE STUDIES Many authorities have described the attrac- tion that the act of suicide holds for some young people in distress. Few researchers have gone directly to the source: to youth themselves. One account, in particular, called Loss and Grief Overload by Judith M. Stillion of Western Carolina University provides such a clear analysis of current youthful problems and attitudes that it is out- lined here to serve as confirmation of the validity of this approach. Ms. Stillion is a teacher who has headed an enrichment project for bright students. Her test cohort consisted of fifty persons each summer for four successive years. Every stu- dent had a measured intelligence quotient (IQ) of 130, had scored on achievement tests at least two grade levels above their current placements, and had been recommended for the course by a school official. The investiga- tion was a part of a project for gifted students in the ninth grade called the Cullowhee Ex- perience, then in its 24th year, the theme of which was to study decisionmaking using the components of feeling, reasoning, valuing, and deciding. They discussed adolescent at- titudes toward suicide for two days. The discussion was important in under- standing the world view of adolescents. These students stated that many of their peers are in a constant state of bereavement and grief. They believed that the suicide rate was already proportional to the increase in loss experienced by young people today. Several sources of loss which they listed are: body changes, the increasing divorce rate, geographical relocation and moving away from friends and support groups, stress and tension associated with living up to parental expectations and academic pressures, drug and alcohol experimentation, and the threat of nuclear destruction. John Mack docu- mented similar feelings of anger and grief regarding the nuclear threat in his later studies with young people which he presented through videotape. Mrs. Stillion concluded her study with many good suggestions, including: 1. Stop considering kids as innocents. They’re not. 2. Think of them as veterans of grief and loss who have not yet learned to cope. 3. See that they find adults who can teach them about coping and who will give them permission to express their sorrow and pain, talk out their anger, and resolve their grief. 4. Consider applying grief counseling models to adolescents. S. Help youth acknowledge and understand their feelings and teach them skills to cope with the overload of loss. 3-265 Report of the Secretary’s Task Force on Youth Suicide Stillion believes that many adolescents have images of a safe, predictable, consistent world only in their fantasies (if at all). She concurs with Erik Erikson’s theory that men- tally healthy people must have a basic sense of trust in the world around them. She con- cludes with optimism that we must communi- cate to youth that, in spite of the stress, loss, television newscasts and media headlines, there still exists a rewarding world. In March 1986, a special edition of Life magazine examined America’s teenagers, their passions and problems. They put together a nationwide team of 22 teenage reporters and assigned them the task of inter- viewing their high school peers. They looked at what the 25.5 million youngsters born be- tween 1967 and 1973 were doing. The sur- vey concluded that "the highs are the highest, the lows the lowest; it’s the first time and the last time and forever." Adolescence has never been tougher with all-time-high num- bers of teen runaways, pregnancies, im- prisonments, and suicides. Although the Life survey was criticized for containing shallow and vapid questions relat- ing to current sayings, hairstyles, eating habits, etc., many of the questions allowed us a composite view of the more serious pres- sures facing today’s adolescents. For in- stance, it was reported that "marijuana smoking is as regular as breathing." When asked, "What do you want to be when you grow up?" most teens responded "rich." Regretfully, the Life report missed the op- portunity to deal with the teens’ attitudes about suicide, depression, and stress. HIGH SCHOOL STUDENT RESPONSES TO THE QUESTIONNAIRE In February and March of 1986 question- naires were given to high school students across the country. They were asked to respond to four questions relating to suicide. The first question was as follows: 3-266 "In your opinion, why are so many young people between the ages of 15-24 attempt- ing suicide today?" In reviewing the responses, it is clear that high school youth feel that they are living ex- traordinary stressful lives, due in part to pres- sure from family, school, peers, and others in authority. Some thought that they are being pushed too hard to achieve. A student from Pennsylvania said: "society expects too much from us like ... get good grades, go to a good college, get a good job, make a lot of money, be a success.” Other students felt pressure to be perfect and never make mistakes. They felt they could not live up to the many expec- tations which were placed on them by parents, and society as a whole. They spoke of competition and giving up when it became clear that they couldn’t "measure up." Many students attributed suicide attempts to wanting attention from parents or friends or "to see if anybody cares." There was a near consensus that people who attempt suicide are lonely, sad, depressed, have low self-es- teem, feel unloved, rejected, unwanted, or feel they are in the way. A student from Georgia said "if I killed myself, my friends wouldn’t have to put up with me anymore." They talked of the cruelty of "friends" who “criticize, call you names, or make fun of you." The high school students spoke to the feel- ings of hopelessness when too many problems piled up and they didn’t feel com- fortable sharing that with anyone. Being overwhelmed with pain and sadness and "not being able to trust anyone" was a common statement. One student lamented, "they don’t know where to turn so they turn to the easy way out." A 16-year- old girl speaking of those who think of suicide said, "if people don’t try to do something to stop them, they (the youngsters) think that they (the adults) really don’t care." Some students believed that most people really don’t mean to go through with suicide; they leave hints and hope somebody will stop them. A number of respondents wrote about the meaninglessness of life in terms of feeling I.Bolton: Perspectives of Youth... bored. "There is nothing for us to do ... there is nothing to live for ... there is no purpose to life." A student from Washington wrote "I don’t feel needed, important, valuable ... so what’s the point of living?" A common issue for teens attempting suicide was discussed as "confusion about sexual identity." Comments were written as follows: "You have guilt about your sexuality, your ac- tions .... you have fears of being gay." Also, "due to religious upbringing, guilt and fear of being evil or a sinner is terrifying. You're afraid you will be punished by God, so you might as well go ahead and punish yourself." Drug and alcohol abuse was viewed by stu- dents as an important reason for some kids attempting suicide. "There is a lot of peer pressure to do drugs and they are used for recreation ... people can’t handle it ... if you're already depressed, using drugs exag- gerates and magnifies your feelings." Punk music and rejection by friends was also men- tioned frequently as causes for attempted suicide. Other issues discussed included kids who are abused physically or sexually, kids who are angry and have no outlet for their anger, teen pregnancy and the associated guilt and shame, as well as getting revenge on others. "In your opinion why are so many young people between the ages 15-24 completing suicide today?" A number of responses from high school stu- dents repeated and emphasized the same reasons as for attempting, with special em- phasis on stress and pressure to achieve. Ad- ditional relevant comments include the following: "Suicide is an escape from too many problems at once" "They don’t actually see themselves DEAD" "People don’t take the warning signals seriously” "Everyone would be better off if I was dead” "It’s so easy to get a gun or other weapon" "Revenge, to make everyone sorry and guilty" "Breakup of a romance" Adolescents get depressed because they tell themselves things that aren’t true, such as "I'm stupid ... nobody likes me ... I'll never stop hurting ... it’s hopeless ... I'm hopeless." Another comment was "sometimes you believe you are crazy and you're better off dead." The third question was stated as follows: "What do you think can be done to prevent suicide?" Answers ranged from things individuals can do to suggestions for society in general. Typi- cal responses were: "Use students as counselors because they re- late better to their peers” "Have telephone counseling lines" "We need more speakers in high school and junior high to educate students, teachers and parents." "Relieve some of the pressures” "Take all threats seriously” "Teen clubs would help kids have something to do" "Tell teens bad times won’t last" "Have dinner together as a family--we never see each other anymore” It is apparent, from the responses, that these high school students want to be included in problem solving and helping to save lives. The last question in the questionnaire ad- dresses the issue of why people do not com- mit suicide. The question was posed as follows: "Why do you think most young people do not commit suicide?" Most of the respondents concluded that the reasons most people do not commit suicide is that life is going well for them; they have friends; they know how to ask for what they want; they realize the pain or the problem 3-267 Report of the Secretary's Task Force on Youth Suicide won't last forever; and they have goals for the future. Others felt that these teens had people to talk with who understood them and who listened to them without trying to solve their problems. Some thought that many kids are taught that suicide is morally wrong and their religious beliefs kept them from doing it. A few teens felt that their peers did not com- mit suicide because they were afraid to die and frightened of the pain involved. They were also frightened that they might be dis- abled if they did not succeed. A few teens ex- pressed the thought that if no weapons or pills could be found in the house, the depres- sion and urge to kill themselves might pass. On a positive note, several respondents stated, "They have a love of life in spite of trouble" "They are not afraid to tell someone they are in trouble or to ask for help" "Most teens value their life, even in spite of present circumstances” COLLEGE STUDENT RESPONSES TO THE QUESTIONNAIRE College students in the some geographical areas were given the same questionnaire as the high school students. The first question was as follows: "In your opinion why are so many young people between the ages of 15-25 attempt- ing suicide today?" Their responses were similar to those of the .high school students, citing pressure to achieve beyond their abilities, high pressure from family, peers, school, themselves, and society to perform, drug and alcohol involve- ment and abuse, family problems, loneliness, and hopelessness. They addressed the issues of low self-esteem, inability to communicate, acombination of problems, inadequate social skills, losing a boyfriend/girlfriend, and not knowing how to cope. There were numerous responses from students who believed that 3-268 suicide attempts were for attention-getting and were a cry for help which had gone un- heard. Some of the specific comments were as follows: "Possibly we lack the mental toughness that it takes to face the reality of the real world." "As an animal scientist, a very important prin- ciple was constantly drilled into my head-- genotype (internal) x environment = phenotype (the outward expression of a trait). Environmental factors are numerous: breakdown of the traditional family, lack of responsibility, pressure to achieve beyond their ability, a means of drawing attention." "Ihave been involved in three different situa- tions where people wanted to commit suicide. In all three cases, I felt the major problem was that the persons did not like themselves, mainly because of a hard situa- tion that had happened to them (abortion, break-up of a relationship). In all cases it seemed the only way out." The second question presented to college students was as follows: "In your opinion why are so many young people between the ages of 15-24 complet- ing suicide today?" Responses in this category again reflected the thinking of the high school students with a number of additions as follows: "Younger people are committing suicide be- cause they feel inadequate or they aren’t given enough attention. More should be taught about the psychological thoughts of children" wrote a 21-year-old female who in- dicated that a friend had committed suicide. "Too many pressures to think and act like adults. When I was 15 my mother decided to remarry. I was expected to understand and hold back all my anger." "Unemployment is a problem ... now ... in col- lege, part-time ... and later. Maybe Iwon’t be able to get a job to support myself." "People don’t realize the permanence of suicide." I.Bolton: Perspectives of Youth... "A lot of kids are selfish and don’t think of anybody but themselves." "If you can’t equal your parents’ success, why try? Take the easy way out and kill yourself." "A lot of kids only meant to attempt but end up dead because they didn’t know how to save themselves." "Feeling lonely and trapped--cornered." "The media show life’s problems solved quickly and painlessly. Youth expect to lead such happy lives, quickly resolving their problems. You agonize over decisions and you fear failure." "Both people I know killed themselves be- cause of drugs." "I don’t know who I am and what my purpose on earth is. Finding my identity is hard--I might as well give up now." Question 3 of the survey asks: "What do you think can be done to prevent suicide?" College students discussed a variety of preventive measures involving the in- dividual, the family, schools, and society. Pertinent responses are as follows: "We need more education about suicide for students, teachers and parents. We need to know the warning signs of depression." "Teach kids a more realistic view of life." "Allow for failure and help kids realize you can learn from your mistakes." "Let students know that others have the very same problems and they just hide it better." "Tell them it isn’t okay to die." "I don’t like my problems, but I know they are temporary." "Let kids create plays and improvisations about suicide for other kids to see--kids lis- ten to kids." "Encourage kids to plan conferences them- selves on suicide. Adults can advise, but let kids get involved in actually doing it." "Teach the consequences of suicide on fami- ly and friends." "Deglamourize suicide. Kids need to know the gory details and that it is not romantic. A 24-year-old male from Athens, Georgia summed up his suggestions for preventing suicide with these comments: a. "Parents taking more time to really talk with their children and being courageous enough to openly discuss with them their problems." b. "Giving children tasks to perform to estab- lish early on their own personal impor- tance in the scheme of things." c. "Don’t spoil children by constantly giving them material things, but instead, give them your time." d. "Encourage children to do their best, but make them understand that their own per- sonal best may be less than someone else’, and that’s okay." e. "Encourage children to spend more time in creative activity and athletic activity rather than just sitting and watching television for hours on end." The last question for college students posed in the questionnaire was: "Why do you think most young people do not commit suicide?" The essence of comments from college youth is as follows: "Feelings of responsibility or guilt." "Too scared to do it." "Hopeful about the future." "Have few problems." "Strong support network." "They enjoy life." "I thought of how others would feel ... I couldn’t do that to them." "Mental toughness." 3-269 Report of the Secretary's Task Force on Youth Suicide "They have goals and dreams." "They let their anger out in other ways." "They know that all problems can be solved and that nothing is so bad that they have to die over it. Things just feel that way some- times." "The knowledge that I do have something to offer ... maybe not now, but some day." "It’s okay to fail and to be imperfect, if I do my best and learn from my mistakes." "Since God made me I must be valuable and okay." "My religious teachings keep me from suicide." "I'm afraid of what I might miss if I killed myself." SUMMARY AND RECOMMENDATIONS Through a sampling of high school and col- lege students’ attitudes about suicide, it is clear that many of their lives are full of com- plexities, pressures, stress and frustrations. A 22-year-old female from Minnesota spoke for a number of her peers when she wrote: "Attempting suicide is a cry for help. Too many people take life for granted and too many people take love for granted. Without expression of such feelings, people of ages 14-24 can feel neglected and useless. On a more personal basis, in the early twenties or late teens there are a lot of transitions, walls of financial un- certainty, too many paths to follow and decisions to make with no one to really help. If you're standing on shifting sand, lacking control of life, what better way to control it than to end it? Completed suicides, I feel, are attempted suicides that went too far. People who attempt suicide want to be helped, but are too hasty and sometimes the result is a com- pleted suicide. I think that when suicide is actually completed, it has been an ac- tion taken because of a gnawing growth 3-270 of self-disappointments over many years. Some people just feel that life just isn’t satisfying enough. I just don’t under- stand why these people don’t speak louder or more obviously to get the help they need. Sometimes people can’t lis- ten because they aren’t spoken to open- ly enough." The energy and enthusiasm of the youth who responded to the task of suggesting solutions to the problem of suicide is encouraging and gratifying. They took their charge seriously and gave thoughtful ideas with mature in- sights. Suggestions covered four main areas including preventive measures for in- dividuals, families, schools, and society. Recommendations for Individuals (Ages 14-24) It was probably no accident that major em- phasis was given to the area of individual responsibility and initiative. Although the three other areas were important to them, it is observed here that young people are aware that it is their own coping skills and abilities which can save lives. They do not hide from the reality that they themselves have the power to be in charge of their own lives, even though at times they may feel powerless. They observe that they may not be able to control events in their lives, but that they can take charge of how they respond to those events. A general theme among all the students was learning sufficient coping skills to handle crises, especially a suicidal one for themsel- ves, or for someone else. To learn how to cope and how to solve problems was a priority. It was suggested that when some- one had suicidal thoughts and images, they could consciously replace them with more positive concepts and thoughts. To go fur- ther, young people wanted to change their own attitudes about counseling so that there would be more freedom to seek counseling without fear of being labeled "crazy" or "in- adequate." I.Bolton: Perspectives of Youth... Students wanted to find ways to decrease the pressure they felt from parents, peers, teachers, others in authority, and their own self-imposed pressure to perform and achieve. It was felt that if they could open lines of communication with others and take responsibility for giving feedback about the kind of pressure they were feeling, they at least would not feel like victims. A major insight and recommendation from youth was that if they were to risk sharing their humanness and vulnerability with others, then everyone would begin to realize that most people have the same problems and the same feelings. One student told of a 7-year-old boy who found a Playboy magazine and looked at the pictures. When asked what he thought of the naked people in the pictures, he thought a moment and then said, "Well, under clothes, everybody is naked!" A 20-year-old from California said: "Letting down the barriers and taking off our masks is a way to begin to understand that we are not alone, and that others experience the same feelings. Self-esteem is important to youth and finding ways to feel valued and worthwhile is an im- portant task. They suggest loving yourself unconditionally, just because you were born, and not because of what you do. If you can do that, it doesn’t matter who else loves you because you always have yourself and your self-respect. The students were concerned with finding a purpose in life, some meaning or goal for the future. They suggested that a skill for coping with depression was to find something you wanted to do in the future that if you died you would never have the opportunity to do ... a real "missed opportunity." They suggested picturing what the future would be without you and without your contribution. To an- ticipate something in the future is to have hope. The fast-paced, complex life of school, relationships, family, cars, drugs, alcohol, peer pressure, and the accumulation of problems can lead to wanting to end one’s life. A 19-year-old female from New Jersey said, "I didn’t want to end my life, I wanted to end what was going onin my life ... and there’s a difference.” This insight is rare and needs to be translated to others, according to many college students. The wisdom of youth tells us that we must learn patience in an impatient world; that pain does not last forever, nor does joy. It is important to know that depression doesn’t last a lifetime, even though it may feel that way sometimes. Additionally, when dis- couragement and pressure to achieve are ex- perienced, it is important to remember that it took parents a long time to be successful too. Individuals need to understand that suicide is a permanent solution to a temporary problem. They need to take responsibility to ask for help when they need it. They need to find outlets for anger so it doesn’t get bottled up. They need to know they are not helpless even when they are abused. Many students suggested getting help from their minister or rabbi or from their religious beliefs. The students discussed the need to deal with relationships in terms of learning how to get into one, how to keep it together, and how, when appropriate, to get out of the relation- ship without destroying one another. Students recommended that everyone have a support system of several people whom they trust so they may confide in them during stressful times. Since most young people realize that they ex- perience many losses in their lives, they recommend learning about the process of grief and how to handle loss. They suggest that the individual give him/herself permis- sion to learn from mistakes without being devastated and especially that they learn to laugh at themselves. One 24-year-old stu- dentstated, "to be able to minimize the heavi- ness and seriousness of life with laughter is to be able to survive." Finally, young people wanted their peers to take responsibility to plan conferences ad- 3-271 Report of the Secretary's Task Force on Youth Suicide dressing the problem of depression, stress, and suicide. They believe that peers listen to peers and that they can have the greatest im- pact on one another. They advised that youth work alongside adults in an education- al preventive effort at educating other youth in terms of warning signs, interventions, and counseling resources. They are 99 percent in favor of teaching peer counselors to help identify and counsel troubled youth, perhaps one day teaching peer counseling skills to the whole student body. Recommendations for the Family Youth are asking for the support and en- couragement of parents, rather than pres- sure to succeed. They are asking for quality time with parents, including occasionally eating dinner together instead of everyone in the family being on different schedules. They suggest helping young people to feel better about themselves by talking to them, not at them. They want their feelings to be taken seriously; they want their lives to mat- ter and to feel that parents value them as people, not as puppets to perform and to be controlled. They suggest that parents pay closer attention to the feelings of their children and that they take threats of suicide seriously, especially in those who have at- tempted previously. They suggest that parents take parenting courses to learn to communicate, listen, and hear, so they can teach kids to communicate effectively also. They want parents to set limits, but to be fair in so doing. They suggest removing guns and large doses of medication from the home so that, in moments of desperation, it would be difficult to find the means with which to kill oneself. Young people are concerned about child abuse and wish that parents would get help with the stress and pressure of parenting, and know how to have more control of their anger. They also want parents to help children understand that "someone can be angry at you and still love you." Many kids 3-272 expressed concern that a parent’s anger means the child is unloved and unloveable. Additional recommendations involve issues of drinking and parental alcoholism. They suggest that parents seek help with their problems, so they can set a better example at home. A final suggestion for parents is that parents and children should enjoy one another more without getting into power struggles. Youngsters would like to do things with parents occasionally, but they want their privacy respected and their need to be with their friends understood. They hope that parents know that learning to become inde- pendent is difficult, and finding their identity is a major task. They need help with that, and with the fact that sometimes they feel like kids, and the next minute they feel like adults. They are asking for patience. Recommendations for Schools Almost unanimously, young people recom- mend that suicide prevention programs be conducted in the school, but that they be geared to teaching skills such as communica- tion, dealing with stress, building self-esteem and problem solving. They want the word "suicide" minimized so that it is not always in front of them. They hope that schools will provide this kind of training which includes learning the warning signs of depression, in- tervention skills, and what to do when suicide occurs. They want students, teachers and parents to have the same training. A popular suggestion was to have peer coun- selors in the schools so that everyone would know someone his or her own age with whom she/he can confide. Respondents to the questionnaire emphasized teaching coping skills and helping skills. They wanted to learn how to cope themselves, and they wanted helping skills to be able to help some- one else in trouble. Most students stated that they had no idea of where to go for help outside the school. They recommended posting phone numbers of I.Bolton: Perspectives of Youth... outside counselors and centers in the halls and bathrooms of the school. A young man in Georgia suggested that coping skills be listed on a wallet-sized card so that if you felt really suicidal, you could pull that card out of your wallet and read something like: 1. Pain doesn’t last forever. 2. Go talk to a friend or someone you can trust. 3. Your feelings are normal. 4. Give yourself a break. S. Suicide is a permanent solution to a tem- porary problem. 6. Call 256-9797 to talk with someone who can help. Students recommended that drama clubs and classes write and perform theatrical presen- tations dealing with the subject of stress and depression, so that youth will be more recep- tive to dealing with the issue of suicide. It is their thinking that, in an age of fast-moving video frenzy, one of the best ways to get the attention of young people is through live entertainment. The original student musical entitled, "Dim Lights Need More Current" performed by a cast of students from Griffin High School in Griffin, Georgia, was cited as an example. This kind of presentation, done under qualified adult supervision "benefits the student audiences, the cast, and the com- munity," according to a 17-year-old student. Recommendations for Society The primary focus for action in this area is the changing of attitudes of our culture to be more accepting of counseling and of asking for help. Young people want to change at- titudes about discussing problems so that "you know you are not alone and you know that everybody has problems." Students hope that society will value young people more in the future and will help them understand that their life does matter. They do not want society to dictate what success is; they believe success is an individual matter. They hope that people will become less judgmental and more loving, allowing young people to succeed and fail in their own time, without the pressure of a materialistic culture judging everything in terms of money, things, prestige and power. Students suggest more community centers, supported privately or publicly, so kids have a place to go and to belong. They hope society will encourage more physical activity in this age of the computer revolution and the age of television. They suggest that doctors receive more training in suicide prevention so that patients who have attempted suicide will not be released too soon from the hospi- tal. They want the clergy to have more train- ing in prevention and learn how to help the attempter and the family of a completed suicide, so that they don’t commit suicide themselves. They expect our society to teach people the consequences of suicide to families, friends, and communities throughout the country. REFLECTIONS It is the attitude of the young people in this country who believe that they can make a dif- ference that will help those who feel that they cannot. It is their hope that under our masks we are indeed one and, if not the same, at least similar. It is their vision for the future that provides them with a sense of purpose and direction. It is their attitude, their hope, and their vision that will change this world to effect not only the problem of suicide in this country, but that will allow them to become healing enablers in a society which needs them. As they feel needed, as they feel their power, as they feel their value, our families, schools, and communities will feel their presence. We have been privileged to experience their energy and their enthusiasm in writing this paper. Their voices have been heard and duly recorded, with respect and admiration. A 22-year-old young woman from Minnesota wrote these simple words: 3-273 Report of the Secretary’s Task Force on Youth Suicide "Most people have the strength to hang on to tomorrow, knowing that sunshine and rain- bows follow the rains." PARTICIPATING COORDINATORS Mr. & Mrs. Jay Mossman; Baltimore, Maryland Sparky Cook; Marietta, Georgia Jennifer Dimmick; University of Georgia, Atlanta, Georgia Shirley Bolling: Logansville, Georgia (videotape) Jan Olds; Overland Park, Kansas LaRita Archibald; Englewood, Colorado Charlotte Ross; San Mateo, California Anthony J. DelPercio; St. Paul, Minnesota Shirley Cooper; Marietta, Georgia Brian Jung & Vicki Grossman; Bothell, Washington Jim Walkup; Bronxville, New York Habersham County High School; Cornelia, Georgia (videotape) REFERENCES errr ’ 1. Stillion JM: Death Education, Supplement: Suicide. 8:1-55, 1984. 2. Stillion JM, McDowell ED, and Shamblin JB: The suicide attitude vignette experience: A method for measuring adolescent attitudes toward suicide. Death Education. 8:65-79, 1984. 3. Holinger PC: Adolescent suicide: An epidemiological study of recent trends. American Journal of Psychology, 135:754-756, 1978. 4. Erickson EH: Childhood and Society. New York, W.W. Norton, 1950. ACKNOWLEDGMENT --eeeeeeeeeeeeeennnnn. - The author wishes to acknowledge the assis- tance of Kathleen Gildea in the preparation of this document. Address correspondence or requests to the author at: The Link Counseling Center, 218 Hiderbrand Avenue, Atlanta, Georgia 30328 3-274 I.Bolton: Perspectives of Youth... APPENDIX A DATE: QUESTIONNAIRE For youth 14 to 24 years old ____High School __ College ____Other: Please explain Age ___Large City __Medium City _Suburb __Small Town ____Other: Please explain, sok kk ok k ok %k sk k %k ok kk ok ok ok %k k kk k k %k dk ok ok k k k ok k k k k k *k k k k *k k k k *k k k *¥ k k k *k k Anonymous: OPTIONAL Name: Address: City, State, Zip: Phone: Office Home Person responding to questionnaire: Friend or relative of someone age 24 or younger who has attempted suicide. Friend___ Relative Friend or relative of someone age 24 or younger who has completed suicide. Friend___ Relative I have seriously considered suicide myself. I have attempted suicide myself in the past. ____ Ido not have any personal experience with suicide in this age group of anyone age 14-24 Bok ok sk sk ok kk kk kk ok ok kk %k ok sk ok ok ok ok %k %k 5k 3k sk k k k ok k kk k k k k k k k kk k kk k k k kx *k * PLEASE ANSWER THE FOLLOWING QUESTIONS: 1. In your opinion why are so many young people between the ages 14-24 attempting suicide today? 2. In your opinion why are so many young people between the ages 14-24 completing suicide today? 3. What do you think can be done to prevent suicide? 4. Why do you think most young people do not commit suicide? My response will be: Written (limit: 2 pages) Audio cassette recording (limit: 5 minutes) Video recording (limit: 5 minutes) 3-275 MASS MEDIA AND YOUTH SUICIDE PREVENTION Alan L. Berman, Ph.D., Department of Psychology, American University, Washington, D.C. INTRODUCTION The stereotype of the mass media is that of an omnipotent sculptor of attitudes, inter- ests, and behaviors of a highly malleable and responsive public. After all, if the advertis- ing it presents can effectively merchandise products as diverse as detergents and politicians, then surely it must be responsible for stimulating and controlling the behavior of large numbers of people. Social scientists who have investigated the ef- fect of the mass media on human behavior have been primarily concerned with the im- pact (essentially negative) of television. Of special interest here are those studies sug- gesting media influences on aggressive-- specifically suicidal--behavior among young people. If such a negative impact can be documented, then preventive efforts can focus on attenuating these effects. TELEVISION AND YOUTH Among adolescents, television is the preeminent medium and a trusted source of information. There is clear evidence that television advertising aimed at the youth market is successful in influencing purchas- ing behavior (1). Television as a whole, has been described as a significant source of socialization (2) and, by the National In- stitute of Mental Health (NIMH) as "a sig- nificant part of the total acculturation process" (3, p.87). Estimates of viewing habits of adolescents, ranging from 18 to 21 hours (4) to nearly 28 3-276 hours per week (5), have led to the observa- tion that typical American youth spend more time watching TV than at any other single ac- tivity, including school (6). Among the more negative stimuli to which the typical young viewer is exposed during these hours are fre- quent depictions of people drinking alcohol and a barrage of violent images and acts. Content analyses of television programming have found that alcohol use, both casual and heavy, occurs twice as often as the drinking of coffee or tea (7). Greenberg (8) estimates that the average viewer who is too young to drink will see about 3,000 drinking acts per year, although Wallack, et al. (9) argue that heavy or irresponsible drinking is shown only infrequently. Similarly, very little licit or il- licit drug use is depicted on television (10). In contrast, televised acts of violence have been described as "so pervasive that, by graduation day, the average high-school stu- dent has seen 18,000 murders" (5, p.46) and 800 suicides (11). Violence on Television and Aggressive Behavior The most widely publicized conclusion of the NIMH-sponsored review of the relevant scientific literature on television and be- havior was that "violence on television does lead to aggressive behavior by children and teenagers who watch the programs” (3, p.6). Although the link was not seen as enduring, it was described as "causal" (3, p.6). The re- searchers proposed several theories, all of A.Berman: Mass Media and Youth Suicide Prevention which are common to suicidology, to account for the observed effects: (a) observational learning (imitation and modeling), (b) disin- hibition, (c) attitude change, (d) desensitiza- tion and heightened arousal, and (e) justification of preexisting aggressive be- havior. Representatives of the broadcast industry (12,13) and others (14) have questioned the evidence and conclusions presented in the government review and criticized the studies as methodologically inadequate and proving correlation but not causation. However, while noting that research never yields une- quivocal interpretations, Rubinstein con- cluded that "the convergence of evidence from many studies (of television and aggres- sion) is overwhelming" (7, p.821). MEDIA AND IMITATIVE SUICIDE If violence on television promotes imitative aggressive behavior, can it be demonstrated that media depictions of suicide promote suicidal behavior among its viewers? Bolen and Phillips found significant increases in suicides in the United States just after televised news stories about suicides with ef- fects lasting about 10 days (15). Phillips also reported evidence suggesting that both suicides and fatal and nonfatal auto accidents increased after fictional (soap opera) suicides appeared on television (16). However, Kessler and Stipp criticized his methodology and further analyses in- validated this finding (17). The effect of publicized suicides on imitative behavior has been most consistently docu- mented in studies of the print media. In an early paper, Phillips found statistically sig- nificant increases in suicides just after front- page suicide stories (18). Phillips termed this phenomenon the "Werther effect," referring to an alleged rash of imitative youth suicides that followed the publication of Goethe’s The Sorrows of Young Werther in 1774 (19). Phillips showed that the Werther effect in- creased proportionally to the amount of publicity devoted to the suicide and that it oc- curred primarily in the geographic area where the suicide story was published. Was- serman extended Phillips’ data set and reexamined his findings (20). His analysis revealed that only stories of celebrity suicides appeared to elicit imitative behavior. Recently, Stack reported an extension of this work, demonstrating that entertainers had a more significant effect on imitative suicides than other celebrities (e.g., politicians, criminals) and that this effect on suicide rates was as profound as that found for unemploy- ment (21). Moreover, Stack demonstrated that the effect was specific to those in a similar social role--stories of young male suicides most affected suicides of young males. Other work by Phillips has linked publicized suicides to transient increases in other forms of violent death that might serve to disguise suicidal intent, namely, motor vehicle acci- dents and noncommercial airplane crashes (22-24). Davidson and Gould have summarized this research and concluded that nonfictional, media-reported suicides do serve as models for imitative behavior (25). The evidence linking fictional models with imitative be- havior is, however, more controversial and less conclusive. Anecdotal reports of imitative suicide follow- ing presentations of fictional suicides on television and in movies have appeared (22,27). Radecki (11) has documented 37 deaths by Russian roulette world-wide be- tween 1978 and April, 1985 attributed to im- itations initiated by viewing the movie "The Deerhunter." Even rock music has been blamed for stimulating suicide (28). Until recently, however, no one had presented em- pirical evidence for a relationship between fictional suicides and imitative suicide. Gould and Davidson have reported the first such data (29). They examined completed youth suicides for metropolitan New York City, southwestern Connecticut, and all of New Jersey as well as admissions at six New 3-277 Report of the Secretary's Task Force on Youth Suicide York area hospitals for attempts two weeks before and two weeks after each of four televised movies presenting fictional suicides. They found significant increases in both attempts and completions in the fol- lowup period. Unanswered by this data is the question of whether the cases following the televised fic- tional suicides were new or merely ac- celerated, i.e., they would have occurred anyway at some other time, precipitated by some other event. Also, no data is presented that directly link the event to the film, i.e., did the adolescent even see the film? Also, there is some suggestion that the attempts follow- ing the film were of lower lethality (resulting in a lower proportion of hospital admissions after, rather than before each film). There- fore, it can be construed that an ostensibly negative effect may actually have been posi- tive in that more troubled teens, because of the film, were stimulated to bring themselves to the attention of the helping system. As further evidence of this interpretation, hot- lines and crisis centers reported significantly increased numbers of telephone contacts after these presentations (30). With regard to increased completions, the Gould and Davidson data are difficult to in- terpret. One of the films studied had no sub- sequent youth suicides reported. Another, in which the teenager portrayed actually talked his suicidal father out of killing himself, had fouryouth suicides after the film, versus none prior to its showing. These data argue for more intensive case studies, moving from macro to micro analyses. Television and Suicide: "Surviving": A Case Study - Part | On February 10, 1985, the ABC network aired a three-hour drama about youth suicide. "Surviving" was a fictional portrait of two depressed teenagers who found solace in one another’s arms and escape from pain through a dyadic suicide by carbon monoxide (CO). The New York Times’ review of this production described it as "serious," "enor- mously watchable," and "intelligent" (31). 3-278 ABC conducted its own poll of national prob- ability sample and found that 16 percent of those contacted had watched the film and that 93 percent of the viewers rated the film good to excellent (32). The majority viewed the film with another person (e.g., a parent or child), talked about it after its presenta- tion, and felt that it had increased their awareness and understanding of the problem of teen suicide. Other studies suggest that the film had un- toward effects. Ostroff et al. noted sig- nificant increases in adolescent admissions to a psychiatric emergency service for suicidal overdose in the two-week period after its presentation (33). Gould and Davidson’s data for this film showed four completed suicides after its showing compared with only one before (29). To evaluate further the effects of this film on completed youth suicide, we collected data from nine urban medical examiners’ offices across the United States: Atlanta, Cleveland, Dallas, Ft. Lauderdale, Philadel- phia, San Diego, Seattle, St. Louis, and Washington, D.C., serving a combined population of more than 12 million. The total number of suicides certified by these of- fices in 1985 amounted to 1,843, an estimated 7 percent of all suicides in the United States that year. We then compared two-week and four-week data sets, before and after the February 10, 1985 air date of "Surviving." As noted in Table 1, there were no differences for either study period in: (a) total suicides, (b) youth suicides, and (c) carbon monoxide suicides. However, there was a noticable shift in the proportion of youth suicides by carbon monoxide. This shift might be ac- counted for by a process of identification with and imitation of characters in the film. All five of the youth suicides by carbon monoxide after the movie were male (the movie depicted both a male and a female CO suicide). Further investigation of each of these suicides revealed that only two of them were known to have watched "Surviving." Each had seemingly positive reactions to the film: one wrote in his diary, "I loved that A.Berman: Mass Media and Youth Suicide Prevention film;" the other was reported to have responded to the film more ambiguously as "helpful." Each of these youths had long histories of pathology. One had made a prior suicide at- tempt by overdose six months earlier and was in treatment at the time of his death. The other left a diary spanning the last two years of his life that documented a suicidal gesture two years earlier, frequent suicidal ideations, and at least one other failed attempt by carb- on monoxide. The diary also recorded severe family conflict, rage toward his father, and problems of sexual identity. If anything, he had been more significantly affected by another television movie, "Consenting Adults," which aired prior to "Surviving" and dealt with homosexuality and a lack of fami- ly support. About this film he wrote: "...the movie reminded myself of me 100 percent...I feel exactly about males as he did. I feel like I'm lying to myself." Such psychological autopsies of completed suicides allow for more intensive investiga- tion of idiosyncratic influences not ascer- tainable through aggregate data analyses. From these case studies we can derive no evidence that "Surviving" stimulated new suicides. An imitative effect is suggested only in the shift of method chosen by youth who otherwise were known to be suicidal before the movie aired. The act of viewing a film takes place in the mind. The viewer cannot be reduced to a simple stimulus-response machine. How we participate as observers, with whom we iden- tify, what we perceive selectively, and whether we will be influenced at all depends on a variety of personal and mediating fac- tors: our history, our moods, our predisposi- tion, our needs. Heavy viewers of television have been found to see the world as a mean and scary place (17); those predisposed to violence might be those who prefer to watch Total Suicides 4-Week N 2-Week N Youth Suicides (under age 25) 4-Week N 2-Week N Proportion Youth/Total Suicides 4-Week 2-Week Carbon Monoxide Suicides 4-Week N 2-Week N Youth Carbon Monoxide Suicides 4-Week N 2-Week N 4-Week 2-Week Nine-City Sample of Suicide Completions Before/After Broadcast of "Surviving" Proportion Youth CO/Youth Suicides Before After 165 155 87 74 34 29 23 18 0.21 0.19 .26 0.24 16 18 8 8 2 5 2 3 0.06 0.17 0.09 0.17 Table 1. 3-279 Report of the Secretary’s Task Force on Youth Suicide and who are most readily aroused by viewing violence (34). Those influenced by fictional suicides appear to be so predisposed, rather than molded by the suicidal stimulus. MASS MEDIA AND PREVENTION One approach to the prevention of youth suicide is to limit or inhibit stimulating in- fluences on suicidal behavior. To the extent that the media may present models for imita- tion or where the media depicts behaviors that predispose depressogenic conditions (e.g., makes alcohol consumption seem at- tractive), then reducing these influences would be appropriate. However, the media may also take a more ac- tive and positive role in prevention. Educat- ing the populace could increase early detection of potentially suicidal youth. Using the media to inform viewers about health problems and to provide models that promote health-conscious behavior is one possible way to reach some of these poten- tially suicidal adolescents and thus reduce the population at risk. Public Service Announcements Public service announcements (PSAs) are the most typical mass-media vehicle for cam- paigns to educate the public and promote health. PSAs attempt to increase viewers’ awareness of a specific problem and possible solutions. In addition, some aim at changing beliefs, attitudes, motivations, and behaviors; however, significant change generally fails to occur (6). The reasons for ineffective PSAs are many and varied. Most PSAs consist of a small number of spots of varying quality, aired neither at prime-time nor when the targeted viewing audience is available to watch. In ad- dition, competition between PSA Sponsors limits the frequency of showing, therefore limiting the saturation, dissemination, and reach of any one message. As a result, the average time period during which a PSA is kept active is only three months (35). 3-280 To be effective, public information messages must be factual and specific and must reach the intended target audience. The great majority of those produced do not meet these criteria (35). Only 6 percent are aired during prime viewing time and even then on local stations when most of the viewing audience watches national programming (37). Han- neman and McEwen found that PSAs which were oriented toward youth were often aired during school hours (38). PSAs are usually placed gratis by the broad- caster (in contrast to paid advertisements); therefore, they are placed as space and time permits. Some of the responsibility for this lack of reach rests with those who produce the PSAs. For example, Capalaces and Starr found that station managers were poorly in- formed about a series of anti-drug PSAs, con- sequently, they allotted haphazard energy and effort to their scheduling (39). Good- man also noted that station public service directors lack specific guidelines for attend- ing to these campaigns (35). Evaluations of PSA campaigns have found anti-drug PSAs to be cost-effective (40) and anti-crime campaigns to be effective in changing behaviors (41). Other campaigns considered effective, however, have not been found to affect the young (42) and some may even have "boomerang" effects in that the risky behaviors sensationalized by the media may be glamorized and lead to increased ex- perimentation among youth (43). Flay (44) and Goldstein (45) have suggested several guidelines for effective PSA cam- paigns. PSAs should be novel and use a knowledgeable, credible spokesperson with whom the audience can identify. The con- tent of the PSA should be based on scientific fact and delivered in a manner that minimizes the arousal of fear. Also, clear alternative behaviors should be presented. The PSA needs widespread dissemination, high saturation (frequent exposure), and ex- tended duration of exposure. To be most ef- fective, PSAs need to be supplemented by other media (e.g., print) and community net- working. A.Berman: Mass Media and Youth Suicide Prevention PSAs have been developed that focus on suicide prevention. Highly professional PSAs have been produced by and are avail- able from the Los Angeles Suicide Preven- tion Center and the American Association of Suicidology. However, to date, no evalua- tion of their effectiveness has been con- ducted. Television and Suicide: "Surviving": A Case Study - Parti One possible explanation for the lack of im- itative suicides following the showing of "Sur- viving" in February 1985 was the extensive public information and awareness campaign conducted by ABC's Community Relations Unit. The campaign included both broadcast and print components. Five one-minute news segments featuring interviews with specialists in adolescent suicide and stress management were distributed. PSAs, with room for a local crisis center hotline number, were provided to all local stations. One hundred thirty thousand handbooks on teenage suicide prevention were distributed free to secondary schools, mental health centers, and crisis centers nationwide. Some local ABC affiliate stations went further, sponsoring mini-documentaries and even town meetings. The central element of this campaign was a half-hour educational program, "ABC Notebook: Teen Suicide" hosted by one of the stars of "Surviving". Eighty percent of af- filiated stations carried "Notebook," 80 per- cent of these were aired on the weekend of "Surviving." However, fewer than one in five stations (18.2%) showed "Notebook" during prime time viewing hours. Perhaps for this reason, only 6 percent of the viewers who watched the movie, saw "Notebook" as well (36). PRINCIPLES OF EFFECTIVE MASS MEDIA PREVENTION CAMPAIGNS Historically, mass media public information campaigns have rested on the assumption that problems should be addressed with more and better information. Thus, information has been confused with education, and education with prevention (46). Behavior change does not result from mere exposure to well-designed informational messages. This "hypodermic needle theory" is too simplistic and has not been proven to be ef- fective (47). Effective suicide prevention must rest on the assumption that the target group--those at risk (potentially suicidal adolescents) or those around the person at risk (parents, teachers, peers)--can be reached by, will at- tend to, participate in, and respond to preventive messages. The messages must be informational (e.g., signs and symptoms, cues) and directional (e.g., where to get help); but, also, they must provide skills and incentives to act. How do you get high risk adolescents, those who are acting out, depressed, abusing sub- stances, etc., to pay attention to suicide prevention programming? What have we learned from studying mass media preven- tion efforts that increases the likelihood of having positive impact? Sacco and Silverman have outlined five prin- ciples, inferred from empirical data, for suc- cessful mass media prevention campaigns: (a) information must be readily available to the target audience, (b) communication strategies must be designed to be salient to multiple targets, (c) contradictory informa- tion must be minimized, (d) objectives must be realistic and specific, and (e) desired be- haviors to be pursued by the audience must be made explicit (48). Flay and Sobel (6) suggest that mass media efforts must use multiple sources of informa- tion, extend campaigns over time, and con- vince gatekeepers (e.g., television station managers) of the worth of the campaign, to ensure adequate dissemination. They argue that more persons will pay attention if the message is seen as meeting a salient need (e.g, offers a skill) and is delivered by some- one with whom adolescents identify or on 3-281 Report of the Secretary’s Task Force on Youth Suicide whom they model themselves (e.g., music groups, sports figures). Furthermore, and perhaps most important, they contend that media programs must be both complimented and supplemented by school-based curricula, home/family involvement, or community or- ganization designed to increase interper- sonal communication, discussion and networking. The most successfully designed mass media campaigns that promote changes in health behavior are those that incorporate interper- sonal communication (49). For example, the USC/KABC-TV Smoking Prevention and Cessation Program consisted of five 5- minute news segments, a coordinated 5-day classroom curriculum for junior high stu- dents emphasizing social skills regarding resisting social influence, homework assign- ments requiring adult involvement, a fol- lowup series of five 5-minute news segments, and a written guide provided to all parents. Students involved in the program made sig- nificant gains in smoking cessation and non- initiation (6). To teach specific behavioral skills, the Stan- ford Heart Disease Three Community Study (50) included intensive mass media cam- paigns (PSAs, radio and television features, newspaper articles, bus cards, billboards) and face-to-face clinics for high risk subjects. The program extended over two years and significantly reduced risk for cardiovascular disease. The Crime Prevention Coalition, with the cooperation of the Advertising Council, produced the national "Take a Bite Out of Crime" campaign (41). This program relied heavily on well-produced PSAs designed to induce behavioral change, a coordinated print campaign, and local community projects. Significant changes were ac- complished in six of seven target goals. A similar and impressively coordinated ef- fort, although as yet unevaluated, has been mounted by WQED-TYV in Pittsburgh. "The Chemical People" project (51) involved two PBS television shows, educational print 3-282 resources, and guides. Most importantly, it involved citizen outreach activities which resulted in more than 10,000 town meetings which evolved into continuing task forces to deal with youth drug problems on a com- munity level. RECOMMENDATIONS AND CONCLUSIONS The mass media are not responsible for caus- ing youth suicide nor are they responsible for preventing youth suicide. Yet, as a sig- nificant part of the sociocultural milieu in which our children are raised, they have the potential to profoundly alter the message en- vironment in which children behave. To the extent that publicized news stories about celebrity suicides contribute to the suicide of a youngster predisposed to suicide, concern within the media needs to be raised. Newsworthiness is an appropriate considera- tion in the amount and type of coverage given a news event. Celebrity suicides are newsworthy, but the possibility of imitative suicides as a consequence to their reports suggests some balance needs to be con- sidered. Neither censorship nor prior restraint* are appropriate, but limits may be. Elective guidelines might be established. Consultative discussion between media rep- resentatives and suicidologists might, for ex- ample, achieve some desired balance between the public’s need to know, the media’s right to report, and alternative con- sequences. * There has been but one legal test of prior restraint, that of Weluvs. CPB, No. H-80-1332 (S.D. Tex, Filed 6/16/80). On June 16, 1980 a temporary restraining order precluding defendents from distributing and broadcasting "Choosing Suicide" was denied for lack of federal jurisdiction. The show, scheduled on PBS that date, was a documentary dis- tilled from 19 hours’ recording of the ideas, plans, and dis- cussion by and with Jo Roman regarding her suicide. Roman, age 62, had Stage II breast cancer (80% survival rate), had planned her suicide over 10 years, and died by suicide in June, 1979 by a lethal dose of Seconal. Plaintiff argued that broadcast would cause immediate and ir- reparable harm. Defendants countered that the feared im- pact of the program was "speculative," that there was no definitive evidence that the broadcast would inevitably, directly, and immediately cause harm, and that prior restraint was unconstitutional. A.Berman: Mass Media and Youth Suicide Prevention To the extent that the media may be used to educate, concentrating on prosocial educa- tion in early childhood appears to be the best possible use of the media for prevention. To the extent that children can be influenced by positive models, taught instrumental skills, etc., and to the extent that positive models are presented by the media, there is likely to be some lessening of the multitude of factors that lead to suicidal behavior. Especially im- portant might be the depiction of nonsuicidal solutions to situations of conflict and despair (52). The use of PSAs as vehicles of health promo- tion, information dissemination, and be- havior change needs to be examined closely. If the guidelines discussed above regarding dissemination, targeting, timing, frequency and duration cannot be implemented effec- tively, then, short of government regulation, industry cooperation accomplished through consultative discussion is needed. The em- pirical research strongly supports sup- plemental efforts--the use of schools, home, and community networking--to make a public information campaign effective. With such activities, normative behaviors are es- tablished around the goals of the campaign and peer influences minimize opportunity for non-participation. What is known about the impact of PSAs has come from their use in health promotion campaigns other than that of suicide preven- tion, primarily drug abuse prevention. Whether guidelines derived from these ex- periences are directly applicable to suicide preveniion either awaits further research or requires a leap of faith. While we await fur- ther research, that leap of faith appears worth making. REFERENCES: -++-+eerseseseessesamannannnnes . 1. Roberts DF: Children and commercials: Issues, evidence, interventions. In: Sprafkin J, Swift C, Hees R. Gs) Rx Television: Enhancing the Prevention Impact of . New York: Haworth Press, 1983. 2. Comstock, G: Influence of mass media on child health behavior. Health Ed Quart; 8:32-8, 1981. 3. Pearl D, Bothilet L, Lazar J: Television and be- havior: Ten years of scientific progress and implications for the eighties. Rockville, Maryland: National Institute of Mortal Health, DHHS Publication No. (ADM) 82-1196, 1982. 4. Traub J: The world according to Nielson. Channels 4:26, 1985. 5. Huntley S, Kennedy H: Expert advice: Keep con- io of family fun. U.S. News & World Report, Oct 28:54, 1985. . 6. Flay BR, Sobel JL: The role of mass media in provervhg adolescent substance abuse. In: Glynn TJ, ukefeld CG, Ludford JP. (Eds.) Preventing Adolescent Drug Abuse: Intervention Strategies. Rockville, Maryland: National Institute on Drug Abuse, NIDA research monograph no. 47, 1983. 7. Rubinstein EA: Television and behavior: Research conclusions of the 1982 NIMH report and their policy im- plications. Am Psychol 38:820-5, 1983. 8. Greenberg B: Sroking dngains, and drinking in top rated TV series. J Drug Ed 11:227-33, 1981. 9. Wallack L, Breed W, Cruz J: Alcohol on prime time television: Findings from the fall 1984 season. Berkeley, California: Prevention Research Center, 1985. 10. McEwen W, Hanneman G: The depiction of drug use in television programming. J Drug Ed 4:281-93, 1974. 11. Radecki T: Suicides on television. Champaign, Il- linois: National Coalition on Television Violence, undated. 12. Wurtzel A, Lometti G: Researching television violence. Society Sept/Oct:22-30, 1984. 13. Wurtzel A, Lometti G: Smoking out the critics. Society Sept/Oct: 36-40, 1984. 14. Freedman JL: Effect of television violence on ag- gressiveness. Psych Bull 96:227-46, 1984. 15. Bollen KA, Phillips DP: Imitative suicides: A nation- al study of the effects of television news stories. Am Soc Rev 47:802-9, 1982. 16. Phillips D: The impact of fictional television stories on U.S. adult fatalities: New evidence on the effect of mass media on violence. Am J Soc 87:1340-59, 1982. 17. Kessler RC, Stipp H: The impact of fictional television stories on U.S. fatalities: A replication. Am J Soc 90:151-67, 1984. 18. Phillips D: The influence of suggestion on suicide: Substantive and theoretical implications of the Werther ef- fect. Am Soc Rev 39:340-54, 1974. 19. Weigand HJ: Foreword. In: Goethe JW. The Sor- rows of Young Werther. New York: New American Library of World Literature vii, 1962. 20. Wasserman |: Imitation and suicide: A reexamina- tion of the Werther effect. Am Soc Rev 49:427-36, 1984. 21. Stack S: Effect of media on suicide: Another look. Presented at annual meeting of the American Association of Suicidology, Atlanta, Georgia, 1986. 22. Phillips D: Motor vehicle fatalities increase just after published suicide stories. Science 196:1464-5, 1978. 23. Phillips D: Airplane accident fatalities increase just after stories about murder and suicide. Science 201:748- 50, 1978. 24. Phillips D: Suicide, motor vehicle fatalities, and the mass media: Evidence toward a theory of suggestion. Am J Soc 84:1150-74, 1979. 25. Davidson L, Gould M: Contagion and media. Presented at the National Conference on Risk Factors for Youth Suicide, Secretary's Task Force on Youth Suicide, U.S. Department of Health and Human Services, Bethes- da, Maryland, 1986. 26. Waldon G: Dad blames son's death on TV movie. Gannett Westchester Newspapers pp. Al, A8, Feb. 21, 1985. 27. Archibald L: Turning Jragedy into triumph. In: Cohen-Sandler R. (Ed.) Proceedings of the Eighteenth An- Report of the Secretary's Task Force on Youth Suicide nual Meeting of the American Association of Suicidology, 185-7, 1986. 28. Los Angeles Times, Sec 6:3(col 1), Jan 14, 1986. 29. Gould M, Davidson L: Risk factors for suicide "con- tagion." Presented at annual meeting of the American As- sociation of Suicidology, Atlanta, Georgia, 1986. 30. Gretemeyer J: Community/viewer response to "Surviving." American Broadcasting Company Memoran- dum, Feb 22, 1985. 31. Corry J: "Surviving," movie exploring teen-age suicide. The New York Times Feb 8, 1985. 32. Lometti G, Feig E: Public reaction to "Surviving." New York: American Broadcasting Company, April, 1985. 33. Ostroff RB, Behrends RW, Lee K, Oliphant J: Adolescent suicides modeled after television movie. Am J Psychiatry 142:989, 1985. 34. Gunter, B: Do aggressive people prefer violent television? Bull Br Psychol Soc 36:166-8, 1983. 35. Goodman RI: Selecting public service an- nouncements for television. Pub Rel Rev 7:25-34, 1973. 36. Hanneman GJ: Communicating drug abuse infor- mation Amens college students. Pub Opinion Quart 37:171-91, 1973. 37. Field T, Deitrick S, Hersey JC, Probst JC, Theologus GC: Implementing public education cam- paigns: Lessons from alcohol abuse prevention. Summary report to NIAAA. Washington, D.C.: Kappa Systems, 1983. 38. Hanneman GJ, McEwen WJ: Televised drug abuse appeals: A content analysis. Journ Quart 50:329-33, 1973. 39. Capalaces R, Starr J: The negative message of anti-drug spots: Does it get across? Publ Telecomm Rev 1:64-6, 1973. 40. Hu T, Mitchell ME: Cost effectivenes evaluation of the 1978 media rg abuse prevenion television cam- paign. Final report submitted to Prevention Branch, NIDA, 1981. 41. O'Keefe GJ, Mendelsohn H: "Taking a bite out of crime": The impact of a mass media crime prevention campaign. Was! ington, D.C.: U.S. Department of Justice, National Institute of Justice, 1984. 42. Plant MA, Pirie F, Kreitman N: Evaluation of the Scottish Health Education Unit's 1976 campaign on al- coholism. Soc Psychol 14:11-24, 1979. 43. Kinder BN: Attituded toward alcohol and drug abuse: Il. Experimental data, mass media research, and methodological considerations. Int J Addict 10:1035-54, 1975. 44. Flay BR: Mass media and smoking cessation. Presented at International Communication Association Convention, Chicago, Illinois, 1986. 45. Goldstein HK: Guidelines for drug education through electronic media. J Drug Ed 4:105-110, 1974. 46. Wallack L: Mass media, youth and the prevention of substance abuse: Towards an integrated approach. J Childr in Contemp Soc (in press). 47. DeFluer M, Ball-Rokeach S: Theories of Mass Communication. 3rd ed. New York: David McKay, 1975. 48. Sacco VF, Silverman RA: Crime prevention through mass media: Prospects and problems. J Crim Justice 10:257-69, 1982. 49. Bandy P, President PA: Recent literature on drug abuse prevention and mass media: Focusing on youth, parents, women, and the elderly. J Drug Ed 13:255-71, 1983. 50. Farquhar JW, Maccoby N, Wood P, Alexander J, Brietrose H, Brown B, Haskell W, McAlister A, Meyer A, Nash J, Stern P: Community education for cardiovascular health. Lancet, Jun 4:1192-5, 1977. 3-284 51. Kaiser LK: (Ed.) The Chemical People Book. Pit- tsburgh: QED Enterprises, 1983. 52. Kinzey DA, Bohoutsis JC: Final report on the role of the media in the prevention of violence. Rockville, Maryland: Mental Health Education Branch, NIMH, 1984. Address correspondence to: Dr. Alan L. Berman, Washington Psychological Center, 2139 Wisconsin Avenue, N.-W., Washington, D.C. 20007 INTERVENTION STRATEGIES: ENVIRONMENTAL RISK REDUCTION FOR YOUTH SUICIDE Pamela C. Cantor, Ph.D., Executive Director, National Committee on Youth Suicide Prevention, Norwood, Massachusetts SUMMARY The cost of suicide in terms of mortality, the effects on lives saved and the costs of health care are great. Numerous factors associated with suicide are far-reaching and deeply rooted in the problems of society, family, and in the biochemical problems of the in- dividual. Each suggests a specific set of inter- ventions. The majority of risk for teenagers, however, appears to center on two areas: the cultural pervasiveness of violence, and the negative social factors of neglect and stress. The interventions that would appear to have the greatest impact on youth suicide are decreasing the cultural pervasiveness of violence, limiting the availability of lethal agents such as drugs and alcohol, firearms and medications, and instituting education programs for youth, parents, and the public. What are the theories about the causes or predisposing factors of youth suicide and what interventions could influence the methods of suicide? This paper examines the issues of environ- mental risk reduction as an intervention strategy for reducing suicide. To do this, two issues must be addressed. What predisposes an adolescent to commit suicide? And, what methods do young people use to commit suicide? Regarding the predisposing factors, many theories abound. The factors most frequent- ly cited as related to the rising rate of adoles- cent suicide are: 1. A high level of social and academic com- petition and pressure. 2. Violence that children and adolescents are exposed to--real violence such as rape, murder, and child abuse and created violence on television, in videos, movies, and music. 3. The lack of socially acceptable ways for youngsters to express anger. 4. The lack of connection to religion. 5. The increase in abuse of drugs and alcohol. 6. The special sensitivity of many kids to so- cial isolation. 7. The increase in the absolute number of adolescents in society. 8. The pressures on kids to grow up too quickly. 9. The increasing mobility of the American family. 10. The disappearance of the extended fami- ly, the dissolution of the nuclear family, and the changing role of the family. Which of these causes are amenable to inter- vention? 3-285 Report of the Secretary’s Task Force on Youth Suicide 1. Competition and Pressure: Suicide rates are high in societies where achievement is a major priority and lower where there is less pressure to achieve. In Japan (DeVos, 1968) and in Sweden (Hen- din, 1964), where achievement is important, suicide rates are high. The highest rates among adolescents today are in countries such as Switzerland, Austria, Canada, and the United States (World Health Organization, 1985), all countries with a highly motivated young population. In the United States there is extreme pressure on our young to achieve. Ever increasing numbers of young people obtain college degrees and doctorates (in 1940, 3/100,000 and in 1971, 15/100,000) (U.S. Bureau of Census, 1974). Ours is an achievement- oriented society and children learn it as early as kindergarten. Pressures to achieve academically are felt particularly strongly during the adolescent years. Failure to achieve may be particularly painful to adolescents whose families place heavy emphasis on success. Some adoles- cents may choose to take their own lives rather than to disappoint their parents or see themselves as a failure. In order to combat a high level of competi- tion both socially and academically, we would need to encourage cooperation. This would mean a different ethos of society. I don’t see this happening. Could we impress upon parents the necessity of reducing the pressure on their youngsters? Can we ask parents not to give their children a large number of lessons, not to send them to prestigious schools, not to instill values of competition and achievement? I doubt it. 2. Violence: Our society has become almost immune to violence. The need for social agencies and law enforcement agencies to intervene to reduce the number of murders and violent crimes is imperative but beyond the scope of this paper. Teens are greatly influenced by the media: 3-286 television, rock videos, the movies, and music. Producers will make and air what the public will pay to hear and see. Isuspect the most forceful intervention will come from the buying public. If adolescents and adults refuse to buy products which depict violent behaviors, the industries will stop producing them. I am not hopeful that this will happen rapidly. In July of 1978 it was estimated that the average American child, by his eighteenth birthday, had watched the equivalent of 710 solid days (almost two full years or 17,040 hours) of television. He will have seen 15,000 television murders (McWhirter, 1986). One study reports that the level of television violence in shows specifically designed for children have become increas- ingly violent. In 1967, one hour of cartoons contained three times the number of violent episodes as one hour of adult programming. By 1969, only two years later, violence in children’s television was six times more prevalent (Murray, 1973). It has been suggested that watching violence on television sensitizes the viewer to per- ceive more violence in the world around him and increases the likelihood that the viewer will use violence as a means of resolving con- flicts (Liebert, Neale and Davidson, 1973). The kids who watched the cartoons in the 1960s are today’s adolescents and young adults. Perhaps, we might want to follow the ex- ample of Iceland. Iceland has television-free Thursdays to reduce the disruption in family life. Otherwise transmission is limited to the hours between 8 and 11 p.m. We should check the suicide statistics for Iceland. Violence will be discussed further in the sec- tion on the use of guns as a method of suicide. 3. Aggression: The inability of adolescents to express anger in a socially acceptable way may be contribut- ing to the suicide rate. Insocieties or in situa- tions where there is an acceptable outlet for P.Cantor: Intervention Strategies: Environmental... aggression, the suicide rate appears to be lower. For example, there is a lower rate of suicide in the army during war time. In- dividuals may have the opportunity to dis- charge aggression and hostility toward an actual enemy. It is possible, however, that suicides are hidden under the guise of battle casualties (Yessler, 1968). Young people learn about the limits of ex- pressing aggression in their societies. When they are angry, some kids may be able to risk open expression of aggression against adults. But some have to internalize their feelings. For these adolescents, inward aggression and self- destruction may seem like a reasonable solution for problems which make them angry such as family disruption or school or social failures. It would be healthier if schools and parents would allow for the open expression of anger, frustration, and resentment within socially acceptable limits. Society must move away from encouraging physical punishment for children or any method of intimidating children which blocks their ability to express their own angers. Non-harmful, non-physi- cal, non-combative techniques must be taught to parents and children and adoles- cents. 4. Religion: It is believed that in cultures where the majority of people subscribe to a formal religion, successful suicides are low and where there is no formal religion successful suicides are higher (McAnarney, 1979). Religious attendance in the United States is undergoing change, especially for adoles- cents. While stated church membership has increased at the same rate as the total population, many churches are reporting dif- ficulties in maintaining contact with adoles- cents through teen groups, once popular in the 1950s and 1960s, but no longer popular now. It would be interesting to study whether the adolescent suicide rate is higher among teenagers whose families practiced a religion as compared to those adolescents who were never reared in any religion, or among those who continue to practice their families’ religion as compared to those who do not. As noted by many sociologists, groups in transi- tion experience more suicides than’ stable groups. The presence or absence of a religious belief may not be the important variable, but rather the transition from a religious system to none, which might make an adolescent vulnerable to feelings of guilt and non-belonging. 5. Drug and Alcohol Abuse: The association of drug and alcohol abuse with suicide is well documented and needs no further elaboration here. Drugs and alcohol are used by teens to belong to a peer group, to numb the psychological pain, and to es- cape depression. But drugs and alcohol are depressants. The substances which are taken to alleviate depression cause kids to become more deeply depressed. Whatever can be done by law or social agencies to stop the use of potentially lethal drugs and whatever can be done by schools, parents, and media to educate kids on the dangers of drug and al- cohol abuse will certainly aid in the efforts to curb teen deaths by suicide. 6. Isolation: Social isolation is a factor which appears to contribute to adolescent suicide. We know, for example, that suicide rates are higher in the western part of the country than in the eastern section. This is thought to occur be- cause there are fewer social services in the West and people are more isolated from one another. We know that kids who commit suicide are often loners, withdrawn, and without friends. What can be done to help these lonely kids? We cannot reshape personalities, but we can reshape the places where the kids spend most of their time--their schools. Our elementary, junior high and high schools could have smaller classes. The schools could encourage 3-287 Report of the Secretary's Task Force on Youth Suicide social skills, not just academic skills. We know that college students commit suicide in the fall more often than any other time of year. One theory accounting for this is that the transition from home to college is a difficult one. Kids move from the protec- tion of their families and home town to the anonymity of an unknown town and a big in- stitution. Colleges need to recognize this and make the transition easier. Freshmen need to live in small units and have more in- timate classes where they will get to know their professors and make friends. Large classes could be reserved for upperclassmen where support may be less a matter of life and death. This is probably an issue of economics for colleges and universities, but if mental health counts more than dollars, classes could become smaller. The same is true for our elementary and high schools. We need smaller classes and more teachers. 7. Population: The absolute numbers of teenagers in society appears to be positively correlated with the number of suicides among this age group. The theory is that the more teens there are present in society, the greater the competi- tion for academic honors and employment opportunities and the fewer chances for suc- cess. This appears to be correcting itself with adecreased teen age population predicted by 1990 (Holinger and Offer, 1986). 8. Accelerated Pace: The pressures on kids to grow up too fast and too soon, often tasting the privileges usually reserved for adults such as sex, money, and drugs while they are psychologically still children, can have disastrous consequences (Elkind, 1981). The average age of today’s top fashion models is 12 to 15 (New York Magazine, 1980). The average age of one’s first sexual experience is now 15. This places enormous pressures on kids who are cogni- tively and emotionally still children. Many of our teens have too much material wealth, have nothing to strive for, and are 3-288 bored all the time. They have experienced very little external adversity and yet feel a great deal of internal disconnectedness. Again, I don’t see much hope in reversing this trend. The only solution I can see is to move much of our population to the back woods of Wisconsin. The pressures have to be less and the pleasures simple. This would also help redistribute the population of the U.S. into less populated areas and reduce the crowd- ing in the East. Thus, violent behaviors, such as murders, which are thought to be the result of frustrations and crowding, would be reduced. With the increased population in the West, isolation and the resultant violent behaviors such as suicide would be reduced. Thus, by redistributing the population of the United States we could reduce both the suicide and the homicide statistics. 9. Mobility: On the other hand, redistributing the popula- tion would increase the already high rate of mobility in the United States. Mobility may be a factor in the youthful suicide rate. Again, groups in transition, such as mobile families, have higher suicide rates than those in stable circumstances. Some of the transi- tional members of contemporary society are people living in disorganized portions of big cities, for example, immigrants, and in- dividuals transferred every few years by cor- porations. Studies in Seattle, Minneapolis, and Chicago show extreme concentrations of suicide in the disorganized central sections of the cities (Shneidman and Farberow, 1957). These transitional sections of the cities are characterized by extreme mobility and per- sonal and social disorganization. People be- come isolated and lonely as they move into unfamiliar surroundings. Studies on immigrants produce similar results. The suicide rates for immigrants is substantially higher than the rates in their countries of origin (Bourne, 1973, Burvill, et al,, 1973). We have seen a high incidence of suicide among adolescents in "new" towns such as P.Cantor: Intervention Strategies: Environmental... Plano, Texas in the last few years. It is pos- sible that some of the suicides in the more af- fluent towns such as Scarsdale, New York, may have to do with the families of trans- ferred executives. We would need to look at the background of the families of teens who have committed suicide and consider how transient their histories have been. Some adolescents welcome changes, but others may be frightened by them. For those youngsters who may be uncertain of themsel- ves, reassurances previously provided by a stable home life, a stable religion, and a pre- dictable place to live, may be absent. Rather than face the insecurity of continual changes, some youth may choose to die. 10. Family: The traditional family is disappearing. Half of today’s children will be adolescents in a divorced home. Increasingly, children will be living in single parent families. Even where two parents are present, both parents may be working. Women are entering the work force and men are not electing to stay home and take their place. Families are depending on two incomes. But the question does not appear to be one of divorce or of working parents, but rather one of involvement. American parents spend less time with their children than parents of almost any other na- tion in the world. The attitude of many American parents is that their children are an inconvenience, an impediment to freedom or to success. In societies where family ties are close, the suicide rates are low. Conversely, where families are not close, the suicide rates are high. One study contrasted the suicide rate (all ages) of the city of Edinburgh, Scotland with that of Seattle, Washington. The suicide rate was 15/100,000 in Edinburgh and 20.8 in Seattle (Ripley,1973). The cities were said to be comparable in population composition, colleges and universities, weather, and location on the water. Edin- burgh was characterized as a more tradition- al, less violent society with strong roots in family and school and with a far less mobile society than Seattle. Work by Dizmang (1974) on the Shoshone Indians again suggests that the lack of family stability and a chaotic childhood account for their high rate of suicide. This message is repeated again and again throughout the literature (Finch and Poznanski, 1971; Hen- din, 1964; Toolan, 1968). The model for the traditional nuclear family is changing in twentieth century America. The once stereotypical family--the male head of the household, the female keeper of the hearth, and a home with children emotional- ly and geographically close--is changing for many. The number of divorces is increasing (a 4.1% increase between 1960 and 1965 and an additional 8.8% increment between 1970 and 1973). Exactly what the American fami- ly will be in the future is hard to assess, but there is little doubt that it will not be the traditional family of the 1900s. The changes in the American family have not been shown to be causally related to self- destructive behaviors. The presence of sup- portive family, however, whether living under one roof or two, can help teens pass through the developmental phases of adoles- cence. If adolescents have lost one or both parents and do not have adequate parent substitutes, they may be severely com- promised in their ability to complete this developmental phase without being vul- nerable to impulsive, self-destructive be- haviors. In addition, in families where adults give little time or concern for adolescents, or where parents are not in contact with their children, early symptoms of suicidal behavior may go unheeded. The adolescent who is neglected or unheard may attempt suicide in order to get attention. Occasionally this at- tempt may turn out to be lethal. Thus, the changing American family may compromise the adolescents’ capacity to cope with the stresses of adolescence and may compromise the parents’ ability to recognize their children’s problems before suicide becomes the only alternative to the 3-289 Report of the Secretary’s Task Force on Youth Suicide youngster. While we cannot influence the divorce rate by public interventions, and while we do not wish to encourage parents who are miserable to remain together, there are directions we could take which might be helpful to our children and adolescents. One method is screening programs for early detection of emotional and behavioral problems. Per- haps where parents are not capable of detect- ing such problems, the schools and the community mental health agencies can help. Early detection can mean early intervention. What are the most common methods of youth suicide? What interventions could influence the methods of suicide? How do kids commit suicide? It is well known that guns and medications are the most common methods of suicide. The fre- quency of use of these agents in teen suicides necessitates concentrating on them. Other methods of suicide such as hanging, jumping off bridges, jumping in front of cars and trains are less frequent and less prone to regulation and, therefore, have less prevention poten- tial. Firearms are the leading cause of suicide, and is the one method which has increased sig- nificantly with the increase in suicide rates (Boyd, 1983). Guns now account for more suicides than all other methods combined: 65 percent of teen suicides are committed with firearms. An environmental risk reduc- tion strategy would call for decreasing the availability of handguns. Some 25 million households have handguns and one-half of these keep their handguns loaded (Cantor, 1985). Adolescents are impulsive. Having a loaded handgun around the house is an in- vitation to disaster. Mandatory safety training and public educa- tion on the dangers of handguns in the home would not solve the underlying problems of self-esteem and depression which contribute to suicide, yet it would result in fewer deaths. The analogy with mandatory seatbelts is ap- 3-290 propriate. Seatbelts do not make people bet- ter drivers, but they do improve the chances of surviving a collision. Epidemiologic studies estimate that of those teen suicides committed with a gun, 70 per- cent of those victims could not have obtained handguns or firearms if there had been gun regulations, and some 50 percent of those in- dividuals might have used another method. Thus, it is estimated that the reduction in firearm accessibility would save the lives of approximately 20 percent of our youth (Hol- linger, 1984). Limited gun control, such as mandatory wait- ing period and background check for hand- gun purchasers is one step. Others might include licensing of handgun owners and halting the manufacture and sale of snub- nosed hand guns. The sale of handguns to in- dividuals with a history of psychiatric hospitalizations or previous suicide attempts might be prohibited and regulations might differentiate between handguns and other firearms. While the issue of gun control is a controver- sial political concern, the bulk of the evidence seems to suggest that it would be an effective method of reducing the suicide rate (Westermeyer, 1984; Browning, 1974; Markush and Bartolucci, 1984; Boyd, 1983; Hudgens, 1983; Lester and Murrell, 1980, 1982; Lester, 1983). When highly lethal methods of suicide are less available, evidence shows that people do not necessarily switch to other means. When the English converted their home heating gas from deadly carbon monoxide-containing coke gas to low-lethality natural gas, the suicide rate dropped 33 percent. The low rate has remained constant despite the bleak economic picture in England which might have been expected to lead to an increase in the suicide rate (Kreitman, 1982; Seiden, 1984). While the mere correlation between guns and violent deaths does not indicate causality, it is clear that strict gun control laws in many countries are correlated with a lower P.Cantor: Intervention Strategies: Environmental... incidence of homicide and suicide. For ex- ample, all guns must be registered in the Netherlands; in Italy and Norway, guns are seldom used by the public. In some countries private ownership of a pistol is forbidden to everybody except the police, military person- nel, and a few competitive marksmen. "In Great Britain, most persons, including offi- cials of the British Rifle Association, find it difficult to comprehend the notion of the right to bear arms, as espoused by many per- sons in this country” (Fredericks, 1984). While it is true that other means of commit- ting suicide are used in countries where firearms are not available, little doubt remains that the availability of firearms makes violent acts such as suicide, easier to commit--and the lethality of the act has no peer. If an individual can be deterred from committing suicide, even temporarily, his chances for survival increase, which with gun in hand, would be lost. Poisoning, usually with prescription medicine, is the second most common method of suicide, accounting for 11.3 per- cent of all suicides. The availability of lethal drugs could be limited by restricting the num- ber of tablets permitted for each prescrip- tion. This kind of legislative restriction on sedative and hypnotic drugs is thought to be largely responsible for the decline in the suicide rates in Australia in the 1960s and 1970s (Oliver and Hetzel, 1973). In addition, the tricyclic anti-depressants could be sold with an emetic or antidote. If a teen overdosed and changed his/her mind or was found, an antidote could be given and a life could be saved. Projections claim this method might save approximately 3 percent of teen suicides per year (Holinger, 1985). The first antidote ever developed against Valium and Librium, the drugs most com- monly used in suicide attempts, has under- gone preliminary successful human testing. The antidote, called Anexate, could be used to save hundreds of lives each year (Chicago Tribune, 11/3/85). Jumping from high buildings or bridges is another method of suicide for which inter- vention may be possible. Access to high buildings could be limited, physical bar- ricades such as high glass or steel fences could be required above a specified height and win- dows above a certain story could be made un- openable or unbreakable. However, since the number of suicides among young people which occur by this method are few, the im- pact of the interventions probably would not be great. The causes of youth suicide appear tobe enormously far-reaching and deeply en- meshed in societal problems. The building of barriers on buildings and bridges would not seem to make a dent in the problem. What can we do to reduce these alarming statistics? Perhaps the most important pos- sibility for intervention is to conduct school programs in positive mental health education for students, teachers and parents. These programs must begin with helping children to develop self-esteem, and communication and listening skills. Then kids can learn how to identify a child in trouble and how to reach him. They must know whom to turn to in the school to get help and when to turn to a professional. They need programs in stress management and coping skills as well as programs in suicide prevention. School faculty need to know what to do in the event that someone in the school does attempt or commit suicide to prevent one suicide from becoming multiple suicides. Educational intervention is difficult to evaluate because of the absence of data. Over a period of years, however, education would appear to have the best potential for decreasing self-inflicted mortality. Here I use the analogy of sanitation and its effects on infectious disease. More lives have been saved by preventive sanitation than by an- tibiotics. Based on studies of the effects of public education in the areas of child abuse, discrimination, and drunk driving, self-in- flicted mortality for children and adolescents could be reduced by as much as 20 percent through public education (Holinger, 1984). 3-291 Report of the Secretary's Task Force on Youth Suicide HIGH RISK POPULATIONS Psychiatric patients and suicide attempters The major focus of this paper is on adoles- cents in the general population. However, it is well documented that psychiatric patients and those who have previously attempted suicide have a far greater rate of completed suicide than that of the general population (Pokorny, 1983; Sainsbury, 1982). Thus, they must be considered. Can our knowledge of their high risk be used for prevention? Kreitman (1982) has shown that if we were to screen a large high-risk population we would catch relatively few suicides at high expense. If one concentrates on a smaller high-risk group, the yield is so low that the overall reduction in the total number of suicides is minimal. What other possibilities can be considered? The first is training of psychiatrists, psychologists, physicians, nurses, and other mental health professionals to be increasing- ly aware of the risk of this group. Another fruitful avenue is that of biochemi- cal research. Specific biochemical tests are being developed to establish the potential of patients with biochemical disorders such as decreased serotonin levels for the propensity to impulsive, violent behaviors including suicide (Van Praag, 1982). A third possibility involves the use of the agents of suicide, particularly guns and medications. Stricter regulations of gun sales and medications are needed, especially to previous suicide attempters and persons with histories of psychiatric hospitalization. Juveniles in jails and detention centers Another group of youngsters with a high rate of suicide are adolescents held in jails and juvenile detention centers. One example of practical intervention is the program that the Samaritans of Boston have instituted. Each week, volunteers meet with the inmates of the Charles Street Jail where they have trained thirty inmates to be "barred befrienders" to help identify the suicidal 3-292 among the 5000 new arrivals to the jail each year. Since the program began there have been six suicides at Charles Street, rather than the fifty-six that the National Institute of Justice statistics indicate normally happen in an institution of its size. In addition, of- ficer training in suicide prevention is now re- quired by the State of Massachusetts. How effective this training will be in the preven- tion of youthful suicide should be evaluated. Another area of risk for youth is in juvenile detention centers. Youngsters are brought here, often on a first offense such as a drunk driving charge, to wait for a parent. They are placed in a cell and left alone. Sometimes there is a television monitor for surveillance. I would rather recommend the use of other inmates for surveillance rather than an in- animate object, and further would recom- mend the holding of these youths in cells with another person present rather than alone. Often these kids are humiliated and frightened, and isolation is the last thing we want for them. These procedures are simple to institute and they may help to reduce the death toll for a select high-risk population. REFERENCES 1. Bourne PG: Suicide among the Chinese in San Francisco. Am. J. of Pub. Health 63:744-750, 1973. 2. Boyd JH: The increasing rate of suicide by firearms. New England Journal of Medicine 308:872-874, 1983. 3. Browning CH: Suicide, firearms and public health. Am. J. of Public Health 64:313-317, 1974. 4. Burvill PW, McCall MG, Reid TA, and Stenhouse NS: Methods of suicide of English and Welsh immigrants in Australia. British J. of Psychiatry 123:285-294, 1973. 5. Cantor PC: Testimony before United States Sub- committee on Juvenile Justice. Washington, D.C., 4/30/85 6. Chicago Tribune. 11/3/85; Chicago, Illinois. 7. DeVos GA: Suicide in cross-cultural perspective. In: H.L.P. Resnick's Suicidal Behaviors (Chapter 8) Boston: Little Brown and Co. 105-134, 1968. 8. Dizmang LH, Watson J, May PA, Bopp J: Adoles- cent suicide at an Indian reservation. Amer J. Orthophych. 44:43-49, 1974. 9. Elkind D: The hurried child growing up too fast too soon. Reading, MA: Addison-Wesley, 1981. 10. Finch SM and Poznanski EO: Adolescent suicide. Springfield, lllinois: C.C. Thomas, 1971. 11. Fredricks C: An introduction and overview of youth suicide. In: Peck, M., Farberow, N., and Litman, R. Youth Suicide. Springer Publishing Co., 1985. 12. Hendin H: Suicide and Scandinavia. New York: Grune and Stratton, 1964. P.Cantor: Intervention Strategies: Environmental... 13. Holinger PC: Suicide prevention and intervention. In: Carter Center, Closing the Gap project. Atlanta: Centers for Disease Control, 1984. 14. Holinger and Offer: Suicide, homicide, and acci- dents among adolescents: Trends and potential for prediction. Advances in Adol. Mental Health 1:119-145, 1986. 15. Hudgens RW: Preventing suicide. New England Journal of Medicine 308:897-898, 1983. 16. Kreitman N: How useful is the prediction of suicide following para suicides? Bibliotheca Psychiatrica 162:77- 84, 1982. 17. Lester D: Preventive effect of strict handgun con- trol laws on suicide rates. Am. J. of Psychiatry 140:1259, 1983 18. Lester D, and Murrell, ME: The influence of gun control laws on suicidal behavior. Am. J. of Psychiatry 137:121-122, 1980. 19. Lester D, and Murrell ME: The preventive effect of strict gun control laws on suicide and homicide. Suicide and Life-Threatening Behavior 12:131-140, 1982. 20. Liebert RM, Neale JM, and Davidson E: The early window: Effects of television on children and youth, New York. Pergamon Press, Ltd. p.146-156, 1973. 21. Markush RE and Bartolucci AA: Firearms and suicide in the United States. Am. J. of Public Health 74:123-127, 1984. 22. McAnarney E: Adolescent and young adult suicide in the United States--A reflection of societal unrest? Adolescence 56:765-774, 1979. 23. McWhitter N: Guiness book of world records. New York. Sterling Publ. Co., Inc., 1986. 24. Murray JP: Television and violence. Amercian Psychologist 28:472-478, 1973. 25. New York Magazine, 1980. 26. Oliver RG and Hetzel BS: An analysis of recent trends in suicide rates in Australia. International J. of Epidemiology 2:91-101, 1973. 27. Pokorny AD: Prediction of suicide in psychiatric patients. Archives of General Psychiatry 40:249-257, 1983. 28. Ripley HS: Suicidal behavior in Edinburgh and Seattle. Amer. J. of Psychiatry 130:995-1001, 1973. 29. Sainsbury P: Depression and suicide prevention. Bibliotheca Psychiatria 162:17-32, 1982. 30. Seiden R: Teenage suicide. Address to American Association of Suicidology. Dallas, Texas, 1983. 31. Shneidman ES and Farberow NL: Clues to suicide. New York: McGraw-Hill, 1957. 32. Toolan JM: Suicide in childhood and adolescence, In: H.L.P. Resnick’s Suicidal behaviors (Chapter 16). Bos- ton: Little Brown and Co., p.220-228, 1968. 33. U.S. Bureau of Census. Statistical abstracts of the United States (95th edition). Washington, D.C., 1974. 34. Van Praag HM: Biochemical and peyshopatisiegjesi predictors of suicidality. Bibliotheca sychiatrica 162:42-60, 1982. 35. Westermeyer J: Firearms, legislation, and suicide prevention. Am. J. of Public Health 74:108, 1984. 36. World Health Organization. Manual of the interna- tional statistical classification of diseases, injuries, and causes of death. Geneva, Switzerland, 1985. 37. Yessler PG: Suicide in the milan, In: H.L.P. Resnick's Suicidal Behaviors (Chapter 18). Boston: Little Brown, and Co., p.241-255, 1968. 3-293 SCHOOL-BASED PREVENTION PROGRAMS Barry D. Garfinkel, M.D., F.R.C.P.(C), Director, Division of Child and Adolescent Psychiatry, University of Minnesota Medical School, Minneapolis, Minnesota INTRODUCTION As the statistics on youthful suicide rise dramatically, prevention of youth suicide has become a priority of professionals involved with community-based facilities. Suicide prevention centers, crisis hotlines, family physicians, and teachers are just some of the different community resources available to help the suicidal individual. The efficacy of these various individuals and programs has, however, been subject to question. Con- troversy still continues about whether suicide prevention centers and agencies, psychologists and psychotherapists can ac- tually prevent suicide (Bagley, 1968; Jen- nings, Barraclough and Moss, 1978; Innes, 1980). Other studies, specifically those ex- amining the role of suicide prevention centers, similarly have found conflicting results (Weiner, 1968; Lester, 1974; Miller, Coombs, Leeper, Barton, 1984). Overall, however, suicide prevention centers may have some beneficial effects especially in diminishing the rate of suicide in young white females (i.e., women and girls 24 years of age or younger). Recently, suicide prevention programs and curricula have been developed for junior and senior high schools. These programs have assumed various tasks but have as their over- all goal the prevention of suicide by the stu- dents enrolled in school. All of the current programs utilize some of the ideas of a com- prehensive school-based prevention program. Unfortunately, serious omissions occur in many of these programs. In review- 3-294 ing this field, it is apparent that nine aspects of prevention programs must be developed and integrated into the different school- based programs. They are: 1. Early identification and screening. 2. Comprehensive and thorough evaluation of the depressed, suicidal, and psychiatri- cally disturbed young person. 3. Crisis intervention and case management. 4. Programs immediately following a suicide. 5. Education for students, teachers, com- munity, and professionals on identifica- tion, diagnosis, and management of suicidal youth. 6. Monitoring and followup. 7. Community linkage and networking. 8. Research of epidemiology, causation, and longitudinal followup of attempters. 9. Advocacy. This paper reviews the existing youth suicide prevention programs developed for schools and elaborates on some of the more critical aspects of the nine components of a sys- tematic protocol that would comprise an ef- fective prevention program. The general purpose of a successful school- based program should be to integrate an un- derstanding of the risk factors for youth suicide, an appreciation of the behavioral characteristics and clinical symptomatology of the suicidal individual, and an awareness B.Garfinkel: School-based Prevention Programs of the various psychosocial stressors with which the suicidal adolescent is attempting to cope. The synthesis of these three functions forms the basis of suicide curricula for schools. This review of existing youth suicide preven- tion curricula was undertaken by the Greater Lakes Mental Health Foundation in Tacoma, Washington. The review commit- tee consisted of educators, physicians, psychologists and community workers. A review of 19 programs revealed that only five curricula received an overall positive rating. Some programs had specific deficiencies when rated on a four point scale examining four aspects of the prevention curriculum. The four areas critiqued were: coping skills, prevention, intervention, and postvention. EARLY IDENTIFICATION AND SCREENING The question of "what do we screen for?" arises in training educators and school ad- ministrative staff to screen for suicidal youth. Does the screening include an examination for depression, antisocial behaviors and at- titudes, impulsiveness, suicidal intent, hope- lessness, coping skills, family background, and psychosocial stressors? Should evalua- tion include all these areas? Three general areas of early identification and screening that must be examined are: 1. Identification of depression in young people, 2. Appreciation of the various psychosocial stressors affecting students, and 3. Methods of responding to and handling difficult problems. Early identification is dependent on observ- ing the various risk factors associated with at- tempted and completed suicide in youth. Risk factors identified by Hawton (1982) and Garfinkel, Froese, and Hood (1982) are: 1. older adolescent, 2. male, 3. previous attempts, 4. chemical dependency (alcoholism) in the family, 5. family breakdown, 6. deteriorating school performance, 7. recent antisocial acts (characterized by rage, aggression, and impulsiveness), 8. living away from the family, 9. history of depression. In a 1986 study of suicide attempts in high school students, Garfinkel et al. (1986) showed that attempters had more than twice the number of psychosocial stressors within a six-month period prior to their attempts than normal adolescents. The various stress- ful events were not just more frequent, but were qualitatively distinct from the stressful events in non-attempters. For example, family breakdown, divorce, and school-based difficulties were far more frequent in the attempter’s group. Common stressors experienced by adoles- cents who attempt suicide include: . Breakup with boyfriend or girlfriend. . Trouble with brother or sister. . Change in parents’ financial status. . Parental divorce. . Trouble with teacher. . Changing to a new school. 1 2 3 4 5. Losing a close friend. 6 7 8. Personal injury or other physical illness. 9 . Failing grades. 10Increased arguments with parents. Although more than fifty events could be identified as upsetting to an adolescent, these ten were reported most often by the at- tempters. The stressful life events are ranked in order of frequency that adoles- cents attempting suicide identified as most to least stressful of the ten most troubling events (Garfinkel, et al., 1986). 3-295 Report of the Secretary’s Task Force on Youth Suicide Suicide attempters in a high school setting demonstrate not only depression, but also behaviors characterized by: 1. Angry and explosive outbursts. 2. Passive withdrawal into drinking, smoking, and drug usage. 3. Avoidant types of behavior including hy- persomnia, joyriding, and infrequent com- munication with adults. 4. Recent antisocial behaviors such as fight- ing, violent outbursts, stealing, and van- dalism. 5. Visits to family doctors concerning depres- sion. 6. Deteriorating school work. Recently, Garfinkel, Hoberman, Parsons, and Walker screened 4,267 junior and senior high school students in rural Minnesota. They examined five symptom and behavioral areas: * Depression * Antisocial behavior * Life stressors ° Familial and demographic factors * Coping and adaptive strategies Hopelessness and nihilism were not evaluated. Information concerning suicide attempts, ideation, and impact of role models were also assessed and shown to be important in determining suicide attempts. Instrumen- tation for this type of screening and early identification is relatively simple and brief. Self-report questionnaires and rating scales were the most efficient way of obtaining this information. The Beck Depression Inven- tory; Berilson Rating Scale for Depression; A-COPE, Johnson and McCutchen Life Events Questionnaire, demographic, familial, and antisocial questions were shown to have validity. By applying this information to specific students, it is possible to identify an individual undergoing severe stress. A pattern emerges that often resembles that of 3-296 a suicidal adolescent who has a sufficient number of preexisting risk factors to warrant a further comprehensive evaluation. COMPREHENSIVE EVALUATION Following the screening measures, some stu- dents will be identified as needing further evaluation. The comprehensive assessment and evaluation of a student attending a junior or senior high school must be based on a structured systematic psychiatric protocol. This should include a structured psychiatric diagnostic interview, self and clinician ratings, and psychometric testing. The evaluation must utilize existing instruments that have a high degree of validity and reliability. Because depression reflects a temporary and episodic state, most rating scales and instruments for evaluating depres- sion do not have a high test/retest reliability. There are, however, a number of structured psychiatric interviews that are effective in identifying depression, including: ¢ Kiddie SADS e DICA, (individual and parent) e DISC, (Diagnostic Interview Schedule for Children) (individual and parent) Clinician ratings such as the Children’s Depression Rating Scale by Poznanski demonstrate good psychometric properties. In addition, self-report ratings such as the Children Depression Inventory (CDI) (Kovacs, 1978, Birelson, 1978) and the Beck Depression Inventory, (Beck, 1979) are worthwhile as well as parental ratings for depression. Psychological tests such as the Personality Inventory for Children (PIC), the Minnesota Multiphasic Personality In- ventory (MMPI), and the Million Personality Profile for Adolescents can be usefully ap- plied to the evaluation of suicidal youth. Oftenin aschoolsetting, the evaluation team must obtain both parental and teacher evaluations of the student and examine the difficulties being identified in as many set- B.Garfinkel: School-based Prevention Programs tings as possible. Weinberg (1973) showed that between 40-60 percent of all learning disabled children met diagnostic criteria for depression. It is also important to measure hopelessness and suicidal intent. Suicidal intentions are measurable with instruments such as the Beck (1974) or Pierce Scales (1981). The Beck Hopelessness Scale has been shown to be avery good predictor of completed suicide in individuals who have made a previous at- tempt (Beck et al., 1985). Evaluation should include a thorough assess- ment of an individual who had already made one suicide attempt. A number of charac- teristics of the attempt indicate how serious the attempt was and the probability of an individuals’ ultimately committing suicide. The following characteristics, originally sug- gested by Beck (1974), Hawton and Catalan (1982), and Garfinkel, Froese, and Hood (1982), are good predictors of the serious- ness of a previous attempt. 1. whether others were near by 2. the likelihood of being rescued . precautions taken to avoid discovery 3 4. actions that indicate that death was likely (e.g., giving away one’s most prized posses- sions) Wn . intricate and extensive suicidal plans 6. leaving a suicide note 7. not telling others of the attempt following the self-destructive actions 8. informing others of the attempt before it actually occurs 9. family history of suicide CRISIS INTERVENTION-- A MODEL Barteolucci and Drayer (1973) and Hawton and Catalan (1982) recommend a crisis inter- vention model based on brief, collaborative problem-solving therapy emphasizing the rapid resumption of control over one’s en- vironmental future. Various personnel in- side and outside the school system may be ef- fective in working with crisis intervention teams based within schools. The following individuals may be asked to provide a con- sultation or act as a liaison: ¢ child and adolescent psychiatrist school psychologist ® nurse ® social worker e teacher e principal ¢ speech pathologist ® occupational therapist ® coach e audiologist e pediatric neurologist * clergy Depending on the unique aspects of a case, these diverse individuals may be asked to consult or become a permanent member of the suicide prevention team. The role of psychotropic medication should be examined critically because it may be useful in treating an immediate crisis (Hawton and Catalan, 1982). In general, the purpose of the suicide team is to transform, for the depressed and suicidal adolescent, an environment that, until the crisis, had only emphasized academics and athletics. The team should provide a very supportive, concerned, and empathetic group of individuals prepared to work with the individual to alleviate psychological and social stress. One of the chief responsibilities of school- based suicide prevention teams should be to insure that an adolescent receives the neces- sary psychotherapy and social work following asuicide attempt. It is important to establish immediately an integrated network of parents, community-based professionals, and school-based educators and counselors. Often the early crisis intervention work done with the school will determine how success- ful subsequent community-based counseling 3-297 Report of the Secretary's Task Force on Youth Suicide will be. If the rapport and therapeutic al- liance established with the school-based per- sonnel is effective, it is likely that community work will also replicate that positive pattern. GOALS OF INTERVENTION AFTER A SUICIDE As outlined by Hawton (1982) and Beck, Schuyller, and Herman (1974), a number of specific goals should be set and accomplished during the crisis-intervention work with an adolescent who has attempted suicide. Catalan et al. (1980) and Hawton and Gaths (1979) demonstrated that all disciplines, in- cluding social workers, counselors, psychologists, nurses, and psychiatrists can assess the individual who has attempted suicide. They clearly indicate that the diag- nostic assessment of every adolescent at- tempter need not involve only one discipline. The goals of the assessment are as follows: 1. Establish a therapeutic alliance. 2. Determine the type of psychosocial stres- sors the individual had been experiencing. 3. Rule out the presence or absence of a psychiatric disorder. 4. Identify the adaptive and coping mechanisms that the individual uses to manage stress. 5. Determine all the external resources and support personnel and systems within the individual’s life that can be called on to help. 6. Identify what further help the person is willing to accept in order to stop the suicidal behavior. The first step of this protocol is to determine all the events that immediately preceeded the attempt. In general, a thorough history of the preceeding 48 hours is essential. Events during the two days are reviewed with the individual to determine whether or not some could be viewed as a precipitant, or per- ceived by the individual as a reason for self- destructive behavior. If no psychosocial stressors are identified and the suicide at- 3-298 tempt appears to have no reasonable event causing it, then it becomes very important that a psychiatric disorder be ruled out. Most often, when no psychosocial stressors or ob- vious reasons for the self-destructive actions are apparent, the suicide attempt is a direct outgrowth of a psychiatric disorder. An evaluation of the degree of hopelessness, suicidal risk, and lethality of the actions must occur next. The circumstances of the at- tempt indicate how lethal and serious the be- havior was. Garfinkel, Froese, and Hood (1982) and Beck et al. (1974) described fea- tures of attempted suicides that are impor- tant to identify: e Was the attempt made in isolation? ¢ Were others nearby and likely and able to intervene? This is necessary to find out because efforts to avoid discovery or to conceal the attempt indicate level of severity. * Were there actions that showed the per- son anticipated death? Giving away one’s most valued possessions is informa- tion about severity often provided by friends. e Was a suicide note left? e What was the extent of premeditation and planning? e What method of self-destruction was chosen? Jumping, hanging, and drown- ing are more lethal and serious than drug overdoses and wrist laceration. Following an attempt, it is useful to have the adolescent list with the person doing the evaluation, all his/her difficulties. In order of frequency, identifiable problems emerge and include the following: e problems with boyfriend or girlfriend ¢ problems with parents. ¢ problems of a non-specific nature within the family. ¢ problems that are school-based. B.Garfinkel: School-based Prevention Programs Hawton, O’Grady et al. (1982) demonstrated a similar list of problem areas in 50 adoles- cents who had attempted suicide, including parental, school, peer, social, physical, sexual, and alcohol based difficulties. Itis im- portant to note that such difficulties may, in fact, be a consequence of depression rather than a cause. They may also perpetuate a person’s pessimistic view of himself and his future. Moreover, they may also have precipitated the crisis or the decision to end one’s life. After identifying existing problems and psychosocial stressors, the next step involves the identification of a possible psychiatric disorder. Identifying cognitive problems is as important as identifying affective disorders. For children and adolescents it is known that individuals with learning disabilities, brain dysfunctions, and severe depressions may have marked cognitive changes. It is also known that individuals with depression have cognitive distortions and altered attributes that affect thought content. At this step in- dividuals assessing the patient can do a full mental status examination, emphasizing an evaluation of both mood and thought disor- ders. The next component of the evaluation em- phasizes identifying family psychopathology, family dynamics, and external resources available to the suicide attempter. Research has indicated that significant psychopathol- ogy in family members, especially alcoholism and family breakdown, are frequently as- sociated with adolescent suicide attempts. General psychopathology in other family members should be documented noting whether another family member has tried to commit suicide. Suicide attempts and com- pleted suicide in other family members are associated with suicidal behavior in adoles- cents. Dynamic issues are also important, especially in determining the help young per- sons can expect from individuals in the im- mediate environment. Because family members may have coped with severe problems for a long period of time, they may not be accommodating to the adolescent’s present crisis. They may take a very reject- ing, uncooperative attitude. They may also treat the child as if he/she were "expendable," i.e., they are so tired of the long-standing con- flicts that they have "given up" on the person’s ever getting better. Without suffi- cient resources available to the attempter, closer observation such as day hospital and full hospitalization may be necessary. The adaptive and coping mechanisms of the attempter should be explored. It is impor- tant to determine whether the attempter is showing the commonly observed behavior of individuals who attempt suicide: passive withdrawal, avoidant behavior, irritable and angry responses, and impulsive/explosive an- tisocial actions. At this stage of evaluation, previous suicide attempts, ruminations, and plans should be identified. Finally, all sup- portive relationships must be identified; they can include peers, family members, clergy, educators, and professionals. It is important that the evaluator not keep the suicidal be- havior strictly confidential. If intentions are kept secret, adolescents with suicidal idea- tion may not be adequately monitored or managed in all settings, and could result in a fatality. The last step is to establish a contract. The individual who has attempted suicide agrees to work on specific problems with identified external resources without turning to suicide during a specified period of time, usually three to four months. Often, three to four months is necessary to allow for sufficient mood elevation, cognitive reorganization with new attributes, and the resolution of various psychosocial stressors to deter the in- dividual from further suicidal behavior. SURVIVING A CHILD OR ADOLESCENT SUICIDE Following a suicide, educators and com- munity workers have difficulty managing and counseling relatives, peers, and classmates of the individual who committed suicide. A number of principles should be adhered to within the school setting that would deem- 3-299 Report of the Secretary's Task Force on Youth Suicide phasize the social learning and role modeling that can occur following a suicide. All sub- sequent actions should be handled in a very sensitive manner deemphasizing, but ac- knowledging the presence of guilt, respon- sibility and anger. Only two basic principles underpin this type of work; first, preventing social modeling from occurring, and second, preventing negative feelings of guilt, respon- sibility, and anger from overwhelming the Survivors. A number of tasks should be carried out in the school that are directed to both school friends as well as the general student body. Following an attempt or completed suicide, approximately one-third of the student body will have already heard about it indirectly. Educators therefore, should not think that by discussing what has occurred, they are giving young people the idea that suicide is an op- tion. The students can be encouraged to ex- plore with adults the sense of loss and abandonment regarding the suicide. Herzog and Resnik (1967) indicated that both the parents and peers may have difficulty in com- municating openly about the individual who committed suicide and may need support, direct encouragement, and time set aside to discuss it. Individuals working with peers of the suicide should attempt to stress the psychopathology that the individual was experiencing. Adolescents often believe the individual who committed suicide did not have any problems whatsoever. Occasionally this is true, however, more often the psychopathol- ogy was minimized or not readily apparent. Stressing the psychopathological elements in the individual’s functioning demystifies the suicide, emphasizes emotional and mental disturbances, and makes it more difficult to identify with the dead individual. A major component in working with the survivors is to break down the identification with the in- dividual who committed suicide. Emphasiz- ing family and other problems unique to the individual and deemphasizing the strengths the individual possessed are methods by which identification can be diminished. Em- 3-300 phasizing psychosocial stressors such as academic difficulties, breakup of peer relationships, and physical issues that were stressful events in the individual’s life, places the suicide in the context of a unique set of circumstances. By sharing the unique aspects, other students have a more difficult time in identifying with the decedent. Fur- thermore, professionals and educators work- ‘ing with the students should deemphasize suicide as the cause of death. In fact, some professionals do not mention suicide at all but emphasize depression and various other stressors as leading to the individual’s suicide. Peers, classmates and family members should be encouraged to limit the extent of memorializing the decedent. Although not fully researched in a controlled fashion, some empirical evidence suggests that individuals who memorialize a suicide do so in a much more elaborate fashion compared to a similar age person dying from other causes. Finally, coming to terms with the loss and abandonment must be explored. Survivors have great difficulty in comprehending suicide as permanent and volitional. These two particular aspects (i.e. volitional and per- manent) of the decedent’s actions should be discussed. In general, ventilation of anger with the decedent ("he was no different from the survivors"), responsibility and guilt, and non-specific communication are not the most effective ways of trying to limit the social modeling and the potential for clustering of suicides. EDUCATION FOR STUDENTS AND SCHOOL PERSONNEL Educational programs to develop sensitivity and awareness of the issues of youth suicide abound for students, teachers, and ad- ministrative school personnel. Educational programs vary in duration, content, and per- sonnel. The major components of various programs are four-fold and usually include coping skills, prevention, intervention, and postvention. Some programs have been as brief as one B.Garfinkel: School-based Prevention Programs class period and others have grown to as many as 3 to 5 classes. No one has been able to demonstrate that programs directed to students have any direct benefit. Similarly, no evidence suggests that programs for school personnel are effective. It is possible that general discussions about suicide may have a deleterious effect on the students in that the topic may inadvertantly become idealized and appealing. STUDY OF EDUCATIONAL PROGRAMS IN MINNESOTA Garfinkel, Hoberman, Walker, and Parsons examined suicide educational programs in rural Minnesota high schools which were directed to either students or educational personnel. They wished to determine if educational programs correlated with either the suicide attempt rate for that particular school or the occurrence of severe depres- sion. The ten schools studied had six suicide attempts within the six months prior to the beginning of the study. Beck Depression In- ventories were completed on 200 students designated for this study. Students in grades 9, 10, 11, and 12 in a particular class were asked to fill out the Beck Depression Rating Scale, provide other information such as suicide information, demographic data, and coping and life stress events schedules. Prin- cipals were interviewed to determine the number of educational programs on suicide, depression, or stress that the school provided. They were also asked whether specific personnel within the school were designated to give these educational programs and whether or not experts were brought in. The range of programs varied from 0-4. Whether educational programs were provided did not correlate with either the suicide attempt rate in a particular school or with the occurrence or severity of depression recorded. The type of speaker, (by dis- cipline, or an outside expert, or someone on faculty) also did not have an impact on the rate of suicide or depression. Similarly, the number of suicide education programs or number of staff persons did not have a sig- nificant effect on these two variables. This is a pilot study that will be replicated in forty more schools. The programs may be criticized on the basis that the philosophy or approach was not known. It is encouraging to note that in spite of media attention to suicide and the social learning and modeling that can occur when suicide is discussed generally, these programs did not have an en- hancing effect on the suicide and depression rate within the schools that we studied. Therefore, one can cautiously conclude from these preliminary findings that a nonstandar- dized educational program within a school setting directed to students and teachers is not associated with an increase in suicide at- tempts or severe depression within a par- ticular high school. When the Greater Lakes Mental Health Foundation critically reviewed youth suicide prevention curricula, they identified four components of the student and teacher cur- ricula: coping skills, prevention, interven- tion, and postvention. Nevertheless, the programs that were critically reviewed had multiple omissions from the reviewer’s perspective. The curriculum that received the highest rating was developed by Thomas C. Barrett, Ed. entitled, "Youth in Crisis, Seeking Solu- tions to Self-Destructive Behavior". It con- sists of 255 pages dealing with prevention and intervention models for the school and com- munity. A five-lesson curriculum is provided along with exercise and resource material ac- companying the lessons. My review of this material is more critical and I have identified a number of shortcomings: 1. The overall model is vague and is overly sociological and anthropological with a deemphasis on mental disorders as a cause of youth suicide. 2. The bibliography is far too brief and not current. 3. The mental health team does not include psychiatry. 3-301 Report of the Secretary’s Task Force on Youth Suicide 4. The student curriculum spends most of its time examining the issues and circumstan- ces surrounding suicide itself, with a deem- phasis on depression. 5S. The model of suicide emphasizes societal factors and does not integrate or em- phasize psychology. Herbert of the Fairfax County, Virginia Public Schools has produced "A Guide to Adolescent Suicide Prevention Programs Within the School". This 55-page booklet summarizes the major characteristics of suicide educational programs based within schools and communities. It has a balanced perspective that identifies depression as being a significant factor in youth suicide. It has not been demonstrated, at this time, whether these particular programs have an ameliorative effect on suicide and depression within school settings. It also has not been shown whether these types of programs are inadvertantly providing role models to al- ready depressed individuals and guiding more people to attempt suicide. Many of the educational models for students emphasize more effective adaptation, coping skills, and communication among teenagers. The program that has developed these educational areas most systematically is the Suicide Prevention Center in Dayton, Ohio. They have produced five separate teacher manuals that deal with various aspects of depression and suicide. The most useful of the manuals emphasizes stress management, i.e., coping skills, instead of focusing on the topic of suicide. It is not known whether it is more helpful to discuss depression, coping mechanisms, or more effective communica- tion than to avoid bringing suicide to the classroom directly to the students attention. Most educational programs encourage the development of early self recognition of depression. Self-identification and iden- tification of depression in one’s peers appear to be worthwhile skills to teach students. Emphasizing depression rather than suicide appears to be an effective way of focusing on a phenomenon associated with suicide. Ex- 3-302 amining ineffective coping styles, as well as more effective coping mechanisms is a very useful and practical educational strategy for students. Learning to deemphasize passive withdrawal, avoidant types of behaviors, al- cohol and drug use, anger, and antisocial be- haviors is important. Emphasizing networking and the integration of adult guidance into the youth’s support system are worthwhile skills with which to train and edu- cate our high school students. Reinforcing assertive and clear communication is also a worthy goal of educational programs. Deem- phasizing the rageful forms of communica- tion and the indirect methods that are commonly seen in suicidal youth are also being brought to the attention of the stu- dents. At this time, it is not known whether educational programs emphasizing these is- sues will alter either coping or communica- tion in depressed and suicidal individuals. COMMUNITY LINKAGE AND NETWORKING School suicide prevention teams can become a community link to other school districts, high schools, community mental health centers, hospitals, universities, churches, and private mental health practitioners. The legal system, including truant officers, proba- tion officers, and community police officers frequently can be included. Community net- working includes determining what com- munity-wide educational programs are available. Networking deals with the media and guides them to deemphasize the coverage of suicides when they occur and to establish followup, aftercare management and treatment networks to serve young people after a crisis or suicide attempt. Often the suicide prevention team within the school must interact with parents, siblings, children, and adolescents who are especially vulnerable to depression and suicide. The prevention team should work with youth in health, recreational, and social areas and in the exchange of expertise among all groups addressing this problem. The linkage serves as a prevention, crisis intervention, and re- B.Garfinkel: School-based Prevention Programs search base for the enhancement of scientific and clinical knowledge about youth suicide. RESEARCH Applying the model outlined above, the University of Minnesota Division of Child and Adolescent Psychiatry and the Agricul- tural Extension Program in the Department of Home Economics, in association with the 4-H Clubs of Minnesota, established a broad community-based study of youth suicide at- tempts in rural Minnesota. The study in- cluded 52 counties in three regions of rural Minnesota. It surveyed 82 schools and in- volved 65 agricultural extension agents. Over 4,267 students were surveyed and their information provided data on: e demographic characteristics of suicide attempters. e the prevalence of suicide attempts in rural Minnesota youth. e stressful life events. ¢ coping and adaptive skills. e a self-rating scale for depression. ¢ an inventory of antisocial behaviors. Research of this type was only possible with the help of an extensive network of profes- sionals working with the schools in collabora- tion with community resources. It utilized existing networks and teams within and out- side our junior and senior high schools. Other research taking place within schools that has examined affective disorders in suicide included a study of the prevalence of Seasonal Affective Disorders in high school students. MONITORING AND FOLLOWUP Because of the unique function of schools that keep all children below the age of 16 in school eight hours a day, five days a week, stu- dents in a junior high school and the early grades of senior high school are much more available for monitoring and followup pur- poses. Unlike adults who have made a serious suicide attempt or gesture, and whose diverse vocational and social functioning make it difficult to determine their com- pliance with recommended management programs, children and adolescents can be followed within the schools by a suicide prevention team and the school team can monitor community-based treatment programs. Because individuals who attempt suicides are at high risk for ultimately com- pleting suicide or making multiple attempts, it is imperative to monitor these individuals closely. Monitoring and supervising the progress of individuals who have been iden- tified as individuals at risk or are involved in self-destructive behavior, can be a major function of the school-based team. Similarly this team is in a unique position to monitor community-wide trends regarding suicide, community education efforts, media ex- ploitation of youth suicide, recent advances insuicide research, and specific school trends over a number of years. Suicide in a par- ticular school can be examined as a function of unique local situations and events. CONCLUSIONS: THE YOUNG PERSON’S ADVOCATE The development of suicide prevention programs and teams within our junior and senior high schools results in a number of programs being developed. These include: ¢ Early identification ¢ Comprehensive evaluation e Crisis intervention ¢ Postvention e Education Monitoring Community linkage Research Providing these different functions can be ac- complished while the students maintain their routine and regular activities. The school 3-303 Report of the Secretary's Task Force on Youth Suicide program becomes a resource not only for the school and district, but for the entire com- munity. It becomes a clearinghouse for new research, comprehensive management tech- niques, and the coordination of community education efforts in the area of suicide. Moreover, as this team monitors, follows-up, and establishes a registry of individuals at highest risk for suicide, it can become the children’s most effective advocates. School, peer and parental attitudes may be insensi- tive and unaware of the depressed and suicidal individuals who lack the energy, so- cial skills, and abilities to deal effectively with the usual adolescent developmental demands. Having advocates within the school setting can provide immediate response to difficulties at teacher, peer, classmate, and parental levels. Explaining to both teachers and parents that a suicidal and depressed individual may not be able to con- centrate and complete homework and, there- fore, punitative actions about incomplete assignments, lack of energy, and excessive daydreaming may be harmful to an already depressed individual. Explaining physical ailments and somatic symptoms may also be very helpful. “As the young persons’ advo- cate, the prevention team ultimately can reframe behavioral, physical, and social problems from one perspective to another. Rather than observing antisocial behaviors entirely within a conduct disorder: framework, one can also present them as adaptive behaviors commonly seen in depressed and suicidal individuals. The most important benefit of all is that the suicidal student will have an advocate readi- ly available eight hours a day who will be knowledgeable about the thoughts and feel- ings the student is experiencing, able to inter- pret the individuals’ behavior towards others more effectively, and will accomplish these tasks in an empathetic fashion to the student. It is estimated between 3-6 percent of all high school students require the direct ser- vices of a suicide prevention team. Not only do at-risk students, but all students, teachers, and members of the community benefit from the diverse activities of this team. 3-304 REFERENCES 1. Bagley, C.R. (1968). The evaluation of a suicide prevention scheme by an ecological method. Social Science and Medicine, 2, 1-14, 2. Barraclough, B., Shephers, D., & Jennings, C. (1977). Do newspaper reports of coroner's inquests incite people to commit suicide? British Journal of Psychiatry, 19, 523-527. 3. Beck, A. T., Beck, R., & Kovacs, M. (1975). Clas- sification of suicidal behavior: |. Quantifying intent and medical lethality. American Journal of Psychiatry, 132, 285-287. 4. Beck, A. T., Schuyler, R.D., & Herman, J. (1974). Development of suicidal intent scales. In: A. T. Beck, H.L.P. Resnick, & D. J. Lettieri (Eds.), The Prediction of Suicide. lllinois: Charles Press. 5. Beck, A. T., Ward, C.H., & Mendelson, M. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561-571. 6. Catalan, J., Marsack, P., Hawton, K.E., Whitwell, D., Fagg, J., & Bancroft, J.H.J. (1980). Comparison of doc- tors and nurses in the assessment of deliberate self- poisoning patients. Psychological Medicine, 10, 483-491. 7. Connell, P.H. (1965). Suicidal attempts in childhood and adolescence. In J.G. Howell (Ed) Modern i in Child Psychiatry. Edinburg liver and Boyd. 8. Garfinkel, B.D., Froese, A., & Golombek, H. (1979). Suicidal behavior in a pediatric population. In: Proceed- ings of the 10th International Congress for Suicide Preven- tion and Crisis Intervention, 305.912. 9. Garfinkel, B.D., & Golombek, H. (1983). Suicidal behavior in adolescence. In: H. Golombek & B.D. Gar- finkel (Eds.), The Adolescent and Mood Disturbance. New York: International University Press. 10. Guze, S., & Robins, E. (1970. Suicide and primary affective disorders. British Journal of Psychiatry, 117,437- 438. 11. Hawton, K., & Catalan, J. (1982). Attempted Suicide: A Practical Guide to Its Nature and Management. Oxford: Oxford University Press. 12. Hawton, K., Cole, D., O'Grady, J., & Osborn, M. (1982). Motivational aspects of deliberate self-poisoning in adolescents. British Journal of Psychiatry, 141, 286- 291. - : 13. Hawton, K., O'Grady, J., Osborn, M., & Cole, D. 1982). Classification of adolescents who take overdoses. ritish Journal of Psychiatry, 140, 124-131. 14. Innes, J.M. (1980). Suicide and the Samaritans. Lancet, |, 1138-1139. 15. Paykel, E.S., Prusoff, B.A., & Myers, J.K. (1975). Suicide attempts and recent life attempts: A controlled comparison. Archives of General Psychiatry, 32, 327-333. 16. Sainsbury, P., & Barraclough, B. (1968). Differen- ces between suicide rates. Nature, 220, 1252. 17. Shaffer, D., & Fisher, P. (1981). The epidemiology of suicide in children and young adolescents. Journal of the American Academy of Child Psychiatry, 20, 545-565. 18. Weissman, M.M. (1974). The epidemiology of suicide attempts, 1960 to 1971. Archives of General Psychiatry, 30, 737-746. “U.S. GPO: 1989-239-155 RN nba aE eae apm Owup 49 warren Hail 642-2511 LOAN PFRIOD 1 I2 J3 DEPARTMENT OF HEALTH & HUMAN SERVICES Public Health Service Alcohol, Drug Abuse, and Mental Health Administration Rockville MD 20857 DHHS Publication No. (ADM)89-1623 4a Alcohol, Drug Abuse, and Mental Health Administration Printed 1989 eI Bon < UC. BERKELEY LIBRARIES WD RRRY C03577kkL4DO