National Institute on Drug Abuse TREATMENT RESEARCH REPORT ©) DRUG TAKING ~ AMONG THE ELDERLY Ss, DEPOSITORY DEC 2 1990 U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Public Health Service Alcohol, Drug Abuse, and Mental Health Administration '' The Treatment Research Reports and Monograph Series are issued by the Treat- ment Research and Assessment Branch, Division of Prevention and Treatment Development, National Institute on Drug Abuse (NIDA). Their primary purpose is to provide reports to the drug abuse treatment community on the service delivery and policy-oriented findings from Branch-sponsored studies, innovative service delivery models for different client populations, innovative treatment management and financing techniques, and treatment outcome studies. DUG L yy * if \ 644 3- $060 This study was conducted under National Institute on Drug Abuse Grant Number H8! DA 01953 to the University of Houston, Houston, Texas. Acknowledgement is extended to Professors Roland A. Patry and Carl W. Driever of the Department of Clinical Pharmacy, University of Houston, for their professional advice and assistance in developing this report. All material appearing in this report, except quoted passages from copyrighted sources, is in the public domain and may be reproduced or copied without permission from the National Institute on Drug Abuse or the authors. Citation of the source is appreciated. A portion of the data from this study has appeared in the Journal of Psychoactive Drugs, 1981, Volume 13, No. 2., under the title, "Psychoactive Drug Use and Potential Misuse Among Persons Aged 55 and Older." DHHS Publication No. (ADM)83-1229 Printed 1982 For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, D.C. 20402 ''Preface PUBL Understanding the abuse or misuse of drugs by the elderly has become an area of increasing concern to the drug abuse field. In 1975, the National Institute on Drug Abuse sponsored a conference on drug use by the elderly that resulted in a programmatic thrust to explore the dangers of drug misuse in that population. Sub- sequently, a grant was awarded to the Department of Sociology, University of Houston, for research on the use of psychoactive substances by the elderly and the problems associated with such use. The resulting study, conducted by Dr. Stephens, Dr. Haney, and Ms. Underwood, clarifies the significance of psychoactive drug use in the lives of the elderly and the extent to which the elderly engage in inappropriate use of these substances, as well as alcohol and tobacco. The report that follows forms a part of the authors! larger exploration of drug use and health issues among the elderly. In large part, the study makes the case for a responsible and cautious approach to psychoactive drug taking by the elderly. Nonetheless, the differences found in drug taking and reported healthfulness within sub- groups of the elderly population should interest service delivery and research personnel who work with the elderly. ''Summary In 1979, a study was made of the extent of drug use and misuse by a random sample of 1,101 of Houston's noninstitutionalized elderly population, stratified by age. Individuals had to be 55 years old or older to qual- ify for inclusion in the sample. Overall, 17.6 percent of the Houston elderly were found to be using prescribed psychotropic drugs at the time of the interview. Of the persons sampled, 6.9 percent were found to deviate from their prescriptions by taking either more or less than had been prescribed. Of those persons who deviated from their prescriptions, 86.8 percent stated that they took less than the amount prescribed by their physician. Thus, less than | per- cent (0.9 percent) of the persons sampled reported themselves as taking psychotropics in excess of physi- cians' prescriptions. Not surprisingly, fewer than 2 percent reported using any illicit drug in their lifetime. Moreover, 60 percent of the sample reported that they had not drunk any alcoholic beverages in the month preceding the inter- view, although 6.6 percent reported consuming five or more drinks on those occasions when they did con- sume alcohol. The 5.7 percent of the sample who reported using a psychoactive substance some time in the past month also reported using alcohol at some time in the past month. Thus, the study coincides with ear- lier findings that portray the elderly as responsible consumers of psychoactive substances. ''Preface . . . +. + «+ +e «ce 2« ee © © © © © © © Summary . 2. 2. 2 © © © © © © © © © © © ee ee Review of the Literature .....+++ + ee - General Population Studies - Extent of Drug Use Studies of Drug Use Among Older Persons . . . Summary . 2. 2. «© «© © © «© © © © © © © © © ew Correlates of Use . 2. 2. «© «6 «© © © © © © © © Misuse of Drugs . . « « «© © © © © © © © we ow Use of Alcohol .... 2. © © © © © © © © © Use of Tobacco ... e+ +++ «© © © © e «© « Methodology . . . .- 2. 6 «© © © © «© © © © © © © ow Sampling ... 2. «© © © © © © © © © © we © Data Collection . . 2. 2. 2 «© © «© © © © ee ow Variables . 2. 2. 2. © © © © © © © © we we eo ow ow Use of psychoactive drugs ....+.e«-s.- Compliance with prescription directions Other variables ..... + 6 « © «© «© » Psychoactive Drug Use and Potential Misuse .. . Psychoactive Drug Use . « © 2 2 © © e+ ee © Use of illicit substances ...... . Comparison with other studies .... - Potential Misuse of Psychoactive Substances . Failure to follow prescription directions Comparison with other studies ..... Use of Alcohol and Tobacco .... + +++ «ee Use of Alcohol ...- . « « © © © © «© © © «@ « Comparison with Other Studies ......-.- Use of Tobacco ....-« + «++ 2 2 © © «© « Health and Drugs .... 2 «© «© © «© © © © © © © Self-Assessment of Health . . 2. . «© «© «© «© «© « General Health Scale .... + 6 «© «© © © « « Number of Physicians Contacted .....-. - Conclusions . . . « « « « «© © © © © © © © © © © References ... +. +++ «+ +e © © © © © © © © Page 10 10 10 11 11 LE 11 12 12 12 13 14 14 14 a5 iS 16 16 16 16 16 18 19 20 20 20 21 23 24 ''''Drug Taking Among the Elderly Richard C. Stephens Departments of Sociology and Urban Studies Cleveland State University C. Allen Haney Suzanne Underwood Department of Sociology University of Houston Review of the Literature The literature on the use of psychoactive sub- stances among older persons is divided into two major categories. The first category consists of studies examining psychoactive drug use among the general population, of whom older persons con- stitute a relatively small part. Until very recently, these studies have been the principal source of data in assessing the extent of drug use in older populations. In the late 1970s, however, a second major category of studies emerged, focusing on the older person's drug use patterns. Most studies dealt with a number of issues, in- cluding the extent of drug use in both the general population and selected subpopulations, predictors and correlates of such use, drug misuse, and the use of alcohol and tobacco. This chapter deals with each of these issues sequentially. General Population Studies - Extent of Drug Use Table | summarizes the characteristics of relevant studies. Mellinger et al. (1974) present data col- lected in a national survey by Parry et al. (1970-71) on the use of psychoactive drugs. They found the following distribution of type of drug use (pre- scription or over-the-counter drugs) by sex for respondents aged 60 to 74 for the year prior to interview. Type of drug use Male Female Total Prescription psychotherapeutic drug use only 19% 28% 24% Over-the-counter and prescrip- tion psychotherapeutic drug use 2% 3% 3% Over-the-counter psychothera- peutic drug use only 7% 4% 5% Use of neither psychothera- peutic drug type 72% 65% 68% The use of prescription psychotherapeutic drugs, the type most commonly taken, was distributed as follows: Type of psychotherapeutic drug Male Female Minor tranquilizer/sedative 11% 25% Stimulant 1% 3% Hypnotic 7% 8% Antidepressant 4% 2% Major tranquilizer 0.5% 2% No psychotherapeutic drugs used in the past year 79% 68% The data show that the majority of persons had not used psychotherapeutic drugs of any type in the previous year and that females used psychoactive substances much more frequently than did males. This finding is particularly evident for the minor tranquilizer/sedative category. In a survey of the general adult population con- ducted in San Francisco, Mellinger et al. (1971) found the following percentages of prescription psychotherapeutic drug use in the prior year among persons aged 60 and over. Category of drug Male Female Stimulant -- 3% Minor tranquilizer 5% 10% Sedative 4% 9% Hypnotic 9% 13% No use of the drugs listed above 77% 66% This study corroborated the findings of the Mel- linger study previously cited. In addition to noting that the majority of those 60 and over had not used these classes of psychotherapeutic drugs in the past year and that more females than males used such substances, Mellinger found that virtually all the users 60 and older reported receiving their drugs from medical sources. ''Table 1.--Characteristics of relevant studies Researcher Subjects Type Year Number conducted General population studies D. Fejer and R. Smart (1973) D. Manheimer et al. (1968) G. Mellinger et al. (1971) G. Mellinger et al. (1974) G. Warheit et al. (1976) Studies of older persons K. Back and D. Sullivan. (1978) C. Chein et al. (1978) B. Frank et al. (1979) C. Green (1977) D. Guttmann (1977; 1978) W. Hale et al. (1979) M. Maultsby and R. Fortier (1981) P. Raffoul et al. (1981) A. Vener et al. (1979) Adults in Toronto, Canada Persons aged 21 or older in California Household survey of persons aged 18 or older in San Francisco, California National sample of persons aged 18-74 Household survey of persons aged 18 or older in Florida Insurance company records of whites aged 45-70 Persons aged 60 or older in Washington, D.C. Persons aged 60 or older in New York City Household survey of persons aged 60 or older in Osceola County, Florida Household survey of persons aged 60 or older in Washington, D.C. Patients 65 and older in hypertension clinic in Florida Registered voters aged 60 and older in Fayette County, Kentucky Registered voters aged 60 or older in Fayette County, Kentucky Retired persons aged 60 and older in Central Michigan 1,200 1971 1,026 1967 1,104 1967-68 2,552 1970-71 1,633 1970s 502 242 1977 146 1979 100 1977 447-1976 1,711 1977-78 323 1978(?) 67 1979 55 1978(?) Manheimer et al. (1968) conducted a study of psy- chotherapeutic drug use among a sample of Cal- ifornians aged 21 and older. He found the follow- ing percentages of persons aged 60 and over re- porting "frequent" use of psychotherapeutic drugs (both prescription and over the counter): Category of drug Male Female Total Stimulant 0% 4% 2% Sedative 10% 13% 11% Tranquilizer 7% 10% 8% Any of the three 13% 19% 16% Number (135) (93) (228) The data shed additional light on the relationship between sex and psychotherapeutic drug use. Ap- parently females are more likely to use these drugs than are males, but they also use the drugs more frequently than males. Warheit et al. (1976) found that 17.5 percent of persons aged 60 and older reported current use of stimulants, sedatives, or tranquilizers. About half (50.5 percent) of those 60 and older reported using at least one of these substances during their life- time (34.0 percent had used sedatives, 35.3 percent tranquilizers, and 3.5 percent stimulants). A period of frequent use (daily or often) at some point in their lives was reported by 20.0 percent of the sedative users, 83.1 percent of the tranquilizer users, and 9.1 percent of the stimulant users. Fejer and Smart (1973) reported the prescribed use of tranquilizers, barbiturates, or stimulants over a ''l-year period for a sample of persons aged 56 and older who were residents in Toronto, Canada. Age Category of drug 56-60 61 and over Tranquilizer 12.3% 12.0% Barbiturate 6.2% 14.4% Stimulant 0.0% 1.8% Studies of Drug Use among Older Persons Probably the most comprehensive study of an el- derly group that has been conducted to date is Guttmann's study (1977; 1978) of 447 older res- idents of Washington, D.C. He examined the use of all prescription and over-the-counter drugs in the 24-hour period before the interview was conducted. In addition, he gathered data on the physical con- dition of respondents, as well as on life satisfac- tion and other measures of morale. He found that the three categories of drugs most frequently used were cardiovascular (39.3 percent), sedative/tran- quilizer (13.6 percent), and antiarthritic (9.4 per- cent). Guttmann (1978) presented the following data (left-hand column) on use of prescription psycho- active drugs. The categories were combined (right-hand column) to determine the percentage of the total sample that currently used each type of drug. Percentage of Number _ total sample of who use Type of drug persons drug type Sedative/tranquilizer 16.6 Daily use 28 1 or more times weekly 16 As necessary 30 Antidepressant 1.6 Daily use 6 As necessary 1 Nervous system 2ad Daily use 10 As necessary 2 Analgesic 7.8 Daily use 18 | or more times weekly 1 As necessary 16 Guttmann's subjects showed the following usage patterns. 1. Almost all users of psychotropic drugs (98.7 percent) obtained their prescribed sedative/tran- quilizer and antidepressant drugs from a physician. 2. More users of psychotropics (50.7 percent compared with 39 percent of the users of all pre- scription drugs) could not perform regular daily activities without prescription drugs. 3. The same percentage (90.6 percent) of psy- chotropic and other prescription drug users had no difficulty in obtaining prescription drugs. 4. A considerably higher proportion of users of psychotropics took several kinds of drugs in com- bination than did users of prescription drugs in general (38.6 percent vs. 26.4 percent). 5. Users of psychotropic drugs consulted more frequently with a physician prior to taking drugs in combination than did users of all prescription drugs (75.8 percent vs. 62.4 percent). Frank et al. (1979) found that 72 percent of re- spondents chosen primarily from New York sen- ior citizen centers reported having taken at least one prescription drug (of any type) in the week preceding the interview. They found that 28 per- cent had taken a sedative/tranquilizer, and 29 percent had taken an analgesic/antiarthritic in the week prior to the interview. One-fourth of the subjects had taken some sedative within their life- times, while 44 percent had used a tranquilizer. Four percent reported ever taking stimulants. Al- most three-fourths (71 percent) of the sample cur- rently used at least one over-the-counter prepa- ration. Hale et al.(1979) reported the following data for regular use (respondent's definition) of psycho- active substances by a sample of 1,711 patients of a hypertension-screening program who were aged 65 or older. Drug type Male Female Total Hypnotic 5.1% 6.4% 5.9% Tranquilizer 7.2% 9.5% 8.7% Sedative 0.7% 1.0% 0.9% Antidepressant 0.9% 1.3% 1.1% Analgesic 17.7% 27.8% 24.0% Back and Sullivan (1978) in a study of 502 persons aged 45-70, reported the following rates of current use of psychoactive substances. Tranquilizers 15.7% Sleeping pills 8.8% Pain killers 11.2% Alcohol 46.4% Tobacco 50.0% In a study of retirees aged 60 or older, Vener et al. (1979) found 33 percent had taken no drugs in the week before the interview. Sixty-five percent of the sample had used some over-the-counter pre- paration in the same period. Green (1977) found that 80 percent of a sample of 100 persons aged 60 and older reported using a drugs in the week before the interview. The fol- lowing percentages used prescription psychotropic drugs. Analgesic 6% Antidepressant 3% Sedative/tranquilizer 19% ''Of psychoactive over-the-counter preparations, 38 persons (38 percent) used analgesics, one used a sleeping aid, and three used antihistamines. Chein et al. (1978) studied 242 persons aged 60 and older residing in Washington, D.C. and found the following drug usage. Analgesic 66.6% Antianxiety 22.3% Hypnotic 12.0% Antipsychotic 4.1% Antidepressant 2.5% Antihistamine 1.2% They concluded: "The results of the study did not reveal any significant abuse of substances, includ- ing alcohol and self-prescribed drugs, if abuse is defined as the intentional use of illegally obtained mind- or behavior-altering substances. No nonmed- ical use of amphetamines, barbiturates, or narcot- ics was found, nor was a single instance of mari- juana usage reported. The common suspicion that a significant number of elderly people might be unreported alcoholics was not verified...." Maultsby and Fortier (1981) found that 82 percent of 323 registered voters 60 and above were cur- rently using central nervous system drugs. Over 50 percent of the sample reported current use of two or more psychotropics, including over-the-counter preparations. The most popular psychotropic drugs were nonnarcotic analgesics; minor tranquilizers and barbiturates were the distant second and third most popular. Summary While the studies cited varied widely in sampling, period during which drug use was assessed, and methodology for collecting the data, four gener- alizations can be drawn. First, the great majority of elderly persons do not use psychoactive sub- stances. Second, analgesics and antianxiety agents are the most frequently used substances, and third, the elderly use those drugs with care. Fourth, males are less likely to use psychoactive sub- stances than are females. Correlates of Use Some of the studies described above focused on correlates of all drug use. Both Green (1977) and Vener et al. (1979) found that poor health was a predictor of drug use. Other studies focused directly on correlates of psychoactive drug use. Both Chein et al. (1978) and Guttmann (1978) found that poor health was a correlate of psychoactive drug use. Mellinger et al. (1974) noted that the use of medically prescribed psychotherapeutic substances was associated with psychic distress and the experiencing of a life crisis (death of spouse, etc.) in the past year. Sim- ilarly, Guttmann found that psychoactive drug users scored significantly lower on life satisfaction than did nonusers and looked upon the present as the worst time of their lives. Misuse of Drugs A number of studies touched upon the subject of the potential misuse of drugs. Raffoul et al. (1981) reported that 28 (43 percent) of their subjects committed 36 instances of drug misuse, i.e., use of drugs not in accord with medical prescription. This amounted to noncompliance with 16 percent of all prescriptions. Twenty-one subjects misused one drug each, 6 misused two drugs, and |! misused three drugs. The types and percentages of occur- rences of misuse were: N Underuse 72% 26 Use after need completed 11% 4 Alcohol/drug or drug/drug 11% 4 interaction Prescribed for another person 3% 1 Improper storage 3% l Underuse comprised the overwhelming number of instances of misuse. To understand this, the re- searchers contrasted misusers with users on num- ber of prescribed and over-the-counter drugs, medical condition, mental status, age, sex, race, education, income, living status, self-rating of health, number of dispensing pharmacies, and number of prescribing physicians. Only the latter two variables were found to be significantly re- lated to misuse. Maultsby and Fortier (1981) defined misuse of drugs as "nonprescribed or nonrecommended use." They reported the following: 1. As the total number of drugs taken per per- son increased, frequency of drug misuse increased. This relationship was stronger for prescription drugs than for over-the-counter preparations. 2. As daily alcohol consumption increased, both the number of other drugs taken and the misuse of these drugs decreased. 3. For most of the subgroups in this study (de- fined by age and education), misuse correlated positively with number of psychoactive drugs taken. 4. High levels of psychological discomfort, as measured by the Derogatis Symptom Checklist 90, correlated positively with multiple drug use. More- over, psychoactive drugs were apparently ineffec- tive in lessening psychological discomfort. The authors suggest: "It then seems logical to sus- pect strongly that some of the psychological dis- comforts of our subjects were probably caused by undiagnosed ADR's (adverse drug reactions) sec- ondary to multiple drug use." ''Chein et al. (1978) reported that about 14 percent of their subjects evidenced misuse of drugs, de- fined as use for reasons other than those stated in the Physician's Desk Reference. The researchers re- ported psychoactive drug misuse as follows. Type of drug Use Misuse! Analgesic 66.6% 9.0% Antianxiety 22.3% 4.0% Hypnotic 12.0% 0.0% Antipsychotic 4.1% 10.0% Antidepressant 2.5% 16.7% Antihistamine 1.2% 33.3% Chein et al. and others have pointed to the use of multiple medications as a serious problem. Chein found that 83 percent of his sample took two or more drugs. Fourteen percent were regularly using 7-15 drugs, and only 8 percent were medication free. "Interestingly, many of the most commonly used drugs and substances mentioned in our data are cited to have interaction effects" (Chein et al. 1978, p. 367). Finally, Green (1977) found that her subjects had made the following adjustments to their medica- tion schedules occasionally. Taken more than the recommended dose 16% Taken less than the recommended dose 36% Stopped when started to feel better 47% Stopped when felt medication not working 23% In addition, 30 percent had shared their drugs with someone else (16 percent with a family member, 7 percent with a friend, and 7 percent with a neigh- bor). Sixty-seven percent felt that they had not receive adequate instructions from the physician. Use of Alcohol A large number of researchers have attempted to assess the extent of alcohol use among the elderly. In one study (Chein et al. 1978), 47 percent were nonusers, 36 percent had less than one drink per day, 9 percent had one drink per day, and 8 per- cent had more than one drink per day. Vener et al. (1979) determined that 58 percent of the males and 23 percent of the females in his sample had used alcohol in the week before the interview. One- fourth of another sample (Frank et al. 1979) had used alcohol in the past month, and 17 percent reported use in the past week. Green (1977) found that 25 percent drank alcohol in the week before the interview and 25 percent sometimes drank al- cohol when using prescription medications. Maults- by and Fortier (1981) discovered the following dis- tribution in their study population: abstainers (less than one drink per year), 39 percent; light drinkers (up to 12 drinks per month), 38 percent; moderate (58 per month/2 a day), 11 percent; and heavy (more than 2 a day), 12 percent. Guttmann also looked in ! Percent of misusers relative to all persons using that category of drug. ll some detail at the use of alcohol and he found that "the majority of respondents (56.2 percent) report- ed little or no use of alcohol." Frequent users of alcohol (daily use) comprised less than a fifth (18.6 percent) of the respondents, while close to one- fourth (24.6 percent) were infrequent users (drank a few times a week to a few times a month). He also found that income was positively and signifi- cantly correlated to alcohol use. Users of psycho- tropic drugs indicated somewhat less moderate drinking of alcoholic beverages than did users of nonpsychotropic drugs (40.5 percent vs. 35.1 per- cent). Warheit et al. (1976) determined that 40.3 percent of the respondents aged 60 or older reported ever using alcohol, while 22.2 percent reported periods of frequent alcohol use. Overall, the literature indicates low to moderate use of alcohol among the elderly. Use of Tobacco Finally, a few studies explored the use of tobacco among the elderly. Maultsby and Fortier (1981) found that 23 percent of their sample smoked, and 5.6 percent smoked more than one pack of ciga- rettes per day. Methodology Sampling The basic goal of the sampling procedure was to obtain a representative sample of noninstitution- alized persons aged 55 and older who resided in Harris County (Houston), Texas. A telephone sur- vey was devised that included unlisted telephone numbers, generating a random sample of 20,000 numbers. Each number was called, and it was determined that 9,299 were not residential. The remaining 10,701 numbers were screened to de- termine if anyone 55 years of age or older was resident in that home. The screen question was answered by 8,616 households, and 1,835 contained a person 55 years of age or older. If the household had more than one eligible person, a procedure developed by Kish (1965) was used to select the respondent. Both telephone and household inter- views were completed by 1,101 persons, making the final response rate for completed interviews 60 percent. Table 2 presents the demographic characteristics of the final sample of 1,101 respondents and comparable data from the 1980 census for Harris County. The sample does not appear to be truly representative of the underlying population from which it was drawn. Males, whites, and persons in the age category 55-64 seem to be underrepre- sented whereas females, minorities, and those over 65 seem to be overrepresented. However, the census data include persons who reside in nursing homes, hospitals, and other institutional settings. The study sample contains only persons who live in residences. Nonetheless, the reader should be ''Table 2.--Demographic characteristics of study sample and 1980 Harris County census data for persons aged 55 and older (in percentages) Harris County Study sample Sex Male 44.2 37.9 Female 55.7 62.1 Ethnicity White 80.5 74.04 Black 19.5 25.1 Other -- Pe Age 55 - 64 53.4 46.8 65-74 29.9 DDD 75 and above 16.6 17.9 3Includes Mexican-American cautioned that the sample appears to be some- what biased. Initially, the dependent and principal independent variables were weighted on the basis of the prob- ability of a respondent being chosen when there was more than one eligible respondent in the household. As suggested by Kish (1965), weighted vs. unweighted estimates of the variables were analyzed. The frequency distributions were found to vary by at most a percentage point or two. The only characteristic that varied somewhat more was marital status. Because the unweighted esti- mates were so close to the weighted ones and because of the statistical and analytical diffi- culties of dealing with such weighted estimates, the unweighted estimates were used. This decision should not affect the major substantive conclu- sions of this study. Finally, one additional potential bias in the data lies in the fact that not all residences have tele- phones. The telephone company estimates that fewer than 6 percent of the households in Houston that contain someone in the over-55 age group are without phone service. Data Collection The data were collected from mid-1979 to early 1980 using two different interview instruments. The first was a telephone interview_ that lasted approximately 45 minutes to | hour.“ This inter- 2 It should be noted that throughout this sampling and interviewing procedure, repeated call-backs were made to busy numbers or phones that were not answered. Call-backs were made throughout the day and evening. 12 view focused on general health issues and also contained the questions from which the morale and activity scales (to be discussed later) were built. Drug usage was not addressed in the telephone interview. At the termination of the telephone interview, the respondents were asked if they would permit an additional interview at home (for which they would be paid $5) that would include more detailed ques- tions on health issues and use of medicines. The in-home interview focused on the current use of all prescription drugs and on the respondent's total history of psychoactive drug use. The in-home interview also lasted about an hour. An effort was made to keep refusals to a mini- mum. All interviewers were extensively trained to deal with refusals. In addition, after respondents completed the telephone interview, they were sent a letter that once again outlined the general pur- poses of the study and encouraged the respondent to participate in the in-home interview. The res- pondents were also given the telephone number of the principal investigators so that they could call and verify the legitimacy of the study. If the res- pondent refused to allow a home interview, a mail- gram requesting his or her cooperation was sent. This mailgram also contained a phone number that the respondents could call if they had any ques- tions about the study. Shortly thereafter, another phone call was placed to the respondent in an at- tempt to set up an appointment for a home inter- view. This phone call usually was made by an in- terviewer supervisor who tried to respond to the respondent's reason(s) for refusal. If the respondent still refused, no further attempts were made to contact that person. As noted above, the final res- ponse rate for completed interviews was 60 percent. The primary reason given for refusal was lack of time. There is no reason to believe that respon- dents were trying to hide anything about their use of drugs (or any other subject) since they did not know in detail about the kinds of data to be col- lected in the in-home interview. Thus, it appears unlikely that psychoactive drug users or misusers selected themselves out of the sample. Various quality control measures were used to assure the validity and consistency of the data. These included extensive training of interviewers, editing of interviews, and random call-backs to make certain interviews had been conducted. In- terviewers and interviewees were matched by race and language. (Spanish-speaking interviewers con- ducted a translated version of the interview.) Variables Use of psychoactive drugs.--In this study, psycho- active drugs were defined as those mood-altering substances that affect the central nervous system. For purposes of the study, the drugs were divided into six categories: sedative/hypnotic, antipsy- ''Table 3.--Categories of psychoactive drugs used in this study Drug category Drug Brand name Sedative/hypnotic Phenobarbital Nembutal Secobarbital seconal Ethchlorvynol placidy! Antipsychotic Chlorprothixene Taractan Thioridazine hydrochloride Mellaril Chlorpromazine hydrochloride Thorazine Antidepressant Imipramine hydrochloride Tofranil Amitriptyline hydrochloride Elavil Perphenazine and ametriptyline hydrochloride Triavil Antianxiety Diazepam Valium Chloridiazepoxide hydrochloride Librium Meprobamate Miltown Stimulant Phentermine hydrochloride Fastin Diethylpropion hydrochloride Tenuate Analgesic Propoxyphene hydrochloride Darvon Aspirin, codeine phosphate Empirin with Codeine chotic, antidepressant, antianxiety, stimulant, and analgesic. Table 3 lists these categories along with representative drugs that appeared as the most frequently used in that category. Some commonly prescribed antihistamines were also included in the sedative/hypnotic category as they are often used as mild sedatives for older patients . The only one of these that occurred with any frequency was Benadryl. To ensure that all use of psychoactive drugs was reported, each respondent was asked to show the interviewer the containers of all prescription drugs currently being used. All respondents agreed. The interviewer then recorded all the information from the prescription label onto the interview form. (Interviewers checked with the respondent and made certain that the medicine bottle in fact con- tained the medication shown on the container.) The interviewer then used a code book to determine whether the substance was a psychoactive drug. This determination was necessary because many questions in the interview were dependent on whe- ther a psychoactive drug had been used. In this way, all psychoactive substances currently being used were identified. In addition to current use of psychoactive sub- stances, both recent use (in the last 3 months and in the past year) and lifetime experience with the psychoactive substances were assessed. Psycho- 3Brand names are used throughout this report be- cause most responses were given by brand name. 391-679 O - 82 - 2 active substance picture charts were employed for this part of the study. Compliance with prescription directions.--In this study the definition of potential misuse was keyed to the question of whether the individual followed prescription orders. If a respondent reported that he or she did not always follow the prescription, that person was labeled a misuser. There are, of course, problems with this particular definition. Many persons in the sample may feel that they know their prescription directions, but in actual fact they do not understand--or follow--the di- rections. Thus, the actual number of misusers may be underestimated and limited to only those per- sons who knowingly violate their prescription directions. Available studies report the elderly as showing a high degree of prescription noncompliance.* How- ever, the major error made in these instances is overwhelmingly omission of a medication (Black- well 1972; Clinite and Kabat 1969; Schwartz et al. 1962). Data from the in-home interview itself indirectly reflects on the question of patient knowledge of and compliance with prescription directions. Pre- 4at least one study has shown "that the patients over 70 had the lowest (medication) error rate. This was probably due to fewer daily distractions in their lives, permitting greater attention to their co requirements" (Clinite and Kabat 1969, p. 450). ''Table 4.--Respondents who used psychoactive drugs for selected time periods by demographic characteristics (in percentages) Psychoactive drug use Demographic characteristic Currently In past 3 months In past year Ever N Sex Male 10.3% 17.7* 20.6* 39.1* 417 Female 22.1 32.2 37.4 61.7 684 Ethnicity Anglo 18.7 29.5* 35.5* 58.2* 749 Black 15.9 21.0 26.8 43.9 276 Mexican-American 12.1 15.2 18.2 34.9 66 Age 55 - 64 17.1 26.2 31.3 57.5% 515 65-74 L725. 26.8 DileZ 51.3 388 75 and above 19.3 27.9 30.5 45.6 197 Annual household income Less than $4,481 18.8 25.8 30.6 48.9 229 $4,481 - 9,487 15.4 26.6 29.3 47.9 188 $9,488 - 20,085 13.0 23.8 26.4 5163 193 More than $20,085 14.3 24.3 30.9 60.0 210 Marital status Married 16.8 25.8 2967 Sled 565 Widowed 17.7 28.0 33.3 55.6 372 Other 20.1 26.8 30.5 5252 164 Education Less than high school 18.7 28.6 Dil. 49.6* 433 High school graduate sera | 23.7 29.1 56.5 279 One or more years of college 20.6 31.0 S77 62.4 287 Percent of total sample 17.6 26.7 S1el 53.1! 1,101 * x2 significant at .05 level. sumably, if a client were not following prescrip- tion directions, he or she might report some phys- ical or emotional problem arising from the over-, under-, or non-use of the medication. Yet the data showed that less than 5 percent of the sample associated some physical or emotional problem with any medication (including all the nonpsy- choactive drugs) they were taking. Other variables.--Large amounts of data were also collected on demographics and morale. These in- cluded measures of depression, alienation, and in- volvement in life. These measures were developed and standardized on a Texas elderly population in a previous study (Stephens et al. 1978; Blau et al. 1979). In addition, the number of life crises (loss of spouse, divorce, retirement, death of children, child leaving home, loss of parent, or change of 14 life) experienced by the respondent were deter- mined. Measures of social activity included scales built to assess use of leisure time and interaction with friends (Stephens et al. 1978). General health sta- tus was assessed through a detailed health-status scale developed on a sample of elderly Texans (Haney et al. 1981). Psychoactive Drug Use and Potential Misuse Psychoactive Drug Use Table 4 presents data on the use of prescribed psychoactive drugs among demographically defined subgroups of the sample for four different time periods: currently, in the past 3 months, in the past ''year, and ever in the respondent's lifetime. The data show that 17.6 percent (194) of the sample reported current use,? 26.7 percent reported use in the last 3 months, 31.1 percent reported use in the past year, and 53.1 percent reported having ever used a _ psychoactive substance. Females were significantly more likely to report experience with psychoactive substances than were males. Sim- ilarly, the three ethnic groups differed signifi- cantly; for three of the four time periods signifi- cantly more Anglos reported use than did Blacks or Mexican-Americans. There were no significant differences by annual household income or marital status. However, younger and better educated persons were significantly more likely to report any experience with psychoactive substances. Most persons who reported using prescribed psy- choactive drugs currently, used only one such substance. Of current users, 85.1 percent were taking one psychoactive drug, 11.3 percent were taking two, and 3.6 percent were taking three. Table 5 shows both the types of psychoactive drugs currently being used and the total number of days they were taken in the past 3 months. The anti- anxiety drugs were by far the most frequently used--9.4 percent of the total sample were cur- rently using them. The next most frequently used category--used by less than half as many people (4.2. percent)--was sedative/hypnotic. Then fol- lowed analgesics (3.0 percent), antidepressants (2.7 percent), antipsychotics (1.2 percent), and finally the stimulants (0.2 percent). 2In an additional 22 cases (2.0 percent of total sample) the respondent provided a medicine bottle which contained a psychoactive substance. How- ever, for methodological reasons (possible inter- viewer,. interviewee, or coder error) these persons were not listed as current users. They are counted as using in the past 3 months, in the past year, and of course as ever users. Thus, current use may be 19.6 percent or a possible error of 2.0 percent. Also noteworthy is the U-shaped distribution of number of days drugs were used. Most respondents reported use either for the entire 3-month period or for less than 30 days. Relatively large percent- ages of the users of antipsychotic and antide- pressant drugs, in particular, took them every day in the past 3 months. However, these findings should be moderated to some extent by the fact that small percentages of the total sample were currently using any psychoactive substances at all. The only possible exception to this statement can be found for the antianxiety drugs; nearly one- tenth of the entire sample used this drug cate- gory, which contained the single most frequently used of all the psychoactive substances--diazepam (Valium). Use of illicit substances.--Only 21 persons (1.9 percent) reported any use of illicit substances in their entire lifetimes. Thirteen persons (1.2 per- cent) reported using marijuana at least once, six (0.5 percent) had used cocaine, one used LSD, and one person reported use of methadone. Thus, as with earlier studies, it was found that these illicit substances simply present no problem for this age group. Comparison with other studies.--Caution must be used in comparing these findings to those of other studies. Age groups examined are not identical, different categorizations of drugs are used, and different periods of time are examined in various studies. Table 6 compares the results of this study with those of other studies having roughly comparable prevalence data. In most cases, the findings were within a few percentage points of each other. In general, the estimates from this investigation were slightly higher, possibly because this study, unlike most others, included the analgesics and persons in the 55-60 year range. The present study also corroborated many of the findings of prior studies: Table 5.--Number of days of psychoactive drug use in past 3 months by drug category (in percentages) Percentage of Every day in sample who use Drug category Under 30 days 30-59 days 60-89 days past 3 months N drug category Sedative/hypnotic 43.5 6.5 4.3 45.7 46 4.2 Antipsychotic 23.1 0.0 Te7 69.2 13 1.2 Antidepressant 23.3 13.3 0.0 63.3 30 Za Antianxiety 36.9 17.5 5.8 39.8 103 9.4 Stimulant 0.0 0.0 0.0 100.0 2 0.2 Analgesic 59.4 9.4 6.3 25.0 2 3.0 15 ''Table 6.--Comparison of study findings (in percentages) Current use Used in past 3 months Used in past year Ever used Female Male Total Female Male Total Female Male Total Female Male Total Study Present study 22.1 10.3 17.6 32.2 Mellinger (1974) Mellinger (1971) Manheimer (1968) 19.0 13.0 16.0 Warheit (1976) 17.5 Maultsby (1981) 82.0 17.7 2657 37.4 20.6 31.1 61.7 39.1 5361 31.0 21.0 27.0 34.0 23.0 50.5 1. The majority of elderly persons do not use psychoactive substances. 2. Females are more likely to use psychoactive substances than males. 3. The most frequently used substances are the sedative/hypnotic, antianxiety, and analgesic agents. Potential Misuse of Psychoactive Substances Failure to follow prescription directions.--Report- ed failure to follow prescription directions was explored by asking each respondent, for each psy- choactive drug they were currently taking, "Do you always follow the prescription?" If they an- swered that they did not (even if only once), they were asked why they did not and whether they had taken more or less of the drug than the prescrip- tion called for. Seventy-six persons or 6.9 percent of the overall sample reported that they did not follow the prescription directions. Of these, 86.8 percent said that they took less than that pre- scribed, and 13.2 percent reported taking more. Thus, .9 percent of all respondents reported taking more of a psychoactive drug than was prescribed for them. Given that .9 percent represents a ceil- ing in terms of possible drug abuse, the danger of the elderly using more psychoactive substances than prescribed appears to be small. When asked why they did not follow their pre- scription, the respondents said: I don't like the medication or the prescription dosage 48.6% I take them only when I feel I need them 23.0% The drug is too expensive 9.4% I get better results taking them my own way 6.8% The doctor told me I could deviate from the prescription 5.4% I get bad side effects 4.1% I sometimes forget or am too busy 7% It was clear that the majority of those who did not 16 comply were doing so by deliberate choice and in an effort to limit drug taking. Table 7 presents the demographic correlates for those who followed and did not follow their pre- scription directions. The table shows that more than twice as many women as men did not follow prescription directions, although the differences were not statistically significant. Interestingly, while those aged 75 and above were more likely to be currently using psychoactive substances, they also were least likely to misuse them; although, again, differences were not statistically signi- ficant. There was no pattern of misuse by income group. While also statistically insignificant, mar- ried persons seemed slightly less likely to follow prescription directions. Comparison with other studies.--These findings seem to be consonant with the results of the few studies that explored issues surrounding misuse. Like Raffoul et al. (1981) and Green (1977), the vast amount of prescription noncompliance in the Harris County sample was underuse. And like Raf- foul, no demographic variables were found that predicted misuse at a statistically significant level. Use of Alcohol and Tobacco Use of Alcohol Although this study focused primarily on psycho- active drug use, a series of questions was asked to determine the extent and nature of alcohol use. About three-fifths of the sample (59.6 percent or 656 persons) reported that they had not drunk any alcoholic beverages in the month before the inter- view was conducted. In fact, 43.1 percent (283) of these persons volunteered that they had never been consumers of alcoholic beverages. Of those who did drink, over 70 percent averaged only one or two drinks per day on the days that they did drink. Only a small percentage of persons overall (6.6 percent) consumed five or more drinks on the occasions that they did drink. When asked what type of alcoholic beverage they mainly used in the ''Table 7.--Psychoactive drug prescription adherence among psychoactive drug users by selected demographic characteristics (in percentages) Percent of total sample which Follow Do not follow does not follow Demographic characteristic prescription prescription N prescription Sex Male 65.1 34.9 43 3.6 Female 59.6 40.4 151 89 x2 = 0.23 ns. Ethnicity Anglo 59.3 40.7 140 7.6 Black 65.9 34.1 44 5.4 Mexican-American 50.0 50.0 8 6.1 (Ethnicity collapsed to Anglo vs. other) x2=0.13 ns. Age 55 - 64 55.7 44.3 88 7.6 65-74 61.8 38.2 68 6.7 75 and above Jlel 28.9 38 5.6 x2 = 0.13 ns. Annual household income Less than $4,481 65.1 34.9 43 6.6 $4,481 - $9,487 65.5 34.5 29 Ded $9,488 - $20,085 48.0 52.0 25 6.7 More than $20,085 63.3 36.7 30 52 x2 = 3,03 ns. Marital status Married 54.7 45.3 95 7.6 Widowed 65.2 34.8 66 6.2 Other 69.7 30.3 3D 6.1 x 2 = 3.09 ns. Education Less than high school 60.5 3959 81 7.4 High school graduate 64.3 35.7 42 5.4 | year or more of college 57.6 42.4 59 8.7 x2 = 0.46 ns, Total sample 60.8 39.2 194 6.9 Past month, 41.5 percent (183) of the drinkers re- ported using mainly beer, 16.5 percent (74) wine, and 41.7 percent (184) hard liquor or mixed drinks.® Respondents were also asked whether and to what extent they drank on the day before the interview 6Unfortunately, because of the interview time which would have been needed to ascertain a com- plete assessment of alcohol use, alcohol equiv- alency data (stated in terms of ounces of alcohol) were not computed. Thus, only broad character- istics of drinking behavior are discussed here. 17 was conducted. Only 17.7 percent of the respond- ents reported drinking an alcoholic beverage on the day before their interview. The great majority of the drinkers (70.7 percent) reported using no more than one drink on that day. Table 8 shows the extent of drinking on the day before the interview, cross-tabulated by demo- graphic characteristics. Males were significantly more likely to be drinkers and to drink more heav- ily than females, and Anglos were significantly more likely to be drinkers than other groups. The prevalence and extent of drinking declined with age, and those with lower incomes were less likely ''Table 8.--Number of drinks consumed on day before interview by selected demographic characteristics (in percentages) Demographic characteristics None 1 2 3+ N Sex Male 70.3 9.8 9.1 10.8 408 Female 89.6 5.9 3.1 1.5 682 x 2 = 78.82* Ethnicity Anglo 80.3 8.9 5.9 4.8 743 Other 87.2 3.9 3.9 5.0 337 x2 = 11.23* Age 55 - 64 75.9 8.8 8.8 6.5 510 65-74 85.1 7.1 2.6 5.2 382 75+ 93.9 <4! 1.5 0.5 197 x2 = 43.19% Annual household income Less than $9,488 88.5 3.4 3.6 4.5 358 $9,488 or more 71.9 12.3 8.6 7.2 374 x 2 = 34,37* Marital status Married Tho GS Sd 7.0 556 Widowed 88.7 5.4 4.0 1.9 371 Other 84.7 4.3 6.1 4.9 163 x2 = 34.36* *Significant at .05 level. to drink or drank less. Also, marital status was related to drinking patterns; married persons were more likely to drink and drank heavier than wid- owed or other. How did the drinking patterns of the respondents relate to their use and potential misuse of psycho- active drugs? Table 9 provides data on current psychoactive substance use and alcohol use in the past month. It illustrates a point supported else- where in this report, namely, that the respondents were responsible consumers of psychoactive sub- stances. A large percentage (47.6 percent) of the sample used neither alcohol nor psychoactive sub- stances. And just as importantly, only 5.7 percent of the sample currently using a psychoactive sub- stances had also used alcohol in the past month. Further analysis revealed that only 24 persons (2.2 percent of the sample or 12.4 percent of current psychoactive users) were current users of psycho- active substances and also drank alcohol on the day before the interview. Further documentation of the relationship between extent of alcohol use and psychoactive drug use is found in table 10. This table depicts the number of drinks consumed on the day before the interview and various measures of psychoactive drug use. The variables "current use of psychoactive drugs" and "psychoactive drug use in the last 3 months" show a slight and insignificant inverse relationship between use of alcohol and current psychoactive drug use; nonusers of psychoactive drugs were somewhat more likely to be users of alcoholic beverages. The same relationship holds when the number of psychoactive drugs used in the past 3 months is examined. Overall, the following conclusions can be drawn about alcohol use in the sample. 1. The majority of persons in the sample did not drink any alcoholic beverages in the past month. 2. A majority of those who did consume alco- holic beverages in the past month drank only one or two drinks on the few occasions that they did drink. 3. Less than a fifth of the respondents drank an alcoholic beverage on the day before the inter- view was conducted. 4. Heavier drinking was found among males, younger persons, and those who were either cur- rently married or divorced. 5. Extremely small percentages of persons seemed to be consumers of alcohol while using psychoactive substances. 6. There was no relationship between use of a psychoactive substance and number of drinks con- sumed either in the month before or the day before the interview was conducted. Thus, the general picture that emerges from this analysis is that the subjects of this study either did not use alcohol or, if drinkers, were responsible consumers of alcoholic beverages. Comparison with Other Studies Again, one must use caution in comparing these findings on alcohol use with those of other studies. In the studies cited in this paper, more than half of Table 9.--Current psychoactive drug use by use of alcohol in past month (in percentages) Currently use a Used alcohol in past month psychoactive drug Yes No Total Yes 5.7 12.0 17.7 No 34.7 47.6 82.3 Total 40.4 59.6 100.0 (N = 1,098) ''the respondents did not use alcohol. This inves- tigation found that almost three-fifths of the sam- ple had not drunk any alcoholic beverages in the month before the interview was conducted. Other studies have shown very small percentages of per- sons reporting even relatively moderate use of alcohol. For instance, Chein et al. (1978) found that 17 percent of their sample reported having at least one drink per day, and 23 percent of another sample (Maultsby and Fortier 1981) had two or more drinks per day. In the present study, 11 percent of the total sample reported having two or more drinks on the day before the interview. All of these studies also found males and upper income respondents more likely to be consumers of alcohol. Use of Tobacco Respondents were also asked about their use of tobacco products for the month prior to the inter- view. A sizable proportion of the sample (42.5 percent) reported that they had never used to- bacco products. Another 19.4 percent noted that they had used tobacco in the past, but had not used any in the month before the interview was con- ducted. In fact, most (94 percent) of these former users had ceased their use of tobacco more than a year prior to the interview. This left a net total of 419 persons or 38.1 percent of the sample who had used tobacco in the past month. Analysis of data on these individuals revealed the vast majority (88.8 percent) of smokers used a single type of tobacco product in the last month. Of the tobacco users, 77.0 percent were cigarette smokers. The majority of the cigarette smokers (61.0 percent) smoked one or more packs per day. Other types of tobacco used included cigars--4.4 percent; pipes--3.0 percent; chewing tobacco--3.7 percent; and snuff--2.7 percent. The extent of chewing-tobacco and snuff users may reflect the regional nature of the sample. Table 11 presents the demographic characteristics of those who had never used tobacco, those who were former tobacco users, and current tobacco users (as defined by use in the month before the interview was conducted). Analysis of the percent- ages yielded some interesting if not unexpected findings. First, use of tobacco was significantly linked to sex. Males were much more likely either to be current or former users of tobacco. Females were much more likely to report that they never used tobacco. Anglos were significantly more like- ly than the other ethnic groups to report never having used tobacco, but were also more likely to be current users. Minority groups were more likely to contain individuals who had once smoked, but had since quit. Age was also significantly related to tobacco use. Older respondents were much more likely to report that they had never used tobacco and were also much more likely to report that they were not current tobacco users. Almost half (47 percent) of those under 65 years of age were current users of tobacco products. There were no 19 Table 10.--Number of drinks consumed on day before interview by selected indices of psychoactive use and potential misuse (in percentages) Number of drinks consumed on day before interview None 1 2 3+ oN Current psychoactive drug use Use 87.6 5.7 4.1 2.6 194 Do not use 81.3 7.7. 5.6 5.5 896 2-5.01 ns. Use of psychoactives in last 3 months Used 84.5 7.5 5.1 3.1 293 Did not use 81.7. 7.3. 5.4 5.6 806 x 23.09 ns. Number of psychoactives used in last 3 months None 81.7 7.3 5.4 5.6 797 One 84.5 64 56 3.4 233 Two or more 83.5 11.7. 3.3 1.7 60 25.63 ns. Psychoactive drug prescription adherence Follow 87.35 68 4.2 1.7 118 Do not follow 88.2, 3.9. 3.9 3.9 76 x 2=1.58 ns. significant differences between income groups. Marital status was also related to tobacco use. Widowed and never married respondents emerged as nonusers; they were much more likely to say they had never used tobacco and also were much less likely to be current users than were the mar- ried, separated, or divorced. Of course, this find- ing may be related to the fact that the widowed were more likely to be older. No significant relationships were found between the use of tobacco in the past month and a number of indices of psychoactive substance use, including current psychoactive drug use, use of psycho- actives in the past 3 months, and psychoactive prescription adherence. To summarize tobacco use: 1. A sizable minority of persons reported never having used tobacco. 2. Cigarette smoking was the predominant pattern of tobacco use among tobacco users. 3. Use of tobacco products appeared to be particularly moteworthy among males, younger ''Table 11.--Use of tobacco products in past month by selected demographic characteristics (in percentages) Previously used but not in Demographic characteristic Never used past month Used in past month N Sex Male 19.9 30. 49.6 417 Female 56.3 12.7 31.0 684 x 2=146.26* Ethnicity Anglo 44.5 35.4 20.2 749 Other 38.0 44.2 17.8 342 x 2-7.69* Age 55-64 33.4 19.6 47.0 515 65-74 47.4 19.3 33.2 388 75 and above 56.3 19.3 24.4 197 x 2=44.01* Income Less than $9,488 43.4 40.1 16.6 362 $9,488 or more 37.2 41.2 21.6 379 x 2=4.28 Marital status Married Sled 22.7 40.2 565 Widowed 52.4 16.1 31.5 372 Other 38.4 45.7 15.9 164 x 2-78.68* *Significant at .05 level. respondents, and those who were currently mar- ried, divorced, or separated. 4. There was no statistically significant rela- tionship between tobacco use and use of psycho- actives (either currently or in the past 3 months) or potential misuse of psychoactives as defined in this study. Health and Drugs Self-Assessment of Health National data would indicate that a substantial proportion of persons aged 55 years and older suffer from one or more chronic diseases or health problems. Virtually all older people are beset to some degree by metabolic and/or hormonal chang- es, stresses related to role changes and situational problems of changing life circumstances, and the normal loss of the body's regenerative and/or autoimmune responses. Nonetheless, as depicted in table 12, approximately one-third of the respon- dents rated their own health as very good, another one-third as good, and roughly onethird as either fair or poor. Remarkably, only 8 percent of the subjects rated their own health as poor. 20 The percentage of elderly in the minority ethnic groups reporting their health status as no better than fair was higher than that of the Anglos, while over 45 percent of the Anglo group reported their health as very good, and only 20 percent of the minority group did so. The differences between the sexes were not impressive. Regarding age the results were as expected--a somewhat higher percentage reporting poor health was found among the oldest respondent group. Few clear trends emerged from the data on marital status, The percentage of subjects reporting very good health rose with level of education. General Health Scale The present study revealed that reports of ill health tended to be related to certain demographic characteristics and other health behaviors includ- ing drug taking. Following this reasoning, poorer health as reported on the General Health Scale (GHS) (Haney et al. 1981) could be anticipated among the older, minority, female, and widowed or never-married respondent groups. Table 13 sup- ports this expectation. Two-thirds of those who reported taking four or more drugs were in the poorest GHS category. Of ''those taking no drugs the corresponding percent- age was 15.3. The percentage of cases in the healthiest category decreased fairly consistently as number of drugs used increased, from a high of 24.3 percent for those not using drugs to 7.4 per- cent for those using four or more. This same situation obtained regarding psychoactive drug use; 22.4 percent of those not using psychoactives reported good health as opposed to 13.5 percent of those who used psychoactives. Attention was also focused upon those behaviors and characteristics associated with problems in taking medication. Subjects reported the following problems with medication taking. Percent of all Type of problem subjects N Trouble taking medicine at right time 8.0 86 Trouble remembering to take medicine 22.9 248 Trouble opening bottle/jar/container 36.5 396 Trouble separating or breaking tablets 4.6 50 Trouble mixing/preparing medications 1.1 12 Trouble keeping adequate supply 6.2 67 Trouble reading label 9.0 98 Taking more or less than specified on label 4.3 47 Medications which look alike 1.1 12 Other 2.3 25 Number of Physicians Contacted Fifteen percent of the subjects had seen no physi- cian in the last year. Over 40 percent saw one physician, 25 percent saw two physicians, and nearly 20 percent saw three or more physicians. Five percent (or | percent of the total sample) of those who had seen no physician in the last year reported themselves as currently taking a psycho- active drug. These data were consonant with the findings of Raffoul et al. (1981) that indicated an association of extent of drug use and misuse with the number of physicians prescribing for this group. Polypharmacy has been argued to be a particularly significant potential problem among the elderly. Health status, the use of multiple physicians, reliance upon multiple drug use, and the number of medication problems all seem to be associated and further, these associations appear to be strong among the most disadvantaged and marginal mem- bers of the study group. For this reason, two ques- tions were asked of all subjects: 1. Has your physician asked if you are going to any other physician? 2. Has your physician asked about the other medications you are taking? Table 14 displays these data. The percentage of subjects reporting that their physician had asked if 21 Table 12.--Self-assessment of health and major demographic characteristics (in percentages) Very Characteristic good Good Fair Poor N Sex Male 35.4 33.0 22.7 8.9 415 Female 38.9 30.5 23.3 7.4 679 Age 55 - 64 37.3 34.6 20.7 7.4 512 65-74 40.1 27.9 25.8 6.3 384 75 and over 33.0 30.5 23.9 12.7 197 Ethnicity Anglo 45.5 31.4 16.1 7.0 745 Other 20.0 32.1 38.2 9.7 340 Marital status Married 39.3 32.3 20.5 7.9 560 Divorced 38.0 33.3 23.1 5.6 108 Separated 23.1 30.8 38.5 7.7 26 Never married 44.8 31.0 20.7 3.4 29 Widowed 35.3 29.6 25.9 9.2 371 Educational level Less than high school 28.3 30.9 30.6 10.2 431 High school graduate 42.2 30.3 22.7 4.7 277 Some college 45.2 37.3 9.5 7.9 126 College graduate 55.1 30.4 10.1 4.4 158 Number of prescription drugs currently taking None 51.2 29.2 17.4 2.2, 408 One 41.2 38.2 17.2 3.4 233 Two or three 30.1 32.6 29.7 7.5 279 Four or more 12.6 25.9 33.3 28.2 174 Number of psychoactives currently taking None 40.8 32.4 22.1 4.7 90 One 25.2 27.0 27.6 20.2 10 More than one 7.1 25.0 25.0 42.9 28 Total 37.6 31.4 23.0 8.0 1094 they were seeing other physicians increased from 7.9 percent for those not taking medication to 21.5 percent for those taking four or more medica- tions. Thus, more than three-fourths of the sub- jects taking four or more medications had not been questioned about other physicians they might be seeing. When asked if their physician was aware of the other medications they were taking, a somewhat brighter picture emerged; 42.9 percent of those taking four or more drugs indicated that their ''Table 13.--General Health Scale score by major demographic characteristics (in percentages) GHS score Demographic characteristic No symptoms Low Medium High N Sex Male 24.9 26. 27.7 20.8 390 Female 16.3 16.8 31.3 35.6 649 Age 55 - 64 24.3 24.5 28.1 23.1 494 65-74 18.1 19.2 33.0 29.7 364 75 and over 9.4 12.2 28.3 50.0 180 Ethnicity Anglo 20.6 22. J19 25.1 714 Other 175 16.5 25.7 40.3 315 Marital status Married 24.5 24.5 27.2 23.8 534 Divorced 17.0 23.0 35.0 25.0 100 Separated 21.7 26.1 39.1 13.0 23 Never married 21.4 32.1 14.3 32.1 28 Widowed 12.4 12.4 33.3 41. 354 Educational level Less than high school 17.0 15.8 29.6 37.6 399 High school graduate 19.9 20.3 35.8 24.0 271 Some college 15.3 23.4 29.8 31.5 124 College graduate 30.9 28.9 24.8 15.4 149 Number of prescription drugs currently taking None 29.9 24.3 30.4 15.3 391 One 22.9 25.7 29.8 21.6 218 Two or Three LL9 18.7 32.8 36.6 268 Four or More 29 7.4 24.1 66.0 162 Number of psychoactive drugs currently taking None 22.4 22.4 30.2 25.1 858 One 6.4 13.5 28.2 51.9 156 More than one 4.0 0.0 32.0 64.0 23 physician had asked about the medication regimen they were following. This percentage declined to approximately 35 percent for those taking one to three medications. Among subjects using psycho- 22 active drugs, 35.1 percent of those taking one and 44.1 percent of those taking more than one had been queried by their doctors. ''Table 14.--Elements of physician's prescribing patterns by selected variables (in percentages) Conclusions Because of severe space limitations, the mass of Yes. No N data that were presented in this report cannot be discussed fully. However, the goals of the project Has your physician asked if you are have been achieved. going to any other physician? Clearly, the data indicate that the use of psycho- Number of medications active substances does not pose a major public currently being taken health problem to persons aged 55 and older. The None 7.9 92, 390 study showed, within its ability to generalize, that One 12.9 87.1 225 less than a fifth of the sample currently used any Two or three 12.4 87.6 274 psychoactive substance and that only a little over Four or more 21.5 78.5 172 half reported having ever used a_ psychoactive Number of psychoactives substance in their lifetimes. These data were : remarkably similar to data reported by others. taken in last months are tls ha ie If misuse is defined as failure to follow prescrip- One 16.3 83.7 227 ‘ 3 ‘ More than one 18.6 81.4 59 tion directions, almost 40 percent of the users (but less than 7 percent of the total sample) had failed Psychoactive use/misuse at least once to follow prescription directions. No psychoactive use 14.3 85. 483 However, the vast bulk of these persons (86.8 Psychoactive use 18.1 81.9 95 percent) reported that they took less drug than Psychoactive misuse 13.9 86. 62 directed. While this can have important clinical implications for some disorders, for most psycho- Has your physician asked if you active substances, medication error was probably are using other drugs? in the safe direction. Moreover, less than | per- cent of the total sample reported themselves as Number of medications taking more of a psychoactive than was prescribed. currently being taken In short, noncompliance with physicians’ prescrip- None 24.3 75.7 591 tions was far more likely to be in the direction of One 36.5 635.7 223 avoiding drug effects than of seeking those effects. Two or three 35.0 65.0 271 Four or more 42.9 57.1 170 Less than 2 percent reported any lifetime use of an Number of psychoactives illicit drug. Indeed, only about 40 percent of the . sample had drunk any alcoholic beverages in the taken in last 3 months : : ; None 31. 69.0 774 month preceding the interview, and only 17.7 per- O cent reported drinking an alcoholic beverage on ne 35.1 64.9 222 i : More Elvan: one 44.1 55.9 54 the day before being interviewed. Only 5.7 percent of the total sample reported themselves as having Psychoactive use/misuse currently used both a psychoactive substance and No psychoactive use 37.0 63.0 481 alcohol (this does not imply concurrent or simul- Psychoactive use 39.8 60.3 113 taneous use). In short, the ambulatory elderly Psychoactive misuse 37.1 62.9 70 appear to be responsible consumers of psycho- 23 active substances. ''References Back, K.W., and Sullivan, D.A. Self-image, medicine, and drug use. Addictive Diseases: An International Journal 3(3) :373-382, 1978. Blackwell, B. The drug defaulter. Clinical Pharmacology and Therapeutics 13(6):841-848, Nov.- Dec. 1972 Blau, Z.; Oser, G.; Stephens, R.C. Aging, social class and ethnicity. Pacific Sociological Review 22:501-525, Oct. 1979. Chien, C.P.; Townsend, E.J.; and Ross-Townsend, A. Substance use and abuse among the community elderly: The medical aspect. Addictive Diseases: An International Journal 3(3):357-372, 1978. Clinite, J.C., and Kabat, H.F. Prescribed drugs...errors during self-administration. Journal of the American Pharmaceutical Association NS9(9):450-452, Sept. 1969. Fejer, D., and Smart, R. The use of psychoactive drugs by adults. Canadian Psychiatric Assoc- iation Journal 18(4):313-319, Aug. 1973. Frank, B; Marel, R; Handwerker, B; and Twersky, J. "Drug use among senior citizens in New York City." Presented at the 30th annual meeting of the Alcohol and Drug Problems Association of North America (ADPA), Washington, D.C., Aug. 26-30, 1979. Green, C.E. "A Study of Drug Use and Misuse by the Elderly in Osceola County, Florida." Osceola County, Florida: The Door of Central Florida's Training Program for Service Providers to the Elderly. Guttmann, D. A study of legal drug use by older Americans. Rockville, Md.: Department of Health, Education and Welfare; Public Health Service; Alcohol, Drug Abuse and Mental Health Administration, DHEW Pub. No. (77-495), 1977. Guttmann, D. Patterns of legal drug use by older Americans. Addictive Diseases: An Inter- national Journal 3(3):337-356, 1978. Hale, W.E.; Marks, R.G.; and Stewart, R.B. Drug use in a geriatric population. Journal of the American Geriatrics Society 27(8):374-377, Aug. 1979. Haney, C.A.; Stephens, R.C.; Cooper, H.P.; Oser, G.T.; and Blau, Z.S. A measure of health status in an elderly population. Health Values: Achieving High Level Wellness 5(2):61-66, Mar.-Apr. 1981. Kish, L. Survey Sampling. New York: John Wiley, 1965. Manheimer, D.I.; Mellinger, G.D.; and Balter, M.B. Psychotherapeutic drugs: Use among adults in California. California Medicine 109(6):445-451, Dec. 1968. Maultsby, M.C., Jr., and Fortier, R.H. "Patterns of Drug Use by Normal Elderly People." Lex- ington, Ky.: Rational Behavior Therapy Center, University of Kentucky Medical Center, 1981. Mellinger, G.D.; Balter, M.B.; and Manheimer, D.I. Patterns of psychotherapeutic drug use among adults in San Francisco. Archives of General Psychiatry 25:385-394, Nov. 1971. 24 ''Mellinger, G.D.; Balter, M.B.; Parry, H.J.; Manheimer, D.I.; and Cisin, I.H. An overview of psychotherapeutic drug use in the United States. In: Josephson, E., and Carroll, E.E., eds. Drug Use: Epidemiological and Sociological Approaches. Washington, D.C.: Hemisphere Pub- lishing Corporation, 1974. pp. 333-366. Raffoul, P.R.; Cooper, J.K.; and Love, D.W. Drug misuse in older people. The Gerontologist 21 (2) :146-150, 1981. Schwartz, D.; Wang, M.; Zeitz, L.; and Goss, M.E.W. Medication errors made by elderly, chron- ically ill patients. American Journal of Public Health 52(T2):2018-2029, Dec. 1962. Stephens, R. C.; Blau, Z.S.; Oser, G.T.; and Millar, M.D. Aging, social support systems, and social policy. Journal of Gerontological Social Work 1(2):33-45, Fall 1978. Stephens, R.C.; Haney, C.A.; and Underwood, S. Psychoactive drug use and potential misuse among persons aged 55 years and older. Journal of Psychoactive Drugs 13(2):185-193, Apr.- June 1981. Vener, A.M.; Krupka, L.R.; and Climo, J.J. Drug usage and health characteristics in noninsti- tutional retired persons. Journal of the American Geriatrics Society 27(2):83-90, Feb. 1979. Warheit, G.J.; Arey, S.A.; and Swanson, E. Patterns of drug use: An epidemiologic overview. Journal of Drug Issues 6(3):223-237, Sum. 1976. U.S. GOVERNMENT PRINTING OFFICE : 1982 0 - 391-679 25 ''''''DEPARTMENT OF HEALTH AND HUMAN SERVICES PUBLIC HEALTH SERVICE POSTAGE —_ PAID ALCOHOL, DRUG ABUSE, AND tk aad MENTAL HEALTH ADMINISTRATION BULK RATE 5600 FISHERS LANE ROCKVILLE, MARYLAND 20857 OFFICIAL BUSINESS Penalty for private use, $300 NOTICE OF MAILING CHANGE (1) Check here if you wish to discontinue receiving this type of publication. (Check here if your address has changed and you wish to continue receiving this type of publication. (Be sure to furnish your complete address including zip code.) Tear off cover with address label still affixed and send to: National Clearinghouse for Drug Abuse Information P.O. Box 416 Kensington, Maryland 20795 DHHS Publication No. (ADM)83-1229 Printed 1982 ''C. BERKELEY LIBRARIES WL MTN CO2&?4ay1o ''