Annotated Bibliography of Cancer-Related Literature on Black Populations NATIONAL CANCER INSTITUTE Division of Cancer Prevention and Control Special Populations Studies Branch Minority Field Program Claudia R. Baquet, M.D., M.P.H. Nancy K. Simpson, M.Sc. Dyrel L.P. Bartee, B.S. Knut Ringen, Dr.P.H. Joyce A. Heinonen Linda Clayton, M.D. Gary King, Ph.D. Noma L. Roberson, Ph.D., R.N. Arthur Schatzkin, M.D., Dr.P.H. ANY National Cancer insti meeesmrm—————— NIH Publication No. 89-3024 U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Public Health Service National Institutes of Health LTH ACKNOWLEDGEMENTS Vincent T. DeVita, Jr., M.D. Peter Greenwald, M.D., Dr.P.H. John Horm, M.Sc. Tom Kean, M.P.H. Damian Crane Selena Abrams Doris Coleman Clarice Dove Paula Green Judy Musgrave NIH Library Staff Derrick Robinson Chris Truesdel Carol Portis Dora Smith Renee Schick AUTHOR AFFILIATION Claudia R. Baquet, M.D., M.P.H. Knut Ringen, Dr.P.H. Joyce A. Heinonen Special Populations Studies Branch and Arthur Schatzkin, M.D., Dr.P.H. Cancer Prevention Studies Branch Division of Cancer Prevention and Control National Cancer Institute Nancy K. Simpson, M.Sc. Technassociates, Inc. Dyrel L.P. Bartee, B.S. Howard University School of Medicine Linda Clayton, M.D. Meharry Medical College Gary King, Ph.D. Westat, Inc. Noma L. Roberson, Ph.D., R.N. Roswell Park Memorial Institute ii TABLE OF CONTENTS PAGE Introduction ............... .... ... ... .... 1 Annotated Bibliography Section 1. Epidemiology Overview . ...... iii 7 Descriptive/Analytic .................. 9 Case-Control, Cohort, Experimental ....... 142 Policy/Review ...................... 171 Section 2. Knowledge, Attitudes, and Practices (KAP) Overview . ...... iii 199 Descriptive/Analytic .................. 201 Experimental ....................... 214 Policy/Review ...................... 215 Section 3. Health Services Research Overview . ....... iii 219 Descriptive/Analytic .................. 221 Experimental ....................... 224 Policy/Review ...................... 225 Appendix Methodology .......................... 229 Exhibit 1. Article Summary Forms ........ 231 Exhibit 2. General Instructions/Classification Definitions . ...................... 243 Exhibit 3. List of Standard Terms ......... 251 Exhibit 4. Coding Sheet . . .............. 261 Exhibit 5. Coding Sheet Edit Procedures for Database Entry . ................. 263 Indexes Subject Index ...................... ... 267 Publications Based on NCI Data for U.S. Index . ................ 275 Geographic Arealndex ................... 277 First Author Index ...................... 279 iii Introduction In August 1985, the Secretary’s Task Force on Black and Minority Health of the Department of Health and Human Services published its Report." The Report sought to document the magnitude and causes of “. . . a persistent and distres- sing disparity in key health indicators among certain subgroups of the popula- tion.” The disparity was particularly pronounced for blacks. Additionally, the Report noted “. . . a national paradox of phenomenal scientific achievement and improvement in overall health status, while at the same time, persistent signif- icant health inequities exist for minority Americans . . . It [is] evident that to bring the health of minorities to the level of all Americans, efforts of monumental proportions [are] needed.” As part of the Secretary’s Task Force, the National Cancer Institute (NCI) took the lead in preparing Volume III: Cancer of the Report.? It documented that cancer— the second leading cause of death in the United States—has a particularly severe impact on blacks. Overall cancer incidence and mortality rates are 10 and 11 percent higher for blacks than for whites, and while 50 percent of whites survive for five years following the diagnosis of cancer, only 37 percent of blacks survive this long. At the same time that the Secretary’s Task Force was active, significant develop- ments were taking place at NCI. Although cancer prevention and control ac- tivities in the United States had been evolving since the 1930s, a systematic national program of research and application of research results was lacking.’ In response to this, two concurrent strategies were developed by NCI in the 1980s. First, a philosphy and set of principles known as Cancer Control Science was developed under the leadership of Dr. Peter Greenwald to give stronger direction to national cancer prevention and control research efforts. The essence of Cancer Control Science is scientific excellence. This means that adherence to the scientific method is as important in this area of research as it is in basic laboratory or clinical research.” Second, to improve efficiency in cancer preven- tion and control research and in the application of state-of-the-science methods, the NCI Director, Dr. Vincent DeVita, proposed a national target of reducing cancer death rates by 50 percent by the year 2000. Subsequently, Cancer Control Objectives for the Nation: 1985-2000 was published, detailing strategies by which the year 2000 target could be met.” In this report, significant differentials in U.S. cancer rates were noted between the majority population and certain subgroups, notably blacks. The reduction of these differentials “. . . require[s] aggressive and long-term ‘catch-up’ efforts if we are to achieve the overall goal.” In 1985, NCI’s Division of Cancer Prevention and Control initiated a program of targeted intervention research for U.S. minorities and other subgroups which experience high cancer rates or have unique cancer control needs. Because blacks experience the most severe differentials in cancer rates, their cancer control needs are being initially addressed.® As part of this effort, the most comprehen- sive compendium of available statistical data on the problem was published as Cancer Among Blacks and other Minorities: Statistical Profiles.” This annotated bibliog- 1 raphy is a companion to the Statistical Profiles. Together, they characterize the cancer-related needs of black populations and represent the most complete as- sessment of this problem available to date. The differentials in U.S. cancer rates which adversely affect black populations were first suggested some twenty years ago.® Since then a growing body of literature on this topic has emerged in many areas including epidemiology; surveys of knowledge, attitudes, and practices; health services research; health policy; and cancer prevention and control interventions. This body of literature is included in this bibliography. A few entries which have been especially helpful in the preparation of the current volume should be acknowledged. A review article on cancer mortality trends by Henschke, Leffall, et. al, in 1973 can be credited for drawing major attention to this problem.’ In the late 1970s, the American Cancer Society published an initial statistical overview of cancer in blacks.'® In 1980, a major international scientific meeting was convened in Buf- falo, New York, to address the state of cancer knowledge for blacks and a book containing the presented papers was published in 1981." Although the amount of research in the area of cancer in black populations is expanding rapidly, the published literature collected in this volume reveals three important findings: 1. In every area of research, there are large voids in our knowledge of cancer in black populations. The most complete area of research is descriptive epidemiology—this is attributable to the presence of population-based cancer registries which code and report cancer data by race. The literature on cancer prevention and intervention research is particularly scant. For instance, there is only one experimental health services study involving blacks. 2. No clinical trials studies were identified which specifically refer to treatment results in blacks. This may reflect the practice of not coding or reporting race in clinical trials research. 3. Although the number of entries in this bibliography—571—is relatively large, the quality varies. Some of the studies have methodological or con- ceptual limitations which the reader must assess according to his or her needs by obtaining the complete item and reviewing it critically. This bibliography is a compendium of published literature identified according to the criteria specified in the Methodology section. Some of the articles contained in this document are culturally insensitive. These articles illustrate the ethical and scientific need for culturally sensitive and respectful research. Information regarding methodology, abstract coding, and annotations is pre- sented in the Appendix at the end of the book. It should be noted that the definitions of terms used in the bibliography were flexible in order to reasonably 2 accommodate the published literature. For example, some authors defined their study designs as case-control when the study was actually a case comparison or case study. Some articles contained multiple samples and multiple study designs such as retrospective descriptive presentations of data and survival analysis. In these instances an attempt was made to list each of the major study designs and samples. We intend to update this bibliography periodically and welcome comments on the present edition as well as suggestions for additional entries in future editions. Communications should be directed to the 10. 11. Special Populations Studies Branch Division of Cancer Prevention and Control National Cancer Institute Bethesda, Maryland 20892-4200. Secretary’s Task Force on Black and Minority Health. Report of the Secretary’s Task Force on Black and Minority Health, Vol. I: Executive Summary. Department of Health and Human Services, Washington, D.C. August 1985 Secretary’s Task Force on Black and Minority Health. Report of the Secretary's Task Force on Black and Minority Health, Vol. III: Cancer. Department of Health and Human Services, Washington, D.C. January 1986 Breslow, L. et al. Cancer Control: Implications from its History. J Natl Cancer Inst 59(2 supl):671-686, 1977 Greenwald. P.; Cullen, J. The New Emphasis in Cancer Control. J] Natl Cancer Inst 74(3):543- 551, 1985 NCI Monograph. Cancer Control Objectives for the Nation: 1985-2000. (eds) Greenwald, P. and Sondik, E. Division of Cancer Prevention and Control. NIH Pub No 86-2880, Number 2, 1986 Baquet, C.; Ringen, K. Cancer Control in Blacks: Epidemiology and NCI Program Plans. In: Advances in Cancer Control: Health Care Financing and Research. Vol. 216, p 215-227. New York: Alan R. Liss, Inc., 1986 . Baquet, C.; Ringen K.; Pollack, E.; Young, J.; Horm, J.; Ries, L.; Simpson, N. Cancer Among Blacks and Other Minorities: Statistical Profiles. NIH Pub No 86-2785, Mar 1986 Cornely, P.B. Health Status of the Negro Today and in the Future. Am J Public Health 58(4):647-654, 1968 Henschke, U.K.; Leffall, L.D., Jr.; Mason, C.; et al. Alarming Increase of the Cancer Mortality in the U.S. Black Population (1950-1967). Cancer 31(4):763-768, 1973 Seidman, H.; Silverberg, E.; Holleb, A.I. Cancer Statistics, 1976: A Comparison of Black and White Populations. CA 26(1):2-29, 1976 Cancer Among Black Populations. In: Progress in Clinical and Biological Research. (eds) Mettlin, C.; Murphy, G. Vol. 53 New York: Alan R. Liss, Inc. 1981 SECTION I Bot Io eo RE ERE a al i . h ' ] ; } Section I Epidemiology Epidemiology is defined in this bibliography as the study of the distribution and causes of cancer in populations. The study of the distribution of cancer is generally thought of as “descriptive epidemiology,” and the study of the causes of cancer, “analytical epidemiology.” The epidemiology articles contained in this section are divided into three parts according to article type and reported study design. Part 1 contains 348 articles classified as descriptive/analytic epidemiology research. The majority of these articles are descriptive studies which use survey methods to measure cancer incidence, mortality, or survival in defined populations. In the United States, most cancer incidence and survival data are derived from 11 population-based cancer regis- tries representing about 12 percent of the U.S. population. Data from these registries are reported to the Surveillance, Epidemiology and End Results Prog- ram (SEER) of the National Cancer Institute. United States cancer mortality data are collected by the National Center for Health Statistics as part of the national vital statistics system. Descriptive epidemiological patterns are frequently re- ported as rates in the population. Some articles which may be classified as analytic are included here because they assess or estimate cancer risk within populations, however, their study designs do not fit the commonly accepted definitions of case-control, cohort, or experimental designs. The second part of the epidemiology section, epidemiology: case-control/cohort/ex- perimental, contains 62 articles with analytic study designs. Case-control studies compare defined cases such as cancer patients to controls who are similar to cases except for the absence of the disease. These studies test hypotheses about the etiology of cancer and are retrospective. The estimates of risk are often presented as “odds ratios” or “relative odds.” Cohort studies identify a defined population and follow it forward in time to determine which members develop a disease. For example, all individuals working in a steel plant during the period 1950-1960 may form a cohort which can be disaggregated by specific worksite. Cohort studies enable the calculation of relative risk(s). Experimental studies describe research where a study variable is manipulated in order to affect the study outcome. All three of these study designs are extremely useful in the identification of cancer risk factors because they enable the investigator to calcu- late (or estimate) cancer risk. Part three, labeled epidemiology: policy/review, contains 104 articles which have policy implications or are literature reviews of selected cancers. These articles generally do not present original data, nor do they necessarily contain a critical analysis of the literature. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Abad, R.S.; Kurohara, S.S.; Graham, J.B. Clinical Significance of Adenocarcinoma of the Cervix Am ] Obstet Gynecol 104(4):517-522, 1969 Subjects Treatment; Risk Factor Exposure: reproductive factors; Cancer Site: cervix uteri; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate; Time Frame: 1920-1953 Sample Characteristics 4,572 cervix uteri cancer cases (blacks 80; non-black 4,492): Buffalo, NY Variables SES: religion; Demographic: age, marital status, menopausal status; Cancer: histology, symptoms, stage Data Sources Hospital medical records Abstract This is a report of a statistical study of 4,572 cases of carcinoma of the uterine cervix treated by primary radiotherapy. Analysis of the different clinical features of squamous cell carcinoma, adenos- quamous carcinoma, and pure adenocarcinoma suggests some biologic differences between the squamous and the glandular types. However, no significant differences were found in the sur- vival rates after radiation treatment. The results of other series are presented for comparison. Ahluwalia, B.; Jackson, M.A.; Jones, G.W.; et al. Blood Hormone Profiles in Prostate Cancer Patients in High-Risk and Low-Risk Populations 2 Cancer 48:2267-2273, 1981 Subjects Risk Factor Exposure: endogenous hormones; Cancer Site: prostate Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 225 prostate cancer cases (United States 170; Nigeria 55): black, male; United States, Nigeria; control group size not given Variables Demographic: age, geographic area; Cancer: stage Data Sources Laboratory test Abstract Epidemiologic data reveal that the incidence rate of prostate gland carcinoma among the black population in the United States (US) is several times higher than among Nigerians. A collaborative study between the two countries was undertaken, and blood hor- mone (testosterone [T], dihydrotestosterone [DHT], estrone [El], estradiol [E2], and prolactin [P1]), total acid phosphatase (TP), and prostatic acid phosphatase (PAP) profiles in the two population groups were compared. In the US groups (patients and controls) there were significantly higher levels of T (p< 0.01) and E1 (p< 0.05) compared with the Nigerians. Also, the US patients had significantly higher levels of T (p< 0.05) and E1 (p< 0.01) compared with their matched controls. In the Nigerians T but not E1 levels were significantly lower (p< 0.05) in patients compared with con- trols. DHT, E2, and P1 were not significantly lower in (p< 0.05) patients compared with controls. DHT, E2, and P1 were not sig- nificantly different in patients and controls between and within populations. Nigerian patients had higher levels (p< 0.001) of TP and PAP compared with US patients. It is concluded that differ- ences in blood hormone profiles in the two population groups are based on factors other than the genetic makeup of the populations. Ajagbe, H.A.; Daramola, J.O.; Junaid, T.A.; et al. Adenomatoid Odontogenic Tumor in a Black African Population: Report of Thirteen Cases 3 J] Oral Maxillofac Surg 43:683-687, 1985 Subjects Treatment; Cancer Site: oral cavity; Cancer Measures: incidence Study Characteristics Design: case study; Time Frame: 1961-1982 Sample Characteristics 13 adenomatoid odontogenic tumor cases (males 6; females 7): black, 10-20 years old; Ibadan, Nigeria Variables Demographic: age, sex; Cancer: tumor location, histology, symptoms Data Sources non-SEER hospital cancer registry Abstract The records of the Cancer Registry of University College Hos- pital in Ibadan, Nigeria, were reviewed for the incidence of adenomatoid odontogenic tumor from 1961-1982. Thirteen cases of the tumor were found. Clinical, histologic, and radiologic fea- tures of the 13 cases are reported. Many features were similar to previous reports of this tumor, but an increased ratio of males to females and preference for the mandible was observed. It is suggested that these factors may be peculiar to Nigerians. A few cases showed an unusual posterior location of tumor. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Alexander, M.G.; Purves, L.R.; Kirsch, R.E.; et al. Alpha-Fetoprotein in Liver Disease S Afr Med J 53:433, 1978 Subjects Biologic Marker: alpha-fetoprotein; Cancer Site: liver Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1975-1976 Sample Characteristics 98 liver disease cases (primary liver cancer 35; chronic active hepatitis 12; virus B hepatitis 8; alcoholic cirrhosis 43): 14-77 years old, hospital patients; South Africa Variables Demographic: age, sex, race-ethnicity (black, colored, white) Data Sources Laboratory test, hospital medical records Abstract The serum concentration of alpha-fetoprotein (AFP) was mea- sured by radio-immunoassay in 98 patients with liver diseases including hepatoma, chronic active hepatitis, alcoholic cirrhosis, and acute virus B hepatitis. Raised AFP levels, above 30 ng/ml, were found in 87% of patients with acute viral hepatitis, in 82% of patients with primary liver cell carcinoma, in 58% with chronic active hepatitis and in 14% of patients with alcoholic cirrhosis. However, levels above 1,000 ng/ml were found only in patients with hepatoma and in acute viral hepatitis. Amendola, B.; Hazra, T.A.; Belgrad, R.; et al. Radiation Therapy for Esophageal Cancer: The Medical College of Virginia Experience 5 South Med J 73(11):1481-1483, 1980 Subjects Treatment; Cancer Site: esophagus; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate; Time Frame: 1968-1977 Sample Characteristics 70 esophagus cancer cases (blacks 47; whites 523): 39+ years old, low SES; Richmond, VA Variables Demographic: age, sex, race-ethnicity; Cancer: histology, tumor location Data Sources Hospital medical records Abstract In a retrospective review of 70 patients with carcinoma of the esophagus treated by external beam irradiation therapy from 1968 to 1977 at the Medical College of Virginia, we correlated survival with race, age, sex, histology, and site of tumor. Results of treat- ment were also analyzed in relation to the length of the esophagus treated, the total area of the treatment field, and the total tumor dose. From this analysis it appears that radiotherapy has a place in the palliative treatment of esophageal cancer and that palliation is independent of the total volume and dose. There is indication that with higher dosage the survival rate is slightly higher. Arellano, M.G.; Linden, G.; Dunn, J.E. Cancer Patterns in Alameda County, California Br J] Cancer 26:473-482, 1972 Subjects Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: age-adjusted cancer incidences rate, higher level statistics; Time Frame: 1960-1969 Sample Characteristics Cancer incidence data from Alameda County, CA Variables Demographic: sex, race-ethnicity (Negro, white, Chinese, Japanese) Data Sources non-SEER population-based cancer registry Abstract There has been a general increase in the incidence of cancer of most major sites during the period 1960-69; this is true even when allowances are made for shifts in the age composition of the popu- lation. Improvements in diagnostic procedures may account for some of these increases but it is doubtful that they are solely responsible for the greater incidence recorded. A few sites stand out as being primarily responsible for the increase in the overall cancer incidence. Lung cancer is increasing in both males and females; the rate of increase, however, is much greater among females. It is generally acknowledged that women begin smoking cigarettes at a later point in time and to a lesser extent. The pattern which has emerged indicates that females are experiencing a simi- lar trend in lung cancer incidence to that of males. The increase in the incidence of female breast cancer is also noteworthy, al- though the forces producing this change can only be speculated upon. The high incidence of prostatic cancer among negroes and the increase in the incidence of prostatic cancer in whites are sub- jects which deserve further investigation, especially since the Alameda County experience is not duplicated in data from the Connecticut Tumour Registry. One of the most encouraging find- ings is that the incidence of stomach cancer appears to be declining. 10 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Arthaud, J.B.; Guinee, V.F. Jejunal and Ileal Adenocarcinoma: Epidemiological Considerations 7 Am ] Gastroenterol 72(6):638-646, 1979 Subjects Cancer Site: small intestine; Cancer Measures: survival Study Characteristics Design: prospective; Measures: descriptive statistics; Time Frame: 1944-1976 Sample Characteristics 29 jejunal and ileal adenocarcinoma cases (blacks 2; whites 26; Spanish surname 1): 21-76 years old, hospital patients; Texas Variables SES: religion, occupation; Demographic: age, sex, race-ethnicity; Cancer: tumor location, histology Data Sources Hospital medical records Abstract Twenty-nine patients with jejunal and ileal adenocarcinoma were characterized by the demographic features of age, sex, ethnic group, residence, occupation, religion and associated diseases. Comparison with the overall M. D. Anderson Hospital patients population did not identify any risk factors characteristic of this cancer. Aur, R.J.A.; Simone, ].V.; Pratt, C.B. Successful Remission Induction in Children with Acute Lymphocytic Leukemia at High Risk for Treatment Failure Cancer 27(6):1332-1336, 1971 Subjects Treatment; Cancer Site: leukemia; Cancer Measures: survival Study Characteristics Design: prospective; Measures: descriptive statistics; Time Frame: 1968-1969 Sample Characteristics 11 acute lymphocytic leukemia cases (Negroes 6; Caucasians 5): 1-12 years old; Memphis, TN Variables Demographic: age, sex, race-ethnicity Providers Physician Abstract Eleven children with untreated acute lymphocytic leukemia at high risk for treatment failure were given a combination of pre- dnisone, vincristine, daunomycin, and L-asparaginase for remis- sion induction. Negro children and those with marked leukocytosis, hepatosplenomegaly, symptomatic visceral infiltra- tion, or other evidence of advanced disease qualified for this study. All 11 patients achieved complete remission marrow in a median time of 18 days. Reversible drug toxicity attributable to L-as- paraginase included azotemia, hyperuricemia, hyperglycemia, hyperamylasemia, coagulation defects, and weight loss. Axtell, L.M.; Meyers, M.H.; Shambaugh, E.M. Treatment and Survival Patterns for Black and White Cancer Patients Diagnosed 1955 through 1964 National Cancer Institute, National Institutes of Health, U.S. Department of Health, Education, and Welfare, NIH Publication No. 75-712, 1975 Subjects Treatment; Cancer Site: multiple sites; Cancer Measures: survival Study Characteristics Design: cross-sectional; Measures: relative survival rate, descriptive statistics, higher level statistics; Time Frame: 1955-1964 Sample Characteristics 240,581 cancer cases (blacks 21,088; whites 219,493): End Results Program areas Variables Demographic: age, sex, race-ethnicity; Cancer: stage Data Sources End Results Program Abstract This is the first comprehensive report issued by the Biometry Branch of the National Cancer Institute which evaluates the end results of cancer among black and white patients. Data are pre- sented for 18 cites of primary cancer for which there were sufficient numbers of black patients for a meaningful survival analysis. These 18 forms of cancer accounted for 76 percent of invasive cancers found in white patients and 81 percent in black patients. The data for each site cover cases diagnosed during the 10-year time period 1955-1964, enabling 10-year survival rates to be computed. Survival tables for black and white patients are presented for each sex and for all patients, and for those with localized disease in sites where extent of disease was pertinant. In addition, the tables present treatment and extent of disease distributions. Over three-fourths of the black patients were seen in hospitals in the state of California and in Charity Hospital in New Orleans. Survival for black patients was not as favorable as for whites. Differences in the extent of disease at time of diagnosis could explain part of this difference in survival. However, in many primary sites, survival for white patients with localized disease was more favorable than that for black patients with the same extent of disease. Thus, other factors must have been involved in this survival advantage. 11 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Axtell, L.M.; Myers, M.H. Contrasts in Survival of Black and White Cancer Patients, 1960-73 J Natl Cancer Inst 60(6):1209-1215, 1978 10 Subjects Treatment; Diagnosis; Cancer Site: multiple sites; Cancer Measures: survival Study Characteristics Design: time trend; Measures: descriptive statistics, higher level statistics, relative survival rate; Time Frame: 1960-1973 Sample Characteristics 298,859 cancer cases (blacks 28,179; whites 270,680—Period 1: 143,425; Period 2: 155,434): Connecticut; California; Iowa; New Orleans, LA Variables Demographic: age, race-ethnicity, sex; Cancer: stage Data Sources SEER cancer registry Abstract The summary data contrasting survival of black and white cancer patients whose disease was diagnosed during 1960-66 and 1967-73, presented for 18 forms of cancer, were derived from case material assembled by the Cancer Surveillance, Epidemiology and End Results (SEER) Program, sponsored by the National Cancer Institute. The average black patient was younger than the average white patient during both periods. About 30% of the cancers in black patients were diagnosed as localized neoplasms, compared to 40% of the cancers in whites. With few exceptions, survival for black patients with each form of cancer has continued to be less favorable than for white patients. It was generally closer to that for white patients when the disease was localized, but whites generally still had higher survival rates. In both races, survival for females was better than for males. The greatest differential in survival between the two races was for cancer of the bladder and cancer of the uterine corpus. Multiple myeloma was one type of cancer for which survival for black patients was more favorable than for white patients during 1967-73. For both races, survival rates improved slightly between 1960-66 and 1967-73. Lung cancer was the leading cancer site for both black and white men for both periods; the proportion increased from 18 to 21% for whites and from 20 to 24% for blacks. Among both black and white patients, the greatest improvement in survival between 1960-66 and 1967-73 occurred for patients with Hodgkin's disease. Improvements were also notable for black patients with cancer of the prostate gland, uterine corpus, and stomach and for those with multiple myeloma. Improvements were less dramatic among white patients. The poorer survival results for black patients may have been associated with socioeconomic factors. Ayangade, O. Gestational Trophoblastic Disease in Nigeria — A 10 Year Review 11 East Afr Med J 56(6):278-282, 1979 Subjects Risk Factor Exposure: reproductive factors; Cancer Site: gestational trophoblastic disease; Cancer Measures: incidence Study Characteristics Design: retrospective; Measures: descriptive statistics; Time Frame: 1966-1975 Sample Characteristics 35 gestational trophoblastic disease cases: black, female; Nigeria Variables Demographic: parity, age Data Sources Hospital medical records Abstract A ten-year review of gestational trophoblastic disease in a com- munity specialist hospital is discussed. Of such 35 cases were 12 identified out of a total of 13,182 deliveries. An incidence of a 1 in 375 seems to reflect a national average. Most of these cases were followed up for a period of two to four years. Our default rate of 17% in molar pregnancy follow-up seems acceptable. Patients with metastatic trophoblastic disease were routinely referred to the nearby university hospital for definitive therapy. The mode of presentation and clinical features are discussed. A call is made for the establishment of regional centres on the continent to study the biological behaviour and other socio-cultural determinants for the high prevalence of this disease. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Baquet, C.R.; Ringen, K.; Pollack, E.S.; et al. Cancer Among Blacks and Other Minorities: Statistical Profiles 12 National Cancer Institute, National Institutes of Health, U.S. Department of Health and Human Services. NIH Publication No. 86-2785, Mar 1986 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, mortality, survival Study Characteristics Design: cross-sectional, time trend; Measures: age-adjusted incidence rate, age-specific incidence rate, age-adjusted mortality rate, age-specific mortality rate, relative survival rate; Time Frame: 1973-1981 Sample Characteristics Cancer incidence and survival data from 9 SEER areas (New Jersey and Puerto Rico are excluded); cancer mortality data for the United States Variables Demographic: age, sex, race-ethnicity (black, white, Anglo, Hispanic, Japanese, Chinese, Filipino, Native Hawaiian, Native American) Data Sources SEER cancer registry, death registration Abstract The report presents selected cancer statistics which show com- parisons between blacks, whites, and other groups. Among the racial/ethnic groups for which cancer data are presented, blacks had the highest overall age-adjusted cancer incidence and mortality rates. Native Americans, followed by blacks, had the lowest overall cancer five-year relative survival rate. Differences between black and white cancer incidence, mortality, and survival rates indicate where efforts must be directed to address the cancer needs of blacks. Benarde, M.A.; Weiss, W. A Cohort Analysis of Pancreatic Cancer, 1939-1969 Cancer 39(3):1260-1263, 1977 13 Subjects Cancer Site: pancreas; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: age-specific mortality rate; Time Frame: 1939-1969 Sample Characteristics Pancreas cancer mortality data from the United States Variables Demographic: age, sex, race-ethnicity (nonwhites, whites) Data Sources Death registration Abstract Pancreatic cancer mortality in the United States was examined by cohort analysis for the period 1939 to 1969. Birth cohorts at 5-year intervals were studied. White males and non-whites of both sexes showed a shift of the cohort mortality rate curves by age toward younger groups as birth cohorts went from 1870-1874 to 1900-1904. This shift disappeared in white males and diminished in non-whites of both sexes between the cohorts of 1890-1894 and 1900-1904. The shift in non-whites was larger than in white males so that although non-whites had lower rates than whites in earlier birth cohorts, their rates exceeded those of whites in the most recent cohort. In the search for environmental cause(s) of pancrea- tic cancer, exposure characteristics should be compatible with these observations. Berg, ].W.; Downing, A.; Lukes, R.J. Prevalence of Undiagnosed Cancer of the Large Bowel Found at Autopsy in Different Races 14 Cancer 25(5):1076-1080, 1970 Subjects Diagnosis; Cancer Site: colon, rectum; Cancer Measures: prevalence Study Characteristics Design: cross-sectional; Measures: prevalence rate, descriptive statistics; Time Frame: 1953-1959 Sample Characteristics 16,105 autopsy cases (Negroes 3,168; whites 10,724; Mexicans 2,047; others 166): Los Angeles County, CA Variables Demographic: age, sex, race-ethnicity; Cancer: tumor location Data Sources Autopsy records 13 Abstract The prevalence of undiagnosed colon and rectal cancer found in 16,105 autopsies is tabulated by age, sex, and race. The preva- lence was higher in women than in men, lower in Negroes than in Caucasians or Orientals, and lowest of all in Mexicans. In this population, 18 percent of the symptomatic cancers were undiag- nosed premortem. Grossly recognizable latent cancer was no more frequent than undiagnosed symptomatic cancer. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Berg, ].W.; Ross, R.; Latourette, H.B. Economic Status and Survival of Cancer Patients Cancer 39:467-477, 1977 15 Subjects Cancer Site: multiple sites; Cancer Measures: survival; Socioeconomic Factors Study Characteristics Design: time trend, prospective (nonconcurrent); Measures: relative survival rate, crude survival rate, age-adjusted survival rate, descriptive statistics; Time Frame: 1940-1969 Sample Characteristics Cancer data from Iowa Variables SES: public-private patient status; Cancer: stage Data Sources SEER cancer registry, hospital medical records Abstract The relation of economic status to survival was studied for 39 kinds of cancer representing all types for which 60 or more indigent patients were seen in Univeristy of Iowa Hospital for primary care during the years 1940-1969. For every type the indigent patients had poorer survival than non-indigent patients. Quality of care would be eliminated as a major variable since a second group of “ward” patients of higher economic status was available for com- parison and the differences were substantially greater between the two groups of teaching patients than between the “clinic pay” and “private” patients. Age differences and differences in stage of dis- ease accounted for less than half of the survival deficits in the indigents. The two important problems were high mortality from causes other than cancer and excess cancer mortality not accounted for by stage differences, particularly among patients who should have had 5-year survival rates between 40 and 70%. In these pa- tients cancer recurred more often and earlier among the indigent. We postulate host differences associated with poverty that could also account for much of the observed Black-White differences as well as some international differences in cancer survival rates. Bernstein, R.; Pinto, M.R.; Morcom, G.; et al. Karyotype Analysis in Acute Nonlymphocytic Leukemia (ANLL): Comparison with Ethnic Group, Age, Morphology, and Survival Cancer Genet Cytogenet 6:187-199, 1982 16 Subjects Biologic Marker: karyotype; Cancer Site: leukemia; Cancer Measures: survival Study Characteristics Design: prospective; Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1978-1980 Sample Characteristics 60 leukemia cancer cases (blacks 22; whites 38): hospital patients, Johannesburg, South Africa Variables Demographic: age, sex, race-ethnicity; Cancer: histology Data Sources Hospital medical records Abstract The karyotype, leukemic cell morphology (FAB classification), ethnic group, age, sex, and survival were compared in 60 patients with acute nonlymphocytic leukemia (ANLL), to determine their diagnostic and prognostic significance. An ethnic age difference was observed; a significantly greater number of black patients were children. The majority of children were males. A higher frequency of chromosome abnormalities was detected in children, yet they survived longer than adults. A specific, significant association be- tween a t(8; 21) karyotype and M2-ANLL was confirmed; four of ten patients with M2-ANLL showed this translocation. The more mature morphology of M2-ANLL was associated with a longer survival irrespective of karyotype, ethnic group, and age. The specificity of t(15; 17) in M3-ANLL and nonrandom monosomy 7 in preleukemic children was confirmed. Patients, particularly adults, with normal karyotypes tended to survive longer than those with abnormal karyotypes. The patient's age and the dif- ferentiative capacity of the leukemic cell appear to be as important as the karyotype in determining survival. The nonrandom associ- ation of certain chromosome aberrations in ANLL appears to be worldwide. 14 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Bernstein, R.; Pinto, M.R.; Wallace, C.; et al. The Incidence, Type, and Subsequent Evolution of 14 Variant Ph1 Translocations in 180 South African Patients with Ph1 - Positive Chronic Myeloid Leukemia 17 Cancer Genet Cytogenet 12:225-238, 1984 Subjects Diagnosis; Cancer Site: leukemia; Biologic Marker: karyotype Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1977-1982 Sample Characteristics 180 chronic myeloid leukemia cases (blacks 94; whites 82; mixed ancestry 4—Standard translocation 166; variant Philadelphia chromosone 14): South Africa Variables Demographic: age, sex, race-ethnicity Data Sources Laboratory test, hospital medical records Abstract A Philadelphia (Ph-1) chromosome translocation was found in 180 of 198 cases of chronic myeloid leukemia (CML). A standard t(9;22) was present in 166 patients, 83 of whom were black, 79 white, and 4 of “mixed” ancestry; whereas a variant Phl translo- cation was detected in 14 patients (7.8%), 11 of whom were black and only 3 white. There was a higher frequency of a variant Ph-1 among black patients compared with whites. The significantly higher frequency of a variant among our patients compared with surveys from elsewhere could be due to differing environmental agents. Simple variants were detected in four patients. Complex variants were found in eight cases; in one of these patients, only chromosomes #9 and #22 were involved, but a complex rearrange- ment of chromosome #9 had occurred. A “masked” Ph-1 translo- cation was detected in two cases, both of which showed monosomy #22 because the Phl chromosome was incorporated or inter- changed with chromosome #9. Karyotypic evolution of the Ph-1- positive cell line was observed more frequently in the variant group (71.4%) than the standard group (29.5%). This difference was sig- nificant (p< 0.005). There was no difference in the type of clonal changes seen in standard and variant groups. The majority of clonal changes were observed during the acute stage in both groups. The majority of clonal changes were observed during the acute stage in both groups. In the variant group, there was no obvious correlation between the type of variant, type of clonal change, blast morphology, or survival. Their initial survival pattern resembled that of Ph-1-negative cases, but those patients who survived longer than 1 year showed a survival trend similar to standard Ph-1-positive cases. Possible explanations for the specific- ity of chromosome #22 involvement and the constancy of the 22q11breakpoint in all these variant translocations are discussed. Bisceglie, A.M.; Hodkinson, H.]. Multiple Myeloma and Pernicious Anaemia: A Case Report S Afr Med ] 62: 535-536, 1982 18 Subjects Risk Factor Exposure: other diseases; Cancer Site: multiple myeloma Study Characteristics Design: case study; Time Frame: 1980 Sample Characteristics One multiple myeloma case: black, male; Johannesburg, South Africa Abstract A patient with both pernicious anaemia and multiple myeloma is reported. There are now 17 known cases with this association. Most of these patients are elderly and in many the pernicious anaemia developed years before the myeloma. These cases prob- ably represent a chance occurrence as there is insufficient evidence for a causal relationship. Blair, A.; Fraumeni, J.F., Jr. Geographic Patterns of Prostate Cancer in the United States J Natl Cancer Inst 61(6):1379-1384, 1978 19 Subjects Risk Factor Exposure: diet-nutritional factors, occupation; Cancer Site: prostate Study Characteristics Design: time trend; Measures: age-adjusted mortality rate, higher level statistics; Time Frame: 1950-1969 Sample Characteristics Prostate cancer mortality data from the 3,056 contiguous U.S. counties Variables SES: county number of school years completed by adult population, county income; Demographic: age, race-ethnicity (nonwhite, white), geographic area, county population density, county percent urban Data Sources U.S. Population Census Abstract Age-adjusted rates of mortality from prostate cancer during 1950-69 were correlated by race with demographic, industrial, and agricultural data from 3,056 U.S. counties. Mortality among non- whites was 50% higher than that among whites in all parts of the country where blacks comprise most of the nonwhite population. The rising rate associated with population density among non- whites, but nonwhites, suggested that environmental exposures related to urban living may account for the predisposition of Amer- ican blacks to prostate cancer. Despite a clustering of counties with elevated mortality in certain North Central and Northeastern States, the geographic variation among whites with prostate tumors was considerably less than that among whites with other tumors. Mortality was elevated in counties with a high percentage of residents of Scandinavian descent, in counties with metal-using and textile industries, and in regions with high consumption of high-fat foods. 15 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Blattner, W.A.; Jacobson, R.J.; Shulman, G. Multiple Myeloma in South African Blacks Lancet 1(8122):928-929, 1979 20 Subjects Cancer Site: multiple myeloma; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate; Time Frame: 1973-1975 Sample Characteristics 56 multiple myeloma cases (blacks 29; whites 27): Johannesburg, South Africa Variables Demographic: age, sex, race-ethnicity Data Sources Histopathology records Abstract The study investigates whether the low rates of multiple myeloma reported for blacks in Africa are related to underdiag- nosis. Cases of multiple myeloma reported in Johannesburg, South Africa between 1973 and 1974 for whites and blacks were examined. Age-adjusted rates per 100,000 for South African black males were 7.47, in South African females 5.11, and among U.S. blacks the figures were 7.52 for males and 5.17 for females. The data may underestimate the incidence in Johannesburg blacks since some individuals may not seek conventional medical care. The age-ad- justed rates per 100,000 for Johannesburg whites were 7.84 for males and 3.34 for females, higher than the corresponding rates for U.S. whites. Local experience suggests the possibility of over- ascertainment because of non-residents giving a Johannesburg ad- dress while receiving medical treatment in that district. Blacks, because of their residency cards, are more objectively classified as to residence. The results of this study suggest that the high rates for myeloma in blacks is not limited to America. Blattner, W.A.; Kalyanaraman, V.S.; Robert-Guroff, M.; et al. The Human Type-C Retrovirus, HTLV, in Blacks From the Caribbean Region, and the Relationship to Adult T-Cell Leukemia/Lymphoma Int J Cancer 30:257-264, 1982 21 Subjects Risk Factor Exposure: other diseases; Cancer Site: leukemia, lymphoma; Biologic Marker: human T-cell leukemia/lymphoma virus (HTLV) Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 35 individuals (cancer cases 17; non-cancer cases 18): blacks; Caribbean ancestry Variables Demographic: age, sex, place of birth Data Sources Laboratory test, hospital medical records Abstract Type-C RNA tumor viruses have been implicated in the etiology of naturally occurring leukemia and lymphomas of animals. Human T-cell leukemia/lymphoma virus (HTLV) is the first human virus of this class consistently identified in association with a spe- cific type of human leukemia/lymphoma. The isolation of HTLV was made possible by the ability to grow mature T-cells in tissue culture usually with T-cell growth factor (TCGF). We now report a cluster of adult T-cell leukemia/lymphoma among Blacks from the Caribbean in which all eight cases are positive for HTLV virus and/or antibody. These patients have disease that appears indistin- guishable from Japanese adult T-cell leukemia/lymphoma which, as we have also reported, is associated with HTLV in over 90% of cases. The finding of HTLV antibodies in some of the normal population in the Caribbean and Japan, and the clustering of a specific form of T-cell leukemia/lymphoma in these virus-endemic areas, suggest that HTLV infection may be associated with the occurrence of a distinctive clinico-pathologic entity. Bloch, B.; Atad, J. Gestational Trophoblastic Disease: The Significance of Vaginal Metastases 22 Gynecol Oncol 16:169-178, 1983 Subjects Treatment; Risk Factor Exposure: reproductive factors; Cancer Site: gestational trophoblastic disease Study Characteristics Design: case study Sample Characteristics Six individuals (blacks 4; whites 2): female; South Africa Variables Demographic: parity, type of delivery Abstract Six patients with gestational trophoblastic disease whose pre- senting symptom was hemorrhage from vaginal metastases are discussed. The clinical features, management, and response to treatment are outlined and it is suggested that the presence of vaginal metastases should be regarded as a poor prognostic factor and an indication of multiple cytotoxic agent chemotherapy. 16 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Blot, W.].; Fraumeni, J.F., Jr. Geographic Patterns of Bladder Cancer in the United States J Natl Cancer Inst 61(4):1017-1023, 1978 23 Subjects Risk Factor Exposure: industry, geographic area; Cancer Site: bladder; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: age-specific mortality rate, age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1950-1969 Sample Characteristics Bladder cancer mortality data from the 3,056 contiguous counties in the United States Variables SES: education, industry, income; Demographic: sex, race- ethnicity, geographic area, population density, urban-rural Data Sources Death registration Abstract Age-adjusted rates of mortality during 1950-69 from bladder cancer were correlated with demographic and industrial indexes for the 3,056 counties of the contiguous United States. Rates among whites and nonwhites of both sexes rose sharply with urbaniza- tion. A small but positive socioeconomic gradient was observed, and mortality was slightly higher among males in counties with high percentages of British and German residents. Even after con- trolling for demographic variables, the Northeastern excess of blad- der cancer among whites was sizeable, whereas the regional differ- ences among nonwhites were small. The high rates in the North- east were seen in both sexes and in rural as well as urban areas, with mortality in small counties in upstate New York and New England equaling or exceeding those in large metropolitan centers elsewhere in the country. Outside the Northeast, high rates were generally limited to urban areas, but clusters of elevated mortality occurred among white males along the Illinois-Wisconsin border, in parts of lower Michigan, and in southern Louisiana. Industrial factors may explain at least part of the geographic clustering, inas- much as rates among males were significantly higher in U.S. coun- ties where the chemical industry is heavily concentrated. Increases were also associated with the printing industry, but correlations with 16 other major manufacturing industries were near or below expected levels. Blot, W.].; Fraumeni, J.F., Jr. Geographic Patterns of Oral Cancer in the United States: Etiologic Implications J Chron Dis 30:745-757, 1977 24 Subjects Risk Factor Exposure: alcohol, tobacco, geographic area; Cancer Site: oral cavity, pharynx Study Characteristics Design: cross-sectional; Measures: age-adjusted mortality rate, higher level statistics; Time Frame: 1950-1969 Sample Characteristics Oral cavity and pharynx cancer mortality data from the 3,056 contiguous counties of the United States Variables SES: industry, education, income; Demographic: sex, race- ethnicity (nonwhite, white), geographic area, urban-rural Data Sources U.S. Population Census 17 Abstract A survey of buccal cavity and pharynx cancer mortality in United States counties, 1950-1969, revealed elevated rates among males in urban areas throughout the country and among females in the rural South. Indirect evidence suggests that the urban gra- dient in males results largely from alcoholic beverage consumption and tobacco smoking, while the Southern excess in females is related primarily to oral snuff use. Mortality was elevated also in counties with leather, paper, and chemical manufacturing indus- tries (in males), and apparel and textile industries (in females). The industrial correlations in both sexes remained after control for urbanization and other demographic factors, but snuff use by Southern women in the textile and apparel industries may have influenced this association for females. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Blot, W.].; Fraumeni, J.F., Jr.; Stone, B.]. Geographic Correlates of Pancreas Cancer in the United States J Natl Cancer Inst 42:373-380, 1978 25 Subjects Risk Factor Exposure: alcohol, occupation, other diseases, geographic area; Cancer Site: pancreas; Cancer Measures: mortality Study Characteristics Measures: age-specific mortality rate, age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1950-1969 Sample Characteristics Cancer mortality data from the 3,056 counties of the contiguous United States Variables SES: industry; Demographic: sex, geographic area, population size, race-ethnicity (nonwhite, white) Data Sources Death registration, U.S. Population Census Abstract Age-adjusted death rates for pancreas cancer during 1950-69 were correlated by sex and race with demographic and industrial data for the 3,056 counties of the contiguous United States. Only a small fraction of the county-to-county variation in mortality was explained by these variables, in contrast to their strong correlation with other common neoplasms. The only geographic cluster occur- red in an area encompassing parts of Louisiana and Mississippi. Throughout the country, however, the rates for pancreas cancer were higher in urban areas, especially in males, and in counties with many residents of Scandinavian and East European (particu- larly Russian) descent. No associations were found with socioeconomic, industrial, or alcohol-consumption indices. The mortality patterns for pancreas and lung cancers were highly cor- related in males, suggesting the influence of tobacco consumption on both tumors. In females, pancreas cancer was significantly cor- related with diabetes mellitus, consistent with other evidence link- ing these two diseases. Blot, W.].; Fraumeni, J.F., Jr.; Young, J.L., Jr. Letters to the Editor: Left-sided Breast Cancer Lancet I1:762-763, 1977 26 Subjects Risk Factor Exposure: familial factors; Cancer Site: breast; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1973-1975 Sample Characteristics 20,000 breast cancer cases from SEER Program areas Variables Demographic: race-ethnicity, menopausal status; Cancer: histology, tumor location Data Sources SEER cancer registry Abstract This letter to the editor was written to support the observation of left-sided predominance of breast cancer. It was found that the laterality effect was primarily seen around and after menopause, when breast cancer patterns appear to reflect environmental fac- tors. Blumenthal, N.J. Comparative Ratios of Carcinoma of the Vagina and Cervix in Different Population Groups S Afr Med J 59:597-598, 1981 27 Subjects Risk Factor Exposure: reproductive factors, sexual practices; Cancer Site: vagina, cervix uteri; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1960-1979 Sample Characteristics 10 vagina cancer cases: black, female, hospital patients; South Africa Variables Demographic: age, age at menarche, fertility, number of pregnancies Data Sources Hospital medical records, histopathology records Abstract A retrospective study of primary squamous carcinoma of the vagina at Baragwanath Hospital over a period of 20 years shows that the incidence of this condition is much lower than that re- ported in other series. In addition, the ratio of primary carcinoma of the vagina to that of the cervix is much lower than in comparable studies. The age distribution, aetiology and the sexual characteris- tics of the Black population are discussed. 18 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Boles, R.S.; Ipsen, J., Jr.; Ruedy, J. A Comparative Study of the Frequency of Carcinoma of the Stomach 28 Cancer 14(12):1584-1588, 1963 Subjects Cancer Site: stomach; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: descriptive statistics; Time Frame: 1867-1959 Sample Characteristics 675 stomach cancer deaths: Philadelphia, PA Variables Demographic: age, sex, race-ethnicity (Negro, white); Cancer: clinical diagnosis Data Sources Autopsy records Abstract Six hundred seventy-five cases of carcinoma of the stomach verified at autopsy at the Philadelphia General Hospital, Philadel- phia, Pa., between 1867 and 1959 are described in respect to fre- quency among all autopsies performed in that period. These autop- sies present a comparatively large fraction of all deaths occurring in the City of Philadelphia, and the frequency found from 1950-1959 is not markedly different from the proportion of this disease among all deaths in Pennsylvania during the same period. Age-standar- dized frequencies of carcinoma of the stomach show a decline in white men and in all women, while there was an increase in male Negroes that contrasts markedly with the decline in other groups. The correctness of clinical diagnosis increased from 54% in the 1920's to 74% in the 1950's. This significant increase has important bearing on interpretation of mortality statistics from death certifi- cates based on clinical diagnosis. Bradshaw, E.; Harington, J.S. A Comparison of Cancer Mortality Rates in South Africa with Those in Other Countries 29 S Afr Med J 19:943-946, 1982 Subjects Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: age-standardized mortality rate; Time Frame: 1970-1974 Sample Characteristics Cancer mortality data from South Africa Variables Demographic: sex, race-ethnicity (black, colored, white, Asian), urban-rural, geographic area Abstract Standardized mortality rates for eight major types of cancer (as well as for all types of cancer considered together) in four South African population groups are compared with each other and with rates in selected countries. Wide differences among the races in South Africa are displayed, and several rates are shown to parallel those in other population groups abroad. This lends support to the belief that socio-economic factors (rather than solely genetic factors) play the major role in cancer causation. Bras, G.; Cole, H.; Ashmeade-Dyer, A.; et al. Report on 151 Childhood Malignancies Observed in Jamaica 30 J Natl Cancer Inst 43:417-421, 1969 Subjects Risk Factor Exposure: geographic area; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-specific incidence rate, descriptive statistics; Time Frame: 1958-1967 Sample Characteristics 151 cancer cases (males 93; females 58): 0-14 years old; Jamaica Variables Demographic: age, sex; Cancer: histology Data Sources non-SEER population-based cancer registry, country population census Abstract During a 10-year period 1958-67, 151 tumors were noted in children under 15 years of age in Jamaica. They comprised 2.9% of all malignant tumors observed. Leukemia was the most common form of cancer. The incidence of retinoblastomas was high and surpassed incidences of neuroblastomas and of Wilms’ tumors. Kaposi's sarcoma was not observed, and hepatocellular carcinoma was extremely rare. The tumor incidence pattern in Jamaica there- fore largely resembles that in the United States and in Western Europe. It contrasts significantly with that in Uganda and in West- ern Nigeria. This merits attention, since over 90% of the Jamaican population is either pure Negro or of predominantly Negro descent as a result of importation from West Africa in former days. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Brenner, D.E. Case Report: Carcinoma of the Cervix—A Review Am J] Med Sci 284(1):31-48, 1982 31 Subjects Treatment; Diagnosis; Cancer Site: cervix uteri Study Characteristics Design: case study; Time Frame: 1976-1981 Sample Characteristics Three cervix uteri cancer cases (blacks 2; whites 1): female, hospital patients; Baltimore, MD Data Sources Hospital medical records Abstract Carcinoma of the cervix is the fourth most common neoplasm in women. The mortality from this tumor has dropped with the advent of Papanicolaou smears and routine periodic screening, particularly in high risk populations. Diagnosis and staging in- cludes a careful physical examination, the use of colposcopy, di- rected biopsy, intravenous urogram and cystoscopy. Computed tomography and lymphangiography may be helpful for detection of iliac or paraaortic lymph nodes. Early, noninvasive stages of this disease (CIN) may be treated with cryosurgery or laser vapori- zation. Carcinoma in situ (CIS) and microinvasive carcinoma is usually treated with simple hysterectomy for cure. More advanced invasion localized to the cervix may be treated with radical hys- terectomy or radiation therapy with 90% of patients surviving 5 years. More advanced tumors are treated with external and intra- cavitary radiation therapy. For patients with paraaorticlymph node involvement or recurrent tumor, 5-year survival is less than 10%. Chemotherapy may provide some palliation to patients with recur- rent tumors but does not increase long term survivorship. Brinton, L.A.; Hoover, R.; Jacobson, R.R.; et al. Cancer Mortality Among Patients with Hansen's Disease J Natl Cancer Inst 72(1):109-114, 1984 32 Subjects Risk Factor Exposure: treatment modalities; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: prospective (nonconcurrent); Measures: standardized mortality ratio(SMR), descriptive statistics, higher level statistics; Time Frame: 1939-1977 Sample Characteristics 1,678 Hansen's disease cases (blacks 163; whites 564; Hispanics 709; Asian-Pacific Islanders 232; American Indian-Alaskan Natives 3; Unknowns 7): United States Variables Demographic: age, sex, race-ethnicity, birthplace Data Sources Hospital medical records, death registration Abstract For the evaluation of cancer risks associated with immunodefici- ences experienced by patients with Hansen's disease (leprosy) and for the assessment of possible effects of dapsone therapy, a follow- up study was conducted of 1,678 patients admitted to the National Hansen's Disease Center in Carville, La., between 1939 and 1977. Overall, no substantial cancer mortality was observed (standar- dized mortality ratio=1.3), nor was there any excess among pa- tients exhibiting defects in cellular immunity by virtue of lepromat- ous forms of the disease. Notable was the absence of any significant excess of lymphoma (5 observed vs. 2.3 expected), despite the predominance of this tumor in certain other immunodeficiency states. Several cancer sites (oral, bladder, and kidney) occurred excessively, but reasons for the elevations were obscure. Although dapsone has been implicated as a carcinogen in laboratory animals, the use of sulfones, including dapsone, did not appear to affect significantly the risk of any cancers in this population. Browning, D.; Gross, S. Epidemiological Studies of Acute Childhood Leukemia: A Survey of Cuyahoga County, Ohio 33 Am ] Dis Child 116:576-585, 1968 Subjects Risk Factor Exposure: familial factors; Cancer Site: leukemia; Cancer Measures: incidence, mortality Study Characteristics Design: cross-sectional; Measures: age-specific incidence rate, descriptive statistics, higher level statistics; Time Frame: 1955-1965 Sample Characteristics 169 acute leukemia cases (nonwhites 15; whites 153) and 174 leukemia deaths: 0-14 years old; Cuyahoga County, OH Variables SES: religion, socioeconomic background; Demographic: age, sex, race-ethnicity, urban-rural, geographic area Data Sources Death registration, hospital medical records Abstract A ten-year epidemiological investigation of all the cases of child- hood leukemia in Cuyahoga County, Ohio, diagnosed from the newborn through 14 years of age, during the years 1955 to 1965, was undertaken in an attempt to study the following variables: peak incidence, sex, race, age of onset, socioeconomic background, geographical distribution, genetic determinants, seasonal inci- dence, and temporalspatial relationships. Significant findings in- cluded a lower incidence of leukemia among Negroes than white children, a lower incidence of leukemia among urban versus sub- urban dwellers regardless of race, and a lower incidence of leukemia among children of low-income families versus those from upper economic families also regardless of race. Neither seasonal variations nor significant case clusters were observed. 20 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Burbank, F.; Fraumeni, J.F., Jr. U.S. Cancer Mortality: Nonwhite Predominance 34 J Natl Cancer Inst 49:649-659, 1972 Subjects Cancer Site: multiple sites; Cancer Measures: mortality an increasing nonwhite-to-white ratio of age-adjusted death rates. Study Characteristics Non-white predominance began in 1950 for females and in 1956 Design: time trend; Measures: age-adjusted mortality rate; Time for males. In females the shift resulted from a decline in death Frame: 1950-1967 rates among whites, as rates for nonwhites remained unchanged, Sample Characteristics and in males from rates which rose over time in both races, but Mortality data from the United States more rapidly in nonwhites. Each cancer site is analyzed with re- Variables spect to its race differential and change in nonwhite mortality. The Demographic: race-ethnicity (nonwhite, white), sex, age shift to nonwhite predominance in cancer mortality is probably Data Sources due to many factors, but parallel trends in cancer incidence suggest Death registration that nonwhites are more exposed or vulnerable than whites to Abstract carcinogenic factors. Study of cancer mortality trends in the United States revealed Byers, T.E.; Vena, J.E.; Rzepka, T.F. Predilection of Lung Cancer for the Upper Lobes: An Epidemiologic Inquiry 35 J Natl Cancer Inst 72(6):1271-1275, 1984 Subjects Abstract Cancer Site: lung; Cancer Measures: incidence Data from 24,798 cases of primary neoplasm of the lung regis- Study Characteristics tered by the Surveillance, Epidemiology, and End Results Program Design: cross-sectional; Measures: descriptive statistics, higher between 1973 and 1977 were examined for the demographic deter- level statistics; Time Frame: 1973-1977 minants of lobe of origin. Lung cancer occurred predominantly in Sample Characteristics the upper lobes in both sexes, in both blacks and whites, and at 23,730 lung cancer cases (blacks 2,561; whites 21,169): precise all ages. This pattern was observed for squamous cell and small lobe origin specified in cancer register; SEER Program areas cell carcinoma as well as adenocarcinoma. The predominance of Variables lung cancer in the upper lobes was much stronger in the young Demographic: age, sex, race-ethnicity, geographic area; Cancer: and somewhat stronger in blacks. These findings are discussed in histology, tumor location light of current theories of pulmonary carcinogenesis. Current Data Sources theories may not explain entirely these observed patterns, and SEER cancer registry further inquiries into this question, therefore, may be useful. Catovsky, D.; Greaves, M.F.; Rose, M.; et al. Adult T-Cell Lymphoma-Leukaemia in Blacks from the West Indies 36 Lancet 1(8273):639-642, 1982 Subjects Risk Factor Exposure: other diseases; Cancer Site: lymphoma, were lymphadenopathy and a high white blood-cell count with a leukemia; Cancer Measures: survival predominance of pleomorphic lymphoid cells with pronounced Study Characteristics nuclear irregularities prominent at ultrastructural level. The cells Design: case-study; Time Frame: 1977-1981 in all cases formed rosettes with sheep red blood-cells and lacked Sample Characteristics terminal transferase. Analysis with OKT monoclonal antibodies Six adult T-cell lymphoma-leukemia cases (males 2; females 4): in four cases confirmed a mature T-cell phenotype defined as black, 21-55 years old; born in West Indies helper/inducer (T4+, T6-, T8-) in three. Combination chemo- Variables therapy resulted in short-lived remissions; four patients died and Demographic: age, sex; Cancer: histology, hypercalcemia two have survived 3-6 months. The disease in these patients is Data Sources indistinguishable on clinical and pathological grounds from adult Hospital medical records, laboratory test T-cell leukaemia/lymphoma in Japan. Geographical clustering Abstract among certain racial groups suggests common aetiological factors Six Black patients (five born in the West Indies and one in in the pathogenesis of this disease. The finding of high titer anti- Guyana), aged 21-55 years, had adult T-cell ymphoma-leukaemia body against the structural core protein (p24) of a new human diagnosed in the U.K. This disorder is rare in Europe and the C-type leukaemia virus (human T-cell leukaemia/lymphoma virus) U.S.A, but is more common in Japan. Five patients had severe in all tested cases from this series and data from all but one case hypercalcaemia which correlated with disease activity, although from Japan suggest that one such factor may be viral. osteolytic lesions were found in only one. Other clinical features 21 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Cedermark, B.].; Blumenson, L.E.; Pickren, J.W.; et al. The Significance of Metastases to the Adrenal Glands in Adenocarcinoma of the Colon and Rectum Surg Gyencol Obstet 144:537-546, 1977 37 Subjects Cancer Site: colon, rectum; Cancer Measures: incidence Study Characteristics Design: retrospective; Measures: descriptive statistics, higher level statistics; Time Frame: 1959-1974 Sample Characteristics 457 autopsy cases (nonwhites 15; whites 442—Colon cancer deaths 251; rectum cancer deaths 206): Buffalo, NY Variables Demographic: age, sex; Cancer: tumor location, stage Data Sources Autopsy records, hospital medical records Abstract A study to determine the frequency of metastases to the adrenal glands following carcinoma of the colon and rectum was underta- ken. Patients with, or without, adrenal spread were compared regarding age, sex, race and survival time. The over-all metastatic pattern was analyzed to determine its usefulness for predicting the presence of metastases to the adrenal glands. Autopsy reports and clinical records of patients with adrenal spread were reviewed regarding the extent of tumor involvement in the glands and pos- sible adrenal insufficiency. Of 457 patients with adenocarcinoma of the colon and rectum who underwent autopsy, 63 or 14 percent had metastasis to the adrenal glands. Of these, 29 had bilateral involvement. Patients with bilateral metastases had alower median age than did those without adrenal spread. No correlation was found between adrenal metastatic status and sex or race. Although survival time was found to be shorter for patients with bilateral metastases of the adrenal glands, adrenal insufficiency did not seem to be the reason for this shorter survival time. In retrospect, however, the presence of adrenal insufficiency could not be ruled out in several of these patients. Charles, N.C.; Stenson, S.; Taterka, H.B. Epibulbar Malignant Melanoma in a Black Patient Arch Ophthalmol 97:316-318, 1979 38 Subjects Cancer Site: malignant melanoma; Cancer Measures: incidence Study Characteristics Design: case study Sample Characteristics One malignant melanoma case: black, female, 53 years old Abstract A 53-year-old black woman showed increasing epibulbar pig- mentation following recurrent episcleritis at the same location. A biopsy was refused until vision was lost and a large black mass protruded from the globe and invaded the superficial cornea. Exenteration was performed and histological examination dis- closed a malignant melanoma of the cornea with an adjacent intra- epithelial conjunctival melanoma. We believe that the tumor arose in an area of acquired melanosis of the perilimbal conjunctiva and that secondary spread into adjacent superficial cornea occurred. Epibulbar malignant melanomas are uncommon conjunctival tumors in the general population and are especially rare in blacks. Chopra, S.A.; Chopra, F.S. Cancer in the Africans and Arabs of Zanzibar Int J Cancer 19:298-304, 1977 39 Subjects Diagnosis; Risk Factor Exposure: diet-nutritional factors, alcohol; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: time trend, cross-sectional; Measures: descriptive statistics; Time Frame: 1957-1967 Sample Characteristics 392 cancer cases (Africans 305; Arabs 87): Zanzibar Variables Demographic: sex Data Sources Histopathology records Abstract Among 392 cancers histologically diagnosed in Zanzibar during 1957-62 and 1964-67, a definite pattern seems to emerge. Skin and 22 cervix cancers were the most common types in both Africans and Arabs. Skin cancer was predominantly of the squamous cell type. The Zanzibar Arabs thus appear to be protected against basal cell carcinoma which in the Arab desert community has been diag- nosed with about the same frequency as squamous-cell carcinoma. In the same manner, the Zanzibar Arab immigrants seem to have a reduced risk for stomach and oesophageal cancers, which are common in other Arab countries. This is probably because Arabs in Zanzibar have adopted the dietary habits and other customs of the Zanzibar Africans in whom cancer of the alimentary tract seems to be uncommon. On the other hand, unlike Zanzibar Africans, the Arabs have an increased risk for Hodgkin's disease similar to that of the Middle East Arab population. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Christopherson, W.M.; Connelly, P.].; Alberhasky, R.C. Carcinoma of the Endometrium: V, An Analysis of Prognosticators in Patients with Favorable Subtypes and Stage I Disease Cancer 51:1705-1709, 1983 40 Subjects Treatment; Cancer Site: corpus uteri; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1971-1981 Sample Characteristics 634 corpus uteri cancer cases (blacks 36; whites 598): female, Stage I disease; Louisville, KY Variables Demographic: age, race-ethnicity; Cancer: histology, tumor grade Data Sources non-SEER population-based cancer registry Abstract A review of the pathologic material from patients in our popu- lation- based uterine cancer registry in Louisville, Kentucky iden- tified three subtypes of endometrial carcinoma with a favorable prognosis: adenoacanthoma, adenocarcinoma with no specific fea- tures, and secretory carcinoma. Three subtypes with a much less favorable prognosis were papillary carcinoma, mixed adenosquam- ous carcinoma, and clear cell carcinoma. The other important de- terminants of treatment results were: stage of disease, age, race, nuclear grade, and depth of myometrical invasion. This study evaluated these prognostic determinants for the favorable subtypes with Stage I disease that were fully treated. There were 595 patients eligible for five-year and 380 for ten-year vital status evaluation. Only three patients were lost to follow-up at five years and four at ten years. The five-year survival was 88.1% and the ten-year 74.5%. At five years 4.4% were dead of disease and at ten years 6.3%. There were 115 women younger than age 50. Their five-year survival was 98.3% and at ten years 94.6%. None was dead of disease at five years and only one at ten years. Black women had a significantly lower survival than did white women, but no black woman younger than age 50 died of disease. Nuclear grade was the most important histologic determinant of survival, followed by depth of myometrial invasion. Recommendations for treatment planning were formulated based on these prognosticators. Christopherson, W.M.; Mendez, W.M.; Ahuja, E.M.; et al. Cervix Cancer Control in Louisville, Kentucky Cancer 26:29-38, 1970 41 Subjects Screening: Pap smear; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: crude incidence rate, age-adjusted incidence rate, descriptive statistics; Time Frame: 1956-1967 Sample Characteristics 1,470 cervix uteri cancer cases (non-Caucasians 308; Caucasians 1,162): female, 20+ years old; Kentucky Variables SES: census tract income; Demographic: age, race-ethnicity; Cancer: histology, stage Data Sources Death registration, histopathology records, hospital medical records, U.S. Population Census Abstract The results of 12 years of mass cytologic screening of a commu- 23 nity for protection from uterine cancer are reported, and the effects upon incidence rates of the various component cancers were analyzed. A preceding 3-year period during which no screening was performed was used to calculate the expected rates. Total cancers of the uterus, including carcinoma in-situ, had an average annual rate of 125.5 per 100,000 women 20 years of age and older during the 15-year period. During the last 3-year period, carcinoma of the cervix was diagnosed at an earlier stage, and the rates decreased by one third. Carcinoma in-situ increased in proportion to screening activity. There was an increase in the rate for endomet- rial carcinoma. The greatest response to screening was in women under 50 years of age. This group showed the greatest decrease in rates and was, for the most part, diagnosed at an earlier stage of the disease. On the other hand, older women were less inclined to avail themselves of cytologic screening. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Christopherson, W.M.; Mendez, W.M.; Parker, ].E.; et al. Carcinoma of the Endometrium: A Study of Changing Rates over a 15-Year Period Cancer 27(5):1005-1008, 1971 42 Subjects Screening: Pap smear; Cancer Site: corpus uteri; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: crude incidence rate, age-adjusted incidence rate; Time Frame: 1953-1967 Sample Characteristics 707 endometrium cancer cases (Caucasians 643; non- Caucasians 64): 20+ years old; Louisville-Jefferson County, KY Variables Demographic: age, sex, race-ethnicity; Cancer: histology, symptoms Data Sources non-SEER population-based cancer registry, histopathology records, U.S. Population Census Abstract A 15-year study of endometrial carcinoma in Louisville-Jeffer- son County, Kentucky, revealed an increase of 26.3% in average yearly rates when the first 3 years, 1953-1955, were compared with the last 3 years, 1965-1967. An increase of 15.9% persisted when the rates were corrected for aging of the population; this was not statistically significant. Thus the mass cytologic screening program which had a beneficial effect on the rates for invasive cervix cancer, failed to have such an effect on endometrial carcinoma. Christopherson, W.M.; Nealon, N.A. Uterine Cancer: A Comparative Study of Black and White Women 43 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 185-195 Subjects Screening: Pap smear; Diagnosis; Cancer Site: cervix uteri, corpus uteri; Cancer Measures: incidence, mortality Study Characteristics Design: time trend, prospective; Measures: age-specific incidence rate, age-adjusted incidence rate, age-adjusted mortality rate; Time Frame: 1953-1977 incidence data, 1956-1977 screening program data Sample Characteristics 4,504 cervix uteri cancer cases and 74,924 individuals initially cytologically screened (blacks 33,756; whites 41,168): female; Louisville, KY Variables Demographic: race-ethnicity, sex; Cancer: histology Data Sources non-SEER population-based cancer registry, hospital medical records Abstract Blacks are at higher risk than are white women for carcinoma in situ, microinvasive carcinoma, and invasive squamous car- cinomas. Contrary to popular belief, black and low-income women were no more difficult to recruit for screening than were other segments of the population. In spite of the lower incidence rates, the death rates from endometrial cancer were appreciably higher in black women. Deaths from uterine sarcoma were also appreci- ably higher in black women. Christopherson, W.M.; Parker, J.E. Control of Cervix Cancer in Women of Low Income in a Community Cancer 24(1):64-69, 1969 4a Subjects Screening: Pap smear; Cancer Site: cervix uteri; Cancer Measures: incidence, mortality Study Characteristics Design: prospective; Measures: descriptive statistics; Time Frame: 1956-1968 Sample Characteristics 37,209 individuals with satisfactory pap smears (Negroes 18,787; Caucasians 18,018): female, 15+ years old, low income; Louisville, KY Variables Demographic: age; Cancer: histology Data Sources Hospital clinic records Abstract A program to control cancer of the uterine cervix in women of 24 a low-income level in Louisville (Jefferson County), Kentucky over a 12-year period is described. Initial satisfactory examinations were obtained on 37,209 women. Rescreenings at an interval of 1 to 12 years totaled 51,063 examinations. The population was equally divided between Negroes and Caucasians. Prevalence rates of cytologically detected cervix carcinoma were very high—3.79 per 1000 for invasive squamous carcinoma and 4.41 for carcinoma in situ. The first repeat examination based on patient years fell to 0.67 and 1.46 per 1000 women for these 2 lesions. The decrease continued through subsequent screenings. Stage I carcinoma in- creased to 63.93% of the total cases during the years of screening. We conclude from the data presented that mortality rates from cervix cancer will have to reflect the decrease in incidence and the increase in early diagnosed cases. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Cohen, C.; Berson, S.D.; Geddes, E.W. Hepatitis B Antigen in Black Patients with Hepatocellular Carcinoma: Correlation Between Orcein Stained Liver Sections and Serology Cancer 41:245-249, 1978 45 Subjects Risk Factor Exposure: other diseases; Cancer Site: liver; Biologic Marker: hepatitis B surface antigens (HBsAg), hepatitis B virus (HBV) associated antigens Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 50 hepatocellular carcinoma cases: black, male, mineworkers; South Africa Variables Cancer: histology Data Sources Autopsy reports, laboratory tests Abstract Formalin-fixed paraffin-embedded autopsy tissue of liver and tumor from 50 male black mineworkers with hepatocellular car- cinoma were examined by orcein stain for the presence of cytop- lasmic hepatitis B surface antigen. The results were correlated with the serum hepatitis B antigen (HBAg). In 72% serum HBAg was positive. Orcein staining of nontumor liver cell cytoplasm was present in 18 (36%). Sixteen (89%) of these orcein-positive cases were serum HBAg positive. The two false negative serum HBAg results were obtained by immunodiffusion, immunoelec- trophoresis and complement fixation. Serum HBAg, measured by radio-immunoassay and hemagglutination, was positive in 14 or- cein-negative cases. Six other negative orcein results appeared to be due to sampling error. Orcein staining was noted in tumor cells of three serum HBAg sampling error. Orcein staining was noted in tumor cells of three serum HBAg positive patients. Provided the limitations of the technique are realized, orcein staining of liver tissue from hepatocellular carcinoma patients may prove use- ful for retrospective screening surveys to assess the prevalence of HBAg positively in these patients. Cohen, C.; Berson, S.D.; Shulman, G.; et al. Immunohistochemical Ferritin in Hepatocellular Carcinoma Cancer 53(9):1931-1935, 1984 46 Subjects Risk Factor Exposure: other diseases; Cancer Site: liver; Biologic Marker: serum ferritin Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 93 hepatocellular carcinoma cases (southern African blacks 74; Americans 17): South Africa, Pennsylvania Variables Demographic: geographic area; Cancer: histology Data Sources Laboratory test, autopsy records, histopathology records Abstract Serum ferritin concentrations are elevated in 35% to 100% of patients with hepatocellular carcinoma (HCC). With an im- munoperoxidase technique, ferritin was demonstrated in tumor tissue from 32 of 74 (43%) black southern African patients, and from 12 of 19 (63%) American patients with HCC (p> 0.1). Ferritin was present in nonneoplastic liver in 82% of African and 100% of American patients (p> 0.1). Moderate to large amounts of stainable hepatic storage iron (hemosiderin) were present in 76% of African and 67% of American patients (p> 0.1). Fifty-two (70%) African patients had macronodular cirrhosis. In the literature, 80% to 90% of American patients with HCC have cirrhosis. High serum ferritin levels in patients with HCC may be due to ferritin production by the tumor, or related to the associated iron overload and/or cir- rhosis. Cohen, 1.K. Reconstruction of the Nipple-Areola by Dermabrasion in a Black Patient 47 Plast Reconstr Surg 67(2):238-239, 1981 Subjects Treatment; Cancer Site: breast Study Characteristics Design: case study; Intervention Study Sample Characteristics One breast cancer case: black, female, 42 years old; Virginia Data Sources Physician medical records Abstract Based on the biological phenomenon that hyperpigmentation of skin usually occurs in blacks following partial-thickness skin ‘removal, a new method of areola reconstruction for black patients is presented that is a safe, simple, rapid outpatient procedure which avoids a donor site and the possibility of graft loss. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Cole, S. Pelvic Cancer in Cincinnati: A 5-Year Study Obstet Gynecol 23(2):274-278, 1964 48 Subjects Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: crude incidence rate, descriptive statistics; Time Frame: 1957-1961 Sample Characteristics 1,500 cancer cases: female, hospital patients; Cincinnati, OH Variables SES: census tract income; Demographic: age, race-ethnicity (Negro, white, Jewish), geographic area; Cancer: histology Data Sources non-SEER hospital cancer registry, hospital medical records Abstract A 5-year study (1957-1961) was made of all female pelvic cancer admissions into Greater Cincinnati hospitals. All patients consi- dered in this report had a proven primary tissue microscopic diag- nosis. In 1500 patients 1509 cancers were found. Of all cervical malignancies 31.4 percent were detected in the in-situ stage; for white patients, 34.6 percent, and for Negro patients, 20.0 percent of cervical malignancies were detected in the in-situ stage. Adenocarcinoma of the cervix represented 2.9 percent of all cervical cancers. The annual incidence for all women with cervical malig- nancy was 29.7 per 100,000; for white women it was 26.4 per 100,000; for Negro women, 49.4 per 100,000; and for Jewish women, 2.9 per 100,000. Therefore, the incidence of cervical cancer in Negro women was twice that of white women. The incidence of cervical cancer in white non-Jewish women was 10 times that of Jewish women. The ratio of cervical to corporeal cancer for all patients was 1.8:1. The ratio of cervical to corporeal cancer for white women was 1.6:1, and for Negro women 4.6:1. The average annual incidence for genital malignancies is shown and compared with figures for New York State and Connecticut. With respect to cervical malignancies, the higher the socioeconomic status, the lower the incidence of cancer; this held true for both races. Nine patients had 2 or more primary genital malignancies. Connelly, P.J.; Alberhasky, R.C.; Christopherson, W.M. Carcinoma of the Endometrium III: Analysis of 865 Cases of Adenocarcinoma and Adenoacanthoma Obstet Gynecol 59:569-574, 1981 49 Subjects Treatment; Risk Factor Exposure: reproductive factors; Cancer Site: corpus uteri; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 10 years Sample Characteristics 865 endometrium cancer cases (blacks 72; whites 793): female, 20+ years old; Louisville, KY Variables Demographic: age, race-ethnicity, age at menopause; Cancer: histology, stage, tumor grade, symptoms Data Sources non-SEER population-based cancer registry, histopathology records, hospital medical records Abstract Eight hundred sixty-five patients with confirmed adenocar- cinoma of the endometrium were studied. Eight hundred eleven 26 were followed up for at least 5 years or until death, and 537 were followed up for 10 years. The survival rate was 77.1% at 5 years and 61.0% at 10 years. Age at diagnosis was the single most impor- tant clinical determinant of survival. Tumor grade was an impor- tant predictor of the course of disease, and nuclear grade was a significantly more accurate indicator than was histologic grade. Stage and depth of invasion were also important predictors of survival. The 5-year survival rate for black women was only 41.9%. Stage for stage and grade for grade, women undergoing hysterec- tomy alone and with radiation had similar survival rates. Adenoacanthoma had a better prognosis than did adenocarcinoma without squamous metaplasia; there were no deaths at 5 or 10 years in women diagnosed before age 50. The subtype of carcinoma was found to be a highly significant predictor of prognosis. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Cooper, E.H.; Kerruish, S.M.; Gunven, P. Serum Beta-2-Microglobulin in African Burkitt Tumour, Nasopharyngeal Carcinoma and Controls Biomedicine 35:104-107, 1981 50 Subjects Biologic Marker: beta 2-microglobulin; Risk Factor Exposure: other diseases; Cancer Site: Burkitt's tumor, nasopharyngeal, head and neck Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 248 individuals (Africans 125; Caucasians 123—Cases 108; controls 141): Uganda; Gambia; Nairobi, Kenya Variables Demographic: age, race-ethnicity Data Sources Laboratory test Abstract Serum beta-2-microglobulin measurements were made in Bur- kitt’s lymphoma and nasopharyngeal carcinoma in Africans and infectious mononucleosis in Europeans. There was a significant increase of beta-2-microglobulin levels in active Burkitt's tumour compared to those in remission but young West African children have higher levels than that in Caucasians. Nasopharyngeal car- cinoma can be associated with a raised beta-2-microglobulin but the distribution is similar to that seen in other head and neck cancers. In 27 cases of infectious mononucleosis 74% have raised levels of serum beta-2-microglobulin. The mechanism of produc- tion of high serum beta-2-microglobulin levels in rural African children is as yet unknown. The study demonstrates that serum beta-2-microglobulin level is limited as a monitor of the progress of Burkitt's tumour. Coovadia, Y.M. Primary Testicular Tumours Among White, Black and Indian Patients S Afr Med J 54:351-352, 1978 51 Subjects Cancer Site: testis; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1952-1977 Sample Characteristics 70 testis cancer cases (blacks 22; whites 24; Indians 24): male, hospital patients; South Africa Variables Demographic: age, race-ethnicity; Cancer: histology Data Sources Hospital medical records Abstract A review of all testicular tumours examined in the Department of Pathology, University of Natal, over the past 25 years has re- vealed a very low incidence in the Black population. In comparison with White and Indian patients, testicular teratomas are particu- larly uncommon, a surprising feature in view of the frequency of ovarian teratomas in Blacks. Correa, P.; Strong, ].P.; Reif, A.; et al. The Epidemiology of Colorectal Polyps: Prevalence in New Orleans and International Comparisons Cancer 39:2258-2264, 1977 52 Subjects Risk Factor Exposure: other diseases; Cancer Site: colon-rectum Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1970-1975 Sample Characteristics 752 deaths (blacks 212; whites 540): New Orleans, LA Variables : Demographic: age, sex, race-ethnicity, geographic area Data Sources Hospital medical records, autopsy records Abstract Based on a study of the prevalence of intestinal polyps in New Orleans and on international comparisons, the following conclu- sions are reached. 1) There is a close parallelism in the epidemiol- ogy of colon cancer and adenomatous polyps. Both conditions are statistically associated with respect to geography, anatomic locali- zation, socioeconomic class, migration experience and time trends. 2) The strength of the association favors the notion of a direct, positive correlation between multiplicity, size and atypia of polyps, and cancer risk, equivalent to “dose-effect.” 3) The epidemiologic findings are coherent with other biologic facts derived from clinical, morphological and experimental studies. 4) Adenomatous polyps are a good epidemiologic indicator of colon cancer risk and their presence should be helpful in advancing from studies of the epidemiology of colon cancer to the epidemiology of precursor lesions. 27 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Cramer, D.W.; Cutler, S.J. Incidence and Histopathology of Malignancies of the Female Genital Organs in the United States Am ] Obstet Gynecol 118(4):443-460, 1974 53 Subjects Cancer Site: cervix uteri, corpus uteri, ovary, vulva; Cancer Measures: incidence Study Characteristics Design: cross-sectional, time trend; Measures: crude incidence rate, age-specific incidence rate, age-adjusted incidence rate, descriptive statistics; Time Frame: 1947-1971 Sample Characteristics Female genital organ cancer incidence data from the SNCS areas; 20,855 newly diagnosed cancers of the female genital organs from the TNCS areas Variables Demographic: age, race-ethnicity, geographic area; Cancer: histology Data Sources Second National Cancer Survey 1947-1948, Third National Cancer Survey 1969-1971 Abstract Data from the Third National Cancer Survey are presented on the 1969-1970 morbidity from cancers of the female genital organs in the United States. Among all invasive cancers of these sites, 38 per cent originated in the corpus luteum; 30 per cent, in the cervix; and 25 per cent, in the ovary. The incidence of cancers of the corpus and ovary was greater in white compared to black persons, whereas the opposite held for cervical cancer. Data from 1947 indicate a decrease in the crude rate for all invasive malignancies of the genital organs in women over 19, from 118 to 89 cases per 100,000. This reflects a decrease in the incidence of invasive cervical cancer, while no significant changes in incidence were felt to have occurred for the remaining sites. Forty per cent of the cases cur- rently reported are in situ lesions, chiefly of the cervix. Finally, several published findings on the relationship of histopathology to age and race were confirmed. Crawford, M.; Chung, E.B.; Leffall, L.D., Jr.; et al. Soft Part Sarcomas in Negroes Cancer 26(3):503-512, 1970 54 Subjects Diagnosis; Treatment; Cancer Site: sarcoma; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics; Time Frame: 1949-1969 Sample Characteristics 59 sarcoma cases (males 20; females 39): Negro, 15-81 years old; Washington, DC Variables Demographic: age, sex; Cancer: histology, tumor location, symptoms, nodal involvement Data Sources Hospital medical records Abstract Fifty-nine patients with soft part sarcomas were seen at Freed- men’s Hospital during the years 1949-1969. The sarcomas were classified as: 1. angiosarcoma; 2. extraosseous chondrosarcoma; 3. dermatofibrosarcoma protuberans; 4. fibromatoses (extra-abdomi- nal desmoids); 5. abdominal desmoids; 6. fibrosarcoma; 7. liposar- coma; 8. leiomyosarcoma; 9. myxosarcoma; 10. malignant neurilemmoma; 11. rhabdomyosarcoma, and 12. synovial sarcoma. Tumors of fibrous tissue origin were most common (69%). There were 39 females and 20 males, with an age range from 15 to 81 years. The most common symptom was a painless mass. The lower extremity was involved most often (24 patients). Blood borne metastases were most frequent, but lymph node metastases also occurred. Angiography was of adjunctive value in 2 patients. Biopsy (incisional or excisional) and treatment by wide local exci- sion, muscle group excision, or amputation were mainstays in management. With exception of desmoids and dermatofibrosar- coma protuberans, 5-year survival was quite low. Prompt biopsy of suspicious masses with appropriate therapy offers best chance in these patients. The only observed racial differences was the apparent greater number of fibrous tissue tumors. 28 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Crichlow, R.W.; Kaplan, E.L.; Kearney, W.H. Male Mammary Cancer: An Analysis of 32 Cases Ann Surg 175(4):489-494, 1971 55 Subjects Diagnosis; Treatment; Cancer Site: breast; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, prevalence rate; Time Frame: 1940-1964 Sample Characteristics 32 breast cancer cases (Negroes 5; Caucasians 27): male, 36-82 years old, hospital patients; Pennsylvania Variables Cancer: histology, tumor location, symptoms, metastases Data Sources Hospital medical records Abstract Experience accumulated over a 25-year period with 32 men with a mammary cancer was analyzed. Their overall survival was 47% (15/32) at 5 years and 38% (8/21) at 10 years. Of those subjected to radical mastectomy, 63% (15/24) survived 5 years, and 50% (8/16) 10 years. None with axillary lymph node metastases lived past the sixth year, and only one reached the 5-year point. Comparison to women concurrently treated demonstrated that men were older and had a higher incidence of nipple discharge and nipple or areolar abnormalities. Fewer men were found suitable for radical mastectomy for both tumor and non-tumor related reasons. Avail- able data show a poorer prognosis for men than women following radical mastectomy due principally to lower salvage among those with axillary lymph node metastases. Additional factors adversely affecting overall prognosis include treatment at a relatively more advanced stage and the greater age and age-related mortality characteristic of men with mammary cancer. Cuenca, C.R.; Becker, K.L. Klinefelter's Syndrome and Cancer of the Breast Arch Intern Med 121:159-162, 1968 56 Subjects Risk Factor Exposure: other diseases, endogenous hormones; Cancer Site: breast Sample Characteristics One breast cancer case: Negro, male, 63 years old, veteran, hospital patient; Washington, DC Abstract A patient who had Klinefelter’s syndrome with gynecomastia and then developed cancer of the breast is described. A review of the literature revealed five additional patients in whom these two uncommon disease entities coexisted. The pathogenesis of breast cancer in patients with Klinefelter's syndrome is discussed. Perhaps the coexistence of these two diseases is coincidental. How- ever, in view of the number of reported patients with coexistent Klinefelter's syndrome and cancer of the breast, it is suggested that all patients with this chromosomal anomaly have careful yearly examinations of the breasts. Davies, ].N.P.; Tank, R.; Meyer, R.; et al. Cancer of the Integumentary Tissues in Uganda Africans: The Basis for Prevention J Natl Cancer Inst 41:31-51, 1968 57 Subjects Risk Factor Exposure: solar radiation; Cancer Site: skin; Study Characteristics Design: retrospective, Measures: crude incidence rate, age-specific incidence rate, descriptive statistics; Tine Frame: 1947-1960 Sample Characteristics 2,918 skin cancer cases: black; Uganda Variables Demographic: age, sex; Cancer: histology, tumor location Data Sources Histopathology records, survey Abstract Superficial cancers in the Uganda African occur in certain locali- zations which differ greatly from those in the European. Most 29 squamous carcinomas in the Uganda African develop from antece- dent lesions which can be identified, such as tropical phagedenic ulceration, burns, scars, sinuses, and fistulae. There are reasons for suspecting that sunlight is a major factor in producing squam- ous cancers of the cranial region, including the common conjunc- tival cancers. Melanomas are similarly virtually confined to re- stricted areas of the superficial tissues, and they may be induced by local heat. There is every indication that the overwhelming majority of integumentary tissue cancers of Uganda are preventa- ble by measures which for the most part would be acceptable to the population. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Dawson, D.F.; Duckworth, J.K.; Bernhardt, H.; et al. Giant Condyloma and Verrucous Carcinoma of the Genital Area Arch Pathol 79:225-231, 1965 58 Subjects Risk Factor Exposure: other diseases; Cancer Site: penis, scrotum Study Characteristics Design: case study Sample Characteristics Four cancer cases (blacks 3; whites 1): male, 23-54 years old Variables Demographic: age; Cancer: histology, metastases, tumor loction Data Sources Histopathology records, hospital medical records Abstract Until recently all reported cases of giant condylomata have not only originated on the penis, but have undergone a more or less typical evolution. We have observed two cases recently of car- cinoma-like condylomata which did not follow the Buschke- Loewenstein pattern. In fact, one of the cases were characterized by repeated recurrences of superficial condylomatous growths which involved large areas of skin. These lesions ulcerated, prolif- erated, and burrowed through soft tissues. They lined fistulous tracts through the penis, scrotum, perineum, and buttocks. These two cases are presented in detail. Two additional cases are pre- sented in which condylomata acuminata were present initially fol- lowed by squamous cell carcinoma with predominantly superficial growth and lymph node metastasis. It is extremely important to realize that all of the following cases point out that the histologic appearance of these lesions may not be a true reflection of the biological behavior observed clinically. Dayal, H.; Chiu, C.Y.; Sharrar, R.; et al. Ecologic Correlates of Cancer Mortality Patterns in an Industrialized Urban Population J Natl Cancer Inst 73(3):565-574, 1984 59 Subjects Risk Factor Exposure: air pollution; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional, time trend; Measures: standard mortality ratio (SMR), age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1968-1980 Sample Characteristics 45 neighborhoods: Philadelphia, PA Variables SES: census tract SES rank; Demographic: sex, geographicarea, race-ethnicity (predominately black or white neighborhood) Data Sources Death registration, U.S. Population Census Abstract Cancer mortality data for the period 1968-80 were analyzed to examine whether the high cancer burden for the city of Philadel phia was “evenly” distributed spatially and, if not, whether this distribution could be associated with socioeconomic variables and air pollution. Areas with significantly higher-than-expected rates tended to cluster; lung cancer and non-lung cancers showed dis- tinctive cluster patterns, which were evident only for males; females rates for any cancer or groups of cancers were almost uniformly distributed over the neighborhoods. Both the high lung cancer and the high non-lung cancer clusters were characterized by low socioeconomic status, but only the high lung cancer clusters exhibited high levels of air pollution. These neighborhood charac- teristics suggested that socioeconomic variables explain non-lung cancer mortality patterns; socioeconomic variables, together with air pollution, could be responsible for the observed lung cancer patterns. The uniform distribution of female lung cancer rates suggested that air pollution by itself cannot be a prominent factor in lung cancer mortality. Dayal, H.H.; Chiu, C. Factors Associated with Racial Differences in Survival for Prostatic Carcinoma J Chron Dis 35:553-560, 1982 60 Subjects Cancer Site: prostate; Cancer Measures: survival; Socioeconomic Factors Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1968-1977 Sample Characteristics 391 prostate cancer cases (blacks 292; whites 99): male, hospital patients; Richmond, VA Variables SES: census tract SES rank; Demographic: age; Cancer: histology, stage, tumor grade Data Sources non-SEER hospital cancer registry, histopathology records 30 Abstract The survival data on 99 white and 292 black patients with the carcinoma of the prostate from the city of Richmond seen at the Medical College of Virginia between 1968 and 1977 were analyzed. Black patients with prostatic cancer have significantly poorer sur- vival prognosis than whites. The distribution of stage at diagnosis is unfavorable to blacks in comparison to whites. Also, blacks present with less differentiated tumors. Degree of differentiation and clinical stage are highly associated and both are important predictors of survival. The prognosis differential between the two races does not seem to be due to difference in the biology of the disease; it is more likely due to the ‘environment’, defined in the broadest sense. Socioeconomic status is associated with race and explains the racial difference in survival. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Dayal, H.H.; Polissar, L.; Dahlberg, S. Race, Socioeconomic Status, and Other Prognostic Factors for Survival From Prostate Cancer 61 J Natl Cancer Inst 74(5):1001-1006, 1985 Subjects Cancer Site: prostate; Cancer Measures: survival; Socioeconomic Factors Study Characteristics Design: prospective (nonconcurrent); Measures: relative risk, crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1977-1984 Sample Characteristics 2,513 prostate cancer cases (blacks 1,032; Caucasians 1,481) Variables SES: census tract ZIP code education; Demographic: age, race-ethnicity; Cancer: stage Data Sources Hospital medical records, U.S. Population Census Abstract Survival data on prostate cancer patients from 11 comprehen- sive cancer centers contributing data to the Centralized Cancer Patient Data System were analyzed to examine the contribution of various factors to the probability of survival from prostate car- cinoma. Applicaton of a number of exclusion criteria resulted in a series of 2,513 patients (1,032 blacks and 1,481 Caucasians) for whom complete data on variables of interest were available. The stage of disease at diagnosis was a major determinant of survival. The proportion of blacks presenting the disease at advanced stage was substantially higher than that of Caucasians—a difference which was maintained within each socioeconomic status (SES) category. Caucasian patients had a better prognosis than blacks for each disease stage. A dose-response relationship between SES and survival prognosis was observed and its relationship persisted for each stage of the disease. Although both races and SES turned out to be significant in regression models of which one or the other was considered, the model including both race and SES showed only SES to be a significant factor. Hence it can be hypothesized that the racial difference in the survival prognosis for prostate cancer is, to a large extent, due to the differences in the distribution of SES in the two races. Dayal, H.H.; Power, R.N.; Chiu, C. Race and Socio-economic Status in Survival from Breast Cancer J Chron Dis 35:675-683, 1982 62 Subjects Cancer Site: breast; Cancer Measures: survival; Socioeconomic Factors Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1968-1977 Sample Characteristics 903 breast cancer cases (blacks 388; whites 515): female, hospital patients; Richmond, VA Variables SES: census tract SES rank; Demographic: age, race-ethnicity; Cancer: stage Data Sources non-SEER hospital tumor registry, U.S. Population Census Abstract The survival data on 515 white and 388 black female breast cancer patients seen at the Medical College of Virginia between 1968 and 1977 were analyzed to study the effect of age, stage and 31 race on survival prognosis. For a subset of the data representing patients from the city of Richmond (117 white and 206 black), socio-economic status (SES) information was generated on the basis of six predictors of SES and, in addition, the role of social class was studied. Each of these factors has a significant association with survival time. In particular, the probability of surviving a given length of time after diagnosis is ordered according to the socio-economic level and the statistical test for dose response shows a highly significant directional relationship. Age and stage do not explain the difference in survival between the two races. Race and SES are highly associated; a higher proportion of blacks than whites come from the lower end of the socio-economic scale. Moreover, the racial difference in survival becomes insignificant when it is adjusted for the distribution of socio-economic levels. This suggests that the observed difference in breast cancer survival between blacks and whites is, to a large extent, due to the difference between the two races with respect to the distribution of socio- economic status. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC de-The, G. Epstein-Barr Virus Behavior in Different Populations and Implications for Control of Epstein-Barr Virus-associated Tumors Cancer Res 36:692-695, 1975 63 Subjects Risk Factor Exposure: other diseases, geographic; Cancer Site: Burkitt's tumor, nasopharyngeal Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 6,000 sera for Chinese, Indians, Africans, and Caucasians: 0-19 years old; Hong Kong, Singapore, Uganda, France Variables Demographic: race-ethnicity Data Sources Laboratory test Abstract The epidemiology of Epstein-Barr virus (EBV) infection in popu- lations at different risk for EBV-associated diseases indicates signif- icant differences between the populations. EBV infection takes place much earlier in Uganda, where all children are infected before the age of 2 to 3 years, than in Southeast Asia, where nasopharyngeal carcinoma is prevalent. It is proposed that such early infections in Equatorial Africa is related to the risk for Burkitt's lymphoma. Four possible interventions to control EBV-associated diseases are presented:(a) simple hygienic measures to delay na- ture primary infection by EBV; (b) EBV vaccine; (c) intervention against cofactors such as malaria in Burkitt's lymphoma; and (d) characterization of high-risk groups to allow early detection and successful treatment. Decoufle, P.; Wood, D.]. Mortality Patterns Among Workers in a Gray Iron Foundry Am J Epidemiol 109(6):667-675, 1979 64 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: prospective (nonconcurrent); Measures: standardized mortality ratio (SMR), descriptive statistics; Time Frame: 1938-1967 Sample Characteristics 2,861 individuals (nonwhites 1,481; whites 1,380): male, gray iron foundry workers Variables Demographic: age, race-ethnicity Data Sources Death registration, industry records Abstract The long-term mortality experience of 2861 men employed for at least one month in the period of 1938 to 1967 in a gray iron foundry was examined to determine if they experienced unusual death rates for specific causes indicative of exposure to hazardous materials in the work environment. Both white and non-white workers experienced favorable mortality for most major disease categories compared to general population rates, even men em- ployed five or more years. No deaths from pneumoconiosis were observed nor were deaths from other chronic respiratory diseases in excess. Analysis of detailed cancer sites showed no significant departures from expectation overall. However, in the subgroup of men who achieved five or more years employment prior to 1938, a twofold increase in mortality from digestive cancer (14 observed deaths vs. 7.4 expected) and respiratory cancer (8 ob- served deaths vs. 4.0 expected) was seen. Absence of information on specific foundry jobs held by the subjects and associated expo- sures limits full interpretation of the findings. However, the excess observed for respiratory cancer among long-term employees fol- lowed for 30 years is consistent with previous reports. 32 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Delzell, E.; Grufferman, S. Mortality Among White and Nonwhite Farmers in North Carolina, 1976-1978 65 Am ] Epidemiol 121(3):391-402, 1985 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: odds ratio, proportionate mortality ratio (PMR), descriptive statistics, higher level statistics; Time Frame: 1976-1978 Sample Characteristics 74,041 deaths (nonwhites 18,417; whites 55,624—Farmers 12,753; other occupations 61,288): male, 15+ years old; North Carolina Variables SES: occupation; Demographic: age, race-ethnicity, marital status, geographic area Data Sources Death registration Abstract Death certificate information identified 9,245 white and 3,508 nonwhite men who died in North Carolina during 1976-1978 and who had been farmers. The distribution of deaths from various causes among these men was compared to that of other male decedents in the state. For both white and nonwhite farmers, proportional mortality ratios (PMRs) were elevated for tuberculosis (whites, 1.6; nonwhites, 1.7), diseases of the skin and subcutane- ous tissue (whites, 2.5; nonwhites, 1.5), and external causes (whites, 1.2; nonwhites, 1.1) and were decreased for cancers of the esophagus (whites and nonwhites, 0.5) and large intestine and rectum (whites and nonwhites, 0.7). White farmers had an in- creased relative frequency of melanoma (PMR = 1.2) and other skin cancer (PMR = 1.8), while nonwhite farmers had an increased relative frequency of melanoma (PMR = 6.3), brain cancer (PMR = 2.3), and leukemia (PMR = 1.9). In addition, among decedents under 65 years of age, both white and nonwhite farmers had an elevated proportional mortality ratio for prostate cancer (whites, 1.6; nonwhites, 1.3). Many of these results are consistent with observations from other studies. Some of these findings, particu- larly those for nonwhites, warrant further evaluation, including detailed investigation of possibly related farming practices. DeMoor, N.G. Nephroblastoma in the Black Child: A Review of Treatment in the Johannesburg Pediatric Tumor Clinic S Afr Med J 51:194-196, 1977 66 Subjects Treatment; Cancer Site: kidney; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, crude mortality rate; Time Frame: 1972-1975 Sample Characteristics 25 kidney cancer cases: black, 1-4 years old; Johannesburg, South Africa Variables Demographic: age, Cancer: tumor location, stage Data Sources Hospital medical records Abstract Twenty-five Black children with nephroblastoma who were re- ferred to the Paediatric Tumour Clinic during the preceding 4 years are reviewed. Results indicate that in the case of a locally advanced, fixed, primary tumour, initial surgery, even if it is combined with chemotherapy, may be a factor in the rapid postoperative develop- ment of multiple pulmonary metastases. Preoperative irradiation appeared to be advantageous for these patients. Long-term chemotherapy holds no advantage over short-term chemotherapy. Dent, D.M.; Vader, C.G. Malignant Gastro-intestinal Tumours: The Frequency Distribution by Age, Sex, Race, and Site at Groote Schuur Hospital, Cape Town, 1974-1978 S Afr Med J 60:883-885, 1981 67 Subjects Cancer Site: digestive system; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1974-1978 Sample Characteristics 2,023 gastrointestinal cancer cases (blacks 311; coloreds 844; whites 868): hospital patients; Cape Town, South Africa Variables Demographic: age, sex, race-ethnicity; Cancer: tumor location Data Sources Hospital medical records Abstract The observed frequency distribution of 2,023 primary malignant gastro-intestinal tumours recorded during the 5-year period 1974- 1978 is reported. The two most common sites were stomach (29.6%) and colon and rectum (29.5%), and the third most common was the oesophagus (19.8%). Marked differences in the site of the tumours according to race were found; oesophageal carcinoma was the most frequent lesion in black patients (62.1%), gastric carcinoma in coloured patients (44.1%) and colorectal carcinoma in white patients (46.8%). Within the black group, rarely affected sites were the pancreas (5.1%) and colon and rectum (3.5%). EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Dent, R.I.; Fleming, J.B.M.; Wicks, A.C.B. Carcinoma of the Stomach in Rhodesian Africans and a Comparative Review 68 Clin Oncol 3:17-26, 1977 Subjects Diagnosis; Treatment; Risk Factor Exposure: diet-nutritional factors, other diseases; Cancer Site: stomach; Cancer Measures: survival; Biologic Marker: ABO blood group Study Characteristics Design: prospective (nonconcurrent); Measures: crude mortality rate, prevalence rate, descriptive statistics; Time Frame: 1967-1974 Sample Characteristics 158 stomach cancer cases: black; Rhodesia Variables Demographic: sex, age; Cancer: tumor location, symptoms Data Sources Hospital medical records Abstract Despite previous suggestions of its rarity in Africa, carcinoma of the stomach is the fifth commonest malignancy seen in a hospital serving a large black African population in Rhodesia. One hundred and fifty-eight histologically-proven cases, occurring in an 8-year period, are presented and their clinical presentation, investigation, pathology and treatment analysed retrospectively. International publications are briefly reviewed with special mention of the few other reported series of carcinomas of the stomach from Africa. Possible aetiological factors are discussed and comparisons made with series from the U.K., the U.S.A., and Europe. On the whole, the pattern of the disease in Africa would appear to differ little from that seen elsewhere in more developed areas. Devesa, S.S.; Diamond E.L. Socioeconomic and Racial Differences in Lung Cancer Incidence 69 Am ] Epidemiol 118(6):818-831, 1983 Subjects Cancer Site: lung; Cancer Measures: incidence; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, age-specificincidence rate, relative risk, higher level statistics; Time Frame: 1969-1971 Sample Characteristics 20,868 lung cancer cases (blacks 2,354; whites 18,514): TNCS areas Variables SES: income, education; Demographic: geographic area, race-ethnicity, age, sex Data Sources Third National Cancer Survey, 1969-1971 Abstract The association of lung cancer incidence with income and edu- 34 cation and the effect of adjustment for socioeconomic distribution on black-white differences in lung cancer rates were evaluated using data from the Third National Cancer Survey. Included in this study were 20,868 cases of lung cancer (18,514 among whites and 2,354 among blacks) diagnosed among metropolitan residents of the survey during 1969-1971. Median family income and median years of education by census tract of residence were used to indi- cate socioeconomic group. Strong significant inverse trends be- tween lung cancer incidence and both income and education were apparent among white and black males, and the effect of income exceeded that of education. Lung cancer rates among black males compared with white males were significantly higher (p<0.001) before socioeconomic adjustment, nonsignificantly higher after ad- justment for education, and nonsignificantly lower after adjust- ment for income. Strong trends in risk with income or education were not observed for lung cancer among females of either race. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Devesa, S.S.; Diamond, E.L. Association of Breast Cancer and Cervical Cancer Incidence With Income and Education Among Whites and Blacks J Natl Cancer Inst 65(3):515-528, 1980 70 Subjects Cancer Site: breast, cervix uteri; Cancer Measures: incidence, prevalence; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, age-specificincidence rate, relative risk, higher level statistics; Time Frame: 1969-1971 Sample Characteristics 2,914 breast cancer cases (blacks 1,570; whites 19,344) and 4,756 cervix uteri cancer cases (blacks 954; whites 3,802): female, TNCS areas Variables SES: census tract income, census tract education; Demographic: age, sex, race-ethnicity, geographic area Data Sources Third National Cancer Survey 1969-1971, U.S. Population Census Abstract Data from the 1969-71 Third National Cancer Survey were used to study the association of cancer incidence with income and edu- cation as indicated by census tract of residence. Also considered was the effect of adjustment for differences in socioeconomic dis- tribution on the observed excess risk of cervical cancer and lower risk of breast cancer among black women compared to white women. Strong positive associations with both income and educa- tion were found, with the rates based on 19,344 breast cancer cases among white women. These observations were noted in most geographic areas studied. The relative risk showed little apparent relationship to age. The rates based on 1,570 cases among black women indicated a strong positive association with education but not with income. Socioeconomic adjustment reduced by almost one-half the black-white differences in breast cancer rates, and education had a stronger effect than did income. White women continued to have a significantly higher rate after such adjustment. Conversely, the incidence of cancer of the cervix showed strong negative association with each of the two variables among both the 3,802 cases in white women and 954 cases in black women. The negative gradient decreased with age and was apparent in almost all the areas. The excess risk among black women was reduced by two-thirds with socioeconomic adjustment, though the rates remained significantly different. Income had a stronger association than did education with cervical cancer incidence. Devesa, S.S.; Silverman, D.T. Cancer Incidence and Mortality Trends in the United States: 1935-74 71 J Natl Cancer Inst 60(3):545-571, 1978 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, mortality Study Characteristics Design: time trend; Measures: age-adjusted incidence rate, age-adjusted mortality rate; Time Frame: 1935-1974 Sample Characteristics Incidence data from FNCS, SNCS, TNCS, and End Results Program areas; mortality data from the United States Variables Demographic: sex, race-ethnicity (whites, non-whites) Data Sources First National Cancer Survey 1937-1939, Second National Cancer Survey 1947-1948, Third National Cancer Survey 1969-1971, End Results Program, death registration Abstract Using incidence data from three national cancer surveys and mortality data for the entire United States, one can make several observations concerning the trends in cancer occurrence. Rates among males have been increasing, whereas those among females have been decreasing. In the past, cancer of all sites combined occurred more frequently among females; now males have rates higher than do females of the same race. White predominance has been replaced by a nonwhite excess in incidence and mortality rates among males and a nonwhite excess in female mortality; racial differences in incidence rates among females have been di- minishing. Increases have occurred in cancers of the lung, prostate gland, and breast among nonwhite females; pancreas, kidney, and 35 bladder among males; and esophagus among nonwhites; melanomas have increased among whites and lymphomas have increased in each race-sex group. The reported increases may be due partly to improvements in diagnosis, which probably varied for the different primary sites and probably affected nonwhites more than whites. Meanwhile, decreases have occurred in cancers of the uterus (particularly cervix), stomach, and liver. Intestinal cancer has increased, whereas rectal cancer has decreased. Possible problems in specifying the site of origin of tumors arising in the area of the rectosigmoid junction make it difficult to determine how accurately the observed trends reflected the true situation. When intestinal and rectal cancers are considered together, the rates for nonwhites have been increasing toward the more stable level of the whites. The incidence of thyroid cancer has increased, whereas the mortality has decreased. Without lung cancer, the incidence of all cancer types combined among white males would be decreasing in recent years rather than increasing. The increases in overall incidence seen among nonwhite males are due primarily to increases in cancers of the lung, prostate gland, intestine, and esophagus. Among white females, the decrease in overall inci- dence is due to the decreases in cancers of the uterine cervix and the stomach; breast cancer rates are steady and continue to have a major impact. Except for breast cancer, the experience of non- white females is similar, with the declines in uterine cancer being greater and those in the other sites being less than the declines observed for white females. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Dini, M.; Jafari, K. Management of Intraepithelial Neoplasia of the Uterine Cervix in the Black Population 72 J Natl Med Assoc 74(7):671-674, 1982 Subjects Screening: Pap smear; Treatment; Cancer Site: cervix uteri Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics; Time Frame: 1972-1977 Sample Characteristics 1,328 intraepithelial neoplasia cases: black, female, hospital patients; Cook County, IL Variables Demographic: age, parity; Cancer: histology Data Sources Histopathology records, hospital medical records Abstract The records of 2,277 patients referred to the Colposcopy Clinic at Cook County Hospital were reviewed, and those of 1,328 black patients who had a histological diagnosis of dysplasia or carcinoma in situ of the cervix were chosen for this study. Treatment mod- alities of intraepithelial cervical neoplasia are discussed and the conservative management of young patients is emphasized. Be- cause of the effectiveness of outpatient management in cervical preinvasive lesions, health providers should make every attempt to motivate people particularly those at high risk, i.e., black women, to undertake preventive measures. Edington, G.M.; Maclean, C.M.U. Incidence of the Burkitt Tumour in Ibadan, Western Nigeria 73 Br Med J 1:264-266, 1964 Subjects Risk Factor Exposure: geographic area; Cancer Site: Burkitt's tumor; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-specific mortality rate; descriptive statistics; Time Frame: 1960-1962 Sample Characteristics 230 cancer cases (Burkitt tumor cases 161; other cancers 69): black, 0-14 years old; Ibadan, Nigeria Variables Demographic: age, sex Data Sources non-SEER hospital cancer registry Abstract The crude and age-specific rates of the Burkitt tumour in Iba- dan, Western Nigeria, have been assessed. When all age-groups were considered the tumour was found to be second in incidence only to carcinoma of the cervix, and it was by far the most common malignancy seen in childhood, being many times more common than leukemia in the American Negro in comparable age-groups. Elliot, M.S.; Steven, D.M. Carcinoma of the Colon and Rectum in Patients Under 30 Years of Age 74 S Afr Med J 66:129-131, 1984 Subjects Diagnosis; Treatment; Cancer Site: colon, rectum; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics; Time Frame: 1968-1982 Sample Characteristics 47 large-bowel cancer cases (blacks 1; coloreds 39; whites 7): 12-29 years old; Cape Town, South Africa Variables Demographic: sex, race-ethnicity; Cancer: stage, tumor location, histology, symptoms Abstract A 15-year study of large-bowel carcinoma at Groote Schuur Hospital revealed 47 patients who presented at the age of 29 years or younger. Only 1% of the White patients with large-bowel car- cinoma were in this age group (similar to other Western series), whereas a much higher incidence was seen in Coloured patients (9%). We found a large number of young Coloured patients with high-grade tumours and lymph node invasion. In spite of surgery, chemotherapy and radiotherapy the prognosis for this group in particular remains extremely poor. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Emes, J.J.; Redmond, C.K.; Jarmolowski, N. Trends in Cancer Incidence in Allegheny County, Pennsylvania, 1937-71 Public Health Rep 9(4):365-373, 1977 75 Subjects Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: time-trend; Measures: age-adjusted incidence rate, descriptive statistics; Time Frame: 1937-1971 Sample Characteristics Cancer incidence data from Allegheny County, PA Variables Demographic: sex, race-ethnicity Data Sources First National Cancer Survey 1937-1939, Second National Cancer Survey 1947-1948, Third National Cancer Survey 1969-1971 Abstract Cancer incidence rates by race, sex, and cancer site were ob- tained from the Third National Cancer Survey for the years 1969-71 for residents of Allegheny County, Pennsylvania. Male incidence of cancers of the lung, bronchus, and trachea increased steadily between 1937 and 1969-71 both in Allegheny County and the United States. In the county, female incidence rates for these can- cers decreased in the period 1947 to 1957-58 but showed an average annual increase of 9.2 percent in the interval 1957-58 to 1969-71. For stomach cancer, incidence rates for both sexes have decreased sharply in the county and in the United States. There appears to have been a steady decline over time in cervical cancer in Allegheny County, although the average annual rate of decrease of 2.8 percent for the latest interval (1957-58 to 1969-71) is not as large as the decrease of 3.9 percent per year from 1947 to 1957-58. Breast cancer rates for the county appear to have been steadily, although slowly, increasing at an average rate of about 0.6 percent per year, in contrast to almost constant U.S. rates. There have been steady increases in prostate cancer incidence in both Allegheny County and the United States since 1937. For all sites combined, male cancer incidence rates increased, while those for females slowly decreased in both Allegheny County and the United States during the interval 1937 to 1969-71. Emovon, A.C. Sociocultural Factors Associated with Cervical Cancer in Bendel State, Nigeria Int ] Gynaecol Obstet 15:253-255, 1977 76 Subjects Risk Factor Exposure: reproductive factors, sexual practices; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Time Frame: 1973-1976 Sample Characteristics 15,049 gynecologic hospital patients (cervical cancer cases 65; other 14,984): black, female; Benin City, Nigeria Variables SES: education; Demographic: age, age at marriage, parity, age at first pregnancy Data Sources Hospital medical records, survey Abstract This is an analysis of data on 50 patients treated for cervical cancer at the University of Benin Teaching Hospital, Benin City, Nigeria, from March 1973 to December 1976. The data were col- lected through a questionnaire designed to obtain information con- cerning various sociocultural factors about cervical cancer patients. Study results showed that 80% of the patients had no formal edu- cation and came from lower socioeconomic groups. Most of the patients (88%) married before the age of 20, and 76% reported frequent coitus. The patients were high-parity women, but high parity is also typical of the general female population. Since the culture demands that the male population be circumcised, a low incidence of cervical cancer might be expected. The occurrence of 65 cases of cancer of the cervix out of a total occurrence of 15,049 hospital gynecologic cases examined during the study period is possible evidence of such a low incidence in this population. 37 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Ernster, V.L.; Selvin, S.; Sacks, S.T.; et al. Prostatic Cancer: Mortality and Incidence Rates by Race and Social Class Am J Epidemiol 107(4):311-320, 1978 77 Subjects Cancer Site: prostate; Cancer Measures: incidence, mortality; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: age-specific mortality rate, age-specific incidence rate, descriptive statistics; Time Frame: 1968-1972 Sample Characteristics 400 prostate cancer deaths (blacks 71; whites 329) and 750 prostate cancer cases (blacks 141; whites 609): male; Alameda County, CA Variables SES: census tract SES rank; Demographic: age, race-ethnicity Data Sources Death registration, Third National Cancer survey 1969-1971 Abstract Among the most striking epidemiologic characteristics of pros- tatic cancer in the United States is the sizeable excess in rates for disease among blacks as compared with whites. This study attemp- ted to determine whether the higher black rates might be explained by controlling for social class, using mortality data from Alameda County for the pericensal period 1968-1972 (n=400), and the popu- lation-based series of cases for Alameda County included in the Third National Cancer Survey, 1969-1971 (n=750). Reviewed first are previous studies addressed to the relationship of prostatic cancer and socioeconomic status (SES), most of which have been confined to whites. In the present study, each death or case was assigned to a socioeconomic class based on census tract of residence and rates by race and social class determined. Comparison of age- specific mortality and incidence rates by SES reveals no gradient for prostatic cancer in either whites or blacks. Ernster, V.L.; Selvin, S.; Winkelstein, W., Jr. Cohort Mortality for Prostatic Cancer Among United States Nonwhites 78 Science 200:1165-1166, 1978 Subjects Abstract Cancer Site: prostate; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: age-adjusted mortality rate, age-specific mortality rate, higher level statistics; Time Frame: 1930-1970 Sample Characteristics Prostate cancer mortality data: males, 45-84 years old; United States Variables Demographic: age, race-ethnicity (nonwhite, white) In recent decades, age-adjusted mortality rates from prostatic cancer have risen precipitously among blacks, remaining un- changed among whites. It is now the most common cancer among United States black males. When nonwhite mortality rates were examined by age and birth cohort, it was found that peak rates occurred at every age in the cohort of 1896 to 1900, and declined thereafter. This presages an arrest and reversal of the time trend in summary mortality rates as more recent nonwhite cohorts reach the ages of maximum risk. Ernster, V.L.; Winkelstein, W., Jr.; Selvin, S.; et al. Race, Socioeconomic Status, and Prostatic Cancer Cancer Treat Rep 61(2):187-191, 1977 79 Subjects Cancer Site: prostate; Cancer Measures: incidence, mortality; Socioeconomic Factors Study Characteristics Design: cross-sectional, time trend; Measures: age-specific mortality rate, age-specific incidence rate, descriptive statistics; Time Frame: 1958-1972 Sample Characteristics 1,061 prostate cancer deaths (blacks 137; whites 924) and 2,197 prostate cancer cases: male; Alameda County, CA Variables SES: census tract SES rank; Demographic: age, race-ethnicity Data Sources Death registration, non-SEER population-based cancer registry Abstract Using mortality and incidence data from Alameda County, 38 California, this study attempted to determine whether the higher occurrence rate of prostatic cancer among black men as compared with whites in the United States might be explained by racial differences in factors associated with socioeconomic status. Each death or case of prostatic cancer was assigned to a socioeconomic status. Each death or case of prostatic cancer was assigned to a social class based on census tract of residence, and rates by race and socioeconomic status were computed. Comparison of age-spe- cific mortality and incidence rates by socioeconomic status reveals no gradient in either whites or blacks. The higher risk for blacks holds up at almost every age and socioeconomic level. However, the racial differences are less pronounced for incidence than for mortality. Racial differences in the occurrence of deaths appearing in Part II of the death certificate are also examined. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Evans, H.L.; Batsakis, J.G. Polymorphous Low-Grade Adenocarcinoma of Minor Salivary Glands: A Study of 14 Cases of a Distinctive Neoplasm Cancer 53:935-942, 1984 80 Subjects Treatment; Cancer Site: oral cavity Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics Sample Characteristics 14 minor salivary gland cancer cases (blacks 6; whites 8): 27-76 years old; Houston, TX Variables Demographic: age, sex, race-ethnicity; Cancer: histology, tumor size, tumor location Data Sources Hospital medical records Abstract M. D. Anderson Hospital cases diagnosed as adenocarcinoma of minor salivary glands before 1977 were reviewed. Within this heterogeneous group of neoplasms there was identified one clinicopathologic tumor entity, which we have designated “polymorphous low-grade adenocarcinoma.” The 14 tumors in that category were characterized by cytologic uniformity and his- tologic diversity; growth patterns varied (both within and among cases) from solid to tubular to papillary to cribriform (pseudoadenoid cystic) to fascicular, while the cells were always small to medium-sized, regular, and lacking in nuclear atypia. Mitotic figures were infrequent, and tumor necrosis was seen in only one instance (a recurrent neoplasm). Clear cytoplasm, oxyphilic and mucinous metaplasia, and intratubular calcification were sometimes present, and stromal mucinization and hyaliniza- tion were common. The tumors were always encapsulated, and exhibited extension into surrounding tissues including bone. The 14 patients ranged in age from 27 to 76 years (median, 64 years). Eight were male and six were female; eight were white and six were black. The neoplasm was intraoral in all cases, involving the palate in 11, the buccal mucosa in two, and the posterior mandibu- lar area in one. Local recurrence developed in one case, cervical lymph node metastasis in one, and both recurrence and cervical lymph node metastatsis in two. The number of successive recurr- ences ranged up to three, and the interval to recurrence varied up to nine years (the interval to metastasis up to five years). Although radical surgical procedures were necessary for tumor control in some cases, no distant metastases occurred and all patients were clinically tumor-free at latest follow-up. Fan, K.J.; Kovi, J. Metastatic Brain Tumors in Two Predominantly Black Hospitals: A Statistical Analysis J Natl Med Assoc 71(7):671-673, 1979 81 Subjects Cancer Site: brain; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1953-1973 Sample Characteristics 199 metastic brain cancer cases (U.S. 104; Africans 95): black; Africa; Washington, DC Variables Demographic: sex, age, geographic area; Cancer: primary site Data Sources Histopathology records Abstract A retrospective statistical analysis was done on metastatic brain tumors collected from two predominantly black hospitals in Washington, DC. A composite African series of metastatic brain tumors was also constructed for comparison. The results indicate that bronchogenic carcinoma is the predominant metastatic brain tumor (45.2 percent) among American blacks in Washington, DC, and chorioepithelioma, the most common (20.0 percent) among African blacks. In comparing these two series, much dissimilarity in the pattern of tumor distribution between these two genetically related ethnic groups suggests an important environmental role in the genesis of metastatic brain tumors. The present study also reveals a relatively high proportional frequency of prostatic car- cinoma among metastatic brain tumors in blacks (3.8 percent in Washington, DC, and 2.1 percent in Africa). 39 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Fan, K.].; Kovi, ].; Earle, K.M. The Ethnic Distribution of Primary Central Nervous System Tumors: AFIP, 1958-1970 82 J Neuropathol Exp Neurol 36(1):41-49, 1977 Subjects Cancer Site: central nervous system; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1953-1971 Sample Characteristics 16,311 central nervous system cancer cases (Negroes 606; Caucasians 8,264; unknown 7,441): United States and 810 central nervous system cancer cases: Negro; Uganda; Nairobi, Kenya; Transvaal, South Africa; Malawi; Rhodesia; Zambia; Ethiopia; Ivory Coast; Dakar, Senegal; Ibaden, Nigeria Variables Demographic: age, sex, race-ethnicity, geographic area; Cancer: histology Data Sources Hospital medical records Abstract A retrospective ethnic study was made of 16,311 cases of pri- mary central nervous system (CNS) tumors seen at the Armed Forces Institute of Pathology (AFIP), Washington, D.C., from 1958 to 1970. Results showed a considerably higher Caucasian:Negro (C:N) case ratio (13.7:1) than the C:N population ratio (8.4:1), in- dicating a higher relative frequency of primary CNS tumors in American Caucasians as compared to American Negroes. The glioma was significantly more frequent in Caucasians than in Neg- roes (p< 0.005). In contrast, Negroes had an excess of the pituitary adenoma as compared to Caucasians (p< 0.01). The proportional frequencies of the meningioma and the nerve sheath tumor were also higher in Negroes than in Caucasians. When this pattern of the tumor distribution of American Negroes was compared to that of African Negroes (a composite African series), the preponderance of the pituitary adenoma and the meningioma and the relative paucity of the glioma in the Negro race as compared to Caucasians were again confirmed. The differences in the relative frequency and the tumor distribution between American Negroes and Cauca- sians and the considerable similarity of the tumor distribution between American and African Negroes emphasize the importance of genetic factors in the development of at least some primary CNS tumors. Farman, A.G.; Nortje, C.].; Grotepass, F.W.; et al. Myxofibroma of the Jaws Br J Oral Surg 15:3-18, 1977-78 83 Subjects Diagnosis; Cancer Site: jaw Study Characteristics Design: case study Sample Characteristics One myxofibroma case: colored, male, 16 years old; South Africa Abstract The clinical, radiographic and histological features of a large unilateral myxofibroma of the maxilla are described. The literature pertaining to myxofibroma is reviewed. Favrot, M.C.; Phillip, I.; Phillip, T.; et al. Distinct Reactivity of Burkitt's Lymphoma Cell Lines with Eight Monoclonal Antibodies Correlated with the Ethnic Origin J Natl Cancer Inst 73(4):841-847, 1984 84 Subjects Diagnosis; Cancer Site: Burkitt's tumor; Biologic Marker: monoclonal antibodies, Epstein-Barr virus Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 28 cell lines (Africans 5; North Africans 9; Caucasians 13; Reunion Island 1) Variables Demographic: race-ethnicity, geographic area Data Sources Laboratory test Abstract Twenty-eight Burkitt's lymphoma(s) (BL) cell lines were analyzed with anti-human immunoglobulins and monoclonal anti- bodies: Y29/55, B1, and BA1 are slightly different pan-B-reagents; TU1 and BL13 are two discriminating markers of the follicle; RFT1 is a pan-T-reagent expressed on the follicle mantle; AL2 reacts with the common acute lymphoblastic leukemia antigen gp100; and 38:13 recognizes a BL-associated antigen. Those lines were classified into 3 groups according to their membrane phenotype. In the first 2 groups, cell lines were derived from BL of germinal center origin, whereas in the last they were established from BL cells originating in the bone marrow. All cell lines in the last group were from Caucasian BL, whereas lines from African BL of a high- incidence area were in group 1. North African cases were in group 2. Those distinct subgroups were not related specifically to the reactivity with Epstein-Barr virus nuclear antigen, the type of chromosomal translocation, or the clinical features. The variations induced by growth culture as well as the clinical implications were discussed. 40 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Feldman, J.G.; Carter, A.C.; Nicastri, A.D.; et al. Breast Self-Examination, Relationship to Stage of Breast Cancer at Diagnosis 85 Cancer 47:2740-2745, 1981 Subjects Screening: breast self-exam (BSE); Diagnosis; Cancer Site: breast Study Characteristics Design: retrospective; Intervention Study; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1975-1975 Sample Characteristics 996 breast cancer cases (nonwhites 209; whites 787): female; Brooklyn, NY Variables SES: education, medical insurance status; Demographic: age, race-ethnicity, marital status, menopausal status; Cancer: stage, tumor size Data Sources Histopathology records, survey Abstract Data from 996 newly diagnosed breast cancer patients indicated a highly significant association (p< 0.001) between periodic breast self-examination (BSE) and pathologic stage of disease. Among women reporting periodic BSE, only small differences were noted between those those who practiced monthly and those who prac- ticed several times annually. Average maximum tumor diameter and frequency of tumors 4 cm or larger were significantly greater (p< 0.01) among women who rarely or never practised BSE. The relationship between the periodic practice of BSE and the diagnosis of breast cancer before nodal involvement was present even after controlling for a wide variety of variables. The regular practice of BSE was associated with a one-third reduction in the likelihood of diagnosis of disease with positive nodes. This difference trans- lated to a 10% decline in five-year mortality for whites and a 17% decline for nonwhites. Fernandes, C.M.C.; Solomon, A. Pyriform Fossa Carcinoma in South African Negro Patients S Afr Med ] 52:528, 1977 86 Subjects Diagnosis; Cancer Site: larynx, pharynx; Cancer Measures: incidence Study Characteristics Design: case study; Time Frame: 1973-1976 Sample Characteristics Seven pyriform fossa cancer cases: Negro; Johannesburg, South Africa Variables Cancer: stage Data Sources Hospital medical records Abstract Seven patients with pyriform fossa carcinoma who presented at Baragwanath Hospital during the period 1973-1976 are dis- cussed. The value of positive contrast studies with special reference to laryngography is emphasized and the results are reviewed. Festa, R.S.; Meadows, A.T.; Boshes, R.A. Leukemia in a Black Child with Bloom's Syndrome: Somatic Recombination as a Possible Mechanism for Neoplasia Cancer 44:1507-1510, 1979 87 Subjects Cancer Site: leukemia Study Characteristics Design: case study; Time Frame: 1976 Sample Characteristics One leukemia case: black, female, 5% years old Abstract A 5% year-old black child with Bloom's syndrome de- veloped acute lymphoctic leukemia (ALL). Bloom's syn- drome is associated with chromosomal aberrations, and af- fected individuals have an increased incidence of leukemia and solid tumors. The skin on our patient had adjacent areas of decreased and increased pigmentation similar to the “twin-spots” seen in Drosophila. “Twin-spots” are the manifestation of somatic cell DNA recombination and pro- vide evidence that clones of cells in Bloom's syndrome have become homozygous for a particular gene. Somatic cell re- combination is proposed as a mechanism to explain the increased incidence of neoplasia in Bloom's syndrome and supports the hypothesis that cancer may be a recessive dis- order at the cellular level. 41 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Ford, A.B.; Bialik, O. Air Pollution and Urban Factors in Relation to Cancer Mortality 88 Arch Environ Health 35(6):350-359, 1980 Subjects Risk Factor Exposure: tobacco, air pollution, geographic area; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: crude mortality rate, age-specific mortality rate, age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1969-1971 Sample Characteristics Cancer mortality data: 45 years old; United States; cancer mortality data from Cuyahoga County, OH Variables SES: census tract income, census tract employment, census tract education; Demographic: sex, age, race-ethnicity (Negro, white) Data Sources Death registration, U.S. Population Census Abstract Age adjusted mortality rates for the 1969-1971 adult population in the United States reveal a consistent pattern of 13% excess cancer mortality in the metropolitan counties with central cities compared to nonmetropolitan counties. Only 14% of all cancer death exhibit a different pattern. A detailed analysis of cancer mortality in Cuyahoga County, (Cleveland), Ohio, during the years 1969-1971, shows that apart from cancer mortality of the respiratory tract, general air pollution characteristics correlate sec- ondarily with mortality, compared with socioeconomic factors. Forrest, J.B.; Howards, S.S. Survival Following Orchiectomy for Stage D Adenocarcinoma of the Prostate ' 89 J Surg Oncol 25:71-72, 1984 Subjects Treatment; Cancer Site: prostate; Cancer Measures: survival; Socioeconomic Factors Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics; higher level statistics; Time Frame: 1970-1975 Sample Characteristics 89 prostate cancer cases: male, hospital patients; Virginia Variables SES: public-private patient status; Demographic: age Data Sources non-SEER hospital cancer registry, hospital medical records Abstract The survival of nonprivate and private patients following or- chiectomy for stage D adenocarcinoma of the prostate was studied. While the private patients enjoyed a longer survival, their im- proved survival was found to be a function of an earlier presenta- tion. Examination of the survival statistics with linear regression analysis suggested that survival following orchiectomy is a func- tion of the age at the time of orchiectomy. Fragoyannis, S.; Brits, C.; Griessel, P. Age Patterns of Tswana Women with Carcinoma of the Cervix S Afr Med J 52:493, 1977 90 Subjects Screening: Pap smear; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1 year Sample Characteristics 6,756 Pap smears (malignant and dysplasic smears 386; others 6,370): female; Tswana, Pretoria, South Africa Variables Demographic: age; Cancer: stage Data Sources Hospital medical records Abstract We report the results of the cervical cytology studies done on 6,756 initial cervical smears of Tswana women in the newly estab- lished cytology laboratory at Ga-Rankuwa Hospital. We have de- tected a high incidence of positive smears (31 per 1,000), a figure which is among the highest in the recorded literature. We also emphasize that the age patterns (profiles) of Tswana women with carcinoma of the cervix present a shift to the left in comparison with results of similar studies in Black or White population groups, and that there is immense benefit to be gained by the expansion of cytology laboratories. 42 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Franklin, E.W., III; Rutledge, F.D. Epidemiology of Epidermoid Carcinoma of the Vulva Obstet Gynecol 39(2):165-172, 1972 91 Subjects Diagnosis; Risk Factor Exposure: other diseases, reproductive factors; Cancer Site: vulva Study Characteristics Design: retrospective; Measures: descriptive statistics; Time Frame: 1944-1968 Sample Characteristics 249 vulva cancer cases (Negroes 52; whites 174; Latin Americans 9): female, 40+ years old, hospital patients; Houston, TX Variables Demographic: age, race-ethnicity, marital status, number of pregnancies, menopausal status; Cancer: stage, symptoms Data Sources Hospital medical records Abstract Epidemiologic review of 249 patients with invasive and intra- epithelial epidermoid carcinoma of the vulva was undertaken and correlated with data from the literature. Recognizing the disparity of populations studied, common epidemiologic factors were recog- nized as characteristic of agent, host and environment. These in- cluded age distribution, stigmas of veneral exposure, associated systemic disease and patient as well as physician-delay in diag- nosis. A high incidence of associated neoplasia of the anogenital tract, especially the cervix, suggested that common pathogenetic factors, possibly viral, were operative in carcinogenesis in the vulva, cervix and vagina. Franklin, R.; McSwain, B. Carcinoma of the Colon, Rectum and Anus Ann Surg 171(6):811-818, 1970 92 Subjects Diagnosis; Treatment; Cancer Site: colon, rectum, anus; Cancer Measures: survival Study Characteristics Design: time trend, prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics; Time Frame: 1925-1968 Sample Characteristics 1,022 colon, rectum, anus cancer cases (Negroes 172; Caucasians 950): hospital patients; Nashville, TN Variables Demographic: age, sex, race-ethnicity; Cancer: stage, tumor location Data Sources Hospital medical records Abstract Findings in 1,022 patients with carcinomas of the colon, rectum, and anus, and results of follow-up of 5 or more years in 761 are reported. Over a period of more than 40 years (1925-1968), a study of carcinoma of the colon, rectum, and anus shows that there has been a relative decrease in incidence of carcinoma of the rectum with a relative increase in carcinomas elsewhere in the large bowel. There has been an increase in resectability rate, a decrease in operative mortality, and in, 1956-1963, an increase in 5-year survi- val. Fraumeni, J.F., Jr. The Face of Cancer in the United States Hosp Pract 18(12):81-96, 1983 93 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, occupation, tobacco, geographic area; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: age-adjusted mortality rate, relative risk, descriptive statistics; Time Frame: 1950-1970 Sample Characteristics Mortality data from 3,056 counties of the 48 contiguous states Variables SES:industry-occupation; Demographic: age, sex, race-ethnicity, geographic area Data Sources Death registration, U.S. Population Census Abstract Geographic analysis of cancer mortality data in the U.S. reveals a surprising variety of patterns. Among them a southern coastal clustering of high lung cancer rates in white males, high oral cancer rates in rural southern females, a pronounced excess of bladder cancer deaths among whites in certain areas. Field studies reveal sometimes surprising etiologic correlations. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Fraumeni, J.E., Jr.; Blott, W.]. Geographic Variation in Esophageal Cancer Mortality in the United States J Chron Dis 30:759-767, 1977 94 Subjects Risk Factor Exposure: occupation, geographic area; Cancer Site: esophagus; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: age-adjusted mortality rate, higher level statistics; Time Frame: 1950-1969 Sample Characteristics Esophagus cancer mortality data from the 3,056 contiguous counties of the United States. Variables SES: industry; Demographic: sex, race-ethnicity (nonwhite, white), geographic area, urban-rural Data Sources U.S. Population Census Abstract A survey of esophageal cancer mortality among U.S. counties, 1950-1969, revealed higher rates among males in the Northeast and among females in the South. Throughout the country mortality among White males and Nonwhites of both sexes increased mar- kedly with urbanization. A nearly 2:1 excess among Nonwhites was observed, with the largest racial disparity in urban areas. Clustering of exceptionally high rates among Nonwhites was ob- served in a narrow band of counties along the Southeast Atlantic coast, particularly around Charleston, South Carolina. Among Whites, mortality was inversely proportional to socioeconomic in- dices and strongly related to ethnicity, with elevated rates among males in counties with high percentages of residents of Irish, Ger- man, Polish, Czechoslovak, and other East European descent, and among females in counties with Irish and Scandinavian com- munities. Little concomitant variation with county involvement in any of 18 manufacturing industries was observed for males, but slightly higher rates among females were seen in southern counties with textile or paper industries. Fraumeni, J.F., Jr.; Li, F.P. Hodgkin's Disease in Childhood: An Epidemiologic Study J Natl Cancer Inst 42:681-691, 1969 95 Subjects Risk Factor Exposure: familial factors, other diseases, geographic area; Cancer Site: lymphoma; Cancer Measures: mortality Study Characteristics Design: time trend, cross-sectional; Measures: age-specific mortality rate, standard mortality ratio (SMR); Time Frame: 1940-1968 Sample Characteristics 359 Hodgkin's disease deaths (Negroes 46; Caucasians 308; Filipinos 1; Chinese 1; Japanese 1; American Indians 1; Eskimos 1) under 15 years old; United States; 314 Hodgkin's disease cases: under 17 years old, 13 hospitals; National Cooperative Leukemia Survey Variables Demographic: age, sex, race-ethnicity, geographic area Data Sources Death registration, hospital medical records Abstract Study was made of 359 death certificates of United States chil- dren who died of Hodgkin's disease, 1960-64, and of 314 hospital 44 charts of children with this neoplasm. Unlike most childhood neop- lasms, there was no early peak in mortality or relation to specific congenital defects to suggest a prenatal origin. Hodgkin's disease in children increased progressively with age, and the much earlier rise in morbidity among boys resulted in a 3:1 male excess. The sex ratio diminished from 10-12 years of age, indicating that the transition to the pattern of Hodgkin's disease in young adults may occur at this time. Among boys dying of Hodgkin's disease, a higher than expected number were born in July and August, suggesting the role early in life of sex-limited environmental agents with seasonal prevalence. Excessive mortality also occurred in the West-South Central division of the United States, an area with the lowest rate for the disease at older ages. Among children there was no difference in mortality by race, in contrast to the excess among white adults. A number of features of Hodgkin's disease in children and in young adults are consistent with the activity of environmental factors, with certain childhood patterns resulting possibly from exposure or enhanced susceptibility to such factors close to birth. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Fraumeni, J.E., Jr.; Miller, R.W.; Hill, J.A. Primary Carcinoma of the Liver in Childhood: An Epidemiologic Study J Natl Cancer Inst 40:1087-1099, 1968 96 Subjects Risk Factor Exposure: geographic area, familial factors, other diseases; Cancer Site: liver; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: standard mortality ratio (SMR), descriptive statistics; Time Frame: 1950-1964 Sample Characteristics 282 liver cancer deaths (Negroes 27; whites 252; Orientals 2; Indian-Filipinos 1): under 15 years old; United States; 76 liver cancer cases: under 15 years old; hospital patients from Boston, MA; Los Angeles, CA; Chicago, IL; Baltimore, MD; St. Louis, MO; Philadelphia, PA; the National Cooperative Leukemia Survey Variables Demographic: age, sex, race-ethnicity (nonwhite, white), geographic area; Cancer: histology Data Sources Death registration, hospital medical records, survey Abstract In a search for etiologic clues, study was made of 282 death certificates of U.S. children who died of hepatic carcinoma, 1960-64, and of 76 hospital charts of children with this neoplasm. There was an early peak in mortality which, coupled with the occurerence of neonatal liver cancer, suggests that in early childhood the tumor has a prenatal origin. The array of congenital defects recorded in the hospital series indicates that this tumor may share etiologic mechanisms with Wilms’ tumor and adrenocortical neoplasms. Two children with liver cancer had congenital hemihypertrophy and two had extensive hemangiomas, anomalies which occur in excess with Wilms’ tumor and adrenocortical neoplasms; one child had Wilms’ tumor synchronous with the diagnosis of liver diseases: nonspecific cirrhosis (2 cases), neonatal hepatitis, congenital biliary atresia, and de Toni-Fanconi syndrome (1 case each). Additional cases found in the mortality series included von Gierke’s disease. The early peak incidence of liver neoplasia seems related to coexist- ing congenital anomalies such as hemihypertrophy and to liver damage following biliary atresia or neonatal hepatitis, whereas liver disease due to nonspecific or metabolic cirrhosis is a precursor to neoplasia in older children and adults. Geographic variaton in liver cancer mortality occurred in children under 5 years of age, which suggests an environmental influence on the early peak. No evidence was found to implicate genetic factors, other than their role in certain disorders which increase the risk of liver cancer. Fraumeni, J.F., Jr.; Vogel, C.L.; Easton, ].M. Sarcomas and Multiple Polyposis in a Kindred—A Genetic Arch Intern Med 121:57-61, 1968 Variety of Heredity Polyposis? 97 Subjects Diagnosis; Risk Factor Exposure: other diseases, familial factors; Cancer Site: colon, sarcoma Study Characteristics Design: case study; Time Frame: 1956-1964 Sample Characteristics Four cancer cases (males 2; females 2): Negro, 16-50 years old Variables Cancer: histology Data Sources Hospital medical records, autopsy records, histopathology records Abstract A family is presented in which the father and one of his children died, respectively, of pulmonary osteosarcoma and retroperitoneal 45 liposarcoma; another child died of carcinoma of the colon with multiple polyposis; and a third has both a reticulum cell sarcoma and colonic polyps. The familial association of polyposis coli with extra-alimentary sarcomas has not previously been reported, and may be a variant of Gardner's syndrome (multiple polyposis with benign mesenchymal tumors). As in Gardner's syndrome, the fam- ilial pattern was consistent with autosomal dominant inheritance of a single mutant, highly penetrant, pleiotrophic gene. Phenotypic differences from Gardner's syndrome and familial polyposis suggest either that there was a separate gene or that the gene for one of these syndromes was expressed differently by modifiers segregating in this family. The development of fatal metastatic carcinoma of the colon in an 11-year-old boy appears to be the earliest recorded instance of this lesion in hereditary polyposis. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Funderburk, W.W.; Rosero, E.; Leffall, L.D. Breast Lesions in Blacks Surg Gynecol Obstet 135:58-60, 1972 98 Subjects Diagnosis; Cancer Site: breast; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: crude incidence rate, descriptive statistics; Time Frame: 1944-1969 Sample Characteristics 2,552 breast tumor biopsies: black; Washington, DC Variables Cancer: histology Data Sources Histopathology records Abstract Results from a study of 2,552 biopsies of breast lesions in blacks who were diagnosed at Howard University Hospital between 1944- 1969. The authors attempted to substantiate the claim that more benign lesions and less carcinomas have been seen in biopsies of the breast than are quoted in the medical literature. Specifically they suggest a greater incidence of fibroadenomas. Neoplasms comprised 55.1 percent and benign neoplasm made up 32 percent of the biopsies. Eighty-six percent of all benign neoplasms were fibroadenomas. Fibroadenomas occurred almost twice as often in blacks as was documented in whites. Itis concluded that adherence to biopsy of all suspicious mammary lesions is warranted, espe- cially for blacks. Garabrant, D.H.; Peters, J.M.; Mack, T.M.; et al. Job Activity and Colon Cancer Risk Am J Epidemiol 119(6):1005-1014, 1984 99 Subjects Risk Factor Exposure: occupation; Cancer Site: colon, rectum; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, proportional incidence ratio (PIR), descriptive statistics, higher level statistics; Time Frame: 1972-1981 Sample Characteristics 2,950 colon cancer cases (blacks 253; North American Hispanic whites 108; immigrant whites 46; non-Hispanic whites 2,373; not given 170): male, 20-64 years old; Los Angeles County, CA Variables SES: occupation, census tract SES rank; Demographic: age, race-ethnicity; Cancer: tumor location Data Sources non-SEER population-based cancer registry Abstract The authors studied 2,950 population-based colon cancer cases in males in Los Angeles County, California, that were diagnosed between 1972 and 1981. To determine if colon cancer risk is reduced by physical activity on the job in males aged 20-64 years, the authors first rated each occupation by judging the activity level as high, moderate, or sedentary. Men with sedentary jobs had a colon cancer risk at least 1.6 times that of men whose jobs required a high level of activity. Risk increased in a stepwise manner as activity level decreased. This gradient was consistently seen within each socioeconomic stratum, among whites, blacks, immigrant and native Hispanics, and for each subsection of the colon from the hepatic flexure to the sigmoid. The protective effect of physical activity was very strong in the descending colon and diminished in a gradient both proximally and distally. There was no such relationship between physical activity and risk for rectal cancer. Physical activity may play a major, previously unrecognized role in colon cancer occurrence and with our understanding of colon physiology and colon cancer pathogenesis. In addition to the im- plications for prevention, understanding the effects of physical activity on colon cancer risk may allow future studies to evaluate more accurately the role played by diet. Garg, S.K.; Subbuswamy, S.G.; Alabi, J.O.; et al. Dermatofibrosarcoma Protuberans in Northern Nigeria: A Clinico-Pathological Review of 17 Cases Clin Oncol 4:113-122, 1978 100 Subjects Treatment; Cancer Site: sarcoma Study Characteristics Design: case study; Time Frame: 1973-1976 Sample Characteristics 17 sarcoma cases: presumed black, 16-62 years old; Nigeria Data Sources Histopathology records Abstract Seventeen cases of dermatofibrosarcoma protuberans seen in Northern Nigeria during the 4 years 1973 to 1976 are reviewed. There is some evidence that the tumour may be more common in Negroes than in Caucasians. It is a slow-growing mesenchymal tumour of the skin, with liability to local recurrence, but an ex- tremely low tendency to metastasize. The tumour is now consi- dered to be of histiocytic origin and is regarded as a very low grade malignant fibrous histiocytoma. Recognition of the tumour can be difficult. Initial biopsy is essential, followed by wide exci- sion, usually requiring skin grafting or other reconstruction. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Gaskin, H.S.; Hardy, R.E.; Fletcher, R.L. Multiple Primary Malignancies in Black Patients J Natl Med Assoc 73(11):1065-1068, 1981 101 Subjects Cancer Site: multiple sites; Cancer Measures: prevalence Study Characteristics Design: retrospective; Measures: descriptive statistics; Time Frame: 1959-1979 Sample Characteristics 42 multiple primary cancer cases: black, 36-81 years old; Nashville, TN Variables Demographic: age, sex; Cancer: tumor location Data Sources non-SEER hospital cancer registry Abstract Charts of 42 black patients with multiple malignant neoplasms were among 1,953 cancer patients selected for review during the period of 1959 through 1979. The incidence was 2.15 percent. Most patients were female and the breast was the most frequent initial primary carcinoma. With this relatively small number of cases, there was no consistent cluster of initial and second primaries about which to make inferences. In the absence of a regional regis- try of primary tumor incidence, no generalities could be obtained concerning a matched non-black population on the incidence of multiple primaries. However, comparison of this group could be made with data previously recorded. Goldberg, L. To the Editor: The Rising Incidence of Retinoblastoma in Blacks S Afr Med J 51:368, 1977 102 Subjects Diagnosis; Cancer Site: eye; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1975-1976 Sample Characteristics 20 eye cancer cases: black, children; Johannesburg, South Africa Data Sources Hospital medical records Abstract The author reports on an apparent increase in the incidence of retinoblastoma among black children in South Africa. The increase does not appear to represent better diagnosis or selective referral as the patients were from a low socioeconomic group presenting with very advanced tumors. Gordis, L.; Szklo, M.; Thompson, B.; et al. An Apparent Increase in the Incidence of Acute Nonlymphocytic Leukemia in Black Children Cancer 47:2763-2768, 1981 103 Subjects Cancer Site: leukemia; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: age-adjusted incidence rate, age-specific incidence rate, higher level statistics; Time Frame: 1960-1974 Sample Characteristics 292 acute leukemia cases, all types (blacks 47; whites 245): under 20 years old; Baltimore, MD Variables SES: census tract rental value; Demographic: age, race-ethnicity; Cancer: histology Data Sources Hospital medical records, U.S. Population Census Abstract A community-wide study was conducted to examine time trends in incidence rates of leukemia in children. Cases of acute leukemia, both lymphocytic (ALL) and nonlymphcytic (ANLL), newly diagnosed in children ages 0-19 years in the Baltimore area from 1960 to 1974 were ascertained. Over the 15 years, 286 children with acute leukemia were identified, of whom 77% had ALL. In- cidence rates of ALL were two or three times as high in whites as in blacks, and remained virtually unchanged over time. However, the picture was quite different for ANLL. Initially, incidence rates were higher in whites than in blacks, but during the final five years the incidence increased dramatically in blacks so that it actu- ally exceeded the rate among whites. The increase of ANLL in black children occurred primarily in blacks of high socioeconomic status. This suggests that the increase may have resulted from environmental influences associated with upward socioeconomic mobility in the black population in recent years. 47 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Graham, S.; Priore, R.; Graham, M.; et al. Genital Cancer in Wives of Penile Cancer Patients Cancer 44:1870-1874, 1979 104 Subjects Cancer Site: ovary, corpus uteri, vagina, cervix uteri, penis; Risk Factor Exposure: other diseases; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1951-1975 Sample Characteristics 255 penis cancer cases (blacks 22; whites 208; unknown 25): male; New York Variables Demographic: age, race-ethnicity Data Sources non-SEER population-based cancer registry Abstract We identified the wives or ex-wives of 227 males of the 256 reported with cancer of the penis to the New York State Cancer Registry from Upstate New York from 1960-64. Utilizing the Regis- try, death certificates, hospital and physician records, we ascer- tained those wives who developed cancer at any site from 1951- 1975. We generated expected numbers of cases of cancer at each site by applying the age-specific incidence rates experienced by women of a specific age in a specific year designated by the age of the wife of the index case in each year, estimating withdrawals from age-specific death rates. Thus, our expected numbers are based on the experience of the women in Upstate New York with traits like those of spouses of the men in the same population with cancer of the penis. We found significantly more cases of cancer of the cervix than expected. This was not true for other sites of cancer. Gray, G.E.; Henderson, B.E.; Pike, M.C. Changing Ratio of Breast Cancer Incidence Rates With Age of Black Females Compared With White Females in the United States J] Natl Cancer Inst 64:461-463, 1980 105 Subjects Risk Factor Exposure: reproductive factors; Cancer Site: breast; Cancer Measures: incidence, mortality Study Characteristics Design: cross-sectional; Measures: standardized incidence ratio (SIR), standardized mortality ratio (SMR), descriptive statistics; Time Frame: 1937-1976 Sample Characteristics Cancer incidence data: TNCS areas and Los Angeles County, CA; cancer mortality data: United States Variables Demographic: age, race-ethnicity (black, white, nonwhite), age at menarche, age at menopause, age at first pregnancy Data Sources Third National Cancer Survey 1969-1971, non-SEER population-based cancer registry, death registration Abstract Age-standardized breast cancer rates were approximately 30% lower in U.S. black women compared to white women. This obser- vation concealed the fact that black women under age 40 years had a higher incidence of breast cancer than did white women, whereas white women over 40 years had a higher incidence. The known risk factors for breast cancer development (early age at menarche, late age at first full-term delivery, and a late age of menopause) differed in black and white populations, which might explain this difference in breast cancer incidence between blacks and whites at different ages. Greenberg, M. Kupffer Cell Sarcoma of the Liver. Report of 2 Cases in South African Blacks 106 S Afr Med ] 52:244, 1977 Subjects Cancer Site: liver; Cancer Measures: incidence Study Characteristics Design: case study; Time Frame: 1964, 1976 Sample Characteristics Two Kupffer cell sarcoma of the liver cases: black, hospital patients; Johannesburg, South Africa Data Sources Hospital medical records 48 Abstract The first 2 cases of Kupffer cell sarcoma of the liver in South African Blacks are reported from Baragwanath Hospital. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Greenberg, R.S.; Stevens, J.A.; Whitaker, J.P. Cancer Incidence Rates Among Blacks in Urban and Rural Georgia, 1978-82 Am ] Public Health 75(6):683-684, 1985 107 Subjects Risk Factor Exposure: geographic area; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, higher level statistics; Time Frame: 1978-1982 Sample Characteristics Cancer incidence data: blacks; Georgia Variables Demographic: sex, urban-rural Data Sources SEER cancer registry, hospital medical records, physician medical records, histopathology records, autopsy records, death registration Abstract The records of the Atlanta Cancer Surveillance Center were reviewed for all incident cases of cancer diagnosed among Black residents of the catchment areas during calendar years 1978 through 1982. The resultant age-adjusted overall cancer incidence rates for urban Blacks were greater than those for rural Blacks. The largest urban excess was found for cancers of the buccal cavity and pharynx. Urban residence also was associated with lung and prostate cancers among males, and pancreatic cancers among females. Greene, M.H.; Fraumeni, J.F., Jr.; Hoover, R. Nasopharyngeal Cancer Among Young People in the United States: Racial Variations by Cell Type 108 J Natl Cancer Inst 58(5):1267-1270, 1977 Subjects Abstract Cancer Site: nasopharyngeal; Cancer Measures: incidence, U.S. mortality and incidence statistics for nasopharyngeal mortality cancer showed a fourfold excess risk of sarcomas in white children Study Characteristics Design: cross-sectional; Measures: crude mortality rate, relative risk, descriptive statistics, higher level statistics; Time Frame: 1960-1969, 1969-1971 Sample Characteristics 144 nasopharyngeal cancer deaths (blacks 36, whites 108): under 20 years old; United States; 19 nasopharyngeal cancer cases (blacks 6; whites 13): TNCS areas; 155 nasopharyngeal cancer cases (nonwhites 42; whites 113): under 20 years old; End Results Program areas Variables SES: census tract median years of school completed by adult population; Demographic: age, sex, race-ethnicity, geographic area; Cancer: histology Data Sources Death registration, U.S. Population Census, Third National Cancer Survey 1969-1971, End Results Program under age 10, and a fourfold to sevenfold excess of carcinomas in teen-age blacks. Mortality from nasopharyngeal carcinomas in young people was greater in the South than in the North, with the excess mortality in blacks linked to rural residence and low socioeconomic status. These and other characteristics of nasopharyngeal carcinoma in young persons suggested that en- vironmental (perhaps infectious) agents are involved in this age group. These patterns contrasted with nasopharyngeal carcinomas developing after age 25, when the rates predominated in Chinese Americans. Nasopharyngeal cancer in the United States had three age peaks, with racial and epidemiologic distinctions that seemed to reflect different etiologies. Gregorio, D.I.; Cummings, K.M.; Michalek, A. Delay, Stage of Disease, and Survival Among White and Black Women With Breast Cancer Am ] Public Health 73:590-593, 1983 109 Subjects Diagnosis; Treatment; Cancer Site: breast; Cancer Measures: incidence, mortality, survival Study Characteristics Design: prospective; Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1957-1965 Sample Characteristics 185 breast cancer cases (blacks 29; whites 156): female, hospital patients; Buffalo, NY Variables SES: education; Demographic: age, race-ethnicity; Cancer: stage Data Sources Hospital medical records, death registration, survey Abstract Delay in obtaining treatment, stage of disease at diagnosis, and five-year survival were compared for 29 Black and 156 White females treated for breast cancer at Roswell Park Memorial Institute between 1957 and 1965. No statistically significant differences were found between Black and White patients in treatment delay al- though a tendency for longer delay among Blacks was noted. Race had little effect on survival when delay, stage of disease, and age were controlled. 49 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Grillo, I.A.; Bond, L.F.; Ebong, W.W. Cancer of the Colon in Nigerians and American Negroes J Natl Med Assoc 63(5):357-361, 1971 110 Subjects Diagnosis; Treatment; Cancer Site: colon; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1946-1968 Sample Characteristics 304 colon cancer cases (Americans 219; Nigerians 85): black, 11+ years old, hospital patients Variables Demographic: age, sex, race-ethnicity, geographic area; Cancer: histology, tumor location, symptoms Data Sources Hospital medical records Abstract Carcinoma of the colon, rectum and anus has been compared between American and African (Nigerian) Negroes and striking differences have been found only in the pattern of site-distribution within the large bowel and in age range. Nigerians feature twice as many carcinomas in the cecum and anorectal region than pa- tients in the United States. The younger age in which carcinoma of the colon is found in Nigerians as against Americans parallels the younger age distribution of neoplasms in general among Nige- rians. Guinan, P.D.; Wiggishoff, C.C.; Rotkin, I.D.; et al. Prostate Cancer in Illinois Ill Med J 166(5):341-345, 1984 111 Subjects Diagnosis; Treatment; Cancer Site: prostate; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics, crude survival rate, relative survival rate; Time Frame: 1970-1984 Sample Characteristics 4,389 prostate cancer cases (blacks 831; whites 3,431; Orientals 4): Illinois Variables Demographic: age, sex, race-ethnicity; Cancer: stage, histology Data Sources Hospital medical records, non-SEER hospital cancer registry Abstract An analysis is presented of tumor registry data covering 4,389 patients from 50 Illinois hospitals who were diagnosed with pros- tate cancer between 1970 and 1974. Distributions of cases of age, race, stage, method of diagnosis, survival, and modality of treat- ment are given. Findings include increased incidence and more advanced stages at time of diagnosis in black patients. Relative survivals for all patients correlate inversely with age and stage. Trends generally parallel those of U.S. data. Hacking, E.A.; Dent, D.M.; Gudgeon, C.A. Malignant Tumors of the Breast S Afr Med J 65:323-324, 1984 112 Subjects Cancer Site: breast; Cancer Measures: incidence Study Characteristics Design: cross-sectional, time trend; Measures: descriptive statistics; Time Frame: 1971-1981 Sample Characteristics 2,215 breast cancer cases (blacks 66; coloreds 1,063; whites 1,085): female, 13-105 years old; Cape Town, South Africa Variables Demographic: age, race-ethnicity; Cancer: stage Data Sources Hospital medical records 50 Abstract During the 11-year period 1971 - 1981, 2,215 new female patients with breast cancer were seen at the Groote Schuur Hospital Breast Clinic, Cape Town. The mean age at presentation of White patients was similar to that in most Western countries, but the disease was seen at an earlier age in Coloured and Black patients. Coloured and Black patients also presented with more advanced disease than White patients. There was, however, a progressive overall trend over the 11 years towards presentation at an earlier stage of the disease. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Haerer, A.F.; Jackson, ].F.; Evers, C.G. Ataxia-Telangiectasia With Gastric Adenocarcinoma JAMA 210(10):1884-1887, 1969 113 Subjects Diagnosis; Risk Factor Exposure: other diseases; Cancer Site: stomach; Cancer Measures: incidence Study Characteristics Design: case study; Measures: descriptive statistics; Time Frame: 1964-1967 Sample Characteristics Two stomach cancer cases: Negro; Mississippi Data Sources Hospital medical records Abstract This communication concerns a large Negro sibship of which five of 12 members had ataxia-telangiectasia. Two affected sisters also had mucinous adenocarcinoma of the stomach in the second decade of life. More detailed studies in one member revealed de- creased IgA levels, decreased lymphoblastic transformation index, chromosomal abnormalities, and abnormal thymic and lymphoid tissues, in addition to the typical clinical features of this syndrome. The accepted predisposition to development of lymphoid neop- lasms in ataxia-telangiectasia must be extended to include other malignant neoplasms, in view of the occurrence of histologically similar gastric adenocarcinomas in two affected siblings. Haerer, A.F.; Smith, R.R. Neoplasms Involving the Spinal Cord: An Analysis of 85 Consecutive Cases South Med J 61:801-807, 1968 114 Subjects Treatment; Cancer Site: spinal cord; Cancer Measures: survival Study Characteristics Design: prospective; Measures: descriptive statistics; Time Frame: 1961-1967 Sample Characteristics 85 spinal cord cancer cases with spinal laminectomies (Negroes 37; whites 48): Jackson, MS Variables SES: public-private patient status; Demographic: age, sex, race-ethnicity; Cancer: stage, histology, tumor location, symptoms Data Sources Hospital medical records Abstract The authors contrast the onset of symptoms as between metas- tatic tumors of the spinal cord and those of primary origin. Prog- nosis for return of function and duration of life is quite variable in terms of development of paraplegia and decompression of the cord. Hall, P.J.; Levin, A.G.; Entwistle, C.C.; et al. HLA Antigens in East African Black Patients with Burkitt's Lymphoma or Nasopharyngeal Carcinoma and in Controls: A Pilot Study Hum Immunol 5:91-105, 1982 115 Subjects Biologic Marker: HLA antigens; Risk Factor Exposure: other diseases; Cancer Site: Burkitt's tumor; nasopharyngeal Study Characteristics Design: cross-sectional; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1972-1976 Sample Characteristics 141 individuals (Burkitt's lymphoma cases 42; nasopharyngeal cancer cases 47; controls 52): black, 3-64 years old; Kenya, Tanzania Variables Demographic: age; Cancer: histology Data Sources Laboratory test Abstract A pilot study is reported of HLA-A, B, and C antigens in 141 East African Blacks comprising patients with Burkitt's lymphoma or nasopharyngeal carcinoma, either with active disease or in long- term remission, together with comparable controls. This study forms part of a wider program investigating host factors in these 51 diseases. A protocol was selected for optimal testing of cells pro- cessed and cryopreserved between 1972 and 1976, largely under field conditions, which employed a two-color fluorochromasia typ- ing procedure. Antigen distribution and computed haplotype fre- quencies in the total unrelated population are given. New findings include an approximately equal frequency of Aw23 and Aw24, a high (18%) incidence of Bw21, and the gametic associations of Aw36 with Bw44, and Aw30 with Bw45. Of the major group of B15-related antigens reported earlier, SV is the most common, and there are strong linkages of SV with Cw2 and Bu with Cw3. The possible presence of further variants at the A- and B-loci is re- ported. The proportion of B-locus antigen “blanks” in this study is 5.9%. Relationships have been sought between the HLA antigens and diseases studied: the antigen A29, possibly in linkage with Bw42, shows a correlation with disease susceptibility, and associ- ations are suggested between Bw44 (in possible combination with Aw36) and resistance to both BL and NPC, and between Bw45 and long-term remission in NPC. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Hammond, M.G.; Angorn, B. HLA and Cancer of the Esophagus in South African Negroes Tissue Antigens 16:254-255, 1980 116 Subjects Cancer Site: esophagus; Biologic Marker: HLA Study Characteristics Design: cross-sectional Sample Characteristics 1,010 individuals (esophagus cancer cases 254; controls 756): Negro; Natal, South Africa Data Sources Laboratory test Abstract This short communication was written to rectify results of an earlier study. The first study found an increased frequency of HLA-Bw45 in Negroes with cancer of the esophagus. However, the frequency of HLA-Bw45 in a follow-up study of a second group of patients was found not to be significantly different from the frequency in controls, which means that the original observation was probably due to chance. In the first study there was also a significant decrease in the frequency of patients with only one detectable antigen at the A locus but this was not confirmed in the second group. Hammond, M.G.; Appadoo, B.; Brain, P. HLA and Cancer in South Africa Negroes Tissue Antigens 9:1-7, 1977 117 Subjects Diagnosis; Biologic Marker: HLA; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 18 months Sample Characteristics 1,000 individuals (cancer cases 500; controls 500): Negro; South Africa Variables Demographic: age Data Sources Laboratory test Abstract Five hundred patients with cancer were tested for 32 HLA antigens and the antigen frequencies compared with those of 500 control subjects matched for race, sexand age. Although the overall frequencies showed no significant differences, detailed analysis with regard to site of cancer, age and the number of antigens detected at each locus revealed significant differences. Phenotype traces and haplotype frequencies have been included. Hann, H.L.; Lustbader, E.D.; Evans, A.E.; et al. Lack of Influence of T-Cell Marker and Importance of Mediastinal Mass on the Prognosis of Acute Lymphocytic Leukemias of Childhood J Natl Cancer Inst 66(2):285-290, 1981 118 Subjects Cancer Site: leukemia; Cancer Measures: survival Study Characteristics Design: prospective; Measures: descriptive statistics, higher level statistics Sample Characteristics 100 leukemia cases: under 17 years old, newly diagnosed; Philadelphia, PA Variables Demographic: age, sex, race-ethnicity (black, white); Cancer: histology Data Sources Histopathology records, laboratory test Abstract One hundred consecutive new cases of acute lymphocytic leukemia (ALL) were studied in patients that were 4 months to 16 years of age when admitted to The Children’s Hospital of Philadelphia. Various prognostic factors were examined and re- lated to the duration of the first remission. These factors included lymphoblast surface markers, age at diagnosis, sex, race, initial white blood counts (WBC), presence of mediastinal mass, degree of hepatosplenomegaly, and lymph node size. Classification by lymphoblast surface markers showed 22 T-Cell, 71 null cell, and 3 B-cell leukemias; 3 cases were unclassifiable and 1 had both T- and B-cell markers. Statistical analysis indicated that stratification by presence or absence of mediastinal mass was necessary. Most patients with mediastinal masses, 5 of thymus and 4 of the non- thymus type, fared poorly with a median duration of continuous complete remission of less than 12 months as compared with greater than 48 months for those without such masses. The most satisfactory model to estimate remission duration in children with- out mediastinal masses were dependent upon initial WBC, sex, race, and surface markers. Low WBC, white race, female sex, and T-cell markers in patients without mediastinal masses were as- sociated with a favorable prognosis. The findings suggest that patients with mediastinal masses need special therapy and that T-cell ALL without a mediastinal mass does not carry a poorer prognosis than does null cell ALL. 52 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Hansen, O.E. Thyroid Disease in Blacks S Afr Med J 62:851-852, 1982 119 Subjects Diagnosis; Treatment; Cancer Site: thyroid; Cancer Measures: prevalence Study Characteristics Design: cross-sectional; Measures: prevalence rate Sample Characteristics 330 patients with thyroid gland swelling (malignant thyroid tumors 71; nonmalignant cases 259): black; Pretoria, South Africa Variables Demographic: age, sex; Cancer: histology Data Sources Hospital medical records Abstract Three hundred and sixty consecutive patients with thyroid swellings were investigated and operated on at Ga-Rankuwa Hos- pital, Pretoria. Patients were referred from 34 homeland hospitals and from Caprivi, making this series a reasonable reflection of thyroid disease in Blacks. A noteworthy increase in the incidence of malignant thyroid tumours becomes evident when the series is compared with those in previous relevant publications. A more radical surgical approach is defended in the light of current neocon- servatism. Hardy, J.D.; Ewing, H.P.; Neely, W.A_; et al. Lung Carcinoma: Survey of 2286 Cases With Emphasis on Small Cell Type 120 Ann Surg 193(5):539-548, 1981 Subjects Diagnosis; Treatment; Cancer Site: lung; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); time trend; Measures: descriptive statistics; Time Frame: 1955-1980 Sample Characteristics 2,286 lung cancer cases: Jackson, MS Variables Demographic: age, sex, race-ethnicity (black, white); Cancer: histology Data Sources Hospital medical records, non-SEER hospital cancer registry Abstract Lung carcinoma is the commonest major malignancy in men in the United States and its incidence is increasing rapidly in women. It is estimated that there will have been 117,000 new cases and 101,300 deaths in 1980. The 2286 patients with lung carcinoma admitted to the Hospital of the University of Mississippi from 1955 to 1980 were reviewed by decades of chronology and of life, with respect to age, cell type, sex and racial incidence. The greatest age incidence was in the sixty and seventh decades; cell types overall were epidermoid (45% of the patients), adenocarcinoma (12% of the patients), small (oat) cell (21% of the patients), and others (22% of the patients). There was a steady increase in the incidence of disease in females, adjusted for total hospital admissions, and a less certain increase among black patients. Twenty-eight per cent of 250 patients with small cell carcinoma so studied exhibited some feature of the paraneoplastic or paraendocrine syndromes. In 41 patients with small cell carcinoma treated with multiple drug chemotherapy, there was an overall response rate of 50% and an additional “stable disease” rate of 28%. Mean survival period in this group was 52 weeks, compared with 12 weeks in patients whose diseases went untreated. Clearly, definite progress is being made, not only in our knowledge of the biology of lung carcinoma, in general, but in the treatment of small cell carcinoma in particular. Harris, M. Carcinoma of the Female Breast in Jamaica Trop Geogr Med 29:213-222, 1977 121 Subjects Cancer Site: breast; Cancer Measures: incidence Study Characteristics Design: cross-sectional, time trend; Measures: crude incidence rate, age-specific incidence rate; Time Frame: 1958-1970 Sample Characteristics Breast cancer incidence data: female; Kingston and St. Andrew (K.S.A.), Jamaica Variables Demographic: age; Cancer: stage, histology Data Sources non-SEER population-based cancer registry Abstract The incidence of breast cancer in Jamaican women is inter- mediate between the low levels of African countries and the high levels of USA, UK, and other Western industrial societies. The annual incidence in Jamaica has shown an upward trend from 1958 to 1974. Compared with Uganda, breast cancer in Jamaica conforms to a “westernized” pattern with regard to shape of the age-specific incidence curve and stage of presentation. There are no obvious histological differences between breast cancer in Uganda, USA and Jamaica. The incidence of breast cancer in Jamaica appears disproportionately high in view of the high prop- ortion of early pregnancies occurring in Jamaican women. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Hayashi, K.; Bracken, M.B.; Freeman, D.H., Jr.; et al. Hydatidiform Mole in the United States (1970-1977): A Statistical and Theoretical Analysis Am ] Epidemiol 115(1):67-77, 1982 122 Subjects Risk Factor Exposure: reproductive factors; Cancer Site: gestational trophoblastic disease; Cancer Measures: incidence Study Characteristics Design: time-trend; Measures: crude incidence rate, age-adjusted incidence rate, age-specific incidence rate, descriptive statistics; Time Frame: 1970-1977 Sample Characteristics 31,965 estimated hydatidiform mole cases (blacks 2,764; whites 23,945; others 5,256): 15+ years old, female; United States Variables Demographic: age, type of delivery, parity, race-ethnicity, geographic area Data Sources National Center for Health Statistics survey Abstract The incidence of hydatidiform mole and some epidemiologic features of the disease in the United States were studied. Data for the present study were obtained from the Hospital Discharge Sur- vey (1970-1977) of the National Center for Health Statistics. The estimated incidence of hydatidiform mole over the eight-year period was 108.4 per 100,000 pregnancies (SE = 10.3) or one case in every 923 pregnancies. There was an apparent increase in the incidence rates from 1970 to 1975, but the rate dropped slightly after 1975. The distribution of hydatidiform mole by maternal age showed the highest rates among women 35 years and older, and the second highest rate among women aged 15-19. The distribution by maternal age found for the entire sample was also apparent within each race category. Overall, black women appeared to have an incidence rate half that of white women. Regional differences in the United States were observed in spite of race and age adjust- ment. Existing studies are reviewed in light of morphologic and cytogenetic classification schemes and a recent hypothesis for the androgenetic origin of hydatidiform mole. A fresh approach to the epidemiology of hydatidiform mole which combines epdemiologic and genetic studies is indicated. Heath, F.R.; Shalaby, O.F.O.; Abdel-Dayem, H.M. Epidemiologic and Prognostic Factors for Lung Cancer in a County Hospital J Surg Oncol 19:109-113, 1982 123 Subjects Treatment; Risk Factor Exposure: tobacco, occupation; Cancer Site: lung; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: mean survival time, descriptive statistics, higher level statistics; Time Frame: 1973-1978 Sample Characteristics 269 lung cancer cases (blacks 55; whites 214): hospital patients; Buffalo, NY Variables SES: occupation; Demographic: age, sex, race-ethnicity; Cancer: histology, metastases, nodal involvement Data Sources Hospital medical records, non-SEER hospital cancer registry Abstract This is a study of 269 patients with lung cancer at the Erie County Medical Center who were admitted between 1973 and 1978. They were analyzed for sex, race, age, and history of smok- ing, occupation, tumor cell type, cytology, incidence of metastases, changes in liver function, mode of treatment, and survival. The incidence of cancer was highest in white males. Only 1.5% of patients were under age 40. Smoking was a predisposing factor. Not enough information was available to determine the relation- ship of occupation to lung cancer. Squamous cell carcinoma was the most common (53.9%), followed by adenocarcinoma (16.0%) and small-cell carcinoma (12.6%). Sputum cytology was 28.3% sensitive, and bronchial washings were 52.2% sensitive. A greater incidence of bone metastases from a small-cell primary (50%) was found than is reported in the literature. Changes in SGOT and/or SGPT liver enzymes correlated significantly with liver metastases, but not with tumor cell type. Changes in alkaline phosphatase correlated well with bone metastases. Radiation was the most com- monly used mode of therapy. The best survival was achieved in patients treated by surgery (22.6 months), followed by surgery and radiation (16.2 months); those treated by radiation alone had a mean survival time of 8.7 months. Untreated patients had a mean survival time of 2.4 months. Treated patients with adenocar- cinomas had the longest survival (18.5 months), compared to 13.0 months for those with squamous cell carcinomas and 8.4 months for those with small-cell carcinomas. Only three patients survived 5 years, all of whom were treated surgically for adenocarcinoma. No patients with tumors of other cell types survived 5 years. The 5-year survival rates are 2.1% (3 of 141) for all histologic types of lung cancer and 12.5% (3 of 24) for adenocarcinoma. 54 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Heizer, W.D.; Lewison, E.F. Concordant Disease in Identical Twins JAMA 188(3):217-220, 1964 124 Subjects Diagnosis; Risk Factor Exposure: familial factors; Cancer Site: breast Study Characteristics Design: case study Sample Characteristics Two breast cancer cases: Negro, female, 67 years old, identical twins; Baltimore, MD Data Sources Hospital medical records Abstract A unique pair of identical twins were concordant for breast cancer, diabetes mellitus, nontoxic goiter, hypertension, cholecys- titis, obesity, relaxed pelvic outlet, and myomata uteri. The medical literature for concordant breast cancer in identical twins is reviewed and there seems to be a strong tendency for these tumors to occur in the same breast and at the same time. Henschke, U.K; Leffall, L.D., Jr.; Mason, C.H.; et al. Alarming Increase of the Cancer Mortality in the U.S. Black Population (1950-1967) Cancer 31(4):763-768, 1973 125 Subjects Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1950-1967 Sample Characteristics Cancer mortality data from the United States Variables Demographic: sex, race-ethnicity (nonwhite, white) Data Sources Death registration Abstract From 1950 to 1967 the U.S. cancer mortality per 100,000 for both sexes rose for blacks from 147 to 177, an increase of 20%, while it remained unchanged for whites at 150. The female cancer mortality rate declined for blacks from 146 to 142, a decrease of only 3%, while it declined for whites from 139 to 126, a decrease of 9%. The male cancer mortality rate rose for blacks from 147 to 220, an increase of 50%, while it increased for whites from 158 to 181, an increase of only 16%. In 1950, the cancer mortality rate for both sexes was 2% lower for blacks than for whites, but, by 1967, it had become 18% higher. Of the 56 most frequent U.S. sex- specified cancer types, 29 increased slower in whites, 9 decreased faster in whites, 14 showed no significant differences, and only 4 (malignant melanoma and reticulum cell sarcoma in both sexes) increased faster in whites. Environmental factors are the most likely cause for this alarming rise of cancer in U.S. blacks. Henson, D.; Tarone, R. Epidemiologic Study of Cancer of the Cervix, Vagina, and Vulva Based on the Third National Cancer Survey in the United States Am |] Obstet Gynecol 129:525-532, 1977 126 Subjects Cancer Site: cervix uteri, vagina, vulva; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: relative risk, age-adjusted incidence rate, age-specific incidence rate, higher level statistics; Time Frame: 1969-1971 Sample Characteristics Cervix uteri, vagina, and vulva cancer incidence data from the TNCS areas Variables Demographic: age, race-ethnicity (black, white), geographic area; Cancer: stage Data Sources Third National Cancer Survey, 1969-1971 Abstract Cases of invasive and in situ carcinoma of the lower female genital tract as reported in the Third National Cancer Survey in the United States were analyzed according to age, race, and geog- raphic distribution. Results indicate that the incidence rates of in situ and invasive carcinoma of the cervix were greater in black than in white women, with a relative risk rate for black women of approximately two for both types of cervical carcinoma. For white women, the age-specific rates for invasive carcinoma of the cervix remained relatively constant after age 45, while for black women the age-specific rates for invasive carcinoma continued to increase after age 45. For both races, the patterns of age-specific incidence rates for in situ and invasive carcinoma of the cervix were not similar to those for carcinoma of the vagina or vulva. The pattern of age-specific incidence rates of adenocarcinoma of the cervix did not resemble those for in situ or invasive squamous cell carcinoma of the cervix. The rates for adenocarcinoma of the cervix demonstrated patterns similar to those for intraductal car- cinoma of the breast. Results of the study are discussed in relation- ship to the field theory of carcinogenesis as developed for the lower female genital tract. 55 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Herbsman, H.; Feldman, J.; Seldera, J.; et al. Survival Following Breast Cancer Surgery in the Elderly Cancer 47:2358-2363, 1981 127 Subjects Treatment; Cancer Site: breast; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, relative survival rate, descriptive statistics; Time Frame: 1966-1979 Sample Characteristics 780 breast cancer cases (blacks 367; whites 365; Puerto Ricans 36; others 7): female, Stages I, II, III with surgical therapy; Brooklyn, NY Variables Demographic: age, race-ethnicity; Cancer: stage, tumor size Data Sources non-SEER hospital registry Abstract A retrospective analysis was performed on a series of 780 pa- tients with breast cancer who underwent surgery, with emphasis on survival in the 138 patients over 70 years of age (elderly group). Absolute survival rates were determined using life table methods. Relative survival figures were derived from absolute figures by adjusting for expected longevity in each age group. Patients over 70 years of age had overall 5 and 10 year absolute survival rates of 54% and 41%, respectively, which did not differ significantly from survival rates of younger patients. Relative five-year survival rates for the elderly were actually higher than those of younger patients, being 90% for those with local disease and 65% for those with regional disease. Survival in the elderly was comparable to that of younger patients, irrespective of race, type of surgery, histology or tumor size. These findings support the conclusion that there is little justification for avoiding conventional operative treatment in elderly patients with breast cancer solely on the basis of advanced age. Herring, B.D. Cancer of the Prostate in Blacks J Natl Med Assoc 69(3):165-167, 1977 128 Subjects Prevention; Cancer Site: prostate; Cancer Measures: incidence Study Characteristics Design: case study Sample Characteristics Two prostate cancer cases: black, male Abstract The exact etiology of adenocarcinoma of the prostate gland remains an enigma. However, there appears to be a clear-cut man- date to search for, and to eradicate if possible, all infections of the genital tract including prostatitis. Diets with adequate zinc or zinc supplements should be considered. Sexual practices could be more closely scrutinized. Finally, an urgent appeal is made for intense investigation of this malady. Two cases of prostatic carcinoma are presented and pertinent literature is reviewed. Heydt, D.; Thompson, S.H.; Shakenovsky, B.N. Transition of Apical Periodontal Cysts to Intramedullary Osteomyelitis: A Clinicopathological Analysis J Endod 11(2):67-70, 1985 129 Subjects Risk Factor Exposure: other diseases; Cancer Site: bone; Cancer Measures: prevalence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1968-1983 Sample Characteristics 140 intramedullary osteomyelitis cases (blacks 98; whites 42): Johannesburg, South Africa Variables Demographic: sex, age, race-ethnicity; Cancer: tumor location, symptoms, histology Data Sources Histopathology records Abstract One hundred and forty cases of intramedullary osteomyelitis 56 were reviewed and comparatively analyzed with reported informa- tion on apical periodontal cysts (radicular) in essentially the same patient population. This was done with a view toward assessing the probability of developing osteomyelitis as a progression from apical periodontal (radicular) cyst formation. The most common predisposing etiological factor in the development of osteomyelitis was the presence of local inflammatory/immunological changes (apical periodontitis) related to carious teeth. In addition, tooth extraction was found to be a major precipitating cofactor. Clinical and/or histological evidence of apical periodontal cyst or granuloma formation were not identified in this sample. A com- parative analysis of clinical information on osteomyelitis and apical periodontal cysts in the same patient population shows that the apical periodontal cyst is probably infrequently responsible for the progression to intramedullary osteomyelitis. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Higgins, ].E.; Wilkens, L.R.; Chi, I; et al. Hospitalizations Among Black Women Using Contraceptives Am ] Obstet Gynecol 153(3):280-287, 1985 130 Subjects Risk Factor Exposure: contraceptive devices, exogenous hormones; Cancer Site: multiple sites Study Characteristics Design: prospective; Measures: descriptive statistics, higher level statistics; Time Frame: 1968-1976 Sample Characteristics 26,507 individuals (interview cases 290; non-interview cases 26,217): black, female, enrolled in family planning clinic; Atlanta, GA Variables Demographic: age Data Sources Family planning clinic medical records, hospital medical records, survey Abstract We analyzed hospitalizations among 26,507 young black women who attended a large metropolitan family planning clinic between 1968 and 1976. Age-adjusted hospitalization rates were compared for women using oral contraceptives, intrauterine con- traceptive devices, and depot medroxyprogesterone acetate. Over- all, intrauterine contraceptive device and depot medroxyproges- terone acetate users were hospitalized at about the same rate, while women using oral contraceptives were hospitalized 30% less often. The oral contraceptive users were not hospitalized at a higher rate for circulatory disease. Compared to women using intrauterine contraceptive devices, users of oral contraceptives and depot medroxyprogesterone acetate were less likely to be hos- pitalized for benign breast disease (rate ratios = 0.5and 0.2, respec- tively, with 95% confidence limits of 0.3 to 0.7 and 0.1 to 0.5) and for pelvic inflammatory disease. Women were four times as likely to be using depot medroxyprogesterone acetate when they were hospitalized with carcinoma in situ of the cervix, but depot med- roxyprogesterone acetate users with at least 3 years of use were less likely to be hospitalized for carcinoma in situ than comparable groups of oral contraceptive and intrauterine contraceptive device users. Hill, P.; Garbaczewski, L.; Walker, A.R.P. Age, Environmental Factors and Prostatic Cancer Med Hypotheses 14:29-39, 1984 131 Subjects Risk Factor Exposure: diet-nutritional factors, reproductive factors, endogenous hormones, geographic area; Cancer Site: prostate Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 270 individuals (South African blacks 90; North American blacks 80; North American whites 100): male, 11-18 years old; New York, South Africa Variables Demographic: age, race-ethnicity, geographic area, puberty, fertility Data Sources Laboratory test, survey Abstract Although prostatic cancer is evident late in life, pathological evidence suggests this disease is initiated earlier in life. As prostatic cancer is an endocrine associated disease and as adult hormone profiles are established during puberty, it was of interest whether difference in pubertal hormonal levels occurred in populations at low or high risk for prostatic cancer. Accordingly we have inves- tigated the hormone profiles in rural Black South African and urban white boys during puberty. It has been suggested that the timing of puberty is modified by environmental factors and that there is a concomitant control of gonadotrophin release and food intake by CNS-peptide hormones. It is therefore postulated that dietary factors during puberty modify the gut-CNS peptide hor- mones which in turn control the hypothalamic-pituitary-testicular axis. Distinct difference in plasma androgen and gonadotrophins between the two races are in part concordant with a modification of CNS-peptide hormones by environmental factors during pub- erty. 57 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Hill, P.; Wynder, E.L.; Garbaczewski, L.; et al. Response to Luteinizing Releasing Hormone, Thyrotrophic Releasing Hormone, and Human Chorionic Gonadotrophin Administration in Healthy Men at Different Risks for Prostatic Cancer and in Prostatic Cancer Patients Cancer Res 42:2074-2080, 1982 132 Subjects Risk Factor Exposure: diet-nutritional factors, endogenous hormones; geographic area; Cancer Site: prostate; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics Data from healthy black and white males from North America; data from healthy black males and black males with prostatic cancer from South Africa Variables Demographic: age, race-ethnicity, geographic area Data Sources Hospital medical records, laboratory test Abstract A comparative study of the pituitary and testicular response to luteinizing releasing hormone (LHRH), thyrotrophic releasing hormone (TRH), and human chorionic gonadotrophic (HCG) ad- ministration was carried out in (a) low-risk young South African black men and high-risk North American black men for prostatic cancer and (b) healthy elderly South African men and South Afri- can black men with prostatic cancer. A comparable HCG response occurred in young South African and North American black men, while a greater release of prolactin, but a lesser release of luteiniz- ing hormone in response to LHRH:TRH occurred in South African black men. The response to HCG was comparable in elderly and young South African black men, although the prolactin release in response to TRH was greater in elderly men. A more prolonged release of luteinizing hormone was evident in men with prostatic cancer. Higher estradiol and estrone but lower androstenedione levels occurred in men with prostatic cancer. Data suggest that, in the elderly South African black men with prostatic cancer, estro- gen metabolism is modified and that either the estrogen level or the higher estrogen:androgen levels modify the pituitary response to LHRH:TRH. A Western diet enhanced the changes in hormone profiles evident in black South African men with prostatic cancer. Hill, P.; Wynder, E.L.; Helman, P. Nutrition and Hormone Levels in Relation to Breast Cancer and Coronary Heart Disease Clin Oncol 4:35-45, 1978 133 Subjects Risk Factor Exposure: diet-nutritional factors, endogenous hormones; Cancer Site: breast Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 275 blood samples (umbilical cord 27; 9-year old girls 56; premenopausal women 20-30 years old 75; postmenopausal women over 50 years old 25; Caucasian women 35-45 years old 92) Variables Demographic: age, race-ethnicity (Bantu, Japanese, Caucasian), menopausal status, age at menarche Data Sources Laboratory test Abstract Epidemiological data have shown that Western women have a higher incidence of breast cancer and coronary heart disease (CHD) than Asian women, and have implicated hormonal factors associated with differences in incidence of breast cancer. Whether hormones play a detrimental or beneficial role in CHD is inconclu- sive. Comparison of the serum lipid and hormone level in Bantu and Japanese women (low risk populations) compared with Cauca- sian women (high risk population) showed higher oestrogen levels in young Bantu and Japanese women, but comparable levels in postmenopausal women from each ethnic group. Comparison of subgroups of Caucasian women at different risk for breast cancer showed marked difference in the hormone profile, while the serum lipid levels were similar except in women with late pregnancy. Although many studies may have failed to consider subgroups of women at risk for CHD, data suggests that premature castration causes little change in total oestrogens or serum lipids and would suggest that oestrogens, at least during this time of life, play little part in coronary heart disease. 58 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Horm, J.W.; Asire, A.]. Changes in Lung Cancer Incidence and Mortality Rates Among Americans: 1969-78 J Natl Cancer Inst 69(4):833-837, 1982 134 Subjects Cancer Site: lung; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: age-specific incidence rate, age-specific mortality rate, age-adjusted incidence rate, age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1969-1978 Sample Characteristics Cancer incidence data from 4 common TNCS-SEER areas (Iowa; Atlanta, GA; Detroit, MI; San Francisco-Oakland, CA) plus Connecticut; cancer mortality data from the United States Variables Demographic: age, sex, race-ethnicity (black, white) Data Sources Third National Cancer Survey 1969-1971, SEER cancer registry, death registration Abstract Cancer incidence data from two large cancer surveys and cancer mortality data for the entire United States were used to estimate trends in lung cancer incidence and mortality rates among whites for specific age groups during the 10 years 1969-78. Data for the Third National Cancer Survey were from 1969 to 1971; data for the Surveillance, Epidemiology, and End Results Program were from 1973 to 1978. For both incidence and mortality, the rates for women increased much more rapidly than the rates for men, except in the oldest age group (85+). For both men and women, there was a striking similarity between the average annual percent changes for age-specific incidence rates and those for age-specific mortality rates. For the first time women and men had a similar incidence rate for the youngest (35-44) age group in 1978. If these trends continue, the data suggest that the total incidence rates for men and women will be equal by the year 2000. Horm, J.W.; Kessler, L.G. Falling Rates of Lung Cancer in Men in the United States Lancet 1(8478):425-426, 1986 135 Subjects Cancer Site: lung; Cancer Measures: incidence, mortality Study Characteristics Design: time trend; Measures: age-adjusted incidence rate, age-adjusted mortality rate, age-specific incidence rate, age-specific mortality rate; Time Frame: 1973-1983 Sample Characteristics Cancer mortality data from the United States; cancer incidence data from SEER Program areas Variables Demographic: age, sex, race-ethnicity (black-white) Data Sources SEER cancer registry, death registration Abstract Lung-cancer incidence and mortality rates in the United States were calculated for the years 1973 to 1983. The historically increas- ing age-adjusted rates for white men levelled off in the late 1970s and fell between 1982 and 1983. These falls were seen for white men only. Both the incidence and mortality rates for women con- tinued to rise with no hint of a reduction. Howard, J.; Lund, P.; Bell, G. Hospital Variations in Metastatic Breast Cancer Med Care 18(4):442-455, 1980 136 Subjects Diagnosis; Cancer Site: breast; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1960-1975 Sample Characteristics 25,535 breast cancer cases (blacks 1,891; whites 23,644): female; California Variables SES: public-private patient status; Demographic: age, race- ethnicity; Cancer: stage Data Sources Hospital medical records, non-SEER population-based cancer registry 59 Abstract Data from the tumor abstracts of 25,535 female breast cancer patients in California were analyzed to measure hospital differ- ences in the rates of metastatic disease at diagnosis. For white patients the metastatic rates varied from 38 to 68 per cent as hospital affiliation varied, and for blacks they ranged from 48 to 76 per cent. Tests for normality showed that the distribution of rates for white patients in private facilities was not merely due to random variation. The metastatic rates for both racial groups tended to be inversely related to the proportion of whites among the patient populations and positively related to total patient load. Hospitals affiliated with medical schools had relatively high rates for white private patients. Several explanations of the data are suggested. They consider the possible effects of patient and physician health care practices and host vulnerability to the spread of breast cancer. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Huben, R.; Mettlin, C.; Natarajan, N.; et al. Carcinoma of Prostate in Men Less Than Fifty Years Old Urology 20(6):585-588, 1982 137 Subjects Diagnosis; Treatment; Cancer Site: prostate; Cancer Measures: survival, incidence Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics, crude survival rate Sample Characteristics 20,156 prostate cancer cases (blacks 2,364; whites 17,321; others 227—Under 50 years old 168; 50+ years old 19,988): male Variables Demographic: age, race-ethnicity; Cancer: stage, histology Data Sources Hospital medical records Abstract The data of the long-term survey on prostate cancer of the Commission on Cancer of the American College of Surgeons were analyzed to assess differences in stage of disease, diagnostic and therapeutic procedures, and survival in men with prostate cancer less than fifty years of age compared with older men. Of 20,156 patients, 168, 0.8 percent of the survey group, were less than fifty years old. There was significantly greater incidence of metastatic disease in the younger age group, with a concomitant deficit in Stage A or occult prostate cancer in those less than fifty years old. Diagnosis was made most frequently by needle biopsy of the pros- tate in younger patients, while the older patients underwent trans- urethral resection of the prostate as a diagnostic procedure in the majority of cases. There was a more frequent use of radical pros- tatectomy and radiation therapy for all stages of disease in the less than fifty-year-old group, while transurethral resection of the pros- tate alone or hormone therapy alone was the predominant therapeutic modality in those fifty and older. Comparison of sur- vival by stage of disease showed significantly better survival for younger patients with Stage B disease at five years (84% vs 68%) and at ten years (70% vs 48%). Survival was slightly better in younger patients at five years for other stages of disease. Although prostate cancer is more likely to be metastatic at the time of diag- nosis in men less than fifty, there does not appear to be any significant biologic differences in prostate cancer in men less than fifty years old when compared with older men. Hulka, B.S.; Chambless, L.E.; Wilkinson, W.E.; et al. Hormonal and Personal Effects on Estrogen Receptors in Breast Cancer Am ] Epidemiol 119(5):692-704, 1984 138 Subjects Biologic Marker: steroid receptors; Risk Factor Exposure: endogenous hormones, reproductive factors, exogenous hormones; Cancer Site: breast Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1977-1978 Sample Characteristics 246 breast cancer cases (blacks 56; whites 190): female, hospital patients; North Carolina Variables Demographic: age, race-ethnicity, menopausal status Data Sources Hospital medical records, survey, laboratory test Abstract Since high estrogen receptor protein in breast cancer tissue has a favorable effect on patient prognosis, factors influencing estrogen receptor levels are the focus of this research. From April 1977 to December 1978, 246 histologically confirmed primary breast cancer patients from three North Carolina hospitals were interviewed and provided with photographic displays of hormonal products. Their tumor tissue was analyzed for estrogen receptor protein by the sucrose density gradient assay. Data were analyzed by multiple regression and logistic regression techniques. No statistically sig- nificant relationship was found between gonadal hormone usage and estrogen receptor levels, although recent users of either oral contraceptives or replacement estrogens had lower receptor levels. Users of progestogen-predominant and -intermediate types of oral contraceptives also had lower estrogen receptor protein. An effect of menopausal status, independent of age, was not evident. Statis- tically significant associations involved age and race among post- menopausal patients. With black breast cancer patients aged <70 years as the reference group, the relative odds of having high estrogen receptor protein (>10 fmol/mg of cytosol protein) were 2.4 times greater among whites aged <70, 1.6 times greater in blacks aged >69, and 12.8 times greater in whites >69 years. These differences in estrogen receptor protein may contribute to the black-white differential in breast cancer survival. 60 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Hulka, B.S.; Redmond, C.K. Factors Related to Progression of Cervical Atypias Am ] Epidemiol 93(1):23-32, 1971 139 Subjects Screening: Pap smear; Treatment; Diagnosis; Cancer Site: cervix uteri Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics, higher level statistics; Time Frame: 1962-1968 Sample Characteristics 376 individuals with atypical Pap smears on initial screening (Negroes 241; whites 135) and 168 individuals with atypical Pap smears on second to fifth screening: female, low income neighborhood; Allegheny County, PA Variables Demographic: age, race-ethnicity, number of pregnancies, parity; Cancer: histology Abstract Five hundred thirty-nine women with atypical cervical smears underwent long term follow-up to determine factors associated with cytologic progression. Life table analysis on prevalent atypical smears revealed a cumulative probability of progression of 41.2% after 5 years’ follow-up. The probability of progression was greatest in the first 6 months, 13.6%. After the first year of follow-up, the progression rate was 3-4% in each 6-month interval. For incident atypical smears the cumulative probability of progression was 17.6% after 2'2 years of follow-up with no progression there- after. The semi-annual progression rates for incidence atyp- ical smears was approximately half that of prevalent atypical smears. If the first follow-up smear was negative, lower progression rates ensued that if it was atypical. Progression rates were high among women 30-44 years of age and lowest in Negroes less than 30. Numerous socioeconomic variables were reviewed but no associations with progression were found. Huvos, A.G.; Butler, A.; Bretsky, S.S. Osteogenic Sarcoma in the American Black Cancer 52:1959-1965, 1983 140 Subjects Cancer Site: sarcoma; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1921-1979 Sample Characteristics 1,120 sarcoma cases (blacks 100; whites 1,020): 3-58 years old, hospital cases; New York City, NY Variables Demographic: age, sex, race-ethnicity; Cancer: tumor size, histology, tumor location, metastases, tumor grade, stage, symptoms Data Sources Hospital medical records Abstract This clinicopathologic study of 100 American black patients with osteogenic sarcoma diagnosed and treated at this Medical Center from 1921-1979, inclusive, demonstrates a progressively increasing proportion of black patients admitted for this disease. 61 The relative upsurge became especially pronounced in the 1970s. The ages of the patients ranged from 3 to 58 years (median, 16, mean 19.2 years). The age distribution shows that blacks, on the whole, are younger than whites when they develop these tumors, and this is particularly evident in the black girls. Skeletal locations for osteogenic sarcomas, in general, are similar in both races, except that the tibia and the fibula were involved significantly more fre- quently in the blacks while the humerus was afflicted less com- monly. The clinical stage of disease on presentation, the duration of signs and symptoms, the histologic subclassification of the tumors, and the radiographic appearances closely matched in both races. The numbers of patients with Paget's sarcomas were also evenly distributed. Twenty-nine patients are currently alive with the 2-year and 5-year disease-free survival being 42% and 32%, respectively. There are no differences in the survival of black as compared to white patients, either for the entire duration of the study or for the period after 1974. The poorer prognosis of cancer in blacks does not apply to osteogenic sarcoma patients. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Huvos, A.G.; Rosen, G.; Bretsky, S.S.; et al. Telangiectatic Osteogenic Sarcoma: A Clinicopathologic Study of 124 Patients Cancer 49:1679-1689, 1982 141 Subjects Cancer Site: sarcoma; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics; Time Frame: 1921-1979 Sample Characteristics 124 telangiectatic osteogenic sarcoma cases (blacks 10; whites 114): 3-67 years old, hospital patients; New York City, NY Variables Demographic: age, sex; Cancer: symptoms, histology Data Sources Hospital medical records Abstract One hundred-twenty-four patients with this rare and special variant of osteogenic sarcoma were treated at Memorial Sloan-Ket- tering Cancer Center from 1921 through 1979, representing 11% of all of osteogenic sarcomas. The lesions were predominantly lytic, destructive tumors with only minimal sclerosis on roentgenograms and soft as well as cystic on gross examination. Histologically, aneurysmally dilated spaces lined or traversed by sarcoma cells producing osteoid were noted. The differential diag- nosis both radiographically and histologically included several be- nign lesions like aneurysmal bone cyst and giant cell tumor, among many others. It was found that telangiectatic osteogenic sarcoma is relatively frequent in the femoral diaphysis and in the distal end of the femur. Twenty-nine percent of the patients present with pathologic fracture, or this develops later. Age and sex distribution, or clinical signs or symptoms were those of ordinary osteogenic sarcomas. No differences in survival rates were found in lesions that were purely lytic or those with minimal sclerosis. Similarly, no differences in survival were noted when comparing patients with telangiectatic or ordinary osteogenic sarcoma. As a matter of fact, definite increase in survival was found in patients treated since 1975 with preoperative multidrug chemotherapy employing high-dose methotrexate, Adriamycin, and the combination of bleomycin, cyclophosphamide, and dactinomycin. Imashuku, S.; Labrosse, E.H.; Johnson, E.M., Jr.; et al. Tyrosine Hydroxylase in Neuroblastoma Biochem Med 5:22-29, 1971 142 Subjects Diagnosis; Cancer Site: neuroblastoma; Biologic Marker: catecholamines, tyrosine hydroxylase Study Characteristics Design: case study Sample Characteristics Two neuroblastoma cases: Negro, 1-2 years old Variables Demographic: age Data Sources Laboratory test Abstract Tyrosine hydroxylase activity was assayed in two neuroblas- toma tissues. It was found that the neuroblastoma tyrosine hydro- xylase required DMPH4(2-amino-4-hydroxy-6,7-dimethyl-5,6,7,8- tetrahydropteridine) as a cofactor and was stimulated by iron. These properties are quite similar to those reported for bovine and pheochromocytoma enzymes. Catechol compounds and other in- hibitors, at concentrations of 0.000025 to 0.0005M, inhibited the 62 neuroblastoma enzyme to the same extent as bovine adrenal tyrosine hydroxylase. This finding suggests that neuroblastoma tyrosine hydroxylase is normally susceptible to feedback inhibi- tion, in contrast to the pheochromocytoma enzyme which has been reported to be less sensitive to inhibition by catechols. Both neuroblastoma and pheochromocytoma are associated with in- creased synthesis of catecholamines; in the case of pheoc- hromocytoma there are large stores of norepinephrine in the tumor, but catecholamine synthesis still continues at an elevated rate, apparently because of a decreased sensitivity to feedback inhibition of the tyrosine hydroxylase in this tumor. On the other hand, in neuroblastoma the catecholamines appear to exert a “nor- mal” feedback inhibition on tyrosine hydroxylase, and, therefore, we must look elsewhere to explain the high rate of catecholamine synthesis, such as lowering of the intracellular catecholamine levels by rapid metabolism. Further investigation will be necessary to establish the alteration of control mechanisms in neuroblastoma. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Isaacson, C. The Changing Pattern of Liver Disease in South African Blacks S Afr Med J 53:365-368, 1978 143 Subjects Risk Factor Exposure: alcohol; Cancer Site: liver Study Characteristics Design: time trend; Cancer Measures: descriptive statistics; Time Frame: 1959-1976 Sample Characteristics 930 liver biopsies (hepatocellular carcinoma 99; no tumor 831): black, 20+ years old; Johannesburg, South Africa Variables Demographic: sex, age; Cancer: histology Data Sources Histopathology records Abstract Consecutive liver biopsies in a large general hospital for Blacks over the period 1959-1960 were compared with those in the 2-year period 1975-1976. Changes were noted in the histological appear- ances of micronodular cirrhosis. With the liberalization of the liquor laws and the granting of access to hard liquor to Blacks, fatty change, alcoholic hepatitis, alcoholic hyalin and alcoholic cirrhosis are making their appearance in the micronodular cirrhosis of the South African Blacks, features which were never observed when alcohol consumption was confined to homebrewed beverages con- taining large quantities of iron derived from the metallic containers in which liquor is brewed. No change in the histological picture or incidence of macro-nodular cirrhosis and hepatocellular car- cinoma was observed. Jacobs, P. Immunophenotypic Classification of Lymphoblastic Leukaemia and Lymphocytic Lymphoma — An Experience in the South Western Area of the Cape Province of South Africa Leuk Res 9(6):755-763, 1985 144 Subjects Treatment; Cancer Site: leukemia; Cancer Measures: incidence, survival Study Characteristics Design: prospective; Measures: descriptive statistics; Time Frame: 1980-1984 Sample Characteristics 58 previously untreated acute lymphoblastic leukemia cases (blacks 5; coloreds 22; whites 28; Asians 3): hospital patients; The Cape, South Africa Variables Demographic: age, sex, race-ethnicity; Cancer: histology Data Sources Hospital medical records Abstract Adequate tumor material was obtained for phenotypic classifi- cation using a standard library of monoclonal antibodies from 81 previously untreated patients with acute lymphoblastic leukaemia (ALL), chronic lymphocytic leukaemia (CLL), or lymphocytic lym- phoma (LL). Sixty-one individuals were adults and 20 were chil- dren of 14 yr or younger. Fifty-eight of the patients (72%) had acute lymphoblastic leukaemia and the remaining 23 (28%) had chronic lymphocytic leukaemia or lymphocytic lymphoma. Con- sidering only the patients with acute lymphoblastic leukaemia (n=58) the median age was 19 yr (range 3-69 yr): 9% were black, 43% were coloured, 48% were white, and the distribution between 63 adult (n=38) and paediatric patients (n=20) was comparable. Com- plete remission rate in the adults was 58% and in the paediatric group 85%. For the total group (n=58) median duration of survival was 59 weeks for common, 39 weeks for null, 63 weeks for T-ALL, and 13 weeks for B-ALL subtypes. In both the common and the null groups overall and disease-free survival was superior in the children. In contrast, no difference was evident in the T-ALL group, which was also notable for its high incidence in young coloured males. The 15 patients with CLL and eight with LL were adults and all the cells were phenotypically of B lineage: in view of the small numbers no comments are possible about ethnic dif- ferences. A multi-centre collaborative study is needed to define the epidemiology of haematologic malignancy in South Africa, with emphasis on differences among ethnically distinct subpopu- lations. Three points are emphasized by this study: 1) irrespective of race, age emerges as a favorable prognostic factor, 2) T-ALL in young colored males is in keeping with other evidence that ethnic- ity can exert an important and distinctive influence on disease expression, and 3) apparently homogenous tumor morphology in lymphoproliferative disorders can be misleading and immunologic techniques provide a suitable methodology to more reliably define phenotypically distinct subpopulations. Ethnicity is a powerful prognostic factor influencing the natural history and response to treatment of patients with hematologic malignancy. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Jensen, R.D.; Miller, R.W. Retinoblastoma: Epidemiologic Characteristics N Engl J] Med 285(6):307-311, 1971 145 Subjects Treatment; Cancer Site: eye; Cancer Measures: incidence, mortality Study Characteristics Design: cross-sectional; Measures: crude mortality rate, descriptive statistics; Time Frame: 1914-1969 Sample Characteristics 269 retinoblastoma deaths (Negroes 67; whites 196; American Indian 4; Japanese 2): under 15 years old; United States and 1,623 retinoblastoma cases (Negroes 171; whites 1,397; other nonwhites 10; unknown 45): hospital patients; United States Variables Demographic: age, sex, race-ethnicity Data Sources Death registration, hospital medical records Abstract Nationwide study of 269 death certificates for United States children with retinoblastoma, 1960-67, revealed a peak in mortality at two to three years of age that was 22 times greater in Negroes than in whites. Nationally, an estimated 81 per cent of chil- dren with the neoplasm survived. Study of 1623 hospital records of children with retinoblastoma revealed an excess of mental retardation: 21 cases observed vs. 4.7 expected. Associated malformations in several cases suggested D-de- letion syndrome, which has been described by others in six children with retinoblastoma. In accord with previous studies, radiotherapy was associated with an excess of os- teosarcoma, soft-tissue neoplasms or carcinoma of the skin. In addition, 11 children had second primary cancer not at- tributable to radiotherapy, a seeming excess over normal expectation. The cell types were diverse, but three were osteosarcomas, suggesting an innate susceptibility that may add to the risk of radiotherapy. Joffe, B.I.; Kew, M.C.; Heding, L.G.; et al. Nonspecific Binding of Insulin to Gamma Globulin in the Serum of Black Patients with Hepatocellular Carcinoma and Other Forms of Liver Disease J Lab Clin Med 100:445-450, 1982 146 Subjects Diagnosis; Cancer Site: liver; Biologic Marker: insulin antibody Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 54 individuals (blacks 30; Danes 24): liver cancer cases 9; noncancerous liver disease cases 11; African healthy controls 10; Danish controls 24: Johannesburg, South Africa; Denmark Data Sources Laboratory test Abstract The study was prompted by the apparent detection of insulin antibodies in a black patient with HCC and recurrent hypoglycemia who had never received insulin. It consisted of two parts. Initially the sera of 30 individuals (six normoglycemic HCC patients, three with HCC and recurrent hypoglycemia, 11 patients with noncan- cerous liver diseases, and 10 healthy black controls) were analyzed for the presence of insulin (and glucagon) antibodies by precipitat- ing the bound, labeled hormone with ethanol and also by the technique of radioimmunoelectrophoresis. In nine HCC patients, binding of iodine(125)-insulin averaged 13% by ethanol separation 64 and 0.018 mU/ml with radioimmunoelectrophoresis, levels that were similar to those of patients with noncancersous liver disease and significantly higher than those of the healthy controls. Mean binding of iodine(125)-glucagon was 11% in HCC sera. Serum binding of labeled hormones correlated significantly with IgG con- centrations in the patients. The second part of the study attempted to define the nature of insulin binding in HCC and other forms of liver disease. After confirmation of the increased serum binding of labeled insulin by another method of precipitation, PEG, an attempt was made to compete with the labeled insulin for its serum binding sites by adding a large amount of unlabeled insulin. This binding was not displaceable, however, and was therefore consi- dered nonspecific. When the same procedures were repeated using normal serum to which increasing amounts of gamma globulin were added, the nonspecific binding of insulin increased in a linear fashion. Furthermore, a similar degree of high nonspecific insulin binding occurred in six patients with multiple myeloma and raised serum IgG concentrations. We therefore conclude that in the many clinical situations where hypergammaglobulinemia exists, false positive tests for the detection of antibodies against insulin (and probably other peptide hormones) will emerge unless appropriate methods are used to check for nonspecific peptide binding. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Johnston, G.A., Jr.; Anderson, D.; Herbst, A.L.; et al. Adenocarcinoma of the Vagina and Cervix in Black Females J Natl Med Assoc 74(7):639-643, 1982 147 Subjects Diagnosis; Treatment; Cancer Site: cervix uteri, vagina; Cancer Measures: survival Study Characteristics Design: retrospective; Measures: descriptive statistics Sample Characteristics 28 cervix and vagina cancer cases: black, female, under 29 years old Variables Cancer: histology, stage Data Sources non-SEER hospital cancer registry, histopathology records Abstract Twenty-eight black females with clear cell adenocarcinoma of the vagina and cervix have been accessioned in the Registry for Research on Hormonal Transplacental Carcinogenesis. Most were born in the southern United States and histories of prenatal expo- sure to diethylstilbestrol were less frequent than with white pa- tients. The characteristics of these cases were similar to those among whites in stage, variety of cell type, and patient survival. Kalwinsky, D.K.; Rivera, G.; Dahl, G.V.; et al. Variation By Race in Presenting Clinical and Biologic Features of Childhood Acute Lymphoblastic Leukaemia: Implications for Treatment Outcome Leuk Res 9(6):817-823, 1985 148 Subjects Diagnosis; Treatment; Cancer Site: leukemia; Cancer Measures: survival Study Characteristics Design: prospective; Measures: descriptive statistics, higher level statistics; Time Frame: 1962-1984 Sample Characteristics 1,405 leukemia cases (blacks 149; whites 1,256): Memphis, TN Variables Demographic: age, race-ethnicity, sex; Cancer: Immunophenotype, biologic characteristics, surface phenotype Data Sources Hospital medical records Abstract In this paper, the authors present their experience with racial variation in clinical presentation and biological features of ALL (acute lymphoblastic leukaemia) for children treated on total therapies at St Jude Children’s Research Hospital. The response of black patients to more recent ALL clinical trials at this institution are reviewed. By univariate and multivariate statistics, the authors analysed whether the initial “risk” associated with the black race for these patients can be explained solely by its association with known patterns of clinical or biological “poor risk” features. Black children with ALL were found to present with a constellation of clinical and biologic features differing from those of white children with this disease. Neither clinical features of elevated tumor bur- den (white bloodcell counts), nor biologic prognostic variables (morphology/ploidy) were found to totally account for the poor prognosis in blacks. This suggests that additional yet-to-be-defined host factors, perhaps drug absorbtion or metabolism, may also be contributing to the poor outcome for black children with this dis- ease. Katz, J.L.; Ackman, P.; Rothwax, Y.; et al. Psychoendocrine Aspects of Cancer of the Breast Psychosom Med 32(1):1-18, 1970 149 Subjects Risk Factor Exposure: endogenous hormones; Cancer Site: breast Study Characteristics Design: prospective; Measures: descriptive statistics, higher level statistics Sample Characteristics 30 hospital patients (blacks 7; whites 21; Puerto Ricans 2): female, 33-79 years old; New York Variables Demographic: age, race-ethnicity, menopausal status; Cancer: histology, hydrocortisone production Data Sources Hospital medical records, laboratory test, survey Abstract Thirty women in the hospital awaiting biopsy of a breast tumor were interviewed in order to assess the adequacy of their ego defenses in this presumably threatening situation. The criteria for 65 this psychological evaluation were the patient's affective state, functional disruption, and defensive reserve impairment. Concur- rent with these assessments, daily production rates of hydrocor- tisone and urinary excretion levels of the principal hydrocortisone and androgen metabolites were measured. A rank order correlation significant of the .02 level was found between rating scores for extent of defensive failing and hydrocortisone production rates, and at the .05 level between this psychological variable and the principal hydrocortisone metabolites; however, no correlation existed with the androgen metabolites. Since other studies have indicated a correlation between prognosis in breast cancer and a numerical ratio calculated on the basis of the patient's endogenous corticoid to androgenic steroids, this study raises questions about the determination and nature of such a ratio and about the possible role of psychological phenomena in the course of this neoplasm. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Keirn, W.; Metter, G. Survival of Cancer Patients by Economic Status in a Free Care Setting 150 Cancer 55:1552-1555, 1985 Subjects Abstract Risk Factor Exposure: alcohol, tobacco; Cancer Site: breast, lung, Patients with lung, breast, and colorectal cancer were classified colon-rectum; Cancer Measures: survival; Socioeconomic Factors as to their economic status for comparison of survival. Patients of Study Characteristics the City of Hope Medical Center are admitted and treated without Design: prospective (nonconcurrent); Measures: crude survival regard to their economic status, providing an excellent place to rate, descriptive statistics, higher level statistics; Time Frame: test the effect of low economic status on survival. In none of these 1976-1981 disease sites was an effect of low economic status observed, even Sample Characteristics when adjusting for age, sex, stage, smoking status, and alcohol 1,101 breast, lung, and colon-rectum cancer cases (blacks 33; usage. The only variable that consistently predicted survival was whites 1,013; Chicano 28; other 11—Indigent 383; stage of disease; when accounting for stage, there was no indepen- nonindigent 702; unknown 16): California dent effect of low economic status. In the presence of uniform Variables care, low economic status as defined in this study is not a factor SES: medical insurance status; Demographic: age, sex; Cancer: in patient survival. stage Data Sources Hospital medical records, non-SEER hospital cancer registry Kew, M.C.; Geddes, E.W. Hepatocellular Carcinoma in Rural Southern African Blacks 151 Medicine 61(2):98-108, 1982 Subjects Diagnosis; Cancer Site: liver; Cancer Measures: incidence, usual. An arterial bruit was heard over the liver in 23 percent of prevalence, survival; Biologic Marker: alpha-fetoprotein (AFP), the patients. Ascites and splenomegaly were each present in about hepatitis B virus associated antigens (HBsAg), gamma-glutamyl one-half of the patients, and 28 percent were jaundiced. Progres- transferase sive muscle wasting was almost invariable. The tumor was rarely Study Characteristics resectable and the patients did not respond to radiation therapy Design: prospective (nonconcurrent); Measures: descriptive or cancer chemotherapy: the mean survival from the onset of statistics, higher level statistics; Time Frame: 1965-1981 symptoms was 11.1 weeks and from the time of diagnosis 6 weeks. Sample Characteristics Macronodular cirrhosis was present in the non-tumorous liver 585 hepatocellular carcinoma cases (Shangaans 415; Xhosas 105; tissue at necropsy in 60 percent of the patients. The hepatitis-B others 65): male, 18-70 years old, mineworkers; Johannesburg, surface antigen was detected in the serum of 62 percent of the South Africa patients. It was more common in younger patients, but was not Variables related to the presence of alpha-fetoprotein. HBs antigenemia was Demographic: age; Cancer: histology, symptoms equally common in patients with and without cirrhosis. Alpha- Data Sources fetoprotein was present in high concentration in the serum of 67 Laboratory test, autopsy records, histopathology records percent of the patients and provided an excellent indicator of the Abstract diagnosis of hepatocellular carcinoma. It was more commonly The clinical presentation and natural history of hepatocellular found in younger patients. Biochemical changes were non-specific carcinoma, as well as its association with cirrhosis and the hepatitis- and did not differentiate hepatocellular carcinoma from cirrhosis B virus surface antigen, were studied in 585 Black males living in or other mass lesions in the liver. Chest radiographs showed ele- rural areas of southern Africa. In addition, the value of serum vation of the right hemidiaphragm in 30 percent of the patients alpha-fetoprotein concentrations, serum biochemical changes, and and multiple pulmonary metastases in 25 percent. One or more various radiologic, isotopic and sonographic techniques in the defects were invariably seen on isotope liver scans but the picture diagnosis of hepatocellular carcinoma in this population was re- was not diagnostic of hepatocellular carcinoma. Blood flow scintig- viewed. The mean age of the patients was 35 years and 40 percent raphy usually showed the lesions to be vascular. Echogenic or of them were less than 30 years of age. The most common present- partly echogenic and partly anechoic lesions with ill-defined edges ing complaints were upper abdominal pain, swelling of the abdo- were demonstrated on ultrasonography. Computerized axial to- men, weight loss and weakness. Rupture of the tumor with acute mograms and hepatic arteriography were useful in assessing the intraperitoneal hemorrhage occurred in 10 percent of the patients. extent of the tumor in relation to possible resection. The liver was almost always enlarged and focal tenderness was 66 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Kew, M.C.; Hodkinson, ].; Paterson, A.C.; et al. Hepatitis-B Virus Infection in Black Children with Hepatocellular Carcinoma J Med Virol 9:201-207, 1982 152 Subjects Biologic Marker: hepatitis B surface antigens; Risk Factor Exposure: other diseases; Cancer Site: liver Study Characteristics Design: case study Sample Characteristics Six liver cancer cases: black, 10-16 years old; Johannesburg, South Africa Abstract The hepatitis B virus (HBV) status of six unselected South Af- rican Black children, aged 10 to 16 years, with hepatocellular car- cinoma (HCC) was investigated. The characteristics of the tumor were similar to those seen in Black adults. Hepatitis B surface antigen and antibody to the core antigen were present in the serum of all the children, and hepatitis Be antigen was detected in four of the children. The serum of the two patients without e antigen was positive for e antibody, and all sera were negative for hepatitis B surface antibody. Taken in conjuction with the HBV status of southern African Black adults with HCC, these findings raise the possibility that nearly all, if not all, Black patients with this tumor are or have been infected with HBV. Moreover, if HBV is on- cogenic, tumor formation may occur within 10 years of the initial infection, which occurs in eary childhood in southern African Blacks. Kew, M.C.; Wolf, P.; Whittaker, D.; et al. Tumour-Associated Isoenzymes of Gamma-Glutamyl Transferase in the Serum of Patients with Hepatocellular Carcinoma Br J Cancer 50:451-455, 1984 153 Subjects Biologic Marker: gamma-glutamyl transferase; Cancer Site: liver Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 650 individuals (blacks 391; Caucasians 22; unspecified 237—Liver cancer cases 413; controls 237): South Africa Variables Demographic: race-ethnicity Data Sources Laboratory test Abstract Sera from 391 southern African Blacks with hepatocellular car- cinoma, matched controls, patients with other malignant tumours, and with various forms of hepatobiliary disease were fractionated by polyacrylamide gradient gel electrophoresis to determine the prevalence of tumour-associated gamma-glutamyl transferase isoenzymes in Black patients with hepatocellular carcinoma. One or more tumour-associated isoenzymes (I, I”, II") were present in 58.6% of the patients with hepatocellular carcinoma: I' in 54.5%, I” in 27.1%, and II’ in 34%. These isoenzymes were detected in one patient with prostatic cancer, occasionally in patients with acute viral hepatitis, but in no normal individuals. The presence of tumour-associated isoenzymes was not related to patient age, sex or hepatitis-B virus status or to the tumour burden. Isoenzymes were present in 42 percent of hepatocellular carcinoma patients with a normal serum alpha-foetoprotein concentration and in 50% of those with a non-diagnostic value. Gamma-glutamyl transferase isoenzymes may be supplementary to alpha-foetoprotein in the diagnosis of hepatocellular carcinoma. Keyhani, A. Comparison of Clinical Behavior of Melanoma of the Hands and Feet: A Study of 283 Patients Cancer 40(6):3168-3173, 1977 154 Subjects Treatment; Cancer Site: malignant melanoma; Cancer Measures: survival Study Characteristics Design: prospective; Measures: crude survival rate, descriptive statistics; Time Frame: 1935-1962 Sample Characteristics 283 malignant melanoma cases (Negroes 6; Caucasians 222—Tumors of the feet 228; tumors of the hands 55): hospital patients; New York City, NY Variables Demographic: age, sex, race-ethnicity; Cancer: tumor location, symptoms, stage Data Sources Hospital medical records 67 Abstract The present study includes all cases of melanoma of the hands and feet between 1935 and 1962 which were followed at Memorial Sloan-Kettering Cancer Center, New York for five years or longer after their initial treatment. When the clinical behavioral of malig- nant melanoma of the hands and feet were compared, it was found that melanoma of the feet occurs more frequently and has a poorer prognosis than melanoma of the hands. The presence or absence of regional lymph node metastasis is found to be one of the most reliable clinical factors predicting prognosis. Melanoma is rare in Negroes; only 2.5 percent of the cases of melanoma of the feet were in Negroes. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Kim, K.; Rigal, R.D.; Patrick, J.R.; et al. The Changing Trends of Uterine Cancer and Cytology: A Study of Morbidity and Mortality Trends Over a Twenty Year Period Cancer 42:2439-2449, 1978 155 Subjects Screening: Pap smear; Risk Factor Exposure: reproductive factors; Cancer Site: cervix uteri, corpus uteri; Cancer Measures: incidence, mortality Study Characteristics Design: time trend; Measures: age-adjusted incidence rate, age-adjusted mortality rate, descriptive statistics; Time Frame: 1955-1974 Sample Characteristics 3,344 uterine cancer cases (cervix cancer 2,254; corpus cancer 1,042): female; Toledo, OH; mortality data from Lucas County, OH Variables Demographic: age, race-ethnicity (black, white), number of pregnancies, age at menopause, geographic area; Cancer: histology Data Sources non-SEER hospital cancer registry, U.S. Population Census, death registration Abstract Data on squamous carcinoma of the cervix from a 20 year study period (1955 to 1974) in metropolitan Toledo revealed a 66% reduc- tion of the average annual age-adjusted incidence rate and a 61% reduction in death rate of cervical squamous carcinoma when the first time period (1955 to 1958) was compared with the last time period (1971 to 1974). The decrease for both morbidity and mortal- ity rates was more pronounced in women age 50 years and younger. The age-adjusted death rate during this study period revealed 15.5/100,000 for black women and 8.7/100,000 for white women. The reduction in death rate of 83% in black women is more prominent than 54.5% in white women. The decrease in both morbidity and mortality for cervical squamous carcinoma has a close relation to cytologic screening activity. The factors of age and race, probably related to socioeconomic status, are two known determinants of risk for cervical squamous carcinoma. Data for endometrial carcinoma during this study period revealed 15.5/ 100,000 women in the average age-adjusted incidence and an in- crease of 13.8% in average yearly rates when the first time period (1955 to 1958) was compared with the last time period (1971 to 1974). The peak age was 60 to 64 years old in the first time period and shifted to 70 to 74 years old in the last time period. The trend in metropolitan Toledo is comparable to that of Louisville, Ken- tucky. The mass cytologic screening program which contributed to a remarkable reduction in morbidity and mortality for the cervi- cal squamous carcinoma, did not have any beneficial effect on endometrial carcinoma. Kovi, J.; Heshmat, M.Y. Incidence of Cancer in Negroes in Washington, D.C. and Selected African Cities Am J Epidemiol 96(6):401-413, 1973 156 Subjects Diagnosis; Risk Factor Exposure: geographic area; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: crude incidence rate, age-specific incidence rate, age-adjusted incidence rate; Time Frame: 1965-1969 Sample Characteristics 6,229 cancer cases: Negro; diagnosed at 11 major Washington, DC hospitals; cancer incidence data on Negroes in Bulawayo, Rhodesia; Natal, South Africa; Ibadan, Nigeria Variables Demographic: age, sex, geographic area Data Sources Hospital medical records, non-SEER hospital cancer registry Abstract The results of a 5-year (1965-1969) survey of cancer among Negroes in 11 major hospitals in Washington, D.C., are reported. In males, the commonest sites of cancer, in order of decreasing frequency were prostate, lung, esophagus, and colon. In females, breast, cervix uteri, and colon were the most common sites. Among Negroes the overall incidence of malignant disease in Washington, D.C. was considerably higher than in Africa. The average annual incidence rate for males was 220.4/100,000 and for females 194.3. Incidence rates for both males and females reached a peak in the 70-79 age group, and then leveled off. Negroes in Washington, D.C., and in Alameda County, California, revealed a pattern of cancer risk for leading tumor sites, except cervix uteri, much closer to whites in Alameda County, California, than to Negroes in vari- ous parts of Africa, south of the Sahara. In fact, American Negroes had higher standardized incidence rates for cancer of the esophagus, stomach, pancreas, prostate and cervix uteri, but lower for breast than Caucasians in Alameda County, California. Among Negroes in Africa, only the South African Bantu showed a trend in susceptibility to cancer comparable to American Negroes and whites. These findings emphasize the importance of environmen- tal influences in the development of cancer in man. 68 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Kovi, J.; Jackson, M.A.; Rao, M.S; et al. Cancer of the Prostate and Aging: Autopsy Study in Black Men from Washington, D.C., and Selected African Cities Prostate 3:73-80, 1982 157 Subjects Cancer Site: prostate; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1973-1980 Sample Characteristics 795 autopsy cases (Washington, DC 400; West Africa 345—, prostate cancers 168; cancer free 627): black, male; Washington, DC; Ibadan, Nigeria; Accra, Ghana Variables Demographic: age, geographic area; Cancer: stage, grade Data Sources Laboratory test Abstract Age-related changes of arteries, veins, glands, and stroma in the prostate of black men from Washington, DC, and from Ibadan, Nigeria, and Accra, Ghana, West Africa, were studied in a total of 795 consecutive, unselected prostate specimens removed at au- topsy during a 7-year period (1973-1980). Except for age group 80 and over, aging changes in the prostate were more severe in all age groups in black men from Washington, DC, than in black men from Ibadan and Accra (p<0.01). However, when the intensity of age-dependent alterations was compared in either US or African black men with carcinoma and with no carcinoma, no significant differences were found. These findings do not support the idea that the aging process per se increases susceptibility to cancer. Krain, L.S. The Rising Incidence of Cancer of the Pancreas—Further Epidemiologic Studies J Chron Dis 23:685-690, 1971 158 Subjects Cancer Site: pancreas; Cancer Measures: mortality, incidence Study Characteristics Design: cross-sectional, time trend; Measures: age-adjusted mortality rate, crude incidence rate, age-specific incidence rate, descriptive statistics, higher level statistics; Time Frame: 1942-1964 Sample Characteristics Pancreas cancer incidence data: 57 California hospitals; Alameda County, CA; and 24 countries. Variables SES: public-private hospital patient status; Demographic: sex, age, race-ethnicity (white, Negro, Caucasian, Chinese, Japanese) Data Sources non-SEER hospital cancer registry, non-SEER population-based cancer registry Abstract California Tumor Registry data and special data prepared for the California Regional Medical Program were analyzed with re- 69 spect to socioeconomic class and race for pancreatic cancer. An inverse relationship between socioeconomic class and pancreatic cancer was noted, but there was no significantly different geog- raphic distribution of pancreatic cancer cases in the State of Califor- nia. Negroes showed the highest pancreatic cancer incidence over time whether Alameda County Tumor Registry data or world-wide comparisons were used; but year-to-year comparisons showed no significantly different incidence figures from all races combined (p>0.05). The threefold increase in pancreatic cancer from 1920- 1970 cannot be attributed to diagnosing a higher proportion of such cases in Negroes over time. Negroes showed the greatest percentage increase in pancreatic cancer in the age-specific group- ing under age 50, of all races or ethnic (nationality) groups, with greater than 60 per cent histologic confirmation of diagnosis, from 1950-1963. Broader, more well-controlled epidemiologic studies focusing on the variables of cigarette smoking, industrial chemical exposure and alcoholism are suggested because of the increasing tendency for pancreatic cancer to appear in the ages of occupational exposure and high productivity. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Krain, L.S. The Rising Incidence of Carcinoma of the Pancreas Am J Gastroenterol 54(5):500-507, 1970 159 Subjects Risk Factor Exposure: endogenous hormones, geographic area, tobacco, air pollution; Cancer Site: pancreas; Cancer Measures: incidence, mortality Study Characteristics Design: cross-sectional, time trend; Measures: age-specific mortality rate, age-adjusted mortality rate, age-adjusted incidence rate, standardized mortality ratio (SMR), higher level statistics; Time Frame: 1950-1967 Sample Characteristics Pancreas cancer data: Alameda County, CA; pancreas cancer mortality data: U.S. Veterans; pancreas cancer incidence data and air pollution data: San Francisco, San Jose, San Diego, Los Angeles, CA Variables Demographic: age, sex, race-ethnicity (Negroes, Caucasians, Japanese, Chinese, white, nonwhite) Data Sources non-SEER population-based cancer registry, survey Abstract Genetic (hormonal), and environmental factors were evaluated in terms of an association with pancreatic cancer mortality, which has increased nearly three-fold in the last 40 years. Genetic (hor- monal) factors were discarded and environmental factors such as cigarette smoking and industrial exposure were emphasized in terms of an association with pancreatic cancer. Evidence was cited from national origin and racial comparisons as supportive of en- vironmental association. An air pollution linkage with pancreatic cancer, at least based on comparative California city air pollution levels was dismissed, due to a nonsignificant (p>0.10) association. A broader epidemiologic study focusing on cigarette smoke and industrial chemical exposure constituents is suggested. Krementz, E.T.; Reed, R.].; Coleman, W.P., III; et al. Acral Lentiginous Melanoma Ann Surg 195(5):632-645, 1982 160 Subjects Treatment; Cancer Site: malignant melanoma; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics Sample Characteristics 180 acral melanoma cases (blacks 41; whites 139—Exact ALM histologic classification 67; non-exact ALM histologic classification 113): New Orleans, LA Variables Demographic: age, sex, race-ethnicity; Cancer: stage, tumor location, histology Data Sources Hospital medical records Abstract Acral lentiginous melanoma (ALM) is the fourth clinicopathologic variant of malignant melanoma. It occurs on volar surfaces of hands and feet, subungual sites, and fingers or toes. It is characterized by slow lentiginous radial growth and central plaque-like thickening, heavily pigmented tumor cells, markedly thickened papillary dermis, and diffuse reticular infiltration. Le- sions are unusually large and, in most cases, thick and ulcerated. There were 180 patients with acral melanoma (AM), which includes 67 in whom the specific features of ALM could be documented. One hundred sixty had primary lesions on the foot, and 20 occurred on the hand. There were 104 men and 76 women. There were 41 black patients and 139 whites. Five-year survivals following all modalities of therapy in 122 patients with Stage I acral melanoma is 63% for plantar/palmar lesions, 58% for subungual lesions, and 27% for skin of digits. For the subgroup of Stage I patients with ALM treated by surgery and regional chemotherapy by perfusion, the five-year survival for all sites is 72% and 56% at 10 and 15 years, respectively. Survival in ALM is essentially the same as for all AM lesions. Leffall, L.D., Jr.; Chung, E.B.; Dewitty, R.L.; et al. Familial Polyposis Coli in Black Patients Ann Surg 186(3):324-333, 1977 161 Subjects Risk Factor Exposure: other diseases; Cancer Site: colon, rectum Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 37 hospital patients from 6 families with polyposis coli (polyposis with cancer 7; multiple polyposis without cancer 12; disease free 18): black; Washington, DC Variables Demographic: age, sex; Cancer: tumor location Data Sources Hospital medical records Abstract We have presented our experience with 19 black patients who had familial polyposis coli, seven of whom had associated cancer. Four of the seven had multiple cancers. Seventeen of the 19 patients were female but most of the patients at risk in our series were female. Sixteen of the 19 had extracolonic manifestations which have been described as part of the Gardner's syndrome. Although our series was quite small, larger series do not report as high an incidence of extracolonic manifestations. Follow-up was too short to make any statements concerning the development of cancer in the retained rectal stump. 70 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Leffall, L.D., Jr.; White, J.E. Cancer of the Oral Cavity in Negroes Surg Gynecol Obstet 120(1):70-72, 1965 162 Subjects Treatment; Risk Factor Exposure: alcohol, tobacco; Cancer Site: oral cavity; Cancer Measures: incidence, survival Study Characteristics Design: prospective; Measures: descriptive statistics; Time Frame: 1948-1963 Sample Characteristics 107 oral cavity cancer cases (males 77; females 30): Negro, 17-86 years old, hospital patients; Washington, DC Variables Cancer: tumor location, symptoms, histology, nodal involvement Data Sources Hospital medical records Abstract In a 15 year survey of cancer of the oral cavity in Negroes, the age range was 17 to 86 years and the average age was 53 years. Carcinoma of the floor of mouth was the most common in 107 patients, and its incidence is gradually increasing. Carcinoma of the lip was rare. There were 3 associated factors: chronic al- coholism, prolonged use of tobacco, and poor oral hygiene. A positive serologic result was found in 18 percent of the patients. Squamous cell carcinoma was the most common type of tumor. Fifty-eight radical neck dissections were performed, either alone or in combination with an intraoral procedure. There were 5 opera- tive deaths. Both patients with lip cancer are alive and well 4 and 11 years postoperatively. There were no 3 year survivors with cancer of the base of the tongue or soft palate. If lip cancer is excluded, the overall 3-year survival rate, without evidence of recurrent or metastatic disease, was 23.8 percent. Lepow, G.M.; Lepow, R.M. Superficial Spreading Melanoma in Blacks J Am Podiatr Med Assoc 75(8):425-428, 1985 163 Subjects Diagnosis; Treatment; Cancer Site: malignant melanoma Study Characteristics Design: case study Sample Characteristics One malignant melanoma case: black, male, 49 years old; Houston, TX Providers Podiatrist Abstract A case study of diagnosis and treatment of a malignant melanoma lesion of the right arch in a 49-year old black male patient. Lerer, T.].; Redmond, C.K.; Campbell, J. Factors Related to the Incidence of Stomach Cancer in the Pittsburgh Standard Metropolitan Statistical Area J Natl Cancer Inst 59(4):1065-1071, 1977 164 Subjects Cancer Site: stomach; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-specific incidence rate, age-adjusted incidence rate, descriptive statistics, higher level statistics; Time Frame: 1969-1971 Sample Characteristics 997 stomach cancer cases (blacks 80; whites 916; others 1): Pittsburgh, PA Variables SES: income; Demographic: sex, age, race-ethnicity, birthplace Data Sources Third National Cancer Survey 1969-1971, U.S. Population Census Abstract This report presents an analysis of several demographic discri- minants of stomach cancer incidence for residents of the Pittsburgh Standard Metropolitan Statistical Area (SMSA) with the use of data collected as part of the Third National Cancer Survey, 1969-71. The Pittsburgh SMSA, one of nine areas included in the Third National Cancer Survey, was of particular interest in the study of stomach cancer, since area residents showed the highest age-ad- justed incidence rate for cancer of this site while showing the fifth highest rate for all sites. To examine the data in greater detail, we used the variables of sex, race, nativity, and median income level to classify the cancer patients into homogeneous subgroups of census tracts of residence to allow comparisons of the average annual age-adjusted incidence rates among the groups. The data indicated that Pittsburgh SMSA residents showed stomach cancer incidence patterns that were generally consistent with earlier pub- lished reports; particularly when contrasted with appropriate population subgroups, higher rates appeared for males, for blacks, for lower income areas, and for areas with “large” proportions of foreign-born and foreign-stock residents. Futhermore, when one considered interactions among the factors, the relationship be- tween ethnic composition and incidence appeared to supercede that of income among white males. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Levin, D.L.; Connelly, R.R.; Devesa, S.S. Demographic Characteristics of Cancer of the Pancreas: Mortality, Incidence and Survival Cancer 47:1456-1468, 1981 165 Subjects Cancer Site: pancreas; Cancer Measures: incidence, mortality, survival Study Characteristics Design: time trend, cross-sectional; Measures: age-adjusted mortality rate, age-adjusted incidence rate, age-specific incidence rate, crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1935-1977 Sample Characteristics 9,668 pancreas cancer cases (blacks 973; whites 8,310; other and unknown 385): SEER Program areas; 8,440 pancreas cancer cases: Connecticut; 4,887 pancreas cancer cases (blacks 592; whites 4,295): TNCS areas; pancreas mortality data from the United States Variables SES: income, education, urban-rural; Demographic: sex, race-ethnicity, geographic area Cancer: histology Data Sources Third National Cancer Survey 1969-1971, SEER cancer registry, death registration, U.S. Population census Abstract Mortality and incidence rates for pancreatic cancer in the United States were examined by various demographic characteristics. Dis- ease rates have continued to increase over time but at a much slower pace than in earlier years. Most recently available rates for blacks were significantly higher than for whites and rates for males of each race were higher than for females. Income and education levels had little influence on incidence rates among either blacks or whites. Incidence rates were not significantly higher in urban as compared with rural areas of Iowa and Colorado. The two-year survival rate for pancreatic cancer was about 5% in recent years and did not vary significantly by race or sex. Smoking and diabetes, the two risk factors most consistently associated with the pancreatic cancer, explain only a small proportion of the disease. Much epidemiologic work remains to be done. Levine, P.H.; Kamaraju, L.S.; Connelly, R.R; et al. The American Burkitt's Lymphoma Registry: Eight Years’ Experience Cancer 49(5):1016-1022, 1982 166 Subjects Biologic Marker: HLA; Risk Factor Exposure: familial factors; Cancer Site: lymphoma; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1971-1978 Sample Characteristics 421 Burkitt's lymphoma cases (blacks 12; whites 198; Latin Americans 10; Orientals 3; unknown 33; unspecified 165): United States, Canada, Mexico, Brazil, Peru Variables Demographic: age, sex, race-ethnicity, geographic area; Cancer: stage, organ involvement Data Sources SEER cancer registry, non-SEER population-based cancer registry, histopathology records, autopsy records, physician medical records 72 Abstract Four-hundred twenty-one Americans diagnosed as having Bur- kitts lymphoma (BL), 409 from the United States, were studied by the American BL Registry to obtain information about the cause and control of this disease. Of these 421 cases, 256 were confirmed by our pathologists as being morphologically indistinguishable from African BL. A relationship between age and organ involve- ment was observed; cervical lymph nodes, ileum, and nasopharynx were initial sites of involvement primarily in younger patients. Although the Epstein-Barr virus (EBV) was less frequently associated with American BL than African, a high antibody titer to the EBV capsid antigen was associated with a more favorable prognosis. American BL resembled African BLs time-space cluster- ing, male predominance, and excellent response to chemotherapy. Unlike African BL, however, more patients had involvement of cervical lymph nodes and bone marrow at an early stage of disease. American BL appears to be a more heterogeneous disease than African BL. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Li, E.P.; Fraumeni, J.F., Jr. Rhabdomyosarcoma in Children: Epidemiologic Study and Identification of a Familial Cancer Syndrome J Natl Cancer Inst 43(6):1365-1373, 1969 167 Subjects Risk Factor Exposure: familial factors; Cancer Site: muscle; Cancer Measures: incidence, mortality Study Characteristics Design: retrospective; Measures: age-specific mortality rate, standardized mortality ratio (SMR), descriptive statistics, higher level statistics; Time Frame: 1960-1964 Sample Characteristics 418 rhabdomyosarcoma cancer deaths (Negroes 40; whites 376; Japanese 1; Portugese-Japanese 1) and 280 rhabdomyosarcoma hospital patients, under 15 years old, United States Variables Demographic: age, sex, race-ethnicity, geographic area; Cancer: tumor location Data Sources Hospital medical records, death registration Abstract To study the origins of childhood rhabdomyosarcoma, an examination was made of the 418 death certificates of U.S. children who died of this neoplasm, 1960-64, and of 280 medical charts from 17 hospital centers. Of exceptional interest was the presence in 5 families of a second child with a soft-tissue sarcoma, 3 sibs (vs. 0.06 expected by chance), and 2 cousins. The parents, grand- parents, and other relatives of children in these families had a high frequency of carcinoma of the breast and diverse neoplasms (e.g., acute leukemia and carcinomas of the lung, pancreas, and skin) at relatively young ages, suggesting a new familial syndrome of multiple primary cancers. Additional components of the syn- drome were implicated by the occurrence of adrenocortical car- cinoma and brain tumor in the first-degree relatives of 2 other children with rhabdomyosarcoma. While suggesting the role of inheritance, the familial patterns seen with rhabdomyosarcoma may result from an interaction of genetic and environmental (?viral) factors. The oncogenic agents and mechanisms in human cancer may be identified by the use of such family aggregations for labo- ratory studies and further epidemiologic studies. Like most child- hood neoplasms, rhabdomyosarcoma showed a peak mortality before 4 years of age and occurred slightly more often in males. This neoplasm was diagnosed in 29 children in the hospital series before 1 year of age and in 9 within 1 month of birth; this indicates that rhabdomyosarcoma may arise in utero. Unlike most neop- lasms of early inception, however, no association with congenital defects was detected. Rhabdomyosarcoma also showed no vari- ations in time and/or space that might reflect environmental influ- ences, such as viral or chemical agents which can induce this neoplasm in laboratory animals. Liang, A.P.; Levinson, A.G.; Layde, P.M.; et al. Risk of Breast, Uterine Corpus, and Ovarian Cancer in Women Receiving Medroxyprogesterone Injections JAMA 249(21):2909-2912, 1983 168 Subjects Risk Factor Exposure: exogenous hormones; Cancer Site: breast, ovary, corpus uteri Study Characteristics Design: prospective; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1967-1981 Sample Characteristics 5,003 individuals: black, female; Atlanta, GA Variables Demographic: age Data Sources Family planning clinic medical records Abstract Animal studies have yielded conflicting results on the car- cinogenicity of long-acting progestins. Since more than 1.5 million women worldwide are currently receiving injections of a contracep- tive progestin, depot medroxyprogesterone acetate, this is poten- tially an important public health problem. We obtained information on the occurrence of breast, uterine, and ovarian cancer among 5,000 black women attending a metropolitan hospital's family plan- ning clinic who had received injections of medroxyprogesterone for contraception (1967 and 1976). The women were followed up for four to 13 years after their initial medroxyprogesterone injec- tion. We compared the observed number of cancer cases in these women with the expected number based on annual age-, race-, and sex-specific rates derived from National Cancer Institute data. During more than 40,000 woman-years of observation, we found no evidence of an increased risk of developing cancer of the breast, uterine corpus, or ovary in these women. After adjusting for pos- sible underascertainment of cancer because of incomplete follow- up, we found the relative risk for medroxyprogesterone users to be 0.7 for breast cancer (95% confidence limits, 0.3 to 1.4), 1.2 (95% confidence limits, 0.1 to 6.7) for cancer of the uterine corpus, and 0.8 (95% confidence limits, 0.1 to 4.6) for ovarian cancer. 73 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Linden, G. The Influence of Social Class in the Survival of Cancer Patients Am | Public Health 59(2):267-274, 1969 169 Subjects Methodology; Cancer Site: multiple sites, breast; Cancer Measures:survival; Socioeconomic Factors Study Characteristics Measures: relative survival rate, descriptive statistics; Time Frame: 1942-1962 Sample Characteristics Cancer registry data: California Variables SES: public-private hospital status; Demographic: sex, age Data Sources non-SEER population-based cancer registry Abstract Cancer patients in private hospitals (higher social class) survive their cancer better than cancer patients in public hospitals (lower social class) for almost every site, sex, and stage of disease group. Additional analysis of breast cancer cases, taking stage, age, race, and treatment into account, produced similar results. A method of competing risks is used to confirm the earlier findings, which were based on the use of a life table to adjust for noncancer mor- tality. Survival rates computed by the two methods are compared with each other and with a third method. The data indicate that a single life table should not be used to compare the survival of groups of cancer patients which differ in social class. More study is needed to evaluate the adequacy of the three methods of adjust- ing cancer survival rates for noncancer mortality. Linden, G.; Bragg, K.; Dunn, J.E., Jr; et al. Cancer in California, 1942-56 Public Health Rep 79(4):346-350, 1964 170 Subjects Cancer Site: multiple sites; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent), time trend; Measures: relative survival rate, higher level statistics; Time Frame: 1942-1956 Sample Characteristics 110,229 multiple cancer cases: California Variables SES: public-private hospital patient; Demographic: age, sex, race-ethnicity (Negro, white, Chinese, Japanese); Cancer: stage Data Sources non-SEER population-based cancer registry, hospital medical records Abstract This paper highlights data drawn from a monograph published in October 1963, Cancer Registration and Survival in California. Negro men were found to have had higher relative proportions of esophageal, stomach, pancreatic, and prostatic cancer. Negro women had higher than expected proportions of esophageal and cervical cancer. White patients generally had a higher proportion of localized cases than nonwhite patients. Negro women, all stages combined, were lower than the rates for other female groups, but the differences disappeared when analysis was restricted to localized cases. It seems likely that the differences in the proportion of localized cases among white and Negro patients are more the result of differences in social class than in race. Lipkin, A.; Miller, R.H.; Woodson, G.E. Squamous Cell Carcinoma of the Oral Cavity, Pharynx, and Larynx in Young Adults Laryngoscope 95:790-793, 1985 171 Subjects Diagnosis; Treatment; Risk Factor Exposure: alcohol, tobacco; Cancer Site: pharynx, larynx, oral cavity; Cancer Measures: survival Study Characteristics Design: retrospective, case study Measures: descriptive statistics Time Frame: 1964-1983 Sample Characteristics 39 head and neck cancer cases (blacks 26; whites 12; Latin Americans 1): 28+ years old, low SES; Houston, TX Variables Demographic: age, race-ethnicity; Cancer: age at presentation of first primary tumor, stage, location Data Sources Hospital medical records Abstract Thirty-nine adults (34 male, 5 female) age 40 or under with squamous cell carcinoma of the head and neck (excluding nasopharyngeal carcinoma) were seen at the Baylor College of Medicine affiliated hospitals from 1964 to 1983. The average age 74 of the patients was 36.3 years and they were nearly uniformly heavy smokers and drinkers. Twenty-six of the 39 patients were black. Lesions of the oral cavity and oropharynx were most fre- quent, followed by laryngeal lesions. Four patients presented with second primary tumors, and one patient had three separate lesions. Of the 39 patients, only 8 had cancers small enough to be consi- dered curable by radiotherapy or organ-sparing surgery. The re- mainder of the patients required radical surgery (26), were consi- dered incurable at the time of presentation (8), or refused treatment (3). Also noted was a relatively high incidence of prior trauma (gunshot wounds, laparotomies for trauma). Of 30 patients avail- able for follow-up more than one year after treatment, 19 have died of their tumors. Only one patient had a documented immune disorder (systemic lupus). In contrast to previous authors, we conclude that the development of squamous cell carcinoma at a young age can be related to heavy smoking and drinking and that the poor survival in many patients is due to self-neglect and failure to seek medical care early in the course of the disease. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Lipworth, L.; Abelin, T.; Connelly, R.R. Socio-Economic Factors in the Prognosis of Cancer Patients J Chron Dis 23:105-116, 1970 172 Subjects Cancer Site: multiple sites; Cancer Measures: survival; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: relative survival rate, higher level statistics Sample Characteristics Cancer registry data: Boston, MA; Connecticut Variables SES: census tract income; Demographic: sex, geographic area Data Sources non-SEER population-based cancer registry, non-SEER hospital cancer registry Abstract The purpose of this study was to test whether the difference in cancer survival reported from the State Registries of Mas- sachusetts and Connecticut could in part depend upon socioeconomic (SES) factors. A preliminary comparison of the 3- year relative survival rates of the Boston cancer patients in the Massachusetts Registry with those of Connecticut confirmed the less favorable outlook of the former patients (SES) for cancers of most sites. Lloyd, J.W.; Ciocco, A. Long-Term Mortality Study of Steelworkers J Occup Med 11(6):299-310, 1969 173 Subjects Methodology; Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: prospective; Measures: age-adjusted mortality rate, descriptive statistics; Time Frame: 1953-1961 Sample Characteristics 59,072 individuals (nonwhites 7,367; whites 51,705—Nonwhite deaths 633; white deaths 4,083): male, steelworkers; Pennsylvania Variables Demographic: age, race-ethnicity Data Sources Industry records, death registration Abstract This report is the first of a series concerned with chronic disease mortality and prior occupational history. In this article methodological problems encountered in designing long-term studies of industrial population mortality are reviewed. Findings on mortality rates and frequency of causes of death among white and nonwhite steelworkers and the population of the county in which the seven iron and steel producing plants under study are located are presented. Among the major findings reported, non- whites had a higher total mortality than whites. The lower mortal- ity of whites occurred for all causes of death except for deaths attributed to arteriosclerotic and degenerative heart disease. Lloyd, J].W.; Lundin, F.E., Jr.; Redmond, C.K; et al. Long-Term Mortality Study of Steelworkers: Mortality by Work Area J Occup Med 12(5):151-157, 1970 174 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: prospective (nonconcurrent); Measures: crude mortality rate, standardized mortality ratio (SMR), descriptive statistics, higher level statistics; Time Frame: 1953-1961 Sample Characteristics 496 multiple site cancer cases (nonwhites 85; whites 252): male, steelworkers; Pennsylvania Variables Demographic: race-ethnicity Data Sources Death registration Abstract This paper, the fourth of a series concerning the mortality of 75 steelworkers, examines the variation in all-causes mortality by work area for the 58,828 men employed in 1953. Age, race, nativity, and residence-specific rates for the total steelworker population were applied to the number at risk in each area to determine the expected number of deaths. Four of 53 work areas showed a statis- tically significant excess of deaths; and one area, a significant deficit in mortality. One work area had an excess of mortality limited to nonwhite steelworkers. Data are also presented for men employed five years or more in each work area. Six additional areas of excess or deficit mortality were identified in this way. For work areas exhibiting significant deviation in total mortality, under either def- inition of population at risk, cause-specific patterns of mortality were evaluated in terms of health effects on job mobility. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Love, L.; Neumann, H.A.; Szanto, P.B.; et al. Malignant Renal Tumors in Adolescence Radiology 92(4):855-860, 1969 175 Subjects Diagnosis; Cancer Site: kidney; Cancer Measures: incidence Study Characteristics Design: case study; Time Frame: 1965-1967 Sample Characteristics Five renal cancer cases (males 1; females 4): Negro, under 21 years old; Illinois Variables Cancer: histology, clinical diagnosis Data Sources Hospital medical records Abstract Five cases of malignant renal tumors in the ten-to-twenty-year age group are presented. Three were renal-cell carcinoma, one was an undifferentiated nephroblastoma, and in one instance renal metastases simulated a primary renal tumor. All 5 cases presented diagnostic findings on renal angiography. The importance of an- giographic studies in addition to conventional diagnostic work-up for a preoperative diagnosis should be stressed. Lynch, H.T.; Guirgis, H.; Lynch, J.; et al. Cancer of the Colon: Socioeconomic Variables in a Community Am | Epidemiol 102(2):119-127, 1975 176 Subjects Risk Factor Exposure: familial factors, occupation; Cancer Site: colon; Cancer Measures: incidence, prevalence, survival Socioeconomic Factors Study Characteristics Design: retrospective; Measures: crude incidence rate, crude mortality rate, descriptive statistics, higher level statistics; Time Frame: 1964-1969 Sample Characteristics 154 colon cancer cases: Omaha-Douglas County, NE Variables SES: census tract income; Demographic: age, sex Data Sources Physician medical records, hospital medical records, histopathology records, survey Abstract Carcinoma of the colon was studied in Omaha-Douglas County, Nebraska (population 345,000). A total of 154 cases of colon cancer were diagnosed in 1964 (44.7/100,000). The frequency distribution of these patients in specific census tracts of this community was determined. Statistical analysis of the data showed a greater fre- quency of colon cancer in patients living in census tracts with higher average income. Colon cancer appears to be nonrandomly distributed with respect to the income and socioeconomic status of its victims, suggesting that hypotheses consistent with environ- mental variables—particularly those characterizing extremely high versus extremely low socioeconomic groups, including occupation, diet and other life patterns—should be pursued. All of these data have implications for cancer epidemiology, cancer control, and carcinogenesis. Mabogunje, O.A.; Nirodi, N.S.; Harrison, K.A.; et al. Teratomas In Adult Nigerians Afr] Med Sci 9:151-158, 1980 177 Subjects Cancer Site: multiple sites; Cancer Measures: prevalence Study Characteristics Design: prospective (nonconcurrent), case study; Measures: prevalence rate, descriptive statistics; Time Frame: 1972-1977 Sample Characteristics 105 teratoma cases: hospital patients; Nigeria and four teratoma cases (males 3; females 1): 19-25 years old; Nigeria Variables Demographic: age, sex; Cancer: histology, tumor location, symptoms Data Sources Hospital medical records Abstract Over a six-year period, 132 teratomas were seen at the Ahmadu 76 Bello University Hospital in Zaria, Nigeria. Twenty-seven of these, comprising most of the extragonadal tumours were seen in infants and children. Of the 105 tumours found in adults, ninety-four were benign cystic teratomas of the ovary occurring mostly in the third and fourth decades of life. There were three instances of malignant transformation of a previously benign cystic ovarian teratoma and four cases of struma ovarii, one of which was ad- mixed with trabecular carcinoid but none of which was associated with hyperthyroidism or carcinoma. This study confirms the rarity in adult Nigerians of the following tumours each of which occurred only once in the series: mediastinal, solid ovarian teratoma, testicu- lar and sacrococcygeal teratoma. The latter two were malignant and the sacrococcygeal tumour has been reclassified as an endod- ermal sinus tumour. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC MacDougall, L.G. Acute Childhood Leukaemia in Johannesburg Leuk Res 9(6):765-767, 1985 178 Subjects Cancer Site: leukemia; Cancer Measures: incidence, survival Study Characteristics Design: time trend; Measures: descriptive statistics, higher level statistics; Time Frame: 1974-1983 Sample Characteristics 219 acute leukemia cases (blacks 90; whites 129): Johannesburg, South Africa Variables Demographic: race-ethnicity; Cancer: clinical diagnosis Data Sources Hospital medical records Abstract In the Johannesburg area of South Africa, significant differences have been noted between Black and White children in the incidence of acute leukemia, the morphologic cell type, response to treatment and survival. The annual incidence of acute leukemia in Johannes- burg was much lower in Blacks than Whites, due to a dispropor- tionately low incidence of acute lymphoblastic leukemia (ALL) in Blacks. Blacks with ALL had significantly shorter median survival than Whites which could not be attributed to an increased preva- lence of “traditional” poor prognostic features (e.g. high WBC, mediastinal mass). MacMahon, B.; Trichopoulos, D.; Brown, ].; et al. Age at Menarche, Probability of Ovulation and Breast Cancer Risk Int J Cancer 29:13-16, 1982 179 Subjects Risk Factor Exposure: endogenous hormones, reproductive factors; Cancer Site: breast; Cancer Measures: incidence Study Characteristics Design: retrospective; Measures: descriptive statistics, higher level statistics Sample Characteristics 431 individuals: female, 15-19 years old; United States, Canada, Denmark, Netherlands, Norway, Sweden, Greece, Finland, Yugoslavia, Estonia, Hong Kong, Japan Variables Demographic: age, race-ethnicity, geographic area, age at menarche Data Sources Laboratory test Abstract An analysis is undertaken of the frequency of ovulation in 17 groups of women aged 15 to 19 who had been the subjects of other studies. A urine specimen of at least 8 h accumulation had been provided on the 20th or 21st day of a menstrual cycle of 681 women. Analysis is restricted to 431 specimens which had been collected between 11 and 3 days prior to the onset of the subsequent menstrual period. A pregnanediol concentration of less than 1 mg per litre in such a specimen was taken as evidence that the cycle was anovular. The probability of a cycle being anovular was in- versely and significantly related to the number of years since menarche, and, with years since menarche held constant, was positively but not significantly associated with age at menarche. This observation indicates that women with early menarche do not have a longer duration of exposure to anovular cycles than do those whose menarche is delayed, and that variation in the dura- tion of exposure to post-menarcheal anovular cycles does not ex- plain the association of breast cancer risk with early age at menarche. The 17 groups of women were classified into four categories according to ethnic origin and breast cancer incidence in the populations from which they derived. Anovular cycles were not more common in the high-risk groups; indeed, the two centers in the lowest risk category had the highest proportion of anovular cycles. Macnab, G.M.; Urbanowicz, ].M.; Kew, M.C. Carcinoembryonic Antigen in Hepatocellular Cancer Br J Cancer 38:51-54, 1978 180 Subjects Cancer Site: liver; Biologic Marker: carcinoembryonic antigen Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 172 individuals (72 liver cancer cases; nonsmoking controls 80; smoking controls 20): presumed black; South Africa Variables Demographic: sex Abstract Serum carcinoembryonic antigen (CEA) concentrations were 77 found to be raised in 28 of 72 black patients (39%) with hepatocel- lular cancer (HCC). The degree of elevation was slight or moderate, except in 3 patients in whom values >20 ng/ml were recorded. No significant correlation could be demonstrated in individual patients between the serum CEA concentration and various tests of liver function. The mean CEA value in the patients with cirrhosis in the non-tumorous liver was slightly higher than that in those without cirrhosis, but the difference did not reach statistical signifi- cance. There was no correlation between serum CEA and alpha- fetoprotein (AFP) levels. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Maisto, O.E.; Bremner, C.G. Cancer of the Colon and Rectum in the Coloured Population of Johannesburg S Afr Med J 60:571-573, 1981 181 Subjects Treatment; Risk Factor Exposure: diet-nutritional factors, tobacco, alcohol; Cancer Site: colon, rectum; Cancer Measures: incidence, survival Study Characteristics Design: time trend, cross-sectional; Measures: crude incidence rate, descriptive statistics; Time Frame: 1970-1980 Sample Characteristics 16 colon cancer cases and 13 rectum cancer cases: coloured; Johannesburg, South Africa Variables Demographic: age, sex, geographic area; Cancer: tumor location, histology Data Sources Hospital medical records, histopathology records Abstract In the Coloured population of Johannesburg, the incidence of carcinoma of the colon and rectum estimated over an 11-year period (1970 - 1980) is 1,33 and 1,08/100,000 respectively per year. This is lower than the incidence reported in Whites and appears to be even lower than that reported in South African and Zimbabwean Blacks. A survey of dietary factors in this population showed a diet which was high in protein, low in fats and high in bulk. The frequency of bowel action was higher than in both Whites and Blacks. These factors support Burkitt's hypothesis which states that races with a high-bulk diet have a lower incidence of large- bowel cancer. Mancuso, T.F. Lung Cancer Among Black Migrants: Interaction of Host and Occupational Environment Factors J Occup Med 19(8): 531-532, 1977 182 Subjects Risk Factor Exposure: occupation, geographic area; Cancer Site: lung; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: age-adjusted mortality rate; Time Frame: 1953-1967 Sample Characteristics Lung and bronchus cancer mortality data: adults, 35-69 years of age; Ohio and lung cancer mortality data: male coke plant employees; Pennsylvania Variables Demographic: sex, race-ethnicity (black, nonwhite, white), birthplace Data Sources Death registration, industry records Abstract A study of Ohio residents by place of birth identified an in- 78 creased cancer risk (100% excess) among black migrants born in the South compared with blacks born in Ohio. The migration of blacks from South to North for the purpose of industrial employ- ment was extensive between the years 1940-1970. The hypothesis is that biological and social imprints (poverty, malnutrition, gene- tics), endemic of the early years of life, provide a physiological imbalance which provides the basis for increased susceptibility to adverse environmental stresses subsequent to migration. Findings consistent with this hypothesis were obtained in a prospective study of steel workers in which 2,543 men in coke plant operations were included, in which a prior high rate for lung cancer had already been established. A strong association was found between place of birth and place of death from lung cancer in the nonwhite coke plant workers. Examination of these death certificates re- vealed that 33 of 35 deaths of lung cancer occurred among men born in the South. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Mancuso, T.F.; Sterling, T.D. Relation of Place of Birth and Migration in Cancer Mortality in the U.S.: A Study of Ohio Residents (1959-1967) J Chron Dis 27:459-474, 1974 183 Subjects Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: age-adjusted mortality rate, descriptive statistics; Time Frame: 1959-1967 Sample Characteristics Cancer mortality data: Ohio Variables Demographic: sex, age, race-ethnicity (nonwhite, white) Data Sources Death registration, U.S. Population Census Abstract The epidemiological method and findings identify and em- phasize the research potential of studying migrants within the U.S., for specific causes of death. The study has demonstrated the marked differences in cancer mortality which become manifest for the nonwhite and native-born white when resident male and female deaths of a state are subclassified by region of birth and state of residence, which would not otherwise be apparent in the comparisons by broad classifications as currently coded in vital records. The age adjusted rates for all cancers (ages 35-69) for the nonwhite males born in the South were 363.67 vs 180.43 for those born in Ohio. The same ratio among the white males born in the South was 225.73, and for those born in Ohio, 197.14. For cancer of the lung and other specific cancer sites, the nonwhite males showed a higher rate for those born in the South compared to those born in Ohio. The excess for the white males was most notable for cancer of the lung. Statistical evidence is presented for the conclusion that the observed differences in cancer mortality among the nonwhite are not due to race per se. The importance of endemic factors associated with place of birth in the U.S. and the influences of the early years of life, particularly in the South in the development of social and biological imprints, and following migration the subsequent exposure to industrial environments, has been considered. Mandal, A.K.; Shaman, I.N.; Bier, R.K.; et al. Outcome of Thoracic Esophageal Carcinoma in Blacks in the Inner City 184 Cancer 54:924-928, 1984 Subjects Treatment; Cancer Site: esophagus; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, crude mortality rate, descriptive statistics Sample Characteristics 45 esophagus cancer cases (males 30; females 15): black, 26-89 years old, hospital patients, residing in a socioeconomically depressed inner-city area; South-Central Los Angeles County, CA Variables Demographic: age, sex; Cancer: stage, symptoms, tumor location Data Sources Hospital medical records Abstract This report is based on 45 cases of cancer of the thoracic esophagus in blacks residing in a socioeconomically depressed area of the inner city of South-Central Los Angeles. There were 30 men and 15 women, ranging in age from 26 to 89 years. Preoperative management, contraindications of surgery, and over- all results are specified. The operative mortality rate was 14%. As a means of restoring continuity of the alimentary tract after resec- tion of esophageal carcinoma, the stomach was preferred, although the colon was used in two instances. Five-year survival in this series was 14% in the operative group and 13% in the radiotherapy group; one of the surgical patients is still alive, and another one survived for 72 months. In the radiation group, no such incidence occurred. It is therefore concluded that surgery is the optimal palliation if there is an acceptable risk of mortality. When operative risks are unacceptable radiation therapy is used. Manton, K.G. Sex and Race Specific Mortality Differentials in Multiple Cause of Death Data 185 Gerontologist 20(4):480-493, 1980 Subjects Abstract Methodology; Cancer Site: multiple sites; Cancer Measures: The black/white mortality crossover is studied by comparing mortality its presence in underlying cause of death data, where one cause Study Characteristics Design: cross-sectional; Measures: age-specific mortality rate Sample Characteristics Mortality data: United States Variables Demographic: age, sex, race-ethnicity (black, white) Data Sources Death registration, U.S. Population Census 79 is assigned to each death, to its presence in data where deaths may have multiple causes. A number of observations are made which suggest that the basic biomedical aspects of aging and chronic disease are better represented in the multiple cause of death data. The constraint of assigning only one cause to a death distorts the picture of age variation in chronic disease mortality and limits the use of underlying cause of death data in the study of certain aspects of population aging processes. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC McDonald, M.W.; Johnson, D.E.; Guinee, V.F. Testicular Tumor in Blacks Urology 23(6):543-546, 1984 186 Subjects Treatment; Diagnosis; Cancer Site: testis; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics; Time Frame: 1944-1980 Sample Characteristics 19 testis cancer cases: black, male, 22-64 years old; Houston, TX Variables Cancer: histology, tumor location, symptoms, stage Abstract Although testicular tumors are less common in blacks than in whites, they do occur and must be considered in making a diffe- rential diagnosis of scrotal masses. At the University of Texas M. D. Anderson Hospital and Tumor Institute at Houston, 19 black men were treated for testicular tumor between 1944 and 1980, representing 1.4 per cent of the total of 1,380 patients treated for testicular tumor. Since blacks made up 7.9 per cent of all patients treated at UT M.D. Anderson Hospital, blacks with testicular tumor were seen about one-fifth as often as would be expected if there were no racial variation in the incidence of the disease. Twelve patients had a diagnosis of seminoma, five embryonal carcinoma, one teratocarcinoma, and one interstitial cell tumor. The majority of patients had metastatic disease at the time of diagnosis. Five of 12 patients with seminoma and 5 of 6 with nonseminomatous cancer died of disease. Prognosis today, however, is no worse for blacks than for whites. McGlashan, N.D.; Bradshaw, E.; Harington, ].S. Cancer of the Esophagus and the Use of Tobacco and Alcoholic Beverages in Transkei, 1975-6 Int J Cancer 29:249-256, 1982 187 Subjects Risk Factor Exposure: alcohol, tobacco; Cancer Site: esophagus; Cancer Measures: incidence, mortality Study Characteristics Design: cross-sectional; Measures: relative risk, crude incidence rate, descriptive statistics, higher level statistics Sample Characteristics 1,161 individuals: black; Nqamakwe, Butterworth, Mqanduli, Libode, Bizana and Lusikisiki regions, South Africa Variables SES: westernization; Demographic: age, sex, race-ethnicity (Mfengu, Xhosa, Pondo, Thembu), geographic area, urban-rural Data Sources Survey Abstract A large questionnaire-based sample survey over a period of 2 years has collected information on tobacco and alcohol use among young and old, male and female respondents in three regions of Transkei selected on the basis of having proved contrasts of inci- dence experienced from oesophageal cancer. Results show signif- icant gradients of social fabric and of customary usage of tobacco and alcohol across the three incidence regions. In particular the concept is supported of a synergistic relationship between tobacco and alcohol when both are used by one individual, with the major role being played by tobacco, especially when smoked in pipes. Stress is also laid on contrasts between old and young on the assumption that dynamic shifts of mortality may be expected to follow as the young of today alter the patterns of intake established by their elders. McGlashan, N.D.; Harington, ].S.; Bradshaw, E. Geographical Distribution of Certain Cancers in South Africa, 1968-1972 S Afr Med J 65:795-804, 1984 188 Subjects Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: age-adjusted mortality rate, age-specific mortality rate, crude mortality rate, descriptive statistics, higher level statistics; Time Frame: 1968-1972 Sample Characteristics Cancer mortality data: South Africa Variables Demographic: age, sex, race-ethnicity (colored, white, Asian), geographic area Data Sources Death registration 80 Abstract The geographical distributions of six major causes of cancer mortality in South Africa have been analysed for three population groups: White, Coloured and Asian. Because of the lack of compar- able data, the majority Black population is not included. Geog- rahpical patterns are compared and discussed in the light of find- ings from other countries. In several cases the potential for aetiolog- ical follow-up is demonstrated. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC McGlashan, N.D.; Harington, ].S.; Bradshaw, E. Eleven Sites of Cancer in Black Gold Miners From Southern Africa: A Geographic Inquiry Br J] Cancer 46:947-954, 1982 189 Subjects Risk Factor Exposure: occupation, geographic area; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: time trend, cross-sectional; Measures: crude incidence rate, standardized incidence ratio, descriptive statistics, higher level statistics; Time Frame: 1964-1979 Sample Characteristics 1,807 cancer cases: male, 18-40 years old, gold miners; South Africa Variables Demographic: age, geographic area Data Sources Industry records, histopathology records Abstract A 5-year study of cancer in black gold miners, 1964-68, has now been extended for a separate 8-year period, 1972-79. This allows analyses of all cancers together and of 6 less common sites of cancer severally: lymphosarcomas, colon and rectum, leukemia, stomach, pancreas and buccal cavity and also of those too rare to classify. The malignancies are considered by territory of origin of the gold miners. Lesotho miners have significantly fewer (p<0.05) tumours of the lymphatic and haemopoietic tissues and Natal min- ers have the highest incidence rates for 5 of the 6 sites (excluding leukaemia). A simple grouping method is applied to determine which of the 11 sites of cancer in the miners have similar distribu- tions in their 10 territories of origin. The aetiological implications of clusters over space of certain sites of cancer are discussed. Fi- nally, temporal change over the years 1964-79 shows a significant decreased overall (p<0.01) in cases of lymphosarcomas and colo- rectal cancers and an increase (p<(0.05) in stomach cancer. The rare tumour, Kaposi's sarcoma, has also decreased significantly between the 2 periods studied. McLendon, R.E.; Robinson, ].S., Jr.; Chambers, D.B.; et al. The Glioblastoma Multiforme in Georgia, 1977-1981 Cancer 56:894-897, 1985 190 Subjects Diagnosis; Cancer Site: brain; Cancer Measures: incidence, mortality Study Characteristics Design: retrospective; Measures: age-specific incidence rate, observed survival rate; Time Frame: 1977-1981 Sample Characteristics 299 brain cancer cases (blacks 48; whites 251): Georgia Variables Demographic: age, sex, race-ethnicity Cancer: histology Data Sources non-SEER hospital cancer registry, death registration Abstract To characterize the incidence and mortality of the glioblastoma multiforme as it is seen at the primary care level, a retrospective study was done through the Georgia Tumor Registry on 299 pa- tients with histologically confirmed supratentorial glioblastoma multiformes. Specifically, the age incidence distribution, sex ratio, and the postoperative survivals were studied. The Registry also provided the opportunity to define the relative racial frequencies of this neoplasm in a large biracial population. The mean age of the patients at the time of surgery was 58.6 years and the age group having the highest incidence was 60 to 74 years. The overall ratio of affected males to females was 1.04 to 1. The postoperative survival periods were considerably shorter than those reported from large cooperative clinical trials. The overall postoperative survivals at 12 and 24 months were only 22% and 4%, respectively. Advancing age had a strong negative relationship with the length of survival such that the 50% survival for patients less than 30 years of age was 17 months, for those age 30 to 44 years it wa 10.2 months, for those age 45 to 59 it was 5.8 months, and for those older than 60 it was only 2.2 months. The study confirmed previous suggestions that the incidence of the neoplasm is considerably higher in whites. There were 2.3 glioblastomas in whites for every one in blacks. No difference was noted in the length of survival between whites and blacks. 81 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC McPhedran, P.; Heath, C.W., Jr.; Garcia, J.S. Racial Variations in Leukemia Incidence Among the Elderly J Natl Cancer Inst 45:25-28, 1970 191 Subjects Cancer Site: leukemia; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, age-specific mortality rate, descriptive statistics; Time Frame: 1956-1967 Sample Characteristics 947 leukemia cases (nonwhites 136; whites 811): Atlanta, GA; leukemia mortality data from the United States Variables Demographic: age, race-ethnicity Data Sources Hospital medical records, physician medical records, death registration Abstract Negroes over age 70 showed a striking reduction in overall leukemia incidence when compared with whites of the same age group. Incidence data for metropolitan Atlanta indicated that this reduction involved all types of leukemia and was not restricted to any single variety. This deficit of leukemia cases among elderly Negroes seems more likely to reflect socioeconomic factors, includ- ing reduced access to medical care, than any purely racial factor. Megafu, U. Cancer of the Genital Tract Among the Ibo Women in Nigeria Cancer 44:1875-1878, 1979 192 Subjects Risk Factor Exposure: reproductive factors; Cancer Site: cervix uteri, corpus uteri, vulva, ovary, vagina; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1971-1976 Sample Characteristics 225 cancer cases (genital cancers 82; breast cancers 87; other cancers 56): black (Ibo), female; Enugu, Nigeria Variables Demographic: age, parity; Cancer: histology, symptoms Data Sources Hospital medical records Abstract An analysis of female genital tract cancer among the Ibos of Nigeria has proven to be preponderantly cervical cancer. Itis there- fore obvious that the practice of ritual circumcision among the Ibo men does not protect the women from cervical cancer. The most important etiological factors contributing to cervical cancer among the Ibo women are early marriage, early age of coitus, and gland multiparity. Corpus cancer is relatively rare among the Ibo women. Menck, H.R.; Mack, T.M. Incidence of Biliary Tract Cancer in Los Angeles Natl Cancer Inst Monogr 62:95-99, 1982 193 Subjects Cancer Site: biliary tract; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1972-1977 Sample Characteristics 1,062 biliary tract cancer cases: Los Angeles, CA Variables SES: social class, occupation, religion; Demographic: age, sex, parity Data Sources non-SEER population-based cancer registry, hospital medical records, U.S. Population Census 82 Abstract Incidence data reported for biliary tract cancer in 4 ethnic groups of Los Angeles County are consistent with earlier reports of a similarity between the ethnic-gender distribution of gallbladder cancer and bile lithogenicity, and between that of other biliary cancer and free bilirubin. The female excess of biliary cancer proved to be highest in those ethnic groups with highest average parity, and risk was related to parity correlates such as marital status and religion (Catholic). A migration effect in Spanish-surnamed and other white females was present for both diseases. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Mettlin, C.; Natarajan, N. Epidemiologic Observations From the American College of Surgeons’ Survey on Prostate Cancer Prostate 4:323-331, 1983 194 Subjects Diagnosis; Cancer Site: prostate; Cancer Measures: incidence, mortality, survival Study Characteristics Design: cross-sectional, time trend; Measures: age-specific incidence rate, age-specific mortality rate, crude survival rate; Time Frame: 1974 or earlier, 1979 Sample Characteristics 34,245 prostate cancer cases (Period 1 cases 20,166; Period 2 cases 14,079): male; United States Variables Demographic: age, race-ethnicity (nonwhite, white); Cancer: stage Data Sources Hospital medical records Abstract Data on the incidence and mortality from prostate cancer are available in this country and others, but there are few sources of information relating to widely prevalent clinical practices. The long- and short-term prostate cancer surveys conducted by the Commission on Cancer of the American College of Surgeons offer a system of assessing patterns of cancer patient care in a large number of hospitals in the United States. Data from the Surveil- lance, Epidemiology, and End Results (SEER) System indicate that at every age blacks experience greater rates of incidence and mor- tality than whites. Comparisons of modes of diagnosis reported for black and white patients for 1974 and 1979 from the American College of Surgeons surveys show little difference between these groups, indicating that the excess in incidence among blacks is not attributable to differential patterns of diagnosis. The survey results also indicate that white patients appear to have a slight survival advantage in stages B, C, and D. The lower survival ap- pears to be the product of relatively advanced stage of the disease rather than the result of biologically more malignant nature of prostate cancer in the blacks. Mettlin, C.; Natarajan, N.; Croghan, I.; et al. Patterns of Care Survey Data on Family History and Ulcerative Colitis in Rectal Cancer Cancer Genet Cytogenet 12:121-128, 1984 195 Subjects Risk Factor Exposure: familial factors, other diseases; Cancer Site: rectum Study Characteristics Design: retrospective; Measures: descriptive statistics, higher level statistics; Time Frame: 1978 Sample Characteristics 6,857 rectum cancer cases (blacks 398; whites 6,139): hospital patients; United States Variables Demographic: age, sex, race-ethnicity; Cancer: stage, tumor grade Data Sources Survey Abstract Data from a short-term survey on rectal cancer, sponsored by the Commission on Cancer of the American College of Surgeons, were analyzed in order to assess the association between family history of bowel cancer, ulcerative colitis, and rectal cancer. Over- all, 7.4% of rectal cancer patients had a positive family history of bowel cancer and 2.5% reported personal history of ulcerative colitis. The association between these two factors was statistically significant. Patients with a positive family history were signific- antly younger than those without a family history of bowel cancer. Significantly more white patients than black patients had positive family histories. Significantly greater proportions of patients with earlier stage disease and well differentiated tumors were observed among persons with a family history of bowel cancer. No such association was observed among patients with a history of ulcera- tive colitis. Miller, B.; Abrahams, C.; Cole, G.C.; et al. Ocular Malignant Melanoma in South African Blacks Br J] Opthalmol 65:720-722, 1981 196 Subjects Cancer Site: eye; Cancer Measures: incidence Study Characteristics Design: case study, cross-sectional; Measures: descriptive statistics; Time Frame: 1954-1978 Sample Characteristics 161 primary ocular malignant melanoma cases (blacks 8; whites 153): South Africa and one primary ocular malignant melanoma case: black, female, 49 years old; South Africa Variables Demographic: age, sex, race-ethnicity; Cancer: tumor location Data Sources Histopathology records 83 Abstract The incidence of primary ocular malignant melanoma in blacks during the last 25 years in a large part of the Transvaal Province of South Africa is reported. Only 8 cases were diagnosed, 1 uveal, 4 conjunctival, and 3 orbital. During this same period 153 cases of primary ocular malignant melanoma were diagnosed in whites. The rarity of these tumours among blacks is stressed and the reasons for this infrequent incidence discussed. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Miller, C.W.; McLaughlin, R.E. Osteosarcoma in Siblings: Report of Two Cases J] Bone Joint Surg 59A(2):261-263, 1977 197 Subjects Risk Factor Exposure: familial factors; Cancer Site: sarcoma; Cancer Measures: incidence Study Characteristics Design: case-study; Time Frame: 1964-1977 Sample Characteristics Two osteosarcoma cases: female, 15and 17 years old; Virginia Abstract Precise data on the incidence of osteosarcoma in the population of the United States are not available. In one study of children in the United States who died from neoplasms, the death rate from osteosarcoma in those under fifteen years old was estimated at 1.51 per million and in those between fifteen and nineteen years old, 5.80 per million. These data confirm the clinical observation that osteosarcoma is a reportable event. The following case reports concern two sisters, both of whom had osteosarcoma. Miller, G.].; Beckles, G.L.A.; Byam, N.T.A.; et al. Serum Lipoprotein Concentrations in Relation to Ethinc Composition and Urbanization in Men and Women of Trinidad, West Indies Int J Epidemiol 13(4):413-421, 1984 198 Subjects Biologic Marker: serum lipoprotein concentrations Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1977-1981 Sample Characteristics 2,488 individuals (Africans 990; Indians 747; Europeans 177; mixed 574): 35-69 years old, urban area; Trinidad and 72 individuals (males 39; females 33): 35-54 years old, rural area; Trinidad Variables Demographic: age, sex, race-ethnicity Data Sources Laboratory test, country census Abstract A total population survey of serum lipoprotein concentrations was undertaken in an urban community in Port-of-Spain, Trinidad, and the results compared with a rural survey on the same island and a study of healthy adults in Bristol, England. Lipoproteins were separated with identical techniques and lipid determinations made in a common laboratory. In Trinidad, response rates for men and women were 93 and 88% respectively in the urban survey, and 89% in the smaller rural study. Mean concentration of HDL cholesterol was significantly lower and LDL cholesterol signific- antly higher in urban men than rural men. No urban-rural differ- ences were found in women. In urban men under 55 years, HDL and LDL cholesterol concentrations were similar in Port-of-Spain and Bristol while VLDL triglyceride was relatively high in Trinidad. Distinct ethnic differences in lipoprotein concentrations were found in Trinidad. Indian men and women tended to have a low HDL cholesterol relative to other ethnic groups, while African men and women were characterized by relatively low concentrations of LDL cholesterol and VLDL triglyceride. The reduction in HDL cholesterol of Indian men appeared to be due mainly to a relatively low HDL concentration. The results are consistent with reported regional, sex and ethnic differences in CHD incidence in Trinidad, and accord with statistics which show cardiovascular disease to have emerged as the major cause of death in this community. Miller, J.M.; Chapman, T.W. Reviewing Cancer in American Blacks: A Baltimore Study J Natl Med Assoc 73(2):127-132, 1981 199 Subjects Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: crude mortality rate, age-adjusted mortality rate; Time Frame: 1977 Sample Characteristics Cancer mortality data: Baltimore, MD Variables SES: income; Demographic: race-ethnicity (black, white), geographic area Data Sources U.S. Population Census, SEER cancer registry Abstract Rates for the incidence and mortality of cancer have greatly increased in American blacks as they have been assimilated into Western culture. There are many apparent causes for this change, and perhaps one factor is the mixing of genetic backgrounds to produce a single population. Heightened rates may be stemmed, however, by directing greater attention to prevention of cancer by elimination of carcinogens from the environment. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Mohla, S.; Sampson, C.C.; Khan, T.; et al. Estrogen and Progesterone Receptors in Breast Cancer in Black Americans: Correlation of Receptor Data with Tumor Differentiation Cancer 50(3):552-559, 1982 200 Subjects Biologic Marker: steroid receptors; Treatment; Risk Factor Exposure: reproductive factors; Cancer Site: breast; Cancer Measures: incidence, survival Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1977-1980 Sample Characteristics 146 breast cancer cases: black, female, 28 + years old; Washington, DC Variables Demographic: menopausal status; Cancer: histology, stage, hormone receptors, tumor grade Data Sources Hospital medical records, histopathology records Abstract It has been observed that 60-70% of breast cancer patients have estrogen receptors (ER) and that nearly two-thirds of such patients respond favorably to endocrine therapy. Cytosolic ER and proges- terone receptors (PgR) have been evaluated in the current study, among 146 black women with breast cancer in order to determine whether the distribution of ER and PgR differs from the national norm. The results showed following trends that were similar to reports from other institutions: (1) postmenopausal patients and primary tumors showed higher ER positivity than premenopausal patients and metastatic sites, respectively; (2) a significant correla- tion between the ER positivity and tumor grade; and (3) a higher PgR positivity in ER-positive patients than in ER-negative patients. However, statistically significant differences were observed in three parameters when compared to reports from other institutions on white patients: (1) a low incidence of ER-positive (46%) and high incidence of ER-negative (42%) tumors; (2) a higher incidence of poorly differentiated (55.5%) and a lower incidence of well dif- ferentiated (5.5%) tumors; tumor grade was independent of age, menopausal status, histopathology and stage; and (3) a higher percentage of patients discovered at a more advanced stage of the disease. The clinical implications of these results in explaining the relatively poorer survival of black women with breast cancer com- pared to whites is discussed. Whether this high incidence of PD tumors and thus a high incidence of ER negativity is due to ethnic differences and/or environmental and other factors remains to be elucidated. Monk, M.; Warshauer, M.E. A Methodologic Problem in Mortality Studies of Migrant Populations J Chron Dis 31:347-352, 1978 201 Subjects Risk Factor Exposure: geographic area; Cancer Site: colon, stomach; Cancer Measures: mortality; Methodology Study Characteristics Design: time trend; Measures: age-adjusted mortality rate, descriptive statistics; Time Frame: 1950-1972 Sample Characteristics Cancer mortality and census data: black, males; New York City, NY Variables Demographic: age, geographic area Data Sources U.S. Population Census, death registration Abstract Using U.S. census reports of state of birth for 1950, 1960 and 1970, we performed a cohort-analysis of black men born in New York State and in three Southern states. There were large increases from one census year to the next in the number of middle-aged men reported to have been born in New York State. Conversely, there were larger than expected decreases from one census to the next in the number of men reporting the Southern states as their birth place. We believe these changes represent errors in reporting state of birth in the census. Such reporting errors will result in erroneous death rates for migrant and non-migrant populations when death certificate data and census data are used to calculate the rates. In New York City, death rates for black migrants were higher than was actually the case and those for non-migrants lower. 85 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Monson, R.R.; Nakano, K.K. Mortality Among Rubber Workers: II. Other Employees Am J Epidemiol 103(3):297-303, 1976 202 Subjects Risk Factor Exposure: occupation; Cancer Site: all sites combined, multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: standardized mortality ratio (SMR), descriptive statistics; Time Frame: 1933-1971 Sample Characteristics 10,529 individuals (black males 986; white males 3,727; nonwhite females 315; white females 5,501): rubber plant employees; Akron, OH and 574 deaths: white, male, non-Akron rubber plant employees Variables Demographic: age, sex, race-ethnicity Data Sources Industry records Abstract Mortality among female, black male and white male salaried employees in Akron, Ohio, is described. Standard mortality ratios for all causes of death are: females, 78: black males, 62: white salaried males, 65. Excess deaths from cancer occurred in females: uterus, bladder, brain and multiple myeloma; in black males: Hodgkin's disease; and in white salaried males: bladder and lym- phatic. Also, proportional mortality among white male employees of six non-Akron plants is reported. Excess deaths from cancer include brain and lymphatic and hematopoietic. Mora, R.G.; Burris, R. Cancer of the Skin in Blacks: A Review of 128 Patients with Basal-Cell Carcinoma Cancer 47:1436-1438, 1981 203 Subjects Risk Factor Exposure: solar radiation; Cancer Site: skin Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1948-1979 Sample Characteristics 128 basal cell carcinoma cases (males 59; females 69): black, 5-92 years old; New Orleans, LA Variables Demographic: age, sex; Cancer: tumor location, histology Data Sources non-SEER hospital registry Abstract Basal cell carcinoma in black persons is generally believed to be rare, with cases only sporadically reported. Information about 148 basal cell carcinomas found in 128 black patients who were seen at Charity Hospital of Louisiana, New Orleans, between 1948 and 1979, was reviewed. As in whites, most of the tumors occurred on the sun-exposed areas of the head and neck. Morgan, ].E.; Martyak, S.N. Risk Factors and Early Diagnosis in Pancreatic Carcinoma Am ] Gastroenterol 67(1):257-260, 1977 204 Subjects Diagnosis; Risk Factor Exposure: tobacco, other diseases; Cancer Site: pancreas; Cancer Measures: incidence Study Characteristics Design: case study Sample Characteristics One pancreas cancer case: black, female, 60 years old Abstract A case is reported of pancreatic carcinoma recognized four years after bile peritonitis complicated cholecystectomy. The pa- tient possessed all the known significant risk factors associated with pancreatic carcinoma. The wider recognition of risk factors for pancreatic adenocarcinoma might lead to closer follow-up of patients with multiple factors and possibly earlier diagnosis and treatment leading to increased survival in patients with the disease. 86 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Morton, W.E.; Baker, H.W.; Fletcher, W.S. Geographic Pathology of Uterine Cancers in Oregon: Risks of Double Primaries and Effects of Socioeconomic Status Gynecol Oncol 16:63-77, 1983. 205 Subjects Cancer Site: cervix uteri, corpus uteri; Cancer Measures: incidence; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, descriptive statistics; Time Frame: 1968-1972 Sample Characteristics 5,595 cancer cases: female; Oregon Variables SES: census tract SES rank; Demographic: age, race-ethnicity (black, Hispanic, other white, Oriental, Amerindian, other), geographic area, urban-rural; Cancer: stage, histology Data Sources non-SEER hospital cancer registry, U.S. Population Census Abstract An intensive statewide study of uterine cancer incidence for the period 1968-1972 revealed that women with one histologic type of uterine cancer were much more susceptible to the simultaneous presence of a second histologic type than were the rest of the residents of the state. Squamous cell carcinoma was the most com- mon uterine cancer; peak number and rate were found in the age group 25-29 years, indicating the need for periodic screening of young adult women and the need for discussion of neoplasia risk in adolescent sex education classes. Risks of squamous cell car- cinoma and leiomyosarcoma were excessive among blacks. Socioeconomic status had a strong and consistent inverse relation- ship with uterine squamous cell carcinoma incidence, implying that its major risk determinant, sexual behavior, could have a similar socioeconomic pattern. Uterine adenocarcinoma incidence had a weaker but consistent direct relationship with socioeconomic status for which the cause is unknown. Mosley, E.L.; Chung, E.B.; Anderson, ].E.; et al. Thyroid Carcinoma in Black Patients J Natl Med Assoc 70(12):915-918, 1978 206 Subjects Treatment; Cancer Site: thyroid; Cancer Measures: survival Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1950-1975 Sample Characteristics 31 thyroid cancer cases: black; Washington, DC Variables Demographic: age, sex; Cancer: histology, symptoms Data Sources Hospital medical records Abstract From 680 surgical specimens of thyroid disease, 31 cases of thyroid carcinoma found at Howard University Hospital, from January 1950 to December 1975, are reviewed. Eighteen (58 per- cent) were females and 13 (42 percent) were males. Of the 31 patients, there were 11 patients with follicular carcinoma, 11 with papillary carcinoma, seven with mixed papillary and follicular car- cinoma, one with Hurthle cell carcinoma, and one with medullary carcinoma. Thyroid carcinoma accounts for only .001 percent of all admissions during the period of study, and is indeed a rare cause of disease among blacks at this institution. Recommenda- tions for surgery and follow-up data are presented. Myers, M.H. Survival From Cancer by Blacks and Whites 207 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 151-165 Subjects Methodology; Cancer Site: multiple sites; Cancer Measures: survival Study Characteristics Design: time trend; Measures: relative survival rate, descriptive statistics; Time Frame: 1960-1973 Sample Characteristics 185,393 cancer cases (blacks 18,030; whites 167,363): California; Connecticut; Iowa; New Orleans, LA Variables Demographic: sex, race-ethnicity Data Sources End Results Program Abstract This article indicates that survival for black cancer patients is generally less favorable than that for white but in a few instances black-white differences were not observed. Survival trends, some encouraging, some discouraging, viewed along with incidence trends suggest that there are major needs for research efforts in both treatment and prevention and that one should not take pre- cedence over the other. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Naik, K.G. Cervical Carcinoma in Zambia Int Surg 62(2):110-111, 1977 208 Subjects Diagnosis; Risk Factor Exposure: reproductive factors, sexual practices; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Measures: age-specific incidence rate; Time Frame: 1971-1973 Sample Characteristics 274 cervix uteri cancer cases (interviewed cases 59; non- interviewed cases 215): female; Zambia Variables Demographic: age, age at marriage, age at first pregnancy, parity; Cancer: histology Data Sources Histopathology records, survey Abstract A series of 274 cases of cervical carcinoma in Zambian Africans is presented. Cervical carcinoma formed 29.3% of all female can- cers. Poorly differentiated squamous carcinoma constituted the most common histologic type. Concomitant bilharziasis was found in 24.4% of cases. In 59 cases causal factors such as coitus, early marriage, age at first pregnancy and parity were studied and dis- cussed. No correlation was found between circumcision and cervi- cal carcinoma. Naraynsingh, V. Gastric Carcinoma in the West Indies: A Trinidad Study Cancer 56(8):2117-2119, 1985 209 Subjects Biologic Marker: ABO blood group; Risk Factor Exposure: diet-nutritional factors; Cancer Site: stomach; Cancer Measures: incidence, mortality Study Characteristics Design: time trend, cross-sectional; Measures: age-adjusted mortality rate, descriptive statistics; Time Frame: 1968-1981 Sample Characteristics 1,052 stomach cancer deaths and 100 stomach cancer cases (blacks 80; East Indians 12; mixed 8): Trinidad Variables SES: income; Demographic: age, sex, race-ethnicity; Cancer: stage, tumor location Data Sources Hospital medical records Abstract Gastric carcinoma, the commonest cause of death from malig- nant disease in Trinidad, has decreased steadily over the period 1968-1977. Of the 1052 deaths during this time, females accounted for a relatively high percentage (45%) of cases. There is a predomi- nance of antral tumors, and most cases present with very advanced disease (88% with Stage III & IV). This results in a low resectability rate (32%). Its racial predilection for blacks of the lower income group suggests that both genetic and environmental factors may be important. The possible role of dietary factors is discussed. Nasca, P.C.; Burnett, W.S.; Greenwald, P.; et al. Population Density As An Indicator of Urban-Rural Differences in Cancer Incidence, Upstate New York, 1968-1972 Am J Epidemiol 112(3):362-375, 1980 210 Subjects Risk Factor Exposure: geographic area; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-specific incidence rate, higher level statistics; Time Frame: 1968-1972 Sample Characteristics Cancer incidence data: New York Variables Demographic: sex, urban-rural Data Sources non-SEER population-based cancer registry, U.S. Population Census Abstract This report presents cancer incidence data (1968-1972) for cities and towns in upstate New York (New York State exclusive of New York City) with various levels of urbanization as measured by population density. A direct association between population de- nsity and incidence was observed for a number of anatomic sites. A statistically significant linear trend of increasing incidence with increasing population density was observed among males and females for cancers of the buccal cavity and pharynx, esophagus, bronchus and lung, stomach and colon. For carcinomas of the liver, gallbladder, pancreas, bladder, larynx and rectum, this as- sociation was observed only among males, while, for malignant neoplasms of the brain and nervous system, only females de- monstrated a statistically significant relationship between these two variables. The geographic distribution of incidence rates for the remaining sites appeared not to be related to population de- nsity. The most striking feature of these data, however, was the U-shaped pattern the authors observed for carcinomas of the uterine cervix. The observed geographic patterns tended to be consistent with data from previously published urban-rural studies and with the distribution of suspected etiologic factors. 88 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Natarajan, N.; Nemoto, T.; Mettlin, C.; et al. Race-Related Differences in Breast Cancer Patients Cancer 56(7):1704-1709, 1985 211 Subjects Biologic Marker: steroid receptors; Risk Factor Exposure: endogenous hormones, familial factors, reproductive factors; Cancer Site: breast; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); cross-sectional; Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1977, 1981 Sample Characteristics 27,355 breast cancer cases (blacks 2,296; whites 24,263; Oriental and other races 405; race not reported 391): admitted to hospital on or before 1976; United States and 19,841 breast cancer cases (blacks 1,571; whites 17,701; Orientals and other races 366; race not reported 203): admitted to hospital in 1981; United States Variables Demographic: age, race-ethnicity; age at first pregnancy, number of pregnancies, age at menopause; Cancer: tumor size, tumor location, nodal involvement, histology Data Sources Hospital medical records Abstract Data from the 1982 breast cancer survey of the American College of Surgeons were used to evaluate factors related to clinical, epidemiologic, and survival differences between black and white patients. Breast cancer in blacks was not discovered as early as in whites. Distribution of pathologic types of tumors were similar for both races with the exception of medullary carcinoma, which was more frequent in blacks than in whites. Estrogen receptor-positive tumors were found significantly less frequently in blacks compared with whites. Survival was better for whites compared with blacks within each axillary nodes group 0, 1 to 3, and 4+. Black women with negative or positive estrogen receptors had lower survival rates than white women of the same receptor status, age at diag- nosis, and estrogen receptor status as significant predictors of survival. Significant differences between black and white patients were also observed with respect to the report of family history of breast cancer, age at first pregnancy, number of pregnancies, and age at cessation of menses. Nelson, J.F.; Ship, LI Intraoral Carcinoma: Predisposing Factors and Their Frequency of Incidence as Related to Age at Onset J] Am Dent Assoc 82:564-568, 1971 212 Subjects Risk Factor Exposure: alcohol, tobacco, other diseases; Cancer Site: oral cavity Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1957-1967 Sample Characteristics 191 oral cavity cancer cases (nonwhite 86; white 105): Philadelphia, PA Variables Demographic: age, sex, race-ethnicity; Cancer: tumor location Data Sources Hospital medical records, death registration 89 Abstract Results of a study of 191 patients with primary intraoral car- cinoma during a ten-year period are reported. The age of the patient at the time cancer was diagnosed was used as the basis for comparing the frequency of predisposing factors that may bear a relationship to intraoral malignancy. The data presented provide evidence that the frequency of these predisposing variables is nega- tively associated with age at the time of diagnosis, for example, that persons with a higher frequency of predisposing factors develop malignancy at an earlier age, and that the relationship is consistent regardless of sex and racial groupings. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Nemoto, T.; Natarajan, N.; Smart, C.R.; et al. Patterns of Breast Cancer Detection in the United States J Surg Oncol 21:183-188, 1982 213 Subjects Screening: mammography, clinical breast exam, breast self- exam (BSE); Cancer Site: breast Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1978 Sample Characteristics 12,315 breast cancer cases (blacks 1,060; whites 10,801; others 279): female, hospital patients; United States Variables Demographic: age, race-ethnicity; Cancer: histology, stage, tumor size, tumor location, nodal involvement Data Sources Hospital medical records, survey Abstract The American College of Surgeons Commission on Cancer Short-term Survey of Breast Cancer in 12,315 patients showed that 73% of malignant tumors are found by patients, 23% by physicians, and 4% by mammography. It also indicates that younger women are most likely to discover tumors than older women, and that mammography is more likely to detect small tumors with negative axillary nodes. The effectiveness of mammography is most evident in women 50 to 74 years of age, although in women 45 to 49 years, the frequency of tumors detected by mammography nearly equals that for other groups. In black women, mammography may not be currently fully utilized. Analysis of the survey data would indi- cate that patients appear to demonstrate adequate skill in detecting tumors, as compared to physicians. A delay in diagnosis of longer than 3 months is associated with larger tumors and increased likelihood of axillary metastasis. Newell, G.R.; Krementz, E.T.; Roberts, J.D. Multiple Primary Neoplasms in Blacks Compared to Whites. IV. Further Cancers in Patients with Cancer of the Digestive Organs J Natl Cancer Inst 54(2):331-334, 1975 214 Subjects Cancer Site: digestive system Study Characteristics Design: prospective (nonconcurrent); Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1948-1970 Sample Characteristics 6,272 digestive system cancer cases (blacks 4,105; whites 2,167): New Orleans, LA Variables Demographic: sex, race-ethnicity; Cancer: tumor location Data Sources Third National Cancer Survey 1969-1971, non-SEER hospital cancer registry Abstract Second primary cancers found among whites and blacks with initial cancer of the digestive organs were reported based on data from the Charity Hospital Tumor Registry. Observed second pri- mary cancers were compared to expected numbers to obtain a direct estimate of risk. Both white and black men had about a twofold risk of developing a second cancer. For white men, the excess was limited to a subsequent skin cancer, but this finding was probably an artifact of reporting and lacked biologic signifi- cance. Among women, both white and black, large excesses of invasive cancer of the cervix and ovary were found after an initial cancer of the large intestine and anorectum was discovered. No excess of breast cancer was found. Newell, G.R.; Krementz, E.T.; Roberts, ].D.; et al. Multiple Primary Neoplasms in Blacks Compared to Whites. I. Further Cancers in Patients with Hodgkin's Disease, Leukemia, and Myeloma J Natl Cancer Inst 52(3):635-638, 1974 215 Subjects Cancer Site: lymphoma, leukemia, multiple myeloma; Cancer Measures: incidence Study Characteristics Design: prospective (nonconcurrent); Measures: relative risk, descriptive statistics, higher level statistics; Tine Frame: 1948-1970 Sample Characteristics 350 Hodgkin's disease cases (blacks 172; whites 178); 993 leukemia cases (blacks 517; whites 476); and 261 myeloma cases (blacks 192; whites 69): New Orleans, LA Variables Demographic: sex, race-ethnicity; Cancer: tumor location Data Sources non-SEER hospital cancer registry, hospital medical records Abstract Incidence of multiple primary neoplasms in patients with Hodgkin's disease, leukemia, and myeloma was determined from 23 years’ experience of the Charity Hospital Tumor Registry. Ex- pected cancers were calculated by application of age-, sex-, and race-specific incidence rates to the person-years experience in the registry; these ratios were compared to observed cancers. Whites with both Hodgkin's disease and leukemia had an increased risk for subsequent development of skin cancer, but this was thought due to artifacts of reporting. Blacks with leukemia had a signific- antly increased risk of later lung cancer. White males with myeloma had a 6.6-fold increased risk of subsequent cancer, and, although not statistically significant, this might be accounted for by sub- sequent cancer of the stomach. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Newell, G.R.; Krementz, E.T.; Roberts, J.D. Multiple Primary Neoplasms in Blacks Compared to Whites. II. Further Cancers in Patients with Cancer of the Buccal Cavity and Pharynx J Natl Cancer Inst 52(3):639-642, 1974 216 Subjects Cancer Site: oral cavity, pharynx Study Characteristics Design: prospective (nonconcurrent); Measures: relative risk, descriptive statistics, higher level statistics Sample Characteristics 1,848 oral cavity cancer cases (blacks 663; whites 1,185) and 587 pharynx cancer cases (blacks 269; whites 318): New Orleans, LA Variables Demographic: sex, race-ethnicity; Cancer: tumor location Data Sources non-SEER hospital cancer registry, hospital medical records Abstract Incidence of multiple primary neoplasms in patients with an index cancer of the buccal cavity and pharynx was determined from 23 years’ experience of the Charity Hospital Tumor Registry. Expected cancers were calculated by the application of age-, sex-, and race-specific incidence rates to the person-years experience in the registry; these rates were compared to observed cancers. White males with primary cancers of the lip and mouth had a significantly increased risk of developing further cancer of the skin and lung. Black males with primary cancer of the salivary gland had a threefold increased risk, which was explained by subsequent ad- ditional cancers of the buccal cavity and pharynx and lung. The fivefold increased risk among black males with primary cancer of the pharynx was accounted for by subsequent cancers of the esophagus, lung, and prostate. White females with cancer of the lip and mouth had an increased risk of further skin cancer. Black females had no increased risk for subsequent cancers. Newell, G.R.; Rawlings, W.; Krementz, E.T.; et al. Multiple Primary Neoplasms in Blacks Compared to Whites: III. Initial Cancers of the Female Breast and Uterus J Natl Cancer Inst 53(2):369-373, 1974 217 Subjects Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1948-1970 Sample Characteristics 8,773 breast, corpus uteri, and cervix uteri cancer cases (blacks 5,859; whites 2,914): female; New Orleans, LA Variables Demographic: race-ethnicity Data Sources non-SEER hospital cancer registry Abstract The incidence of second primary cancers among white and black women with initial cancers of the breast, cervix, and corpus uteri was determined from the experience of the Charity Hospital of Louisiana Tumor Registry. Observed second primary cancers were compared with expected numbers to obtain a direct estimate of risk. Both white and black females with initial breast cancer had an excess risk of developing cancers of the buccal cavity and pharynx and an additional breast cancer. White females with breast cancer had a slightly increased risk for developing cancer of the lung and ovary and black females for developing cancer of the corpus uteri and leukemia. Both white and black women with initial cancer of the cervix had an excess risk for subsequent cancers of the buccal cavity and pharynx and lung. Women with cervical cancer had a small increased risk for subsequent bladder cancer. Newell, G.R.; Roberts, J.D.; Baranovsky, A. Retinoblastoma: Presentation and Survival in Negro Children Compared with Whites 218 J Natl Cancer Inst 49(4):989-992, 1972 Subjects Treatment; Diagnosis; Cancer Site: eye; Cancer Measures: incidence, mortality, survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics Sample Characteristics 533 retinoblastoma cases (Negroes 51; whites 482—Charity hospital patients 37; other hospital patients 496): under 15 years old; New Orleans, LA; United States Variables Demographic: age, race-ethnicity; Cancer: stage Data Sources End Results Program Abstract Higher mortality rates from retinoblastoma among Negroes, as compared to whites, are unique among childhood cancers and thus may have important epidemiologic implications. Data suggest that this excess mortality among Negroes is due to delayed diag- nosis resulting in more advanced disease presenting in older Neg- roes leading to less favorable response to treatment. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Norris, H.J.; Taylor, H.B. Carcinoma of the Breast in Women Less than Thirty Years Old Cancer 26(4):953-959, 1970 219 Subjects Treatment; Risk Factor Exposure: reproductive factors; Cancer Site: breast; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics Sample Characteristics 135 breast cancer cases (Negroes 10; Caucasians 113; non-Caucasian 12): female, under 30 years old; records on file at the Armed Forces Institute of Pathology Variables Demographic: age, race-ethnicity, pregnancy status; Cancer: symptoms, histology, stage, tumor size, tumor location, nodal involvement Data Sources Histopathology records, hospital medical records Abstract Mammary carcinoma in women less than 30 years old appears to have a slightly poorer prognosis than in older women. Our study of 135 women in this age group suggests several explanations for this: 1. about 10% of the patients were pregnant or lactating — conditions associated with a higher incidence of axillary lymph node metastasis and lethal outcome; 2. incidence of axillary metas- tasis was slightly higher in younger women, regardless of pre- gnancy; 3. carcinoma frequently (17%) developed in the opposite breast, and 4. patients with 1 or 2 positive axillary lymph nodes did poorer than expected. Despite the overall unfavorable outlook, crude survival is improved by the relative infrequency of deaths from causes other than cancer, as contrasted with that of older women. Also, the prognosis for women under 30 is improved by the relatively high proportion (24%) of low-grade, infrequently metastasizing tumors, such as medullary, intraductal, juvenile, papillary, and well-differentiated carcinomas. When only infiltrat- ing duct carcinoma was considered, 5-year survival in young women was 50% at 5 years and 37% at 10 years. Patients without axillary metastatis did as well as older women with breast cancer, since 81% were living at 5 years and 74% at 10 years. Norris, H.].; Taylor, H.B. Carcinoma of the Male Breast Cancer 23(6):1428-1435, 1969 220 Subjects Diagnosis; Treatment; Cancer Site: breast; Cancer Measures: prevalence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, prevalence rate, descriptive statistics Sample Characteristics 108 breast cancer cases (Negroes 21; Caucasians 86; Japanese 1): 21-80 years old, male veterans; United States Variables Demographic: age; Cancer: histology, tumor location, tumor size, metastases, symptoms Data Sources Hospital medical records Abstract Men with mammary carcinoma have a worse prognosis than do women, even after noncancer deaths are excluded. This can be explained partly on the basis of the high proportion of lesions that lie beneath the nipple area (78%), as centrally located lesions have a greater incidence of metastasis to internal mammary lymph nodes with its consequent adverse effect on patient survival. The central location and the absence of significant encompassing breast tissue also allow small mammary cancers of men to involve the pectoral fascia and the dermal and subareolar lymphatic channels simultaneously. On the basis of size, smaller mammary cancers had only a slightly greater frequency of axillary metastasis in men than women but had considerably greater lethality, presumably because of higher frequency of internal mammary lymph node metastasis. 92 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Oettle, A.G. Regional Variations in the Frequency of Bantu Oesophageal Cancer Cases Admitted to Hospitals in South Africa S Afr Med J, April 1963, p 434-439 221 Subjects Risk Factor Exposure: geographic area; Cancer Site: esophagus; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: crude incidence rate, descriptive statistics; Time Frame: 1962 Sample Characteristics 1,568 esophagus cancer cases: Bantu; Johannesburg, South Africa Variables Demographic: sex, geographic area Data Sources Survey Abstract A questionnaire to South African hospitals revealed differences in the relative frequency of Bantu cases of cancer of the oesophagus which varied more than one-hundredfold between the regions of highest and lowest frequencies. The maximum frequency occurred in the Southern Transkei, but the disease was also common in all large cities, being commonest in Bloemfontein. This patchy distri- bution suggests a carcinogenic exposure that is neither peculiarly rural nor peculiarly urban, to which males are usually but not invariably more heavily exposed than females. The disease is com- mon in other regions in Africa, and the epidemiological features suggest a cause or causes of recent origin and wide but somewhat haphazard distribution. Among the non-Negroes an increase in cancer of the oesophagus would appear to be occurring among Coloured males, Asian females, and possibly the Sudanese. Among South African Whites provincial differences in mortality from oesophageal cancer have been noted. Ogunbiyi, T.A.].; Adedeji, S.A. Malignant Melanoma of the Skin in Lagos. Experience in an Urban African Community Mat Med Pol 2(35):146-149, 1978 222 Subjects Cancer Site: malignant melanoma; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics; Time Frame: 1963-1975 Sample Characteristics 25 malignant melanoma cases (males 8; females 17): black; Lagos, Nigeria Variables Demographic: age, sex; Cancer: stage, tumor location Abstract 25 cases of malignant melanoma admitted to Lagos University Teaching Hospital over a 13 year period from 1963-1975 were re- viewed. Malignant melanoma constitute 1.1% of all malignant tumors treated during the period. 64% of tumors occurred in the sole of the foot. Clinical staging of tumors revealed 48% of tumors in stage 1, 48% in stage 2 and only 4% in stage 3. Some of the tumors in stage 1 attained large sizes with no lymph node involve- ment. The rarity of malignant melanoma of the skin in negroes and the relationship of trauma to the large number of cases occur- ring in the sole of the foot was discussed. A crude 2 year survival rate figure of 60% was obtained after treatment by surgery and sometimes with adjuvant chemotherapy of radiotherapy. The dif- ficulty of obtaining a 5 year follow up figures in an urban African population is highlighted. Olisa, E.G.; Chandra, R.; Jackson, M.A.; et al. Malignant Tumors in American Black and Nigerian Children: A Comparative Study J Natl Cancer Inst 55(2):281-284, 1975 223 Subjects Risk Factor Exposure: geographic area; Cancer Site: multiple sites Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1960-1973 Sample Characteristics 1,487 cancer cases (American blacks 162; Nigerians 1,325): under 15 years old; Washington, DC; Ibadan, Nigeria Variables Demographic: sex, race-ethnicity, geographic area; Cancer: histology Data Sources Hospital medical records, histopathology records Abstract Results of a study on the relative frequencies of tumors in American blacks and Nigerian children were compared with data from the Childhood Cancer Registries in Manchester, United King- dom, and Kampala, Uganda. The American black child living in Washington, D.C. and the Caucasian child living in Manchester had similar high frequencies for leukemia and glioma, whereas the incidence of lymphoma and retinoblastoma was low. African children living in Nigeria or Uganda had the opposite frequency patterns. These differences in frequencies of tumors between two ethnologically related population groups, American blacks and Nigerian, suggested the influence of environmental factors in the etiology of these tumors, even though exposure to environmental carcinogens was short. The rarity of Ewing's sarcoma and testicular tumors in American blacks and Nigerian children suggested a genetic influence. 93 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Oluwasanmi, J.O.; Daramola, J.O.; Ajagbe, H.A. Malignant Melanoma of the Oral Cavity Plast Reconstr Surg, Oct 1977, p 607-610 224 Subjects Diagnosis; Treatment; Cancer Site: malignant melanoma; Cancer Measures: incidence Study Characteristics Design: case study; Time Frame: 10 years Sample Characteristics Three primary malignant melanomas of the mouth cases: black, male, 22-28 years old, hospital patients; Ibadan, Nigeria Variables Cancer: histology, stage Data Sources Hospital medical records Abstract The occurrence of malignant melanoma in different populations has been the subject of several reports, but there is little information on this type of tumor in the oral cavity of black people, particularly Africans. Although it is a relatively rare tumor when compared with other oral tumors, 150 cases were compiled by Chaudhry, Hampel, and Gorlin; more recently, Liversedge has added 53 cases from various sources. The oral mucous membrances of the Africans have more melanin than is usually seen in other races, but there is no evidence to suggest that this racial group is more prone to development of melanotic tumors of the mouth. (Interestingly, melanomas are more commonly found in the relatively unpig- mented soles of the feet.) We report 3 cases of primary malignant melanomas of the mouth which we have seen during the past 10 years. The patients were Nigerians, and the lesions were treated surgically. Oluwole, S.F.; Freeman, H.P. Analysis of Benign Breast Lesions in Blacks Am J Surg 137:786-789, 1979 225 Subjects Diagnosis; Risk Factor Exposure: exogenous hormones, reproductive factors, familial factors; Cancer Site: breast; Cancer Measures: incidence Study Characteristics Measures: age-specific incidence rate, descriptive statistics; Time Frame: 1975-1977 Sample Characteristics 282 breast lesions (malignant lesions 80; benign lesions 202); black, female; New York City, NY Variables Demographic: age, number of pregnancies; Cancer: histology Data Sources Histopathology records Abstract A clinicopathologic analysis of 202 benign breast lesions in black women is presented. The study shows that the peak inci- dence of fibroadenoma occurs at an earlier age in black than in white patients. Fibrocystic disease is most frequent in both black and white patients between 25 and 45 years of age. It is noteworthy that nulliparous adolescent blacks have a higher risk of fib- roadenoma developing. The low incidence of fibrocystic disease in our patients does not reflect the clinical incidence of the disease because most patients with fibrocystic disease do not undergo biopsy. We are unable to draw any conclusions about the relation between the use of oral contraceptives and the incidence of benign breast disease. In conclusion, a review of the literature and an analysis of our data suggest a relatively higher incidence of fib- roadenoma among black patients. In contrast to the finding in the white population, it appears that in blacks fibroadenoma is more common than carcinoma of the breast. Onuigbo, W.I.B. Lip Lesions in Nigerian Igbos Int J Oral Surg 7:73-75, 1978 226 Subjects Prevention; Risk Factor Exposure: solar radiation; Cancer Site: oral cavity; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1970-1977 Sample Characteristics 16 individuals (carcinoma lip leisons 2; non-carcinoma lip leisons 14—Males 9; females 7): 3-47 years old, Igbos; Enugu, Nigeria Variables Demographic: age; Cancer: histology, tumor location Data Sources Histopathology records 94 Abstract Lesions of the lip seen over a 7-year period at a central labora- tory serving the Igbos of Nigeria are reviewed. In approximately 8,500 surgical specimens, there were 63 jaw tumors but only 16 lip lesions. Two of the lip lesions alone were carcinomas. Both were (a) squamous celled, (b) associated with sunlight-induced degeneration of the dermal collagen, and (c) found in young adult albinos. Accordingly, albinism should be recognized as one of the etiologic factors in lip cancer. The appropriate preventive action is the wearing of wide-brim hats by albinos. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Onuigbo, W.I.B. Short Reports: Relevance of Race in Management of Testicular Tumours British Med J 1(6052):22-23, 1977 227 Subjects Treatment; Diagnosis; Cancer Site: testis; Cancer Measures: prevalence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Tine Frame: 6 years Sample Characteristics 7,500 surgical specimens (testis samples 21; other 7,479): Igbos, male; Nigeria Variables Cancer: histology Abstract Seventy-five thousand specimens from Nigerian Iglos were examined. Of these, 21 were testicular samples - five biopsy speci- mens and 16 orchidectomy specimens - sent in by 16 doctors from 12 hospitals. Of the five biopsy specimens, 2 appeared to be nor- mal. Of the 16 orchidectomy specimens, three were normal; nine showed evidence of granulomatous orchites - 5 idiopathic, 2 tuber- culous, and 2 schistosomal; two embryonal carcinomas; one metas- tic lymphoma; and one fibrous pseudotumour of the testicular tunic. The author suggests that limited biopsy should be used in cases of suspected malignancy of the testis instead of orchidectomy when the patient is from an ethnic group in which the prevalence of testicular cancer is low. Onuigbo, W.1.B. Cancer of the Oral Cavity in Nigerian Igbos ORL 39:247-250, 1977 228 Subjects Cancer Site: oral cavity; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1970-1977 Sample Characteristics 31 oral cavity cancer cases (males 20; females 11): 4-70 years old, Igbos; Nigeria Variables Demographic: age, sex; Cancer: histology, tumor location Data Sources Histopathology records Abstract An unselected biopsy material of 31 Nigerian Igbo patients with cancer of the oral cavity is presented. The descending order of localization was gingiva, palate, tongue and floor of mouth. Bur- kitt’s tumour was limited to the jaw and affected older rather than younger children. There was scarcely any preponderance of squamous cell carcinoma in males. It would appear that cancer of the oral cavity is not a great problem in the Igbos and that potent carcinogens concerned with the causation are lacking in the com- munity. Onuigbo, W.].B. Biopsy Study of Hepatocarcinomas in Nigerian Igbos Acta Hepato-Gastroenterol 24:24-26, 1977 229 Subjects Risk Factor Exposure: other diseases; Cancer Site: liver; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1970-1976 Sample Characteristics 55 hepatocarcinoma cases (males 41; females 14): 0-60 years old, Igbos; Nigeria Variables Demographic: age, sex; Cancer: symptoms Data Sources Histopathology records 95 Abstract Fifty-five hepatocarcinomas were found in a review of approx- imately 7,500 surgical biopsies done on Nigerian Igbos during a period of 6 years. The male:female ratio was 2.9:1 and the age peak was between 20 and 49 years. The main symptoms were abdominal swelling, pain, emaciation, jaundice, fever, anorexia, and diarrhea. Physical examination revealed a palpable liver in nearly all patients. Two patients presented acutely with hemoperitoneum due to rupture of necrotic tumor nodule. Cir- rhosis was found in 60% of the adequately sized specimens. In comparison with published data, this series from an ethnic group in Nigeria, West Africa, reveals both similarities and dissimilarities which are noteworthy. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Ory, HW.; Rubin, G.L.; Jones, V.; et al. Mortality Among Young Black Women Using Contraceptives 230 JAMA 251(8):1044-1048, 1984 Subjects Risk Factor Exposure: exogenous hormones, contraceptive devices; Cancer Site: all sites combined; Cancer Measures: mortality Study Characteristics Design: prospective (nonconcurrent); Measures: crude mortality rate, descriptive statistics, higher level statistics; Time Frame: 1967-1977 Sample Characteristics 30,580 individuals: black, female 15-44 years old, inner-city; Atlanta, GA Variables Demographic: age, marital status, parity Data Sources Family planning clinic medical records, death registration Abstract Data on the risk of death associated with various contraceptive methods are incomplete. Therefore, we analyzed the mortality rates for young, black inner-city women who used one of four methods of contraception-oral contraceptives, depomedroxyp- rogesterone acetate, intrauterine (contraceptive) devices, and bar- rier methods. The subjects were 30,580 15- to 44-year-old women who enrolled at a family planning clinic between 1967 and 1972 and who were observed by monitoring death certificates through the end of 1977. Forty percent of the 218 deaths observed were from accidents and violence. Use of this planning clinic greatly reduced the risk of death from childbearing; only two deaths were associated with pregnancy and childbirth, compared with 24 deaths expected. Overall, users of the four methods died at similar, low rates. Given that this study involves considerable loss to fol- low-up, possible acute effects of contraceptives (e.g., infections or thrombosis) are more accurately estimated than possible long-term effects (e.g., cancer). Page, W.F.; Kuntz, A.J. Racial and Socioeconomic Factors in Cancer Survival: A Comparison of Veterans Administration Results with Selected Studies Cancer 45:1029-1040, 1980 231 Subjects Cancer Site: multiple sites; Cancer Measures: survival; Socioeconomic Factors Study Characteristics Measures: relative survival rate, descriptive statistics, higher level statistics; Time Frame: 1958-1963 Sample Characteristics 45,000 cancer cases: male, VA Hospital patients Variables Demographic: age, race-ethnicity (black, white); Cancer: stage Data Sources non-SEER hospital cancer registry Abstract The survival experience of 46,000 Veterans Administration (VA) male cancer patients was analyzed and compared with the results of three other studies. In the VA data, no significant differences were found between white and black patient cancer survival rates except for bladder cancer; this observation differs from those in other studies. In the VA, all patients receive the same treatment with no distinctions whereas most U.S. hospitals place their pa- tients into categories based on ability to pay. This factor probably accounts for the lack of racial differences in survival rates in the VA and the existence of racial or socioeconomic differences in survival rates in the other studies. Park, I-].; Jones, HW., Jr. Glucose-6-Phospate Dehydrogenase and the Histogenesis of Epidermoid Carcinoma of the Cervix 232 Am J Obstet Gynecol 102(1):106-109, 1968 Subjects Biologic Marker: G6PD; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: cross-sectional Sample Characteristics Eight invasive epidermoid carcinoma of the cervix cases: Negro, female Data Sources Laboratory test Abstract The result of vertical starch gel electrophoresis of blood, cancer- ous, and noncancerous epithelium of the cervix of 18 patients who were heterozygous for G-6-PD variants is presented. Among 10 patients without cancer of the cervix, examination of the cervical epithelium showed 9 with 2 bands and one with 1 band. Among 8 patients with cancer, examination of the cancer tissue showed 5 with a single band (1A and 4B) and 3 with double bands. Is is proposed that the evidence is consistent with the concept that epidermoid carcinoma of the cervix arises from a single cell. It is also concluded that the Lyon phenomenon is also functioning in at least some tumors. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Pathak, D.R.; Samet, ].M.; Howard, C.A.; et al. Malignant Melanoma of the Skin in New Mexico 1969-1977 Cancer 50(7):1440-1446, 1982 233 Subjects Risk Factor Exposure: solar radiation; Cancer Site: malignant melanoma; Cancer Measures: incidence, mortality Study Characteristics Design: time trend; Measures: age-adjusted incidence rate, age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1969-1977 Sample Characteristics 495 malignant melanoma cases (blacks 1; Anglos 447; Hispanics 42; American Indians 5) and 164 malignant melanoma deaths: New Mexico Variables Demographic: age, sex, race-ethnicity, menopausal status; Cancer: location Data Sources SEER cancer registry Abstract We assessed the occurrence of malignant melanoma of the skin in New Mexico from 1969-1977. Incidence data, collected by the New Mexico Tumor Registry, were supplemented with mortality data supplied by the State Bureau of Vital Statistics. These data were analyzed for variation by site, sex, and ethnic group. Malig- nant melanoma occurrence varied with ethnicity. Incidence rates for non-Hispanic whites (Anglos) exceeded comparison US rates, and were approximately six times higher than for other ethnic groups. Annual incidence rates for Hispanics, American Indians, and blacks of both sexes ranged from 0.0-1.8 cases per 100,000. As anticipated from other studies, the lower extremities were the most common site in Anglo women, and the trunk was the most common site in Anglo men. In contrast, the trunk was the most common site for both Hispanic men and women. A statistically significant trend of increasing incidence was demonstrated only for the Anglo women. Mortality rates varied widely during the study period and did not correlate with incidence rates. Pegoraro, R.].; Joubert, S.M. Letters to the Editor: Cytosolic Oestrogen Receptor Content of Breast Cancer Tissue in Blacks and Whites S Afr Med J 60(6):226, 1981 234 Subjects Biologic Marker: steroid receptors; Diagnosis; Cancer Site: breast Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 99 breast cancer cases (61 blacks; 38 whites): Durban, South Africa Variables Demographic: race-ethnicity Data Sources Laboratory test Abstract In breast tumors which were examined for nuclear estrogen receptors in addition to cytoplasmic receptors, there was found to be no significant difference in the incidence of positive receptors between Blacks and Whites. There is evidence that patients whose tumors contain both cytoplasmic and nuclear oestrogen receptors have a better prognosis and a higher response rate to hormonal therapy. Pegoraro, R.]J.; Nirmul, D.; Bryer, J.V.; et al. Clinical Patterns of Presentation of Breast Cancer in Women of Different Racial Groups in South Africa S Afr Med J 68:808-810, 1985 235 Subjects Risk Factor Exposure: endogenous hormones; Cancer Site: breast; Cancer Measures: prevalence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1975-1983 Sample Characteristics 640 breast cancer cases (blacks 252; whites 197; Indians 168; coloreds 23): female; Durban, South Africa Variables Demographic: age, race-ethnicity, menopausal status; Cancer: histology, stage, nodal involvement, grade, tumor size Data Sources Hospital medical records 97 Abstract The ages at presentation of white patients with breast cancer were found to be significantly higher than those of blacks, Indians and coloureds; 73% of white women fell into the postmenopausal group, in marked contrast to only 35% of Indians, while blacks and coloureds had similar proportions of pre- and post- menopausal patients. A significantly higher incidence of poorly differentiated tumours was seen in Indian and black patients. Blacks showed a significant tendency to present with more ad- vanced disease, while whites were generally diagnosed at a much earlier stage. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Peters, R.K.; Mack, T.M. Patterns of Anal Carcinoma by Gender and Marital Status in Los Angeles County Br J] Cancer 48:629-636, 1983 236 Subjects Risk Factor Exposure: sexual practices, other diseases; Cancer Site: rectum; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: relative risk, age-specific incidence rate, age-adjusted incidence rate, descriptive statistics, higher level statistics; Time Frame: 1972-1981 Sample Characteristics 970 anus cancer cases: Los Angeles, CA Variables SES: census tract SES rank; Demographic: age, sex, marital status, race-ethnicity (black, white, Hispanic), histology, stage, tumor location Data Sources non-SEER population-based cancer registry Abstract Marital status and other characteristics of 970 residents of Los Angeles County in whom cancer of the anus (including perianal skin) was diagnosed during the period 1972-1981 were compared with those of all county residents and all other persons in whom cancer was diagnosed during the same period. The incidence rate of anal cancer for single males was 6.1 times that for married males (p<0.001). This excess was limited to squamous and transitional cell carcinomas and was reasonably consistent by age, stage, sub- site, social class and race. Single women were not at increased risk, but separated and divorced persons of both sexes were at increased risk compared to married persons. Anal cancers were more common in males under the age of 35, after which there was a substantial female predominance. This relative excess in older women occurred at all stages, subsites, and social classes of whites and also in blacks, but not in Hispanics, among whom women had lower overall incidence rates compared to both whites and blacks. The findings were consistent with the hypothesis that sex- ual activity involving the anus is related to anal cancer. We could not rule out the possibility that anal cancer is related to the acquired immune-deficiency syndrome (AIDS) since the incidence in young single men appears to have increased in 1980 and 1981. Peters, R.L.; Afroudakis, A.P.; Tatter, D. The Changing Incidence of Association of Hepatitis B with Hepatocellular Carcinoma in California Am ] Clin Pathol 68(1):1-7, 1977 237 Subjects Biologic Marker: hepatitis B surface antigens (HBsAg); Risk Factor Exposure: other diseases; Cancer Site: liver; Cancer Measures: incidence Design: time trend; Measures: crude incidence rate, descriptive statistics; Time Frame: 1949-1974 Sample Characteristics 168 liver cancer deaths: Los Angeles, CA Data Sources Autopsy records, laboratory test Abstract Two hundred seventy-nine patients who died of hepatocellular carcinoma were autopsied at Los Angeles County-USC Medical Center and the John Wesley-USC Liver Unit from 1949 through 1974, and tissues 168 of these cases were available for staining for hepatitis B surface antigen (HBsAg). Twenty-one per cent of the livers had stainable HBsAg. There were prominent increases both in total numbers of hepatic cancers and in the percentages that were HB Ag-positive beginning about 1970, but the numbers of hepatocellular carcinomas arising in noncirrhotic livers also in- creased. From 1969 to 1974, 73% of those who had hepatocellular carcinomas arising in nonalcoholic but cirrhotic livers were HBs Ag-positive. Racial differences in the incidences of cirrhosis, the incidences of hepatocellular carcinomas, and the percentages of hepatocellular carcinomas associated with HGsAg were found. the incidences of cirrhosis were: Caucasian 11%; Mexican 12.2%; Negro 5.7%; Oriental 10%. Hepatocellular carcinomas arose in 3.2% of Caucasians who had cirrhosis; 3.6% of Mexicans; 8.3% of Negroes; 47% of Orientals. Ten per cent of Caucasians who had hepatocel- lular carcinomas in cirrhotic livers were HBsAg-positive; 25% of Negroes; 12% of Mexicans; 47% of Orientals. 98 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Petitti, D.B.; Friedman, G.D. Cardiovascular and Other Diseases in Smokers of Low Yield Cigarettes J Chron Dis 38(7):581-588, 1985 238 Subjects Risk Factor Exposure: tobacco; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: prospective; Measures: relative risk, age-adjusted incidence rate, descriptive statistics, higher level statistics; Time Frame: 1979-1982 Sample Characteristics 58,383 individuals (current cigarette smokers 16,270; never used tobacco products 42,113): Kaiser-Permanente Medical Care Program members; Oakland-San Francisco, CA Variables Demographic: age, sex, race-ethnicity (black, non-black) Data Sources Health maintenance organization medical records, hospital medical records, survey Abstract The association of smoking low yield cigarettes with the risk of cancer, cardiovascular disease, and other disease associated previously with smoking is controversial. In 1979 we began a pros- pective epidemiologic study of this subject. We here report on the first 4 years of follow-up in the 16,270 current, regular cigarette smokers and the 42,113 subjects who never used any form of tobacco enrolled in the study. In multivariate analyses that in- cluded age, sex, race, number of cigarettes smoked per day and other factors related to cardiovascular disease, the risk of cardiovas- cular diseases was consistently higher in smokers of higher than in smokers of lower yield cigarettes, although the magnitude of the difference in risk was very small. The risks of cancer of the trachea, broncus and lung, of all smoking-related cancers as a group, of diseases of the respiratory system, and of peptic ulcer diseases were not significantly associated with yield in smokers. The incidence rates of cardiovascular diseases considered as a group, cancer of the trachea, bronchus, and lung and all smoking- related cancers were higher in smokers of low yield cigarettes than in never users of any form of tobacco. We conclude that the smok- ing of low yield cigarettes is not without associated hazard. On the other hand, the results suggest that smokers who cannot quit might best use the least number of the lowest yield cigarette. Petrakis, N.L. Some Preliminary Observations on the Influence of Genetic Admixture on Cancer Incidence in American Negroes Int J Cancer 7:256-258, 1971 239 Subjects Biologic Marker: Duffy gene; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, higher level statistics Sample Characteristics Cancer incidence data: whites; United States and Negroes; Alameda County, CA; Atlanta, GA; Detroit, MI; and Ibadan, Nigeria Variables Demographic: geographic area Data Sources Survey, non-SEER population-based cancer registy Abstract Despite the relatively small amount of data, a consistent associ- ation was found between the amount of genetic admixture and the incidence of breast cancer in Negro women. Preliminary find- ings suggest that susceptibility to breast cancer in Negro women is related to the extent of Caucasian genetic contribution to the Negro gene pool. 99 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Petrakis, N.L.; Emster, V.L.; Sachs, S.T.; et al. Epidemiology of Breast Fluid Secretion: Association with Breast Cancer Risk Factors and Cerumen Type J Natl Cancer Inst 67(2):277-284, 1981 240 Subjects Risk Factor Exposure: reproductive factors; Cancer Site: breast; Biologic Marker: breast fluid, cerumen type Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 3,929 individuals (blacks 455; whites 2,847; Chinese 329; Japanese 87; Latin Americans 154; Filipino 57): female, nonlactating; San Francisco-Oakland, CA Variables SES:income; Demographic: age, race-ethnicity, age at menarche, age at first pregnancy, menopausal status, parity Data Sources Survey, laboratory test Abstract The epidemiology of breast fluid secretion was studied on the basis of nipple aspirates of breast fluid obtained from 3,929 nonlac- tating women of various racial groups. The results confirmed and extended earlier findings by our group. Variation in the proportion of secretors was related to most breast cancer risk factors, including age, race, age at menarche, age at first pregnancy, age at menopause, clinically diagnosed fibrocystic disease, menopausal estrogen use, and cerumen phenotype. Secretory activity as mea- sured by nipple aspiration appeared to reflect hormonal and gene- tic effects on breast epithelium. Petrek, J.A.; Sandberg, W.A.; Bean, P.K. The Role of Gender and Other Factors in the Prognosis of Young Patients with Colorectal Cancer Cancer 56:952-955, 1985 241 Subjects Treatment; Cancer Site: colon-rectum; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: observed survival rate, higher level statistics; Time Frame: 1968-1978 Sample Characteristics 619 colorectal cancer cases (blacks 496; whites 123—Under 40 years old 50; 40+ years old 569): low SES; Atlanta, GA Variables Demographic: age, sex; Cancer: stage Data Sources Hospital medical records, non-SEER hospital cancer registry Abstract Fifty patients from a socioeconomically disadvantaged popula- tion who were diagnosed when younger than 40 with colorectal cancer between 1968 and 1978 were analyzed. These patients had an increased survival compared with their older counterparts aged 40 years and older who were diagnosed during the same time. The young women had significantly better survival than the young men. Advanced stages, distribution of primary sites, and precan- cerous conditions were not major factors. The fact that the younger patients’ cancers had a higher incidence of extracellular mucin production may have been counterbalanced by their receiving more extensive treatment. At the same time, cultural and social factors related to gender may have more to do with better survival than do factors evaluated in previous studies. In fact, because of the pervasive lack of male/female analysis, it is not known whether the survival difference due to gender found in this report is a universal tendency in young populations. Pillay, S.P.; Angorn, I.B.; Baker, L.W. Colorectal Carcinoma in Young Black Patients: A Report of Eight Cases J Surg Oncol 10:125-132, 1978 242 Subjects Diagnosis; Treatment; Risk Factor Exposure: other diseases; Cancer Site: colon-rectum; Cancer Measures: incidence, survival Study Characteristics Design: case-study; Measures: crude incidence rate; Time Frame: 3 years Sample Characteristics Eight colon-rectum cancer cases (males 5; females 3): black, 10-22 years old; South Africa Variables Demographic: age, sex; Cancer: histology, tumor location Data Sources Histopathology records, hospital medical records 100 Abstract Colorectal carcinoma in the black population of South Africa is very uncommon when compared with incidence among whites. Eight cases of colorectal carcinoma in black patients under 25 years of age are presented. None had any evidence of premalignant lesions of the colon. The prognosis of colorectal carcinoma in young adults is poor in most series and this group of patients was no exception. Reasons for this poor prognosis are mentioned. A more aggressive approach to the early detection and treatment of the lesion in young patients is essential. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Polednak, A.P. Breast Cancer in Black and White Women in New York State: Case Distribution and Incidence Rates by Clinical Stage at Diagnosis Cancer 58(3):807-815, 1986 243 Subjects Cancer Site: breast; Cancer Measures: incidence; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: age-specific incidence rate, descriptive statistics, higher level statistics; Time Frame: 1976-1981 Sample Characteristics 51,641 breast cancer cases (blacks 4,443; whites 47,198): female; New York Variables SES: county income; Demographic: age, race-ethnicity, marital status, geographic area; Cancer: stage Data Sources non-SEER population-based cancer registry Abstract Case distributions and incidence rates by clinical stage at diag- nosis were examined for 47,198 white and 4,443 black female breast cancer cases diagnosed among residents of New York State from 1976-1981 and reported to the population-based New York Cancer Registry. Proportions of cases diagnosed at “regional” and (espe- cially) “metastatic” cancers were slightly higher in blacks in some age groups (<60 years). The proportion of metastatic cancer dif- fered significantly by race for single, married, and widowed cases, with younger (<60 years) black single women showing the highest proportions. The ranking of counties by black-white differences in per capita income was significantly associated with rankings by black-white differences in proportion of metastatic cancers. Possi- ble explanations for this association, in terms of patient delay, and implications for cancer screening and medical care, were discussed. Polednak, A.P. Primary Bone Cancer Incidence in Black and White Residents of New York State Cancer 55(12):2883-2888, 1985 244 Subjects Cancer Site: bone; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, age-specific incidence rate, descriptive statistics, higher level statistics; Time Frame: 1975-1980 Sample Characteristics 639 bone cancer cases (blacks 72; whites 567): New York Variables Demographic: age, race-ethnicity; Cancer: histology, tumor location Data Sources non-SEER population-based cancer registry, U.S. Population Census Abstract Some descriptive epidemiologic characteristics of primary bone cancers were presented for black and white residents of New York 101 State (1975-1980) using data from the population-based New York State Cancer Registry. Average annual race- and age-specific inci- dence rates were calculated for 1975 to 1980 for three histologic types (i.e., osteosarcoma, Ewing's sarcoma, and chondrosarcoma). The significantly lower incidence of Ewing's sarcoma in blacks versus whites was confirmed, whereas lower rates for chondrosar- coma in blacks supported findings from the Surveillance, Epidemiology and End Results (SEER) Program. A higher rate of osteosarcoma in blacks versus whites in the <15-year age group, although not statistically significant, was consistent with findings from another population-based registry in the U.S. This difference was due to osteosarcoma of the leg, and could be related to racial differences in growth rates during childhood. Comparisons with data from Africa suggest certain similarities in patterns as well as some possible differences, which could provide general clues to etiology (i.e., genetic versus environmental factors). EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Pollack, E.S. Tracking Cancer Trends: Incidence and Survival Hosp Pract, Aug 1984, p 99-116 245 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, mortality, survival Study Characteristics Design: time trend, cross-sectional; Measures: age-adjusted incidence rate, age-adjusted mortality rate, relative survival rate; Time Frame: 1948-1980 Sample Characteristics Cancer incidence and mortality data: 4 common SNCS-TNCS- SEER Program areas plus Connecticut; cancer survival data from SEER Program areas Variables Demographic: sex, geographic area, race-ethnicity Data Sources SEER cancer registry, Second National Cancer Survey 1947-1948, Third National Cancer Survey 1969-1971, death registration Abstract The paper describes NCI's Surveillance, Epidemiology, and End Results (SEER) Program which has been collecting and analyz- ing data on cancer incidence and survival in the United States for more than a decade and reviews overall cancer patterns. Survival differences between blacks, whites, and other ethnic groups are described. Potter, ].F.; Dawkins, D.M.; Terry, P.; et al. Breast Cancer in Blacks, Asians and Whites in Birmingham Postgrad Med ] 59:661-663, 1983 246 Subjects Risk Factor Exposure: diet-nutritional factors, familial factors, endogenous hormones; Cancer Site: breast; Cancer Measures: incidence Study Characteristics Design: time trend, cross-sectional; Measures: age-adjusted incidence rate, descriptive statistics; Time Frame: 1970-1981 Sample Characteristics 883 breast cancer cases (blacks 30; Caucasians 815; Asians 17; others 21): female, 30+ years old; Birmingham, United Kingdom and 3,996 individuals attending an obstetric department during 1979: female; Birmingham, United Kingdom Variables Demographic: age, place of birth, parity; Cancer: stage, histology Data Sources Hospital medical records Abstract A retrospective study was mounted into the frequency of breast cancer in blacks, whites and Asians attending Dudley Road Hos- pital in the years 1970-1981 inclusive. This figure was compared with the ethnic distribution of the surrounding population and of admissions for all causes. Breast cancer was uncommon in both ethnic minorities compared with whites, but was commoner in blacks than in Asians. These trends are similar to the World Health Organization figures for Britain, Jamaica and India. Differences in parity and breast feeding do not explain the variability of breast cancer in this study. Quinn, D.; Selah, C. Malignant Melanoma of the Anus in a Negro: Report of a Case and Review of the Literature 247 Dis Colon Rectum 20(7):627-631, 1977 Subjects Treatment; Cancer Site: malignant melanoma Study Characteristics Design: case-study; Time Frame: 1974-1975 Sample Characteristics One malignant melanoma case: black, female, 68 years old Data Sources Hospital medical records Abstract The authors present a case report of malignant melanoma of the anus in a 68 year old black woman and review the medical literature on this disease. They conclude that melanoma of the anus is highly malignant and that treatment has little effect on prognosis. Nevertheless, they recommend surgical extirpation of the primary lesion in hope of obtaining the rare cure and to prevent obstruction. There is no evidence to support one form of therapy over another. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Quinn, R.J.M.; Scragg, ].N.; Rubidge, C.]J. Primary Intracranial Tumours in Black and Indian Children, 1960-1975 248 S Afr Med J 53:211-213, 1978 Subjects Cancer Site: brain; Cancer Measures: incidence, prevalence, mortality, survival Study Characteristics Design: retrospective; Measures: crude incidence rate, crude survival rate, prevalence rate; Time Frame: 1960-1975 Sample Characteristics 60 primary cerebral tumor cases (blacks 42; Indians 18): under 13 years old; South Africa Variables Demographic: age, sex, race-ethnicity; Cancer: histology, symptoms, tumor location Data Sources Hospital medical records Abstract This report of cerebral tumours in 60 children admitted to the medical wards of King Edward VIII Hospital, Durban, shows that cerebral tumour is the commonest solid neoplasm in both Black and Indian children. There is a significantly lower incidence of cerebral tumour in Black children. No difference was apparent in age, sex ratio, site or histological types in our racial groups com- pared with studies in White children from other parts of the world. Ragbeer, M.M.S.; Walrond, E.R.; Bateson, E.M. Kaposi's Sarcoma in the West Indians West Indian Med J] 19(1):54-64, 1970 249 Subjects Cancer Site: sarcoma; Cancer Measures: incidence Study Characteristics Design: case study; Time Frame: 1961-1969 Sample Characteristics Three Kaposi's sarcoma cases (Negroes 2; Caucasian 1): male; Jamaica Data Sources Hospital medical records Abstract This peculiar angiosarcoma affecting predominantly skin and subcutaneous tissues was first described by Moritz Kaposi in Vie- nna in 1872. Subsequently cases were described chiefly in peoples of the Mediterranean litoral and central Europe, mainly among Italians and Jews. Jojot and Laigret reported the first case in a Negro from the Cameroons in 1922, and several cases in other African Negroes were subsequently reported, establishing the fact that there was a fairly high incidence of the disease in native Africans. This high incidence in Negroes has not been confirmed in the American continent, where Kaposi's sarcoma has largely remained a disease of Jews and Italians. We wish to report the only three confirmed cases seen at the University of the West Indies, in Jamaica. The first case was a patient from Guyana, of Caucasian ancestry; the second and third cases occurred in a Jamaican and St. Kitian respectively, both of African origin, and are two of the relatively few cases of Kaposi's sarcoma recorded in Negroes in the Americas. Raju, G.C.; Adam, R.; Ghouralal, S. Primary Neoplasms of the Central Nervous System in Trinidad 250 Trop Geogr Med 36:189-191, 1984 Subjects Cancer Site: central nervous system; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: crude incidence rate, descriptive statistics; Time Frame: 1971-1982 Sample Characteristics 125 central nervous system cancer cases (Africans 61; Indians 38; mixed 22; Caucasians 4): Trinidad Variables Demographic: sex, race-ethnicity; Cancer: histology Abstract The incidence of primary neoplasms of the central nervous system in Trinidad is less than 1 case per 100,000 population per year. There is no significant difference in the sex and age distribu- tion of the cases, but a definite racial difference is seen in the frequency of different CNS neoplasms. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Rambo, V.B.; O’Brien, P.H.; Miller, M.C., III; et al. Carcinoma of the Esophagus J Surg Oncol 7:355-365, 1975 251 Subjects Treatment; Cancer Site: esophagus; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics Sample Characteristics 486 esophagus cancer cases (blacks 357; whites 129): South Carolina Variables SES: public-private patient status; Demographic: age, sex, race-ethnicity Data Sources non-SEER hospital cancer registry Abstract Life tables were computed from the Tumor Registry at the Medical University of South Carolina on 486 patients with histolog- ically proven squamous cell carcinoma of the esophagus. The 5-yr followup was available in only 394 patients, or 81% of the total sample. For purposes of comparison with published literature, the traditional crude 5-yr survival percentages in the life-table analyses figures are presented. The life-table calculation utilizes all follow- up information available, in that patients observed for less than 5 yrs still may contribute to the estimate of the true 5-yr survival rate in contrast to the past when all patients lost to follow-up were presumed to have died. In this method, patients not observed for the full time of the study are assumed to have had the same risk of dying as patients followed to the end of the fifth year experienced in their fifth-year analyses. The group with the most favorable prognosis in this series was that receiving preoperative irradiation. This operation. They were given 4,500 rads of supervoltage X-ray over a period of 3 weeks prior to operation. The differences in surgical or radiation techniques employed by various reporters make comparative data difficult, if not impossible, to evaluate. Supervoltage irradiation has proved itself in some hands to equal surgery. A randomized prospective study would seem appropriate to conclude the best means of managing epidermoid carcinoma of the esophagus. Rawls, W.E.; Tompkins, W.A.F.; Melnick, J.L. The Association of Herpesvirus Type 2 and Carcinoma of the Uterine Cervix Am ] Epidemiol 89(5):547-554, 1969 252 Subjects Risk Factor Exposure: other diseases; Cancer Site: cervix uteri; Biologic Marker: herpesvirus type 2 antibodies Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 491 individuals (Negroes 441; others 50—Invasive cervix uteri cancer cases 65; dysplasia and carcinoma in situ cases 44; carcinoma of other sites 22; prostitutes 43; matched controls 51; healthy persons 266): low-socioeconomic status; Houston, Variables Demographic: age; Cancer: stage Data Sources Histopathology records, laboratory test, survey Abstract Herpesvirus type 2 antibodies were detected in 78% of 41 pa- tients with invasive carcinoma of the cervix, and in 64% of 24 patients who had been successfully treated for the disease 4 to 20 years prior to taking the test serum. Type 2 antibodies were de- tected in 22% of matched control women, 24% of patients with cervical dysplasia and 35% of patients with carcinoma in situ. Antibodies to the virus were not found to be associated with tumors of other sites except possibly carcinoma of the vulva. A seroepidemiologic survey for herpesvirus type 2 antibodies was performed on 266 persons of different ages from the same lower socioeconomic population from which the patients came. Anti- bodies to the virus did not appear until adolescence, and the anti- bodies occurred more frequently among prostitutes (54%) than among the control population (22%). The data support the concept of a venereal mode of spread of herpesvirus type 2 and establish an association between this virus and carcinoma of the cervix; however, further studies are required to determine whether or not the association is a causal one. 104 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Reintgen, D.S.; McCarty, K.M.; Cox, E.; et al. Malignant Melanoma in Black American and White American Populations: A Comparative Review JAMA 248(15):1856-1859, 1982 253 Subjects Cancer Site: malignant melanoma; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics, higher level statistics Sample Characteristics 2,612 malignant melanoma cases (blacks 31; whites 2,581): Durham, NC Variables Demographic: race-ethnicity, sex, age; Cancer: histology, stage, tumor location Data Sources Hospital medical records, histopathology records Abstract Malignant melanoma, a type of cancer that accounts for 1% to 3% of all malignant neoplasms is 20 times more frequent in the American white than black population. During a computer-aided retrospective chart review of more than 2,500 patients with melanoma being followed up at the Duke University Comprehen- sive Cancer Center, 31 blacks have been identified. Primary lesions of the foot were predominant in blacks with melanoma, and a high percentage of these were classified pathologically as acral lentiginous primary lesions. Black patients had a more advanced stage of disease at first presentation and a more deeply invasive primary lesion than their white counterparts. Five-year survival for the total black population was 23%. Blacks had a significantly worse prognosis than the white population when a comparison with the total group was made that was controlled for sex, age, site of primary lesion, stage of disease at presentation, and Clark level of primary melanoma lesion. This emphasizes the aggressive nature of this disease in the American black. Rickard, V.D.; Sampson, C.C. Incidence of Bronchogenic Carcinoma in Negroes: Statistical Analysis of 94 Cases J Natl Med Assoc 63:10-12, 1971 254 Subjects Risk Factor Exposure: tobacco; Cancer Site: lung; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1960-1970 Sample Characteristics 94 bronchogenic carcinoma cases (males 75; females 19): Negro, hospital patients; Washington, DC Variables Demographic: age, sex; Cancer: histology Data Sources Hospital medical records Abstract An analysis of bronchogenic carcinoma of 94 Negro patients is presented. The incidence and relationship of smoking to bron- chogenic carcinoma in Negroes appear to be the same as in other racial groups. Rippey, J.J.; Rippey, E. Epidemiology of Malignant Melanoma of the Skin in South Africa S Afr Med J 65:595-598, 1984 255 Subjects Risk Factor Exposure: solar radiation; Cancer Site: malignant melanoma; Cancer Measures: incidence, survival Study Characteristics Design: cross-sectional, prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics; Time Frame: 1959-1970 Sample Characteristics 746 malignant melanoma cases (blacks 318; whites 428): South Africa Variables Demographic: age, sex, race-ethnicity; Cancer: tumor location, tumor size Data Sources Histopathology records, hospital medical records Abstract Malignant melanoma constitutes about 2% of all cancers occur- 105 ring in South Africa. It is between 2.5 and 6 times more frequent in Whites than in Blacks, but the lowest incidence occurs in the Coloured population. The incidence of malignant melanoma in Whites (as reflected by mortality figures) varies with latitude, being higher nearer the Equator. The disease is more common among females and the older age groups in all sections of the population. Whites present with small lesions (generally less than 20 mm in diameter) which may be situated on any part of the body. In Blacks the tumours are large (sometimes more than 100 mm in diameter) when first seen and are predominantly situated on the sole of the foot and the palm of the hand. The 5-year survival rate is 20% in Blacks and 42% in Whites. The survival rates are more favourable in women. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Rippey, ].].; Rippey, E. Malignant Melanoma with Adjacent Hutchinson's Melanotic Freckle in Black Africans 256 Pathology 9:105-109, 1977 Subjects Abstract Risk Factor Exposure: other diseases; Cancer Site: malignant Hutchinson's melanotic freckle (lentigo maligna) is a well- melanoma; Cancer Measures: survival known pigmented lesion, usually seen on the face of white pa- Study Characteristics tients. It may be associated with invasive malignant melanoma. Design: prospective (nonconcurrent); Measures: crude survival This paper reports the incidence of this lesion in association with rate; Time Frame: 1959-1970 invasive malignant melanomas of the feet and hands of Black Sample Characteristics Africans. In Blacks, as in Whites, malignant melanoma with adja- 100 malignant melanoma cases (clinical and histological cent Hutchinson's melanotic freckle carries a considerably better information 40; histological information only 60): black; South prognosis than other histogenetic patterns of malignant Africa melonoma. Variables Cancer: histology, tumor location, tumor size Data Sources Hospital medical records Robertson, M.A.; Harington, ].S.; Bradshaw, E. The Cancer Pattern in African Gold Miners 257 Br J Cancer 25(3):395-402, 1971 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; mines of South Africa has been studied for the period 1964-68. Cancer Measures: incidence Considering the degree of selection to which they are subjected, Study Characteristics the crude cancer rate was unexpectedly high. The most common Design: cross-sectional; Measures: crude incidence rate, cancers were those of the liver, the oesophagus, the respiratory descriptive statistics; Time Frame: 1964-1968 system and the bladder. Geographical and tribal analysis showed Sample Characteristics that both liver and bladder cancers were predominantly found in 923 cancer cases: black, males, 15+ years old, gold miners; Africans from Mozambique, while most of the oesophageal cancer South Africa occurred in Xhosas from the Transkei. The highest rate for cancer Variables of the respiratory system was found in Africans from Natal, pre- Demographic: age, geographic area dominantly Zulu. The findings of this survey confirm those of Data Sources previous South African surveys. The differences in cancer inci- Industry records, histopathology records dence are linked to both geographical area (physical environment), Abstract and to tribe, which may mean an association with tribal habit and The incidence of cancer among the African workers on the gold custom. These factors need futher investigation. Rose, E.F.; McGlashan N.D. The Spatial Distribution of Oesophageal Carcinoma in the Transkei, South Africa 258 Br J Cancer 31:197-206, 1975 Subjects Abstract Cancer Site: esophagus; Cancer Measures: incidence Data on the incidence of cancer of the oesophagus in the Trans- Study Characteristics kei for years 1965-69 are presented, age specific rates for the sexes Design: cross-sectional; Measures: age-adjusted incidence rate, discussed and the spatial relationship of well-defined regions of descriptive statistics, higher level statistics; Time Frame: high and low incidence demonstrated. 1965-1969 Sample Characteristics 1,995 esophagus cancer cases (males 1,122; females 873): Transkei, South Africa Variables Demographic: age, sex, geographic area Data Sources Population census, survey 106 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Ross, R.K.; McCurtis, ].W.; Henderson, B.E.; et al. Descriptive Epidemiology of Testicular and Prostatic Cancer in Los Angeles Br J] Cancer 39:284-292, 1979 259 Subjects Risk Factor Exposure: endogenous hormones; Cancer Site: prostate, testis; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: age-adjusted incidence rate, age-specific incidence rate, descriptive statistics, higher level statistics; Time Frame: 1972-1975 Sample Characteristics 6,626 prostate cancer cases (blacks 792; whites 5,357; Mexican Americans 477) and 427 testis cancer cases (blacks 11, whites 360, Mexican Americans 56): Los Angeles County, CA Variables SES: census tract SES rank, occupation, Demographic: age, race-ethnicity Data Sources non-SEER population-based cancer registry, hospital medical records, U.S. Population Census Abstract Data from the Los Angeles County Cancer Surveillance Prog- ram (CSP) from 1972 to 1975 were used to study the descriptive epidemiology of testicular cancer and prostatic cancer. The very high black/white ratio and late age peak of cancer of the prostate contrasted sharply with the very low ratio and early age peak of testicular cancer. However, both sites had higher rates among upper occupational and social class groupings. Available descrip- tive and analytical research suggests that the etiology of prostatic cancer is most probably related to hormonal influences rather than to a horizontally transmitted agent, while the etiology of testicular cancer is most probably related to endogenous or exogenous hor- monal influences in utero or in infancy, or to in utero exposure to other exogenous agents. Ross, R.K.; Paganini-Hill, A.; Henderson, B.E. The Etiology of Prostate Cancer: What Does the Epidemiology Suggest? Prostate 4:333-344, 1983 260 Subjects Risk Factor Exposure: diet-nutritional factors, occupation, tobacco, endogenous hormones, other diseases, sexual practices; Cancer Site: prostate; Cancer Measures: incidence, mortality Study Characteristics Design: cross-sectional, prospective (nonconcurrent); Measures: age-specific incidence rate, age-adjusted incidence rate, proportional incidence ratio (PIR); Time Frame: 1946 Sample Characteristics 16,090 prostate cancer cases (blacks 2,008; whites 12,653; Spanish surnamed 1,271; Chinese 48; Japanese 110): male; Los Angeles County, CA; 220 individuals (prostate cancer cases 110; controls 110): male; Los Angeles County, CA and 1,400 Catholic priests: Los Angeles, CA Variables SES: census tract SES rank; Demographic: age, race-ethnicity Data Sources non-SEER population-based cancer registry, survey, death registration Abstract The two most important demographic characteristics of prostate cancer in Los Angeles are the high rates among blacks, which are two times those among whites and four times those among Asians, and the rapid increase in rates with age after age 40. Despite the high rates among blacks, a birth cohort analysis indicates that mortality rates among black men born after 1900 have decreased. In this report, epidemiologic and experimental evidence support- ing each of three etiologic hypotheses—industrial exposure to cad- mium, sexual transmission by an infectious agent, and endocrine factors—are reviewed. Evidence from descriptive data in Los Angeles suggest that only a small portion of cases might be attribut- able to industrial exposures. In a cohort study of Catholic priests, we found no deficit of prostate cancer mortality, strong evidence against sexual transmission of the disease. Experimental evidence and a limited amount of human data support an endocrine hypothesis. Preliminary results of a case-control study of prostate cancer are presented, but these results are unable to distinguish among these hypotheses further. This study finds a substantial protective effect of vasectomy, an event that is accompanied by reduced prostatic function and size, but this result is thus far statistically insignificant. 107 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Rothschild, H.; Voors, A.W.; Weed, S.; et al. Trends in Respiratory System Cancer Mortality in Louisiana: Geographic Distributions in 1950-1969 and 1967-1976 Compared Am |] Public Health 69(4):380-381, 1979 261 Subjects Cancer Site: respiratory system; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: age-adjusted mortality rate; Time Frame: 1950-1976 Sample Characteristics Respiratory cancer mortality data from Louisiana's 64 parishes Variables Demographic: age, sex, race-ethnicity (nonwhite, white), geographic area Data Sources Death registration Abstract Mortality rates for 1967-1976 were determined to update parish- specific rates of mortality attributable to respiratory system cancers (RSC) in Louisiana, and to investigate clustering and the stability of geographic trends over time. Data from each of Louisiana's 64 parishes were obtained from death certificates during the period. When the average mortality from RSC in Louisiana for 1950-1969 was compared with that for 1967-1976, an increase in rate was noted for each of four race-sex groups; nonwhite-male, nonwhite- female, white-male and white-female. Although the age-adjusted rate for males is higher than that for females in both time periods, the rates for females increased more than those for their corres- ponding male groups. The rates increased more for white than for nonwhite groups. Rutledge, F.N.; Galakatos, A.E.; Wharton, J.T.; et al. Adenocarcinoma of the Uterine Cervix Am J Obstet Gynecol 122(2):236-245, 1975 262 Subjects Treatment; Risk Factor Exposure: reproductive factors; Cancer Site: cervix uteri; Cancer Measures: survival Study Characteristics Design: prospective; Measures: crude survival rate, descriptive statistics; Time Frame: 1947-1971 Sample Characteristics 219 cervix uteri cancer cases (Negroes 48; Caucasians 170; Mongolians 1): female, 18-84 years old; Houston, TX Variables Demographic: age, race-ethnicity, parity, menarche, menopausal status; Cancer: stage, symptoms, histology Data Sources Hospital medical records Abstract From January 1, 1947, through December 31, 1971, 219 patients with primary adenocarcinoma of the intact uterine cervix were treated at the M. D. Anderson Hospital and Tumor Institute. Two modes of therapy were primarily used, namely, irradiation therapy alone and irradiation therapy plus operation. The 5 year survival results are 83.7 per cent for patients with Stage I disease, 48.0 per cent for patients with Stage II disease, 28.2 per cent for patients with Stage III disease, and 0.0 per cent for patients with Stage IV disease. The group with irradiation plus operation had a better over-all survival rate. In addition, the incidence of central and pelvic recurrent disease was remarkably lower (fourfold differ- ence). The urologic and bowel complications are discussed. This review lends support for our practice of preoperative irradiation followed by simple (conservative) hysterectomy for selected pa- tients with adenocarcinoma of the uterine cervix.) Saftlas, A.F.; Satariano, W.A.; Swanson, G.M.; et al. Methods of Cancer Case Selection: Implications for Research 263 Am J Epidemiol 118(6):852-856, 1983 Subjects Diagnosis; Cancer Site: multiple sites; Cancer Measures: incidence; Methodology Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1973-1978 Sample Characteristics 63,301 cancer cases: metropolitan Detroit, MI Variables Demographic: age, sex, race-ethnicity (black, white) Data Sources SEER cancer registry Abstract The race, age, and sex differences between clinically and micro- 108 scopically diagnosed cancer cases for a total of 63,301 incident cases for 18 sites are reported. The cases, diagnosed during 1973- 1978, were drawn from the Cancer Surveillance Section in the Division of Epidemiology at the Michigan Cancer Foundation in Detroit, a participant in the Surveillance, Epidemiology, and End Results (SEER) program. Although microscopically diagnosed cases and those diagnosed by other methods differed by age for 15 of the 18 sites examined. Cases without microscopic diagnoses (i.e., clinical diagnoses) tended to be older than microscopically confirmed cases. These findings indicate that in studies of certain cancer sites, the method of case selection could affect results per- taining to age differences in exposure to and effects of relevant risk factors. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Salyers, A.A.; Sperry, J.F.; Wilkins, T.D.; et al. Neutral Steroid Concentrations in the Faeces of North American White and South African Black Populations at Different Risks for Cancer of the Colon S Afr Med J 51:823-827, 1977 264 Subjects Risk Factor Exposure: diet-nutritional factors; Cancer Site: colon Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 182 individuals (blacks 84; whites 98): 5-43 years old; Virginia, South Africa Data Sources Laboratory test Abstract To test the hypothesis that a high risk for cancer of the colon might be associated with high concentrations of neutral steroids in the colon and with breakdown of these compounds by intestinal bacteria, the faecal neutral steroid concentrations of 84 rural South African Black adults (a population at low risk for colonic cancer), and of 98 North American White adults (a population at high risk for colonic cancer) were compared. Not only were the concentra- tions of animal steroids higher in the faeces of the North Amer- icans, but the chemical state of their faecal steroids was different from that of the Blacks. A high proportion of plant and animal steroids in the faeces of the Blacks was esterified to long-chain fatty acids, whereas in the faeces of the North Americans, most of the neutral steroids were free (non-esterified). There was consid- erable variation in the extent of cholesterol metabolism by the intestinal bacteria in the North Americans. This was not the case with the South African population, which was much more homogeneous with respect to cholesterol metabolism. Sampson, C.C.; Braxton, E.L. Esophageal Carcinoma in Blacks: Analysis of 88 Necropsied Cases J Natl Med Assoc 68(4):278-280, 1976 265 Subjects Risk Factor Exposure: alcohol, tobacco; Cancer Site: esophagus Study Characteristics Design: time trend, cross-sectional; Measures: descriptive statistics; Time Frame: 1957-1975 Sample Characteristics 88 esophagus cancer deaths (males 71; females 17): black; Washington, DC Variables Demographic: age, sex; Cancer: histology, tumor location Data Sources Autopsy records Abstract Carcinoma of the esophagus in 88 necropsied black patients is presented. A significant number of the patients were tobacco users and alcohol drinkers, either singly or in combination. A clearly defined etiological agent could not be identified. The cause for the rate increases in esophageal carcinoma in blacks remains elusive. Sandler, D.P.; Sandler, R.S.; Horney, L.F. Primary Liver Cancer Mortality in the United States J Chron Dis 36(3):227-236, 1983 266 Subjects Risk Factor Exposure: other diseases; Cancer Site: liver; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: age-adjusted mortality rate, age-adjusted incidence rate; Time Frame: 1958-1975 Sample Characteristics Mortality data: United States Variables Demographic: age, sex, race-ethnicity (nonwhite, white), geographic area Data Sources U.S. Population Census, death registration Abstract We analyzed time trends and regional differences in mortality from primary liver cancer in the United States in order to explore possible effects of several factors which have been suggested as causes for nonwhite males among whom there has been a 45% increase in liver cancer mortality between 1958 and 1975. The rates for nonwhite males are twice those for white males, and geographic trends in liver cancer mortality differ by race. The rates for white males are greatest in two southern regions, whereas rates for non- white males in these same regions are lower than they are else- where. Cirrhosis mortality trends parallel those for liver cancer among nonwhite males but not among white males. The similarities in trends for cirrhosis and liver cancer mortality among nonwhite males suggest that cirrhosis may be a major risk factor in this group. For white males, we must took to some other factor to explain the geographic differences that were observed. 109 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Satariano, W.A.; Belle, S.H.; Swanson, G.M. The Severity of Breast Cancer at Diagnosis: A Comparison of Age and Extent of Disease in Black and White Women 267 Am ] Public Health 76(7):779-782, 1986 Subjects Diagnosis; Cancer Site: breast; Cancer Measures: incidence regional, and remote breast cancer between Black and White Study Characteristics women by age at diagnosis. The analysis is based on 17,361 cases Design: retrospective; Measures: age-specific incidence rate, of breast cancer, newly diagnosed between 1973 and 1982, in Black descriptive statistics, higher level statistics Time Frame: 1973- and White female residents of the Detroit metropolitan area. Cases 1982 with breast cancer classified as local, regional, or remote were Sample Characteristics drawn from the Metropolitan Detroit Cancer Surveillance System 17,361 breast cancer cases (blacks 2,605; whites 14,756): female; of the Division of Epidemiology at the Michigan Cancer Founda- Detroit, MI tion, a participant in the Surveillance, Epidemiology, and End Variables Results (SEER) program of the National Cancer Institute. The find- Demographic: age, race-ethnicity; Cancer: stage ings indicate that older women, particularly Black women, are at Data Sources elevated risk for being diagnosed with the most advanced form SEER cancer registry, hospital medical records of breast cancer. Abstract A comparison was made of the incidence rates of localized, Satariano, W.A.; Schwartz, A.G.; Swanson, G.M. Screening for Cervical Cancer: Results from Several Intervention Strategies 268 In: Issues in Cancer Screening and Communications. New York: Alan R. Liss, 1982, p 275-283 Subjects Screening: Pap smear; Cancer Site: cervix uteri reach clinics; and employee clinics. For each type of clinic, the Study Characteristics age-specific cancer detection rates (total number of cancers de- Design: cross-sectional; Intervention Study; Measures: prevalence tected in a specific race-age group/the total number of women rate, age-specific cervical cancer detection rates, descriptive screened in that race-age group X 1,000) were compared. This rate statistics; Time Frame: 1976-1979 can be used to provide an assessment of the relative effectiveness Sample Characteristics of each intervention modality for blacks and whites in a high-risk 34,135 individuals (blacks 16,755; whites 17,380): female, area. Moreover, it provides an assessment of the effectiveness of resident in districts of low socioeconomic status; Detroit, MI the modalities for each age group. This study indicates that selec- Variables tive screening in target populations is both feasible and effective. Demographic: age, race-ethnicity; Cancer: stage In the MDCCP study, the detection rates were higher for black Data Sources females than for white females. Furthermore, among black females SEER cancer registry screened, a higher proportion of cancers were diagnosed as in situ Providers than is observed in the annual incidence of cervical cancer among Outreach mobile screening unit, standing clinics, employee black females in Metropolitan Detroit. The outreach clinics were screening particularly effective among black females. In addition to screening Abstract the largest number of black females the outreach clinics accounted In this paper, the results of a mass screening program for cer- for the highest cancer detection rates. Their effectiveness varied vical cancer in the Detroit Metropolitan Area administered through by age at first screen- ing. The outreach clinics were particularly the Metropolitan Detroit Cancer Control Program (MDCCP) are efficient in screening black females in their 40s. While only 10.2 reported. For this program, a high-risk group was defined as those percent of black females screened in those clinics were in their women living in districts with the highest incidence of invasive 40s, the cancer detection rate was 7.27 per 1,000. The relatively disease. Incidence rates from the Michigan Cancer Foundation high detection rates for older females indicate clearly that cervical Registry, a member of the National Cancer Institute's SEER prog- cancer screening programs should not be limited to younger ram, were used to identify these districts. Residents of these dis- females. Finally, as was true of the standing and employee clinics, tricts were primarily black and of lower socioeconomic position. cancers were not detected in females under 20. This may suggest Three intervention strategies were utilized by the Metropolitan that screening programs should be limited to females 20 years and Detroit Cancer Control Program from June, 1976 through De- over. In conclusion, community outreach is an effective approach cember, 1979 to determine the most effective method of community for reaching high-risk black females at all ages from 20 to 80. screening for cervical cancer: standing, central clinics; mobile out- 110 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Savage, D.; Lindenbaum, J.; Van Ryzin, ].; et al. Race, Poverty, and Survival in Multiple Myeloma Cancer 54:255-264, 1984 269 Subjects Cancer Site: multiple myeloma; Cancer Measures: survival; Socioeconomic Factors Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1965-1980 Sample Characteristics 144 multiple myeloma cases (blacks 98; whites 46—Harlem Hospital cases 52; Columbia-Presbyterian Medical Center Controls 92): 41-97 years old; New York City, NY Variables SES: census block income, census block education; Demographic: age, sex, race-ethnicity Data Sources U.S. Population Census, hospital medical records Abstract To identify possible interracial differences in the behavior of multiple myeloma, the records of 52 black myeloma patients at Harlem Hospital Center (HHC) and 46 black and 46 white patients at Columbia-Presbyterian Medical Center (CPMC) were reviewed. In addition to clinical variables such as tumor burden, azotemia, and hypoalbuminemia, the effect of poverty on prognosis was examined, using socioeconomic indices from the United States census block group data of each patient. The median survival of CPMC black and white patients was comparable (34 and 29 months, respectively) whereas that of the HHC group was 12 months (Breslow test, p< 0.0001). Overcrowding and hypoal- buminemia were the most significant prognostic factors by mul- tivariate regression analysis on all 144 patients (p = 0.001); for HHC patients, overcrowding was the single significant variable affecting survival (p = 0.004). By all socioeconomic indices, HHC patients were more impoverished than CPMC patients (p< 0.001); they also presented with more advanced disease. Race is not a significant prognostic factor in myeloma, whereas the effect of socioeconomic status on survival appears to equal that of previ- ously described clinical features. Sawyer, D.R.; Mosadomi, A.; Page, D.G.; et al. Racial Predilection of Ameloblastoma? A Probable Answer from Lagos (Nigeria) and Richmond, Virginia (U.S.A.) J Oral Med 40(1):27-31, 1985 270 Subjects Diagnosis; Cancer Site: oral cavity; Cancer Measures: incidence Study Characteristics Design: retrospective; Measures: descriptive statistics; Time Frame: 1969-1979 Sample Characteristics 73 oral cavity cancer cases (American blacks 17; American whites 10; Nigerian blacks 46): 9+ years old; Nigeria, United States Variables Demographic: age, race-ethnicity; Cancer: histology, location, clinical diagnosis Data Sources Histopathology records Abstract Forty-six cases of ameloblastoma from a Nigerian population were compared and contrasted with seventeen cases in Black Americans and ten cases in White Americans. Non-major differ- ences were observed using a variety of clinical parameters. The most significant finding is the establishment by T and P values of probable differences in frequency of occurrence of the lesion be- tween Black and White populations. 111 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Saxinger, W.; Blattner, W.A.; Levine, P.H.; et al. Human T-Cell Leukemia Virus (HTLV-I) Antibodies in Africa Science 225:1473-1476, 1984 271 Subjects Cancer Site: leukemia, lymphoma, skin; Biologic Marker: HTLV-1 Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 1,890 individuals (Egyptians 101; Tunisians 278; Ghanians 746; Ugandans 86; Nigerians 9; South Africans 670): Africa Variables Cancer: histology Data Sources Laboratory test Abstract Antibodies specific for human T-cell leukemia-lymphoma virus type I (HTLV-I) were demonstrated in serum samples from various groups of people in South Africa, Uganda, Ghana, Nigeria, Tunisia, and Egypt. The samples had been collected for other purposes and were presumably selected without bias toward clin- ical conditions associated with HTLV infections. Regional differ- ences in antibody positivity were observed, indicating widely dis- tributed loci of occurrence of HTLV on the African continent in people of both black and white ancestry. Two patients with high titers of antibody to HTLV-I had some signs of adult T-cell leukemia-lymphoma. In several groups a high frequency of false positive serum reactions was indicated when specific confirmation steps were included in the assay. Further characterization of these sera revealed highly elevated immunoglobulin levels, possibly due to polyclonal activation of immunoglobulin synthesis in these sub- jects. The possibility that related cross-reactive human retroviruses coexist in the same groups was not eliminated. Schmaugz, R.; Jain, D.K. Geographical Variation of Carcinoma of the Penis in Uganda Br J Cancer 25(1):25-32, 1971 272 Subjects Prevention; Risk Factor Exposure: circumcision, geographic area; Cancer Site: penis; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-standardized incidence rate, age-specific incidence rate; Time Frame: 1964-1968 Sample Characteristics 458 penis cancer cases: black, male; Uganda Variables Demographic: age, race-ethnicity, geographic area; Cancer: histology Data Sources Histopathology records Abstract A series of 458 cases of carcinoma of the penis occurring in Ugandan Africans is analysed. These were derived from the re- cords of a country-wide biopsy service over the 5-year period 1964- 68. Where circumcision is practised the incidence of this tumour is very low. However, the geographical variation also showed marked differences in the uncircumcised, regardless of tribal an- tecedents and sometimes over quite small distances. It is suggested, therefore, that in Uganda other aetiological factors apart from circumcision are operative and that these factors vary with geographical location rather than with tribal affiliation. Schneiderman, M.A. Time Trends: United States 1953-1973 Laryngoscope 88(1pt2):44-49, 1978 273 Subjects Cancer Site: all sites combined, multiple sites; Cancer Measures: incidence, mortality Study Characteristics Design: time trend; Measures: crude incidence rate, crude mortality rate; Time Frame: 1935-1970 Sample Characteristics Cancer incidence and mortality data: United States Variables Demographic: sex, race-ethnicity (nonwhite, white) 112 Abstract The paper examines trends in cancer incidence and mortality by sex and race for the period 1935 - 1970. Increases in cancer of the lung, pharynx, larynx, and esophagus are thought to be related to increases in smoking. The increases in cancer are greatest for non-white (mostly black) men. In order to better understand the etiology of these diseases, the authors suggest that clinicians help gather occupational exposure and drinking behavior information from cancer patients. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Schoenberg, B.S.; Bailar, ]J.C., III; Fraumeni, J.F., Jr. Certain Mortality Patterns of Esophageal Cancer in the United States, 1930-67 J Natl Cancer Inst 46(1):63-73, 1971 274 Subjects Risk Factor Exposure: alcohol, tobacco, geographic area; Cancer Site: esophagus; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: relative risk, age-specific mortality rate, age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1930-1967 Sample Characteristics Esophagus cancer mortality data: United States Variables Demographic: age, sex, race-ethnicity (Negro, nonwhite, white, Chinese, Japanese), geographic area, urban-rural Data Sources Death registration Abstract Cancer of the esophagus shows substantial racial, geographic, and temporal variation within the United States. In a cohort analysis of mortality during 1930-67, the nonwhite population ex- perienced steadily and rapidly increasing rates, while those of the white population remained relatively stable. The rising nonwhite mortality occurred at all ages and all areas of the country and was more pronounced among males than females. Death rates in the Northeast were consistently higher than in other regions of the country, with the geographic differential being more prominent among nonwhites (threefold) than among whites (twofold). De- tailed analysis of the nonwhite statistics for 1950-66 showed that Negro males and females, together with Chinese and Japanese males, contributed to the excess in mortality. However, only Neg- roes had a steady increase in mortality over time. To study factors implicated in the origin of esophageal cancer, mortality rates by States for 1950-66 were compared with the following variables for each State: the percentage of population living in urban areas, per capita cigarette sales, and per capita alcohol sales. Mortality showed significant geographic correlations with each of these three variables, but the associations with cigarette and alcohol consump- tion seemed to result from some other, still unidentified, hazard in the urban environment. The striking racial variations in esophageal cancer suggest that Negroes have been increasingly exposed, and are more susceptible, to these carcinogenic influ- ences. Schulte, P.A.; Ringen, K.; Hemstreet, G.P.; et al. Risk Assessment of a Cohort Exposed to Aromatic Amines: Initial Results J Occup Med 27(2):115-121, 1985 275 Subjects Screening: bladder cytology; Risk Factor Exposure: occupation; Cancer Site: bladder; Cancer Measures: incidence, mortality, prevalence Study Characteristics Design: prospective (nonconcurrent); Measures: crude incidence rate, age-specific incidence rate, descriptive statistics, higher level statistics; Time Frame: 1940-1982 Sample Characteristics 1,385 individuals (black invasive cancer cases 6; white invasive cancer cases 5; black carcinoma in situ cases 2; bladder cancer free individuals 1,372): chemical company employees; Augusta, GA Variables Demographic: age, race-ethnicity; Cancer: stage Data Sources Industry records, survey, death registration Abstract A cohort of 1,385 workers potentially exposed to carcinogenic amines was evaluated to determine the extent of its risk for bladder cancer. The cumulative incidence of bladder cancer was deter- mined from death certificates, from interviews with community urologists, and from a screening program. A total of 13 confirmed cases of bladder cancer were identified at the conclusion of the first year of study. The entire cohort has approximately a fourfold excess risk of bladder cancer; however, black workers with more than 10 years of employment had a risk ratio of 111 (based on three cases). The onset of disease occurred, on the average, 15 years earlier in these black workers than in the general U.S. popu- lation. The cumulative incidence of bladder cancer increased with the duration of employment, ranging from 0.4% for workers with five or fewer years of employment to 36% for those with greater than 20 years. No significant differences were found between cases and noncases for cigarette smoking, coffee drinking, use of artificial sweetners, or prior employment in high-risk occupations. More cases of bladder cancer are expected in this cohort because many members have not yet achieved the average latency found for the confirmed cases. 113 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Schupbach, J.; Kalyanaraman, V.S.; Sarngadharan, M.G.; et al. Antibodies Against Three Purified Proteins of the Human Type C Retrovirus, Human T-Cell Leukemia-Lymphoma Virus, in Adult T-Cell Leukemia-Lymphoma Patients and Healthy Blacks from the Caribbean Cancer Res 43:886-891, 1983 276 Subjects Cancer Site: leukemia, lymphoma; Cancer Measures: incidence; Biologic Marker: anti-HTLV antibodies Study Characteristics Design: cross-sectional; Measures: descriptive statistics Time Frame: 1973-1982 Sample Characteristics 361 cases (normal blood donors 337; healthy family members 18; adult T-cell leukemia-lymphoma patients 6): black; Caribbean Data Sources Histopathology records Abstract Six black patients of Caribbean origin with adult T-cell leukemia-lymphoma, 18 of their healthy family members and rela- tives, and 337 healthy black individuals from the Caribbean were investigated for the presence of serum antibodies against human T-cell leukemia virus (HTLV). Three distinct structural proteins of this virus with molecular weights of 24,000, 19,000, and 15,000 were purified, radiolabeled, and used in radioimmune precipita- tion assays. Five of the patients, three of the family members (two of them spouses), and 11 of the normals had specific antibodies against at least the proteins with molecular weights of 24,000 and 19,000. High antibody titers against these proteins were often as- sociated with antibodies against the protein with a molecular weight of 15,000. In all cases, antibody titers against this protein were considerably lower than those against the proteins with molecular weights of 24,000 and 19,000. HTLV is highly associated with Caribbean adult T-cell leukemia-lymphoma and is also en- demic among the normal Caribbean population. By comparison of the frequencies of anti-HTLV positives among family members of patients and the normal population, we conclude that infection by HTLV occurs in a horizontal way and at least in the West Indian black and Japanese population probably without a requirement for uncommon genetic factors. Schwartz, M.R.; Randolph, R.L; Panko, W.B. Carcinoembryonic Antigen and Steroid Receptors in the Cytosol of Carcinoma of the Breast: Relationship to Pathologic and Clinical Features Cancer 55:2464-2471, 1985 277 Subjects Diagnosis; Risk Factor Exposure: reproductive factors; Cancer Site: breast; Biologic Marker: steroid receptors, carcinoembryonic antigen (CEA) Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1980-1981 Sample Characteristics 159 breast cancer cases (blacks 43; whites 106; Hispanics 10): female; Houston, TX Variables Demographic: age, race, menstrual status, parity; Cancer: histology, stage, metastases Data Sources Hospital medical records, laboratory test Abstract Cytosols of 159 primary breast carcinomas were assayed for 114 estrogen and progesterone receptors and 60 of these for carcinoem- bryonic antigen (CEA). The relationships of pathologic and clinical features previously shown to be of prognostic significance and the biochemical parameter measured were examined. It was found that degree of histologic differentiation was significantly related to receptor status. There were significant inverse relationships between estrogen receptor and lymphocytic infiltrate as well as tumor necrosis. Relationships between estrogen, but not proges- terone, receptor and patient age, menstrual status, and race existed. Cytosol CEA, on the other hand, related only to pathologic stage and not to histologic differentiation and other pathologic variables. Therefore, steroid receptors and CEA appear to provide information about different biologic characteristics of a carcinoma. Furthermore, the information provided by CEA appears to be in- dependent of many known pathologic, clinical, and biochemical prognostic indicators. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Segal, I.; Cooke, S.A.R.; Hamilton, D.G.; et al. Polyps and Colorectal Cancer in South African Blacks Gut 22:653-657, 1981 278 Subjects Risk Factor Exposure: diet-nutritional factors, occupation, Cancer Site: colon-rectum; Cancer Measures: incidence Study Characteristics Design: retrospective; Measures: crude incidence rate, crude mortality rate, descriptive statistics; Time Frame: 1969-1978 Sample Characteristics 328 colon-rectum cancer cases (blacks 205; whites 123): Johannesburg, South Africa Variables SES: occupation; Demographic: age, sex, race-ethnicity; Cancer: tumor location, clinical presentation Data Sources Hospital medical records, physician medical records, survey Abstract This study reflects the fact that cancer of the large bowel is uncommon in South African Blacks, and that colorectal polyps do not appear to constitute a precursor to most colorectal cancers. Furthermore, it is shown that dietary factors associated with this cancer in Western populations are not evident in the Black popu- lation. In the South African White population, however, the dis- ease behaves in a similar way to that observed in Western coun- tries. Other definitive differences found were the absence of mul- tiple synchronous cancers and diverticular disease in the Blacks with colorectal cancer. It is thus postulated that dietary factors are absent, or have not been present for a sufficient length of time to influence the development of polyps or polyp-cancer sequence in this population. It is also possible that the adenoma-carcinoma progression observed in Western countries may not be relevant to the development of all colorectal carcinomas in communities such as those reported here. Seggie, ].; Gelfand, M. Serum Alpha-Feto Protein (alpha-FP) and Hepatoma in Rhodesian Africans Trans R Soc Trop Med Hyg 69(2):209-211, 1975 279 Subjects Biologic Marker: alpha-fetoprotein; Cancer Site: liver Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1972-1973 Sample Characteristics 30 liver cancer cases: black, 15-70 years old; Rhodesia Variables Demographic: age, sex Data Sources Hospital medical records, laboratory test Abstract Alpha-fetoprotein is now accepted as a very useful serum marker of hepatocarcinoma, and since its discovery in 1963, sur- veys have been carried out in various regions of the world, where hepatoma is particularly prevalent, which have shown that alpha- fetoprotein may be detected in 50 to 80% of patients with liver cancer. The sera of 30 Rhodesian African patients with histologi- cally proven hepatoma were tested in this study for the presence of alpha-fetoprotein. In only 47 percent was alpha-fetoprotein de- tected. Thus it is our experience that a positive result is strong evidence in favour of a diagnosis of hepatocarcinoma whilst a negative result is of little value, necessitating that liver biopsy be carried out in all patients with a clinical suspicion of liver cancer. Seidman, H. Cancer Death Rates by Site and Sex for Religious and Socioeconomic Groups in New York City Environ Res 3:234-250, 1970 280 Subjects Cancer Site: multiple sites; Cancer Measures: mortality; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: age-adjusted mortality rate, standardized mortality ratio (SMR); Time Frame: 1949-1951 Sample Characteristics 24,083 cancer deaths (Negroes 2,056; whites 21,629; Puerto Ricans 294; Chinese and others 104): New York City, NY Variables SES: census tract SES rank, religion; Demographic: sex, race-ethnicity Data Sources Death registration, U.S. Population Census 115 Abstract Age adjusted death rates by sex are presented for the various sites of cancer for subgroups of the 25-64 year old non-Puerto Rican white population of New York City, 1949-1951, classified jointly by religious group and socioeconomic status. Rates by sex and site of cancer are also given for those of Puerto Rican descent, non-Puerto Rican Negroes, and for the other nonwhites. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Selzer, G.; Kahn, L.B.; Albertyn, L. Primary Malignant Tumors of the Thyroid Gland: A Clinicopathologic Study of 254 Cases Cancer 40:1501-1510, 1977 281 Subjects Cancer Site: thyroid; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics; Time Frame: 1952-1975 Sample Characteristics 254 thyroid cancer cases: South Africa Variables Demographic: age, sex, race-ethnicity (coloreds, whites); Cancer: histology, metastases, clinical presentation Data Sources Histopathology records Abstract Of 254 primary tumors of the thyroid gland diagnosed in the pathology department of the University of Cape Town, 212 (83.5%) were differentiated carcinomas. In this differentiated group, tumors composed exclusively of papillary structures exhibited the most favorable prognosis; local recurrences and/or spread were noted in tumors with a mixed papillary and follicular pattern and, more commonly, in those with a pure follicular pattern. A chronic lymphocytic thyroiditis occurred in associaion with 21 tumors (8.3%) but an associated multiodular goitre was found in only nine patients (3.5%). In addition to the well-known preponderance of thyroid carcinoma in women, a majority of lesions occurred in white patients in this series despite a higher hospital admission rate for coloreds (mulattos). Sengupta, B.S. Carcinoma of the Vulva in Jamaican Women Acta Obstet Gynecol Scand 60:537-544, 1981 282 Subjects Diagnosis; Treatment; Risk Factor Exposure: other diseases, reproductive factors; Cancer Site: vulva; Cancer Measures: incidence Study Characteristics Design: time trend, retrospective; Measures: crude incidence rate, crude survival rate, descriptive statistics; Time Frame: 1958-1975 Sample Characteristics 198 vulva cancer cases (Period 1: 98; Period 2: 100): female; Jamaica and 119 vulva cancer cases (blacks 109; others 10): female, hospital patients; Kingston, Jamaica Variables SES: social class; Demographic: age, parity; Cancer: histology, nodal involvement, tumor location, stage Data Sources non-SEER cancer registry, histopathology records, hospital medical records Abstract Clinical and pathological analysis of vulvar carcinoma in 119 Jamaican women, treated at the University Hospital of the West Indies, over the past 18 years stresses that the tumor was more frequently seen in women under the age of 50 — contrary to the experience of authors from the Western metropolitan countries. The natural history of the tumor showed that in 51% of cases, vulvar carcinoma co-existed or followed previous chronic vulvar granulomatous disease and showed a rapid tumor progression in these post-granulomatous cases. Features of other epidemiological and clinical importance are discussed. Special reference has been made concerning the problem of applying FIGO staging of vulvar carcinoma. Sengupta, B.S.; Chatterjee, D.; Persaud, V.; et al. Primary Neurological Manifestation of Choriocarcinoma Int Surg 61(2):88-94, 1976 283 Subjects Diagnosis; Treatment; Cancer Site: choriocarcinoma; Cancer Measures: incidence Study Characteristics Design: case study; Time Frame: 1958-1973 Sample Characteristics Seven choriocarcinoma cancer cases (Negroes 6; Chinese 1): female, hospital patients; Jamaica Abstract This article describes seven of the 26 cases of choriocarcinoma, treated in the University Hospital, Jamaica, from 1958 to 1973. 116 These cases presented primarily with neurological features. The authors believe that such an incidence of choriocarcinoma is com- mon in the Negro race and presents a special diagnostic problem. A high index of suspicion is necessary to diagnose this condition, especially in young women. The prognosis is not always poor and seems to depend on the host-tumor relationship regardless of widespread metastases. When this host defense is good, chemotherapy may be curative. Further knowledge of host-tumor behavior is necessary to predict the disease pattern and perfect the treatment. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Sheffet, A.; Thind, I.; Miller, A.M.; et al. Cancer Mortality in a Pigment Plant Utilizing Lead and Zinc Chromates Arch Environ Health 37(1):44-52, 1982 284 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: prospective; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1940-1969 Sample Characteristics 1,946 individuals (nonwhites 650; whites 1,296 and Black cohort deaths 110; white cohort deaths 211): pigment plant workers; Newark, NJ Data Sources Industry records, death registry Abstract Several studies of workers exposed to various forms of chromium compounds have suggested an increased incidence of respiratory cancers. Lead and zinc chromates were among the chromium compounds implicated. The Department of Preventive Medicine and Community Health of the New Jersey Medical School undertook a detailed mortality study of a pigment plant in Newark which utilized both of these compounds. We compared observed deaths from each cause among 1296 white and 650 non-white males who were employed at the plant between January 1, 1940 and December 31, 1969, with expected deaths, as computed from cause-, age-, and time-specific standard death rates for the United States. A statistically significant relative risk of 1.6 for lung cancer among white male employees was found as well as among the cohorts of white males employed 10 yr or more. A relative risk of 1.9 was noted for individuals employed at least 2 yr who were at least moderately exposed to chromates. An increased incidence of lung cancer among nonwhite males and stomach and pancreatic cancer among the total cohort was also evident. These data support the validity of the association between chromate pigment exposure and lung cancer, suggesting that pigment workers and others ex- posed to chromates be observed carefully in regard to possible risk of pancreatic and stomach cancers. Simone, ]J.V.; Verzosa, M.S.; Rudy, J.A. Initial Features and Prognosis in 363 Children with Acute Lymphocytic Leukemia Cancer 36:2099-2108, 1975 285 Subjects Treatment; Cancer Site: leukemia; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1962-1971 Sample Characteristics 363 leukemia cases (Negroes 42; whites 321): under 21 years old; Memphis, TN Variables Demographic: age, sex, race-ethnicity Data Sources Hospital medical records Abstract The relationship of a variety of initial features and the outcome of therapy was analyzed for 363 children with acute lymphocytic leukemia (ALL). All had entered “total therapy” studies between 1962 and 1971. The standard for comparing outcome of therapy was whether patients with a given feature attained or exceeded 117 the median duration of complete remission, hematologic remisson or survival for the group. The results showed that, in general, the more massive or extensive the disease at diagnosis, the poorer the outcome. Factors associated with a significantly poorer prognosis included: initial leukocyte count above 100,000/mma3; spleen en- largement greater than 5 cm; mediastinal involvement and early CNS involvement. Children over 10 years old at diagnosis and Negro children also had a poorer prognosis. From another view- point features were examined for patients who attained at least 3 years of continuous complete remission. This confirmed some ear- lier findings and, in addition, showed that children under 2 years of age at diagnosis or with hepatomegaly over 5 cm were less likely to attain this goal. With the exception of early CNS involve- ment, however, patients with excellent responses to therapy were found with each factor of poor prognosis. Two major factors were not analyzed because their relationship to prognosis is generally accepted: therapeutic differences and acute nonlymphocytic leukemia. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Sizaret, P.; Tuyns, A.; Martel, N.; et al. Alpha-Fetoprotein Levels in Normal Males From Seven Ethnic Groups with Different Hepatocellular Carcinoma Risks 286 Ann NY Acad Sci 259:136-155, 1975 Subjects Cancer Site: liver; Biologic Marker: alpha-fetoprotein age-groups. In addition, there was a systematic age-dependency Study Characteristics of AFP levels particularly evident in the groups from Singapore- Design: cross-sectional; Measures: descriptive statistics, higher Lyon, in which there was a 50% AFP increase between the ages level statistics of 20 and 40. Comparison between Africans on the one hand and Sample Characteristics people from Singapore-Lyon on the other hand revealed highly 1,335 individuals (blacks 616; caucasians 174; Malays 199, significant differences (p<0.001), especially in the younger groups, Chinese 197; Indians 149): male; 15 + years old, blood donors; whereas Chinese, Malays, and Indians from Singapore had very Singapore; Nairobi, Kenya; Ivory Coast; Lyon, France similar AFP patterns; this suggests an important role for environ- Variables mental factors in the regulation of AFP levels. The age dependency Demographic: age, geographic area of the presumed effect of environmental factors is in keeping with Abstract experimental data showing that young animals respond more vig- Alpha-fetoprotein (AFP) levels of 1,335 males (15 years and orously to AFP-stimulating factors. Although the incidence of older) of seven ethnic groups (Chinese, Indians, and Malays from hepatocellular carcinoma (HCC) differs in the three Singapore Singapore, Caucasians from Lyon, and Blacks from Nairobi, forest, groups (the highest in Chinese and the lowest in Indians), no and the savanna region of the Ivory Coast) were determined by relationship was observed in this study between mean AFP level radioimmunoassay. A few elevated levels (up to 30 nanounits/ml) and HCC incidence in Singapore. were detected in some normal individuals, especially in the older Skinner, R.L.; Davenport, W.D., Jr.; Weir, J.C.; et al. Oral Malignancy: A Regional Incidence Observed Over a 13-Year Period 287 South Med ] 78(6):652-656, 1985 Subjects Abstract Cancer Site: oral cavity; Cancer Measures: incidence In this study slightly more than two out of every 100 biopsy Study Characteristics specimens from the set population 10,414 specimens showed Design: cross-sectional; Measures: descriptive statistics; Time malignant oral lesions. The most common sites for oral malignancy Frame: 1969-1982 were the floor of the mouth and the alveolar ridge, respectively. Sample Characteristics Lesions were most common during the fifth decade of life, followed 252 oral cancer cases (blacks 28; whites 209): 19+ years old; by the sixth. Eighty-eight percent of the malignant lesions occurred southeast Louisiana in whites and 12% in blacks. This does not deviate significantly Variables from the percentage in the total biopsies reported. Of the categories Demographic: age, sex, race-ethnicity; Cancer: histology, tumor of neoplasms, surface ectodermal neoplasms predominated, fol- location lowed by salivary and mesenchymal lesions, respectively. There Data Sources did not appear to be a significant difference in sex incidence. Histopathology records Smith, D.B.; Jenkins, R.S. Cervical Cancer in a Housing Project: Report of a Cluster of Cases 288 Am ] Obstet Gynecol 104(4):603-605, 1969 Subjects Abstract Risk Factor Exposure: sexual practices; Cancer Site: cervix uteri; This paper reports on the finding of a cluster of cervical cancer Cancer Measures: incidence cases in a single housing project containing predominately black Study Characteristics women. The authors suggest that an oncogenic agent which is Design: retrospective; Measures: crude incidence rate, spread by sexual contact may be present among these women. descriptive statistics; Tine Frame: 1965-1966 Sample Characteristics 199 cervix uteri cancer cases (Perry Homes housing project residents 10; others 189): Negro, female; Fulton County, GA Variables Demographic: geographic area; Cancer: stage Data Sources Hospital medical records, survey 118 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Smith, E.B. Carcinoma of the Esophagus in Blacks J Natl Med Assoc 77(1):44-48, 1985 289 Subjects Diagnosis; Treatment; Cancer Site: esophagus; Cancer Measures: incidence, survival Study Characteristics Design: retrospective; Measures: crude incidence rate, crude survival rate, descriptive statistics; Time Frame: 1957-1982 Sample Characteristics 721 esophageal cancer cases (blacks 114; whites 607): Pittsburgh, PA Variables SES: education; Demographic: age, sex, race-ethnicity; Cancer: histology, stage Data Sources Hospital medical records Abstract The incidence of esophageal carcinoma in the United States in white males is 6 per 100,000, compared with 20 per 100,000 in black males. The average 5-year survival for both groups is 5 per- cent. This study is a comprehensive analysis of esophageal car- cinoma in blacks in the Pittsburgh area from 1957 to 1982. Family history, sex, age, ethnic background, social history, occupation, educational level, symptoms and signs, diagnosis, histopathology, cancer staging, treatment, and survival are reviewed. This informa- tion may expand our knowledge of esophageal cancers in blacks as well as all patients afflicted with this dreadful disease. Smith, ].W.; Townsend, D.E.; Sparkes, R.S. Genetic Variants of Glucose-6-Phosphate Dehydrogenase in the Study of Carcinoma of the Cervix 290 Cancer 28(2):529-532, 1971 Subjects Treatment; Biologic Marker: G6PD; Cancer Site: cervix uteri Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 13 individuals (cervical dysplasia cases 5; carcinoma in situ case 1; invasive cervical carcinoma cases 7): Negro, female, neoplastic cervical disease Variables Demographic: age, Cancer: histology Data Sources Laboratory test Abstract Electrophoretic variants of the X-linked enzyme glucose-6- phosphate dehydrogenase were used to study the possible single cell or multicellular origin of human cervical neoplasia. Normal and affected tissue from Negro women with cervical dysplasia, carcinoma in situ, and invasive cervical carcinoma were evaluated. The results with pre-invasive disease were consistent with a single cell origin. Findings in invasive carcinoma in five patients are compatible with a unicellular origin, while in two patients a mul- ticellular origin seems likely. This raises the possibility of a heterogeneous etiology of invasive cervical epidermoid carcinoma. Sequential studies of one patient with invasive carcinoma, studied while undergoing radiation therapy, suggest the utility of G-6-PD electrophoresis for determining effectiveness and progress of treat- ment. Soichet, S. Ethnic-Related Incidence of Ovarian Carcinoma in New York City 291 Israel ] Med Sci 14(3):363-369, 1978 Subjects Cancer Site: ovary; Cancer Measures: incidence, mortality Study Characteristics Design: time trend, cross-sectional; Measures: descriptive statistics; Time Frame: 1964-1974 Sample Characteristics 874 ovary cancer deaths (nonwhites 79; whites 795—Period 1 cases 487; Period 2 cases 387) and 143 ovary cancer cases: female; New York City, NY Variables SES: religion; Demographic: age, race-ethnicity Data Sources Death registration, hospital medical records Abstract A two-part statistical study of ovarian cancer in the population of New York City was conducted in an effort to explain the striking racial differences in the incidence of death from this disease. The first part of the study comprised an examination of 487 death certificates from case reports in 1964 and 387 in 1974, with emphasis on race and religion, and revealed that there is 1) a lower incidence of ovarian cancer among nonwhite than white women and 2) a greater mortality from cancer of the ovary among Jewish than non-Jewish women. The second part entailed a review of 143 re- cords from hospitals in New York City from which more precise data, such as history and treatment, were available. The review enabled us to confirm diagnoses and to investigate whether specific clinical features are directly related to racial incidence. No etiologic factors could be correlated with the disease, but further epidemiologic studies are warranted. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Solomon, A.; Hurwitz, S.; Conlan, A.A. Bronchogenic Carcinoma with Coexisting Active Pulmonary Tuberculosis in Urban Blacks S Afr Med J 55:979-981, 1979 292 Subjects Cancer Site: lung; Cancer Measures: mortality, incidence Study Characteristics Design: case study; Time Frame: 1978 Sample Characteristics Six lung cancer cases with active pulmonary tuberculosis: presumed black, hospital patients; Johannesburg, South Africa Abstract There is a significant incidence of active pulmonary tuberculosis with coexisting lung cancer. It is likely that the cancer activates dormant tuberculosis. In a population group beset by a high inci- dence of tuberculosis, the heavy smoker who has attained his fifth decade seems particularly vulnerable to the combined diseases. Song, E.; Dusheiko, G.M.; Bowyer, S.; et al. Hepatitis B Virus Replication in Southern African Blacks with HBsAg-Positive Hepatocellular Carcinoma Hepatology 4(4):608-610, 1984 293 Subjects Cancer Site: liver; Biologic Marker: hepatitis B surface antigens (HBsAg), alpha-fetoprotein (AFP) Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 106 hepatocellular carcinoma cases (males 96; females 10): black, 13-72 years old; South Africa Data Sources Laboratory test Abstract Sera from 106 southern African blacks with hepatocellular car- cinoma and hepatitis B surface antigenemia (HBsAg) were tested for hepatitis B viral DNA (HBV-DNA) activity, HBV-DNA polymerase concentrations, and HBV e antigen (HBeAg) and anti- body (anti-HBe) to investigate the state of viral replication in these patients. HBeAg and anti-HBe were detected by radioimmunoas- say, HBV-DNA by molecular hybridization using a p(32)-labeled HBV-DNA probe, and HBV-DNA polymerase was measured by incorporation of H(3)-labeled thymidine triphosphate into double- stranded HBV-DNA. HBeAg was present in 30.2% (32/106) of the patients, almost always in low titer; 63.8% of the patients were anti-HBe positive. Circulating HBV-DNA was detected in 18.8% (20/106) of patients, including 14 of 32 (43.7%) who were HBeAg positive and 6 of 74 (8.1%) who were anti-HBe positive. In most patients, only trace amounts of HBV-DNA were evident. Raised HBV-DNA polymerase activity was found in 5.6% (6/106) of the patients, all of whom were HBeAg positive and 4 of whom had detectable amounts of circulating HBV-DNA. The HBV-DNA polymerase activity was relatively low in these patients. HBV re- plication thus appears to be present in only a minority of southern African Blacks with HBV-related hepatocellular carcinoma, and when present is of low grade activity. Sorahan, T.; Crombie, I.K. Cancer of Cervix and Cancer of Penis Lancet 11(8260):1419-1420, 1981 294 Subjects Cancer Site: penis, cervix uteri; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, higher level statistics Sample Characteristics 62 international populations (nonwhites 25; whites 37) Variables Demographic: race-ethnicity (white populations, nonwhite populations) Abstract The analysis investigates the question of whether there is a common infection agent involved with both cancer of the cervix and cancer of the penis by examining international variation in the incidence of these cancers. Correlation analyses (non-paramet- 120 ric) were done for cancer of the penis and 44 indvidual female sites and for cancer of the cervix (excluding carcinoma-in-site) and the 42 individual male sites. The strongest correlation was for cancer of the penis and cancer of the cervix. The 62 populations were separated into white and non-white registries. Scattergrams for these registries of the incidence of cancer of the penis against cancer of the cervix show that a strong association is observed only among the non-white registries. However, when the correla- tion coefficients were recalculated with rates weighted by popula- tion size, a highly significant correlation was also found for the white registries. There may, therefore, be an association between these two sites among whites but it may be small and overwhelmed by other factors. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Spitz, M.R.; Sider, ]J.G., Johnson, C.C.; et al. Ethnic Patterns of Hodgkin's Disease Incidence Among Children and Adolescents in the United States, 1973-82 J Natl Cancer Inst 76(2):235-239, 1986 295 Subjects Cancer Site: lymphoma; Cancer Measures: incidence Study Characteristics Design: cross-sectional, time trend; Measures: age-specific incidence rate, descriptive statistics, higher level statistics; Time Frame: 1969-1982 Sample Characteristics 1,109 Hodgkin's disease cases (blacks 68; whites 860; Hispanics 156): 0-19 years old; SEER Program areas and Hodgkin's disease incidence data: white, males, 5-19 years old; 4 TNCS areas (Atlanta, GA; Detroit, MI; San Francisco-Oakland, CA; Iowa) Variables Demographic: age, sex, race-ethnicity geographic area; Cancer: histology Data Sources SEER cancer registry, Third National Cancer Survey 1969-1971 Abstract The descriptive epidemiologic findings were summarized on 1,109 patients (white, black, and Hispanic) under 20 years of age who were diagnosed with Hodgkin's disease as reported to the Surveillance, Epidemiology, and End Results (“SEER”) Program of the National Cancer Institute from 1973 to 1982. Across all ethnic strata, incidence rates increased with advancing age at diagnosis, with white adolescents 15-19 years old exhibiting the highest rates (male, 3.67; female, 4.18). Gender difference among children 0-14 years of age was most evident in blacks (male:female ratio: 4.0 for blacks, 1.0 for whites). Highest adolescent:childhood ratios of in- cidence rates were noted for females (5.81 for whites and 8.29 for New Mexico Hispanics) and lowest, for Hispanic males (1.25, New Mexico; 2.15, Puerto Rico). Whites exhibited the highest percentage of the nodular sclerosis histologic subtype (65%) and Hispanics, the lowest (45%). Conversely, Hispanics had higher rates of his- tologic types associated with a poorer prognosis (mixed cellularity and lymphocyte depletion). These differing age and histologic pat- terns were consistent with previously described international pat- terns of disease occurrence. Analysis of secular trends for whites from 1969 to 1982 revealed relatively stable rates for youngest ages and male adolescents. Rates increased over time for white female adolescents, but the trend was not statistically significant. Spivey, G.H.; Perry, B.W.; Clark, V.A_; et al. Predicting the Risk of Cancer at the Time of Breast Biopsy Am Surg 48(7):326-332, 1982 296 Subjects Diagnosis; Cancer Site: breast Study Characteristics Design: time trend, cross-sectional; Measures: relative risk; Time Frame: 1958-1979 Sample Characteristics 7,180 breast tumor biopsies (blacks 182; whites 4,869; Hispanics 256; Asians 80; others 5; race unknown 1,788): 8 major hospitals; Los Angeles, CA; Breast tumor biopsy information from 2 major hospitals; Los Angeles, CA Variables Demographic: age, race-ethnicity; Cancer: histology Data Sources Hospital medical records Abstract The benign to malignant ratio (B:M ratio) among breast biopsies (number of benign breast lesions divided by number of breast cancers) is widely believed to be around 4:1 or 5:1. This belief appears to be based upon experience from the 1950s and 1960s. We investigated possible reasons for the current wide variability of the B:M ratio. A straight line relationship between the log B:M ratio and age exists in our data. The B:M ratio is sensitive to racial differences between patient groups but is only minimally affected by varied histologic inclusion criteria. The overall B:M ratio should not be used to counsel women about their breast cancer risk at the time of biopsy. Age-specific B:M ratios provide a more realistic risk assessment. Stapleton, J.J. Extramammary Paget's Disease of the Vulva in a Young, Black Woman J Reprod Med 29(7):444-446, 1984 297 Subjects Diagnosis; Treatment; Cancer Site: vulva Study Characteristics Design: case-study Sample Characteristics One Paget's disease of the vulva case: black, female, 24 years old; Reno, NV Data Sources Physician medical records Abstract Since the initial report, in 1901, by W. Dubreuilh of a vulvar lesion with the same histopathology as the eczematoid lesion found on the nipple and areola described seven years previously by Sir 121 James Paget, fewer than 100 cases of this uncommon vulvar neop- lasia have been reported. This disease process, with and without an accompanying adenocarcinoma, is usually found in post- menopausal Caucasian women. A 24-year-old, black woman gave a subjective history of having had a pruritic vulvar lesion for 18 months. Histologically this neoplasm proved to be extensive ex- tramammary adenocarcinoma involving the vulva and, unilater- ally, the inguinal-femoral, pelvic and aortocaval lymphatics. Im- munoperoxidase staining for carcinoembryonic antigen is valuable for differentiating this disease from similar vulvar lesions plus assessing surgical. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Stevens, J.; Leffall, L.D., Jr.; Agbatha, A.; et al. Cancer of the Pancreas in Blacks: A Ten-Year Experience J Natl Med Assoc 69(4):249-251, 1977 298 Subjects Treatment; Cancer Site: pancreas; Cancer Measures: survival Study Characteristics Design: time trend, prospective (nonconcurrent); Measures: descriptive statistics, Time Frame: 1965-1974 Sample Characteristics 68 pancreas cancer cases (males 38; females 30): black, 22-84 years old, hospital patients; Washington, DC Variables Demographic: age, sex; Cancer: stage, tumor location, symptoms Data Sources Hospital medical records Abstract A ten-year retrospective study of cancer of the pancreas in black patients at Howard University Hospital is presented. A total of 68 patients was studied. The authors concluded that no form of treatment had significant effect in terms of survival in stages II, III, and IV of disease. Stober, J.; Asal, N. Multiple Myeloma in Oklahoma: Racial, Age, Sex, Geographic and Time Variations South Med J 69(3):298-303, 1976 299 Subjects Cancer Site: multiple myeloma; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: crude mortality rate, age-adjusted mortality rate, age-specific mortality rate, standardized mortality ratio (SMR); Time Frame: 1950-1970 Sample Characteristics 770 multiple myeloma deaths (nonwhites 87; whites 683): Oklahoma Variables Demographic: age, sex, race-ethnicity, geographic area, urban-rural Data Sources Death registration, U.S. Population Census Abstract Oklahoma death certificates listing multiple myeloma as the underlying cause of death during the period 1950-1970 were examined for secular trends, and geographic and demographic characteristics associated with the disease. An upward trend in the death rates for all age, sex, and race groups was found; the most dramatic increases were seen in the oldest age groups. During the 20-year period, the death rate for white males 75 years of age and older increased tenfold. For white females 75 years of age and older, it increased eightfold. White male rates were constantly higher and showed a sharper increase in age-adjusted death rates than white female rates. Because of the low number of nonwhite deaths in Oklahoma from multiple myeloma, the rates showed extreme fluctuations. After adjustments were made for age differ- ences, the nonwhite annual rate was 2.36 per 100,000 compared with 1.62 for the white population. There was not enough informa- tion to draw conclusions about the geographic distribution of mul- tiple myeloma. Findings for the Oklahoma population were consis- tent with the few previously reported studies. Susseelan, A.V.; Gupta, LM. Malignant Melanoma in Nigeria-Pathological Studies Afr J Med Sci 6:209-213, 1977 300 Subjects Cancer Site: malignant melanoma Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1974-1975 Sample Characteristics 21 malignant melanoma cases: Negro, 27-71 years old; Nigeria Variables Demographic: age, sex; Cancer: tumor location, histology Data Sources Hospital medical records, histopathology records Abstract Pathological features of twenty-one cases of malignant melanoma studied in the University of Nigeria Teaching Hospital, Enugu during the period January, 1974 to December, 1975 are presented. Malignant melanoma accounted for 2.4% of all tumours 122 and 4.5% of all malignant tumours, greatest age incidence being in the fifth to seventh decades. The male to female sex ratio was 2:1. 73.2% of cases were of the nodular variety. 81% melanomas occurred on the sole of feet validating the hypothesis that the pigmented skin in Africans is resistant to malignant melanoma. Melanoma in Nigerians would appear essentially to be arising from epidermal melanocytes and not from pre-existing naevus cells. Hence we do not feel prophylactic removal of plantar moles as suggested by Onuibgo (1975) is desirable. Histologically, there was no clear association between the cell types and the kind of malenoma or invasion of the tumour. The difference in behavioural and natural history of malignant melanoma would appear to have a bearing on the local tissue and also general immune mechanisms of the host. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Swanson, G.M.; Young, J.L., Jr. Trends in Cancer Incidence in Metropolitan Detroit, 1937-1977: Leads for Prevention Prev Med 12:403-420, 1983 301 Subjects Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: time trend; Measures: age-adjusted incidence rate; Time Frame: 1937-1977 Sample Characteristics Cancer incidence data: Detroit, MI Variables Demographic: sex, race-ethnicity Data Sources First National Cancer Survey 1937-1939, Second National Cancer Survey 1947-1948, Third National Cancer Survey 1969-1971, SEER cancer registry Abstract Cancer incidence among the black and white male and female population of metropolitan Detroit is presented for the years 1937, 1948, 1969-1971, and 1973-1977. These trends reveal increases in lung cancer among all four groups over time, while stomach can- cers have decreased among each of the four groups. For black and white men, lung cancer has the highest incidence in the two recent time periods, while cancer of the prostate was highest in the two earlier time periods. Among women, uterine cancer had the high- est incidence for both blacks and whites in the earlier time periods, but breast cancer occurred among women more frequently than any other cancer in the most recent time period. These trends reveal the changing incidence patterns over time within an urban, industrialized population. Population-based cancer incidence data provide the most complete and accurate assessment of the occurr- ence of cancer within a population. Such data are particularly useful for proper planning of screening programs and for evaluat- ing the need for preventive measures. As an example, the continu- ous increase in lung cancer in all four groups across all time periods clearly reveals the need for more effective measures to reduce cigarette smoking in this population. Szklo, M.; Gordis, L.; Tonascia, J.; et al. The Changing Survivorship of White and Black Children with Leukemia Cancer 42:59-66, 1978 302 Subjects Treatment; Cancer Site: leukemia; Cancer Measures: survival Study Characteristics Design: time trend; Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1960-1975 Sample Characteristics 287 leukemia cases (blacks 48; whites 239): under 20 years old; Baltimore, MD Variables SES: census tract rental value; Demographic: age, race-ethnicity; Cancer: histology Data Sources Death registration, hospital medical records Abstract A study was conducted in metropolitan Baltimore to examine changes in survival of white and black children with acute leukemia from 1960 through 1975. Two-hundred eighty-seven cases were identified, of which 77% were acute lymphocytic leukemia (ALL). Eighty-three percent of all cases occurred in white children. In white children with ALL, two-year survival rates increased from 32% from 1960-64 to 47% in 1965-69, and to 79% in 1970-75 (p< .005). In the small sample of black children with ALL, two-year survival rates increased from 25% in 1960-64 to 59% in 1965-69 (p<.01), with no further increas in 1970-75. For acute nonlym- phocytic leukemia (ANLL), survivorship was analyzed in white children, among whom one-year survival rates increased from 42% in 1965-69 to 71% in 1970-75 (p<.005). The increasingly better survival over time of white children with acute leukemia probably reflects the increasing efficacy of new therapeutic approaches. 123 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Szpak, C.A.; Stone, M.].; Frenkel, E.P. Some Observations Concerning the Demographic and Geographic Incidence of Carcinoma of the Lip and Buccal Cavity Cancer 40:343-348, 1977 303 Subjects Risk Factor Exposure: solar radiation; Cancer Site: oral cavity, skin; Cancer Measures: incidence Study Characteristics Design: cross-sectional, time trend; Measures: age-adjusted incidence rate; Time Frame: 1947-1970 Sample Characteristics Oral cavity and skin cancer incidence data: TNCS areas Variables Demographic: sex, race-ethnicity (black, white), geographic area Data Sources Third National Cancer Survey 1969-1971, U.S. Population Census, survey Abstract The geographic and demographic data obtained during the Third National Cancer Survey have provided a perspective on etiologic factors and incidence trends for cancers of low frequency. The incidence of cancer of the lip, oral cavity and skin from this survey was compared to similar studies in 1947 and intra-regional patterns in one area of the Third National Cancer Survey (Dallas- Fort Worth Metropolitan SMSA) were evaluated. The average age- adjusted annual incidence of cancer of the lip in white men in this latter area was 11.5 per 100,000 (based on 1950 population stan- dard), 2-fold greater than that geographic area with the second highest incidence and approximately 3-fold greater than in all the other areas. The incidence in white women was only 8% that seen in white men. Intra-regional differences of similar populations were seen with the incidence in Fort Worth men being 50% greater than in a similar population in Dallas. Incidence trends over the past 2 decades reveal a significant decline in the incidence of oral cavity cancer and a slight decrease in lip cancer. Comparisons of the incidence of lip cancer did not correlate with the incidence of skin cancer nor with the geographic latitude in the other survey areas. The studies fail to support the classical implication of actinic radiation as the primary etiologic factor in lip cancer incidence. Talerman, A. Clinico-Pathological Study of Malignant Lymphoma in Jamaica Br J Cancer 24(1):37-47, 1970 304 Subjects Cancer Site: lymphoma; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude incidence rate, descriptive statistics; Time Frame: 1958-1966 Sample Characteristics 260 malignant lymphoma cases (Negroes 248; Chinese 7; Europeans 3; Indians 2): Jamaica Variables Demographic: age, sex, race-ethnicity; Cancer: histology, symptoms Data Sources Hospital medical records, non-SEER population-based cancer registry, histopathology records Abstract A clinico-pathological study of malignant lymphoma in Jamaica was undertaken to examine the disease pattern in a predominantly Negro population of West African origin. During a 9-year period (1958-66) 260 histologically verified cases of malignant lymphoma were encountered. The distribution of the different histological types was as follows: Hodgkin's disease 50.9%, lymphosarcoma 33%, reticulum cell sarcoma 14.2%, giant follicular lymphoma 1.9%. No cases of Burkitt's tumour were encountered. This study indicates that malignant lymphoma is not uncommon in Jamaica, and thatits distribution pattern is similar to that observed in Europe and North America, except for the paucity of giant follicular lym- phoma, and is different from the pattern observed in parts of Africa populated by Negroes, where Burkitt's tumors is the most common type, and where Hodgkin's disease is relatively uncom- mon. The age and sex incidence was in general similar to other reported series, but the duration of symptoms was short. The majority of patients presented with generalised peripheral lym- phadenopathy. Hepatosplenomegaly and anaemia were common on admission. The prognosis was generally poor in comparison with European and North American series due to advanced stage of disease on presentation. 124 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Talerman, A. Clinico-Pathological Study of Multiple Myeloma in Jamaica Br J Cancer 23(2):285-293, 1969 305 Subjects Cancer Site: multiple myeloma; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude incidence rate, descriptive statistics; Time Frame: 1957-1966 Sample Characteristics 101 multiple myeloma cases (Negroes 99; whites 1; Chinese 1): 35-79 years old; Jamaica Variables Demographic: age, sex, race-ethnicity; Cancer: clinical diagnosis, symptoms Data Sources non-SEER population-based cancer registry, hospital medical records, histopathology records. Abstract A clinico-pathological study of multiple myeloma in Jamaica was undertaken in order to examine the disease pattern in a pre- dominantly negro population of West African origin. During a 10-year period (1957-66) 101 cases of multiple myeloma and 3 cases of solitary plasmocytoma were encountered. This indicates that multiple myeloma is more common in Jamaica than in parts of Africa populated by negroes. The sex incidence was nearly equal, and there was no evidence that the disease occurs earlier in the negro than in white populations. The haematological, biochemical, and radiological findings were similar to those reported from Europe and North America. The majority of patients presented with advanced disease, and the prognosis was very poor. Post mortem findings showed a high incidence of extra-medullary in- volvement and amyloidosis. Te Groen, L.H.T.; Rose, E.F. Preliminary Investigation into the Incidence of Cancer of the Cervix S Afr Med ] 48:2341-2345, 1974 306 Subjects Screening: Pap smear; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1965-1972 Sample Characteristics 798 hospital cancer admissions (blacks 517; whites 257; colored and Asiatics 24); 525 cancer bed bookings for blacks; 1,774 histologically confirmed cancers (hospital laboratory cancers 1,451; private laboratory cancers 233); and 93 consecutive black gynecological patients: South Africa Variables Demographic: age, race-ethnicity, parity; Cancer: histology Data Sources Hospital medical records, histopathology records, survey Abstract An investigation was undertaken to estimate the prevalence of carcinoma of the cervix in a population where circumcision is com- monly practised. A high rate of evidence of carcinoma of the cervix and vaginal infection was established. Terris, M.; Hall, C.E. Decline In Mortality From Gastric Cancer In Native-Born and Foreign-Born Residents of New York City J Natl Cancer Inst 31(1):155-162, 1968 307 Subjects Diagnosis; Cancer Site: stomach; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: age-specific mortality rate, age-adjusted mortality rate; Time Frame: 1949-1961 Sample Characteristics Stomach cancer mortality data: Negroes, native-born whites, foreign-born whites; New York City, NY Variables Demographic: age, sex, race-ethnicity, place of birth Data Sources Death registration, U.S. Population Census Abstract This study of mortality from gastric cancer in New York City disclosed that the decline between 1949-51 and 1959-61 occurred not only in the total population but also in males and females in each of 3 population groups: native-born whites, foreign-born whites, and Negroes. Furthermore, the decrease occurred in al- most every age group within each category. It was concluded that the reduction in mortality could not be attributed to the decreasing proportion of a high risk group, the foreign-born, in the total population. The decline was greatest in those population groups with the lowest mortality rates in 1949-51. Examination of available data on accuracy of diagnosis did not support the hypothesis that the lower rate resulted from improvement in diagnostic accuracy. The findings of this and other studies indicate that the lower mor- tality from gastric cancer is probably not an artifact caused by demographic changes or improvements in diagnosis or treatment, but may well reflect a real decline in incidence due to environmen- tal factors that remain unidentified. 125 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Theron, E.].; Middlecote, B.D. Tumours of the Salivary Glands: The Bloemfontein Experience S Afr | Surg 22(4):237-242, 1984 308 Subjects Cancer Site: salivary glands Study Characteristics Design: retrospective; Measures: crude incidence rate, descriptive statistics; Time Frame: 1971-1983 Sample Characteristics 217 salivary glands cancer cases (blacks 124; whites 93): 12-92 years old, hospital patients; Bloemfontein, South Africa Variables Demographic: race-ethnicity; Cancer: histology, tumor location Data Sources Histopathology records, hospital medical records Abstract Tumours of the salivary glands are uncommon, but they pre- sent special problems in diagnosis and surgical management. In a review of 217 cases treated surgically, the age and sex incidence are comparable, but some tumours predominate in different race groups. The incidence of the various tumour types is very similar to that reported in a number of large series in the literature. How- ever, in this series it was found that the clinical and sialographic features were usually not helpful in distinguishing between benign and malignant tumours. Thind, 1.S.; Carnes, R.; Najem, R.; et al. Cancer Incidence and Mortality in Newark, N.J. 1970-1974: A National Comparison Cancer 47(5):1047-1053, 1981 309 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, mortality Study Characteristics Design: cross-sectional; Measures: age-specific incidence rate, age-adjusted incidence rate, age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1970-1974 Sample Characteristics 4,790 multiple site cancer cases (blacks 1,823; whites 2,967): Newark, NJ Variables Demographic: age, sex, race-ethnicity Data Sources Third National Cancer Survey 1969-1971, SEER cancer registry, hospital medical records, U.S. Population Census, death registration Abstract New Jersey has acquired the invidious label “Cancer Alley U.S.A.” based upon a national cancer mortality analysis. However, a cancer incidence survey conducted in Newark, the largest met- ropolitan industrial city in New Jersey, showed that age-adjusted 126 Newark rates for all sites were comparable to the Third National Cancer Survey (TNCS) and Surveillance, Epidemiology and End Results (SEER) populations, except for black males who had statis- tically lower rates compared with the SEER population only. How- ever, Newark did have statistically higher incidence of the follow- ing: a) esophagus cancer among white men, black men, and black women; and b) cervix, uterus, ovary, and bladder cancers among black women. Age-adjusted Newark cancer mortality for all sites was not statistically different from the SEER experience, except for an excessive cancer mortality among white men for stomach and esophagus; white women for stomach, colon-rectum, and uterus; black men for esophagus and colon-rectum; black women for colon-rectum, cervix, uterus, and ovary. An analysis of Newark mortality/incidence ratios suggests that the excessive cancer bur- den for the majority of sites studied resulted from poor end results of therapy, probably due to either late diagnosis, poor compliance, and/or suboptimal therapy. The Newark data cast doubt on the validity of the use of mortality data only in referring pejoratively to New Jersey as “Cancer Alley U.S.A.” EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Thind, I.S.; Najem, G.R.; Richter, R.; et al. Carcinoma of the Cervix in Newark, New Jersey, 1970-76, A Very Low in Situ: Invasive Ratio 310 J Reprod Med 26(10):513-518, 1981 Subjects Screening: Pap smear; Cancer Site: cervix uteri; Cancer Measures: tional Cancer Survey (TNCS) population. In contrast, Newark incidence, mortality blacks’ invasive cancer rates were higher than those found in any Study Characteristics individual geographic area surveyed in TNCS except for Min- Design: cross-sectional; Measures: relative risk, age-adjusted neapolis as compared to Newark whites, who had lower rates than mortality rate, age-specific mortality rate, age-adjusted all individual TNCS areas except Colorado and San Francisco. incidence rate, age-specific incidence rate, descriptive statistics, Newark blacks had a relative risk of 4.0 for invasive cancer as higher level statistics; Time Frame: 1970-1976 compared to Newark whites, whereas the corresponding relative Sample Characteristics risk for blacks versus whites in the TNCS population was 2.0. 739 cervical cancer cases (blacks 556; whites 183): female; Newark blacks and whites together had the lowest in situ:invasive Newark, NJ; cervical cancer mortality data from Newark, NJ; ratio as compared to women in other parts of the United States. cervical cancer incidence data from the TNCS areas; cervical The invasive cancer incidence and mortality rates for both Newark cancer incidence and mortality data from the SEER Program blacks and whites were significantly higher than in the Surveil- areas lance, Epidemiology and End Results (SEER) population, but inci- Variables dence:mortality ratios for Newark and SEER were not different. Demographic: age, race-ethnicity (black, white), geographic The most likely explanation for the low in situ rates and high area; Cancer: stage invasive rates among Newark blacks is their failure to obtain Data Sources Papanicolaou (Pap) smears. The low in situ rate among Newark Death registration, non-SEER hospital cancer registry, Third whites in the absence of a high invasive rate is difficult to explain. National Cancer Survey 1969-1971, SEER cancer registry It seems that the problem among blacks can be alleviated by the Abstract widespread use of Pap smears, which reduce the frequency of Both blacks and whites in Newark had significantly lowered invasive cancer and the associated mortality. incidence of in situ cervical cancer as compared to the Third Na- Thind, I.S.; Najem, R.; Paradiso, J.; et al. Cancer Among Blacks in Newark, New Jersey, 1970-1976: A National and International Comparison 311 Cancer 50(1):180-186, 1982 Subjects Diagnosis; Cancer Site: multiple sites; Cancer Measures: incidence ally similar to other American black women. American blacks, like Study Characteristics those of Newark, had cancer rates higher than the African blacks Design: cross-sectional; Measures: age-specific incidence rate, in three out of four countries studied. Only blacks of Rhodesia age-adjusted incidence rate; Time Frame: 1970-1976 had cancer experience like that of American blacks. Except for Sample Characteristics cancer of the cervix, the African blacks had generally lower rates 2,585 multiple site cancer cases: black; Newark, NJ compared to the American blacks for the ten sites investigated. Variables However, high rates did occur for certain sites, e.g., esophagus Demographic: age, sex, geographic area and bladder cancers in Bulawayo, Rhodesia. Newark blacks also Data Sources had higher rates for certain sites (prostate, lung, breast and cervix) Hospital medical records compared to Washington, D.C., and certain other U.S. cities. Thus, Abstract the total risk of malignancies to develop among Newark blacks Newark black men had age-adjusted cancer rates comparable was no greater than that expected, based upon the four U.S. cities’ to those of Washington, D.C., but lower than the other three experience. American city blacks. Rates for Newark black women were gener- 127 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Thomas, G.E.; Wicks, A.C.B.; Clain, D.].; et al. Hepatocellular Carcinoma in the Rhodesian African Digestive Diseases 22(7):573-581, 1977 312 Subjects Biologic Marker: alpha-fetoprotein, hepatitis B antigen; Risk Factor Exposure: alcohol, other diseases, geographic area; Cancer Site: liver Study Characteristics Design: prospective; Measures: descriptive statistics; Time Frame: 1973-1974 Sample Characteristics 28 primary liver carcinoma cases: black, hospital patients; Salisbury, Rhodesia Variables Demographic: age, sex, geographic area; Cancer: symptoms, histology Data Sources Survey, hospital medical records, laboratory test, histopathology records Abstract We have carried out a prospective survey of 28 primary liver carcinomas over one year. Hepatocellular carcinoma is the com- monest malignancy seen in Rhodesian blacks, which results in a high index of suspicion and accounts for the 96.4% positive diag- nosis before death in this study. The age distribution was evenly spread through adult life with no definite peak incidence. Some were young and without evidence of chronic liver disease, but many had the stigmata of established hepatic disease. This con- trasts with the common assertion that in areas of high incidence for primary liver cancer those affected are mainly young and lack signs of chronic liver disease. The commonest presenting symptoms were abdominal pain and swelling and weight loss. Hepatomegaly, often tender and nodular, was present in all but one. The incidence of alpha-fetoprotein, 46.5% is low compared with other countries where primary liver cancer is common. Hepatitis B antigen was absent in all 28, suggesting that there is no association between the persistence of the antigen and hepatocellular carcinoma in Rhodesia. Liver function tests, al- though abnormal, were never diagnostic of primary liver cancer. We have confirmed the association of high alcohol consumption and cirrhosis with hepatocellular carcinoma. Torres, F.O.; Purchase, [.F.H.; Van der Watt, J.J. The Aetiology of Primary Liver Cancer in the Bantu J Path 102:163-169, 1970 313 Subjects Cancer Site: liver; Risk Factor Exposure: other diseases; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1957-1963 Sample Characteristics 200 patients (males 165; females 35): Bantu, clinical diagnosis of jaundice; Lourenco Marques, South Africa and 278 liver biopsies: Bantu; Lourenco Marques, South Africa Variables Cancer: histology Data Sources Histopathology reports, laboratory test Abstract A survey of the pathological lesions occurring in the liver of the Bantu in Mocambique was undertaken. Primary carcinoma, viral hepatitis and toxic liver damage were the commonest lesions seen in a series of biopsies. The lesions described as toxic liver damage and primary carcinoma were remarkably similar to the lesions produced in rats by the addition of the mycotoxin sterig- matocystin to their diet. It is concluded that toxic liver damage may be a response to environmental carcinogens and represents an early stage of liver damage that progresses to the development of hepatic carcinoma. The hypothesis that primary carcinoma of the liver in the African is produced by a food-borne toxin, possibly a mycotoxin, is supported by these findings. Tsai, S.P.; Lee, E.S. Potential Gains in Life Expectancies by Partial Elimination of Leading Causes of Death in Texas 314 Tex Rep Biol Med 36:185-196, 1978 Subjects Prevention; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1969-1971 Sample Characteristics Mortality data: Texas Variables Demographic: age, sex, race-ethnicity (nonwhite, white) Data Sources U.S. Population Census, death registration Abstract Potential gains in life expectancies among Texas population by partial elimination of 3 major causes of death are examined on the basis of the available statistics from the population census and mortality statistics for 1970. Contrary to the popular anticipation of longer potential gains, the results are not particularly encourag- ing. The number of years of life that would be gained during the working ages of 50% elimination of major cardiovascular diseases in less than 1/2 of 1 year, about 1/4 of 1 year by 50% elimination of malignant neoplasms, and less than 1/4 of 1 year by 50% elimi- nation of motor vehicle accidents. Even with a scientific break- through in combating those causes of death it appears that future gains in life expectancies for working ages will not be spectacular. The implications of the results in relation to the current debate on the national health policy are discussed. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Tsai, S.P.; Lee, E.S.; Kautz, J.A. Changes in Life Expectancy in the United States Due to Declines in Mortality, 1968-1975 315 Am J Epidemiol 116(2):376-384, 1982 Subjects Cancer Site: all sites combined; Cancer Measures: mortality expectancy at age 45 years has occurred for all race/sex groups Study Characteristics combined. The added years of life for the normal working ages Design: time trend; Measures: descriptive statistics; Time Frame: (15-70 years) is only 0.6 years for the total US population, 0.3 years 1968-1975 for white females, 0.6 years for white males, 1.5 years for nonwhite Sample Characteristics males, and 1.7 years for nonwhite females. The relative contribu- Mortality data: United States tion of the five leading causes of death to this gain varies at different Variables ages. For example, more than 50% of the increase in life expectancy Demographic: age, sex, race-ethnicity (nonwhite, white) at age 45 years was due to a lower mortality rate in diseases of Data Sources the heart which is still the leading cause of death among each of U.S. Population Census, death registration the race/sex groups. Other contributions to the increase in life Abstract expectancy at age 45 years are: cerebrovascular diseases, 16% ac- This study examines the gains in life expectancy for four race/ cidents, 6%; influenza and pneumonia, 7%; and all other causes, sex groups of the US population between 1968 and 1975. An in- 16%. The increase in the malignant neoplasms mortality rate had crease of 2.3 years of life expectancy at birth and 1.7 years in life a negative effect, -2%, on the gain in life expectancy. Tucci, P.; Haralambidis, G. Carcinoma of the Scrotum: Review of Literature and Presentation of 2 Cases 316 J Urol 89(4):585-590, 1963 Subjects Treatment; Cancer Site: scrotum; Cancer Measures: incidence tional disease in mule-spinners, paraffin workers and wax Study Characteristics pressmen. The disease is extremely rare in the general population. Design: case study; Time Frame: 1953-1962 It is extremely unusual in Negroes, possibly because of the differ- Sample Characteristics ence in racial susceptibility. This finding is related to the fact that Two scrotum cancer cases: Negro, male, 52 and 62 years old; skin cancer is generally of lower incidence among the colored New York City, NY population. Two cases of non-occupational carcinoma of the Data Sources scrotum in Negroes are presented; to our knowledge, the first Hospital medical records ones reported in this country. The etiology, incidence, pathology, Abstract and the problems related to the clinical picture, diagnosis, prog- Carcinoma of the scrotum, as an occupational disease in chim- nosis, and treatment of the disease are briefly discussed. ney sweeps, was first recognized in England. It is also an occupa- Turgman, J.; Modan, B.; Shilon, M.; et al. Nasopharyngeal Cancer in a Total Population: Selected Clinical and Epidemiological Aspects 317 Br J Cancer 36:783-786, 1977 Subjects Abstract Risk Factor Exposure: geographic area; Cancer Site: A nationwide study of nasopharyngeal cancer in Israel, during nasopharyngeal; Cancer Measures: incidence, survival a 9-year period (1960-68) demonstrated a mean annual incidence Study Characteristics rate of 1.0 per 100,000 in males and 0.4 per 100,000 in females. Design: cross-sectional, prospective (nonconcurrent); Measures: Significantly higher incidence was observed in residents born in age-adjusted incidence rate, age-specific incidence rate, crude North Africa (3.0 in males and 1.1 in females). Survival was rela- survival rate, descriptive statistics; Time Frame: 1960-1968 tively better in females among patients with lymphoepithelioma Sample Characteristics and in those without neurological complications. The results may 150 nasopharyngeal cancer cases: 0-85 years old; Israel support the presence of environmental factors in aetiology, though Variables genetic predisposition cannot be ruled out. Demographic: age, sex, birthplace (Africa, Asia, Europe/ America, Israel, Arabs); Cancer: histology, symptoms, tumor location 129 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Turnbull, A.D.M.; Goodner, J.T. Primary Adenocarcinoma of the Esophagus 318 Cancer 22(5):915-918, 1968 Subjects Abstract Treatment; Cancer Site: esophagus; Cancer Measures: incidence, Forty years of experience with carcinoma of the esophagus at survival the Memorial Sloan-Kettering Cancer Center is reviewed. Of 1859 Study Characteristics patients, 1529 were found to have epidermoid carcinoma and 45 Design: prospective (nonconcurrent); Measures: descriptive had primary adenocarcinoma. Treatment and results in these 45 statistics; Time Frame: 1926-1966 patients are discussed. Because of varying reports concerning the Sample Characteristics incidence and prognosis of adenocarcinoma of the esophagus, all 45 adenocarcinoma of esophagus cases (blacks 1; whites 44): cases of extension into the distal esophagus by a primary carcinoma 38-85 years old; New York of the stomach were carefully excluded. The two forms of Variables esophageal cancer are compared and show a striking similarity in Demographic: sex, race-ethnicity; Cancer: tumor location their clinical behavior and ultimate prognosis. Data Sources Hospital medical records Ueshima, H.; Cooper, R.; Stamler, J. Age Specific Mortality Trends in the U.S.A. From 1960 to 1980: Divergent Age-Sex-Color Patterns 319 J Chron Dis 37(6):425-439, 1984 Subjects Cancer Site: all sites combined; Cancer Measures: mortality 1970s. The ratio of male to female deaths for white teenagers and Study Characteristics young adults (15-24 and 25-34) increased over both decades, al- Design: time trend; Measures: age-specific mortality rate, higher though the mortality rate for white women age 15-24 increased level statistics; Time Frame: 1960-1980 during the 1960s. (2) Non-white men age 15-24 and 25-34 experi- Sample Characteristics enced biphasic mortality trends, with a rise peaking in 1971 and Mortality data: United States then a fall. (3) The mortality ratios, non-white to white men, age Variables 25-34 and 35-44, were far higher than those of other age groups. Demographic: age, sex, race-ethnicity (nonwhite, white) (4) The mortality ratios, non-white males to females, increased Data Sources steadily for all age groups, and those of age groups 15-24 and Death registration 25-34 turned up sharply. Selected causes of death (1960-1977) were Abstract reviewed to uncover reasons for the unfavorable mortality trends Trends in age-specific mortality (10-year age groups, 5-74) for among males. The following causes contributed to rising mortality the U.S. population were reviewed for all causes of death from among males: (a) accidents, suicide, and homicide for white and 1960 to 1980. The age-specific mortality rate generally declined. non-white teenagers and young adults, age 15-24, 25-34 and 35-44; However, the following age-sex-color groups showed divergent (b) suicide and homicide for non-white men age 45-54; (c) cirrhosis mortality trends compared to those of other groups: (1) The death of the liver for white men age 35-44 and for non-white 25-34, 35-44 rates for white male teenagers and young adults (ages 15-24 and and 45-54; (d) malignant neoplasms for white men age 35-44 and 25-34) increased during 1960s and stayed at the same level during for non-white age 45-54. Van Tonder, S.; Kew, M.C.; Hodkinson, J.; et al. Serum Vitamin B,, Binders in South African Blacks With Hepatocellular Carcinoma 320 Cancer 56:789-792, 1985 Subjects Diagnosis; Cancer Site: liver; Biologic Marker: unsaturated than 30 years of age. Eighty-four percent of the patients vitamin B12 binding capacity had a slightly raised UBBC and 86% had a slightly elevated Study Characteristics vitamin B12 value, but in no patient was an exceptionally Design: cross-sectional; Measures: descriptive statistics, high UBBC present. Serum UBBC and vitamin B12 were higher level statistics not higher in younger patients, and raised UBBC values Sample Characteristics were not related to serum alpha-fetoprotein values. Serum 242 liver cancer cases: black, 10 + years old; South Africa UBBC and vitamin B12 concentrations were not significantly Data Sources different in patients with and without coexisting cirrhosis. Laboratory test In none of the patients with a UBBC above 3000 pg/ml was Abstract the fibrolamellar variant of hepatocellular carcinoma pre- Sera from 242 South African blacks with hepatocellular sent. The authors conclude that South African blacks with carcinoma were assayed for unsaturated vitamin B12 bind- hepatocellular carcinoma do not secrete an abnormal vita- ing capacity (UBBC) and vitamin B12 activity. Six patients min B12 binding protein. were younger than 20 years of age, and 24% were younger 130 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Vena, J.E.; Sultz, H.A.; Fiedler, R.C.; et al. Mortality of Workers in an Automobile Engine and Parts Manufacturing Complex Br J Ind Med 42:85-93, 1985 321 Subjects Cancer Site: multiple sites; Risk Factor Exposure: occupation; Cancer Measures: mortality Study Characteristics Design: retrospective; Measures: descriptive statistics, higher level statistics; Time Frame: 1970-1979 Sample Characteristics 769 deaths (nonwhites 134; whites 635): male, automobile workers; Erie County, PA Variables SES: occupation; Demographic: race-ethnicity Data Sources Industry records, death registration Abstract A proportionate mortality ratio (PMR) study was conducted using data on workers from three local unions representing an integrated automobile factory composed of forge, foundry, and engine (machine and assembly) plants. Ninety four percent of the death certificates were obtained for all active and non-active work- ers who died during the period 1 January 1970 to 31 December 1979 and were vested in union and company benefit programmes. Observed numbers of deaths were compared with expected num- bers based on two standards, the proportionate mortality among men in the United States 1970-9 and among men in Erie County 1975. There was close agreement between the number of observed and expected deaths by either standard of comparison among white auto workers in the forge and foundry plants. Valid analyses of cause specific mortality among non-whites could be conducted for the foundry plant only. Although there was raised PMR for deaths due to diseases of the circulatory system using the Erie County standard, none of the other cause specific PMRs was sig- nificant. Although based on small numbers, the risk of cancer of the lung was significantly high in non-whites under age 50 in the foundry (PMR = 2:6; p< 0:05). The cause specific PMRs for whites in the engine plant were statistically significant for malignant neop- lasms (1:2) and all external causes (0:62) based on the U.S. white male standard. Analysis of cancer specific mortality among white men in the machining/assembly plant showed significant excesses for cancer of the digestive system (PMR = 1:5), particularly of the liver (PMR = 2:6) and pancreas (PMR = 1:9); cancers of the respirat- ory system (PMR = 1:4 using the Erie County standard); and cancer of the urinary bladder (PMR = 2:3). Workers employed for more than 20 years showed statistically increased mortality ratios for cancers of the digestive system (1:9), particularly cancer of the pancreas (2:3) and cancer of the rectum (2:8). Individuals whose employment began during or before 1950 exhibited increased PMRs for cancers of the digestive organs (1:8), particularly of the pancreas (2:5) and of the bladder (3:4). Workers whose employ- ment began after 1950, on the other hand, exhibited raised PMRs for cancers of the respiratory system (1:5) and of the kidney (3:2). Since the foundry and forge plants did not start production until 1955, mortality associated with those work settings may be greater in the future. Vernon, S.W.; Tilley, B.C.; Neale, A.V_; et al. Ethnicity, Survival, and Delay in Seeking Treatment for Symptoms of Breast Cancer Cancer 55:1563-1571, 1985 322 Subjects Treatment; Cancer Site: breast; Cancer Measures: survival Study Characteristics Design: retrospective; Measures: descriptive statistics, higher level statistics, observed survival rate; Time Frame: 1949-1968 Sample Characteristics 5,792 breast cancer cases (blacks 912; whites 4,618; Hispanics 262): female; Houston, TX Variables SES: ability to pay; Demographic: age, race-ethnicity; Cancer: stage Data Sources Hospital medical records, U.S. Population Census Abstract This study examined differences in 10-year survival rates from breast cancer among white, black, and Hispanic women controlling for the effects of age, socioeconomic status (SES), stage of disease, and delay in seeking treatment for symptoms. Breast cancer pa- tients (n = 1983) treated at M. D. Anderson Hospital and Tumor Institute in Houston, Texas between 1949 and 1968, were followed for 10 years. Ethnicity, SES, stage of disease, and delay were all found to affect survival when considered separately. Black patients were less likely to survive than either white or Hispanic patients whose survival experience appeared to be similar. Multivariate analysis that used a Cox regression technique showed that ethnic differences remained when age, SES, stage, and delay were in- cluded in the model. In contrast, the authors could not detect an effect of delay on survival when ethnicity and all other variables were included. These data suggest that ethnic differences in breast cancer survival are not mediated by differences in delay in seeking treatment for breast cancer symptoms. 131 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Vianna, N.J.; Thind, 1.S.; Louria, D.B.; et al. Epidemiologic and Histologic Patterns of Hodgkin's Disease in Blacks Cancer 40:3133-3139, 1977 323 Subjects Risk Factor Exposure: occupation; Cancer Site: lymphoma; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics, relative risk; Time Frame: 1950-1974 Sample Characteristics Hodgkin's disease mortality data: Newark, NJ; New York; South Africa Variables SES: census tract income, occupation; Demographic: age, sex, race-ethnicity (black, white) Abstract The epidemiologic pattern for Hodgkin's disease in blacks from two different communities in the United States was characterized by higher childhood rates and significantly lower rates in the young adult and older age groups than for whites. In addition, a signific- antly greater number of black patients belonged to low occupa- tional groups. These observations and the different epidemiologic patterns for blacks in Southern Transvaal, South Africa, and other countries suggest that the natural history of Hodgkin's disease might be strongly influenced by social milieu. The variability in Rye subtype distribution, particularly for whites and blacks in the young adult and other age groups raise the possibility that age related environmental factors might be important in the histologic reactivity of the host. Vos, G.H. Anti-Fab’ Antibodies in Human Sera Vox Sang 33:16-20, 1977 324 Subjects Cancer Site: all sites combined; Biologic Marker: Fab antibodies; Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 1,685 sera (Negro blood donors 245; Negro cancer patients 500; Negro patients 500; Caucasian blood donors 440): 18-75 years old; South Africa Variables Demographic: age, sex, number of pregnancies Data Sources Laboratory test, hospital medical records Abstract Antibodies to the Fab fragments of human IgG are commonly found in human sera. This study shows that the incidence of such antibodies is higher in women, both blood donors and hospital patients, than in men: it is no higher in patients with malignant disease than in hospital patients of the same sex without malignan- cies. The incidence in hospital patients is always greater than in healthy blood donors. The presence of Fab antibodies is not related to pregnancy. The greatest difference in the frequency of such antibodies occurred between males and females under the age of 50. Vos, G.H.; Brain, P. Heterophile Antibodies, Immunoglobulin Levels, and the Evaluation of Anti-T Activity in Cancer Patients and Controls S Afr Med J 60:133-136, 1981 325 Subjects Cancer Site: cervix uteri, breast, esophagus; Biologic Marker: immunoglobulin, ABD blood group Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 418 individuals (esophagus cancer cases 95; controls 323): black, Zulu, male; South Africa—357 individuals (breast cancer cases 103; cervix uteri cancer cases 54; controls 200): Indian ancestry, female; South Africa Data Sources Laboratory test Abstract In a study of human agglutinin response to multideterminant ABH-like antigens, such as sheep red blood cells (SRBC), a signif- icant correlation was found between the presence of high antibody levels to such red cells (titres of 1:20 and higher) and total serum concentrations of IgM and IgA. Individuals with lower levels of SRBC agglutinins (titres of 1:10 and lower) generally have reduced concentrations of these two classes of immunoglobulin. High or low levels of antibody to SRBC did not correlate with the develop- ment of several kinds of malignant tumour (cervix uteri, breast and oesophagus). It was found, however, that producers of high levels of agglutinins for SRBC, whether cancer patients or normal subjects, also more often produce higher levels of antibodies to the cryptantigenic T (Thomsen-Friedenreich) - antigen structure of human red cells. The only significant difference is that cancer patients have persistently depressed levels of anti-T. This provides further evidence that malignant tissues may have exposed T-anti- gen structures on their membrances. 132 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Walker, A.R.P.; Walker, B.F.; Isaacson, C.; et al. Short Duration of Survival Among South African Blacks with Oesophageal Cancer S Afr Med ] 66:877-878, 1984 326 Subjects Cancer Site: esophagus; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics; Time Frame: 1982 Sample Characteristics 183 esophagus cancer cases (males 146; females 37): black, 23-75 years old, hospital patients; Soweto, South Africa Variables Demographic: age, sex Data Sources Hospital medical records, survey Abstract South African urban Blacks are very prone to oesophageal cancer, and mean age at diagnosis is almost a decade lower than that of White patients (male preponderance is equally marked in both ethnic groups.) In recent years the socio-economic position of urban Blacks has improved greatly, with considerable strides being made in literacy and awareness of hospital services. This socio-economic improvement could conceivably improve the ex- tremely short durations of survival among oesophageal cancer patients noted in earlier studies; this, however, has not occurred. Of a series comprising 146 male and 37 females patients, 50% had died after 3,6 and 4,2 months respectively. The corresponding periods noted for White patients are about twice as long. Walker, A.R.P.; Walker, B.F.; Tshabalala, E.N., et al. Low Survival of South African Urban Black Women With Breast Cancer Br |] Cancer 49:241-244, 1984 327 Subjects Treatment; Risk Factor Exposure: diet-nutritional factors, reproductive factors; Cancer Site: breast; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude incidence rate, crude mortality rate, crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1971-1982 Sample Characteristics 127 breast cancer cases (Period 1: 31; Period 2: 96): black, female, hospital patients; Johannesburg, South Africa Variables SES: social class; Demographic: age, geographic location, age first pregnancy, parity; Cancer: stage Data Sources Hospital medical records, death registration Abstract Two series of black South African female breast cancer cases are examined to shed light on the low survival of these patients. Hypotheses, dietary and reproductive, are presented for the lower breast cancer incidence in S.A. black females compared to whites. The time periods covered by the study are: 1971-1972 (N=31) and 1980-May 1982 (N=96). Stage at diagnosis was generally advanced in these patient series, and 50% of patterns in the two series were dead within 1.1 to 1.4 years. Further prospective studies are strongly recommended to shed light on this subject. Walrath, J.; Fraumeni, J.F., Jr. Cancer and Other Causes of Death Among Embalmers Cancer Res 44:4638-4641, 1984 328 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: prospective (nonconcurrent); Measures: proportionate mortality ratio (PMR), descriptive statistics, higher level statistics; Time Frame: 1925-1980 Sample Characteristics 1,109 deaths (nonwhites 44; whites 1,065): embalmers; California Variables Demographic: sex, race-ethnicity Data Sources Death registration, industry records Abstract To evaluate the potential carcinogenic effects of formaldehyde, 133 we examined the proportionate mortality experience of embalmers licensed to practice in California. Mortality was significantly ele- vated for total cancer, arteriosclerotic heart disease, and suicide, whereas significant deficits were noted in mortality from diseases of the respiratory and genitourinary systems. Deaths from cancers of the brain, colon, and prostate and leukemia were significantly higher than expected. No increased mortality was seen for cancers of the respiratory tract, including the nasal passages, where an effect might be expected based on animal studies. A parallel mor- tality survey of embalmers from New York State showed similar findings, with excesses of brain tumors, leukemia, colon cancer, arteriosclerotic heart disease, and cirrhosis. Further investigation is needed to determine whether any of these outcomes is related to formaldehyde exposure. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Walters, T.R.; Bushore, M.; Simone, J. Poor Prognosis in Negro Children with Acute Lymphocytic Leukemia 329 Cancer 29:210-214, 1972 Subjects Treatment; Diagnosis; Cancer Site: leukemia; Cancer Measures: incidence, survival Study Characteristics Design: prospective (nonconcurrent); Measures: descriptive statistics; Time Frame: 1962-1969 Sample Characteristics 334 leukemia cancer cases (Negroes 46; Caucasians 288): under 18 years old; Memphis, TN Variables SES: occupation; Demographic: age, sex, race-ethnicity; Cancer: histology Data Sources Hospital medical records Abstract This study of 334 children with acute lymphocytic leukemia was undertaken to compare the clinical course and response to chemotherapy of Negro children with those of Caucasian children. An initial hematologic remission was attained in 34 of 46 Negro children (74%) with a median duration of the first hematologic remission of 4.9 months and median survival of 14 months. In Caucasian children, a remission occurred in 265 of 288 (92%) with a median duration of 13.1 months and a median survival of 23 months. The poor prognosis in Negro children was related to advanced disease and poverty. However, the natural history of leukemia in the Negro children suggested that other factors related to race also influence host-disease interaction. Because of the dif- ference in response, Negro children should be considered sepa- rately in designing and evaluating chemotheraphy programs. Walther, R.R.; Grossman, M.E.; Troy, J.L. Basal-Cell Carcinomas on the Scalp of a Black Patient Many Years After Epilation by X Rays 330 J Dermatol Surg Oncol 7(7):570-571, 1981 Subjects Treatment; Risk Factor Exposure: treatment modalities; Cancer Site: skin; Cancer Measures: incidence Study Characteristics Design: case study; Time Frame: 1981 Sample Characteristics One skin cancer case: black, male, 45 years old; New York Data Sources Hospital medical records Abstract The development of skin cancers in areas treated with X rays for benign dermatoses has not been common. Moreover, basal-cell carcinomas in blacks are exceedingly rare. We report successive development of two basal-cell carcinomas in the hair-covered scalp of a black man three decades after he had undergone epilation of the scalp by X rays for tinea capitis while a youth. Walton, L.A.; Kernodle, W., Jr.; Hulka, B. Factors Influencing the Occurence of Advanced Cervical Carcinoma 331 South Med J 72(7):808-811, 1979 Subjects Screening: Pap smear; Cancer Site: cervix uteri Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1975-1977 Sample Characteristics 170 cervix uteri cancer cases with advanced disease (blacks 98; whites 70; Indians 2): female, hospital patients; Chapel Hill, NC Variables SES: income; Demographic: age, race-ethnicity, geographic area; Cancer: stage, symptoms Data Sources non-SEER hospital cancer registry Abstract The continued occurrence of advanced cervical carcinoma (stage II, III, IV) was studied. Patients were evaluated with regard to age, racial origin, socioeconomic status, geographic distribution, and presenting symptoms. Particular emphasis was placed on the role of screening in disease detection. The disturbing finding was that of the 170 patients reviewed, disease was diagnosed by screen- ing examination in only 20 patients. Furthermore, 62 patients were not diagnosed even though they were exposed to the health care system. Factors which influence obtaining Pap smears are dis- cussed, with suggestions as to how to enhance use of the Pap smear. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Walzer, P.D.; Perl, D.P.; Krogstad, D.J.; et al. Pneumocystis Carinii Pneumonia in the United States: Epidemiologic, Diagnostic, and Clinical Features Natl Cancer Inst Monogr 43:55-63, 1976 332 Subjects Treatment; Cancer Site: leukemia, lymphoma; Cancer Measures: prevalence Study Characteristics Design: prospective; Measures: descriptive statistics; Time Frame: 1967-1970 Sample Characteristics 194 pneumocystis carinii pneumonia cases (blacks 9; whites 156; other 4) Variables Demographic: age, sex Data Sources Histopathology records, death registration, non-SEER population-based cancer registry Abstract Analysis of 194 patients with confirmed Pneumocystis carinii pneumonia in the United States over a 3-year period revealed that P. carinii pneumonia occurred almost exclusively in the im- munosuppressed host who had a serious underlying disease. The epidemiologic features of pneumocystis pneumonia primarily re- flected those of the underlying disease. P. carinii was identified antemortem in 81% of the cases usually by biopsy or needle aspi- ration of the lung, procedures associated with considerable mor- bidity and mortality. Laboratory identification of P. carinii was usually accurate, but errors resulting from faulty staining technique occurred. Most patients had been ill less than 2 weeks with bilateral diffuse interstitial pneumonia. Leukopenia (white blood cell count <3,100/mm3) and probably severe hypoxia were negative prognos- tic factors. Although treatment with pentamidine was effective, the drug frequently caused adverse reactions, particularly impaired renal function, when given with other nephrotoxic agents. Weaver, 5.M.; Kelly, A.P.; Lopansri, S. Dysplastic Epidermal Keratosis in a Black Woman Arch Dermatol 117:800-803, 1981 333 Subjects Diagnosis; Treatment; Risk Factor Exposure: ultraviolet light, solar radiation, other diseases; Cancer Site: skin Study Characteristics Design: case study; Intervention Study; Time Frame: 1979 Sample Characteristics One Bowen's disease case: black; female, 81 years old; Los Angeles, CA Variables Cancer: histology, stage Data Sources Histopathology records, clinic medical records Providers Physician Abstract Skin cancer is rare in black persons. When it does occur, in most cases it occurs on non-sun-exposed areas. In white persons, most skin cancer occurs in sites that have been exposed to sunlight for long periods. The decreased susceptibility of more darkly pig- mented skin to the sequelae of long-term ultraviolet (UV)-energy exposure probably accounts for the low incidence of actinic keratoses, basal cell carcinomas, and squamous cell carcinomas in black persons. Multiple, dysplastic, epidermal lesions developed in a black woman, which were possibly induced by long-term UV-light or heat exposure. Her condition responded well to topical fluorouracil therapy but had an unusual delayed reaction to treat- ment. Weiss, A.B.; Adams, G.; Brackin, B.; et al. Osteosarcoma: A Review of 50 Patients Treated at the University of Alabama in Birmingham Medical Center Between 1944 and 1975 Clin Orthop 135:137-147, 1978 334 Subjects Treatment; Cancer Site: sarcoma; Cancer Measures: survival Study Characteristics Design: time trend; Measures: crude survival rate, descriptive statistics; Time Frame: 1944-1975 Sample Characteristics 50 osteosarcoma cases (blacks 18; whites 32): hospital patients; Birmingham, AL Variables Demographic: age, race-ethnicity, geographic location; Cancer: tumor location, tumor grade, histology Data Sources Physician medical records, hospital medical records, histopathology records, non-SEER population-based cancer registry 135 Abstract Only 12% of 50 patients lived 5 years or longer. Those who survived the longest were those with the tumor in the distal part of the lower extremity, with a well differentiated or chondroblastic tumor, or those who underwent radical surgery. Patients treated with surgery and chemotherapy lived about 1.5 months longer than those treated with surgery alone and about 1.2 months longer than those who had radiation included in the therapy. For a patient with a diagnosis of osteosarcoma, the prognosis was to be grave, no matter which methods of treatment were used. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Weiss, N.S.; Homonchuk, T.; Young, J.L. Incidence of the Histologic Types of Ovarian Cancer: The U.S. Third National Cancer Survey, 1969-1971 335 Gynecol Oncol 5:161-167, 1977 Subjects Cancer Site: ovary; Cancer Measures: incidence primary ovarian cancer, as well as the variation in that incidence Study Characteristics according to several demographic characteristics, was examined. Design: cross-sectional; Measures: age-specific incidence rate, All types of ovarian cancer except germ cell tumors increased in descriptive statistics; Time Frame: 1969-1971 incidence until the seventh and eighth decades of life, after which Sample Characteristics the rates plateaued. Epithelial tumors were equally common in 4,594 ovary cancer cases: female; TNCS areas young whites and blacks but were substantially more common in Variables whites among middle-aged and older women. The incidence of Demographic: age, race-ethnicity (black, white), marital status; nonepithelial ovarian tumors was similar in the two races. Rates Cancer: histology of epithelial tumors were greatest in single women, whereas Data Sources nonepithelial tumors occurrence showed no correlation with mar- Third National Cancer Survey, 1969-1971, hospital medical ital status. There was little variation among the ovarian epithelial records tumors in their relationship to age, race, and marital status, al- Abstract though true associations may have been obscured because of class- Using the data gathered in the U.S. Third National Cancer ification inconsistencies in the data. Survey, 1969-1971, the incidence of the major histologic types of Weiss, N.S.; Peterson, A.S. Racial Variation in the Incidence of Ovarian Cancer in the United States 336 Am ] Epidemiol 107(2):91-95, 1978 Subjects Abstract Cancer Site: ovary; Cancer Measures: incidence The incidence of ovarian cancer was determined in four US Study Characteristics populations of heterogeneous racial-ethnic composition. Japanese, Design: cross-sectional; Measures: age-standardized incidence Chinese, Hispano, and black women had rates of epithelial tumors rate, descriptive statistics, higher level statistics; Time Frame: that were 19-42% lower than those of white women. Though these 1960-1975 differences were due primarily to lower rates of serous and papil- Sample Characteristics lary tumors, all but Chinese women also had a decreased incidence 3,398 ovary cancer cases (blacks 158; whites 2,799; Hispano 217; of mucinous tumors, and Hispano and black women had low rates Japanese 163; Chinese 61): female, under 75 years old; Hawaii; of endometrioid-clear cell malignancy. The incidence of nonepithe- Los Angeles County, CA; New Mexico; San Francisco-Oakland, lial ovarian tumors in the four populations showed little relation- CA ship to race. Variables Demographic: race-ethnicity; Cancer: histology Data Sources non-SEER population-based cancer registry Weiss, N.S.; Young, J.L., Jr.; Roth, G.J. U.S. Third National Cancer Survey, 1969-71 337 J Natl Cancer Inst 58(4):913-915, 1977 Subjects Abstract Risk Factor Exposure: reproductive factors; Cancer Site: ovary; As determined in the U.S. Third National Cancer Survey, 1969- Cancer Measures: incidence 71, the incidence of ovarian cancer was 60-70% higher in women Study Characteristics who had never married than in those who had. This relationship Design: cross-sectional; Measures: age-adjusted incidence rate; was present in both whites and blacks and in all age groups over Time Frame: 1969-1971 25 years. Among ovarian epithelial tumors, those for which the Sample Characteristics incidence rates between never-married and ever-married women 4,468 ovarian cancer cases: female, 25+ years old, TNCS areas differed most were endometrioid and clear cell tumors. The inci- Variables dence of tumors of germ cell and sex cord-mesenchyme origin, Demographic: age, histology, marital status, race-ethnicity however, showed no relationship to marital status. (black, white) Data Sources Third National Cancer Survey 1969-1971 136 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Wilkinson, G.S.; Edgerton, F.; Wallace, H.]., Jr.; et al. Delay, Stage of Disease and Survival from Breast Cancer J Chron Dis 32:365-373, 1979 338 Subjects Diagnosis; Risk Factor Exposure: reproductive factors; Cancer Site: breast; Cancer Measures: survival Study Characteristics Design: prospective (nonconcurrent); Measures: crude survival rate, higher level statistics; Time Frame: 1969-1976 Sample Characteristics 1,784 breast cancer cases: female; New York Variables Demographic: age, menopausal status, marital status, race-ethnicity (nonwhite, white), urban-rural; Cancer: stage Data Sources non-SEER population-based cancer registry Abstract An investigation of 1784 cases of histologically confirmed breast cancer reported to the Lakes Area Regional Tumor Service Registry was conducted to examine the effect of delay, age, availability of physicians, stage of disease and marital status upon length of survival. Significant relationships were found between length of delay and extent of disease, length of delay and length of survival, extent of disease and length of survival as well as between age and survival. Patients residing in areas having fewer than one physician per 4000 population demonstrated poorer survival than patients from areas having one or more physicians per 4000 popu- lation. A regression analysis employing a Cox modeling technique showed stage of disease, length of delay, and age to be significant predictors of survival. Delay by patients was found to affect survi- val through its influence upon the extent of disease at the time of diagnosis. These findings provide qualified support for efforts di- rected toward early detection and treatment. They imply that re- ducing delay may have a favorable impact upon prognosis for a portion of breast cancer patients. Williams, R.R.; Horm, J.W. Association of Cancer Sites With Tobacco and Alcohol Consumption and Socioeconomic Status of Patients: Interview Study From the Third National Cancer Survey J Natl Cancer Inst 58(3):525-547, 1977 339 Subjects Risk Factor Exposure: alcohol, tobacco; Cancer Site: multiple sites; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: relative odds, descriptive statistics, higher level statistics; Time Frame: 1969-1971 Sample Characteristics 7,518 cancer cases (blacks 60; whites 7,330; other races 128): 9 TNCS areas Variables SES: education, income; Demographic: age, sex Data Sources Third National Cancer Survey 1969-1971 Abstract From personal interviews obtained for 7,518 incident cases of invasive cancer from the population-based Third National Cancer Survey, the quantitative lifetime use of cigarettes, cigars, pipes, unsmoked tobacco, wine, beer, hard liquor, and combined alcohol were recorded, as well as education and family income level. In an initial screening analysis of these data, Mantel-Haenszel 2x2 contingency tabulations and multiple regression analyses were used to compare each specific cancer site with controls from other sites to test for associations with the “exposure variables.” Signif- 137 icant positive associations with cigarette smoking were found for cancers of the lung, larynx, oral cavity, esophagus, stomach, pan- creas, bladder, kidney, and uterine cervix. Other forms of tobacco were associated with cancers of the oral cavity, larynx, lung, and cervix. Consumption of wine, beer, hard liquor, and all combined showed positive associations with neoplasms of the oral cavity, larynx, esophagus, colon, rectum, breast, and thyroid gland. Col- lege education and high income both showed positive associations with cancers of the breast, thyroid gland, uterine corpus, and melanomas in males. These same indicators of high socioeconomic status showed inverse associations with invasive neoplasms of the uterine cervix, lung, lip-tongue, and colon in females. College attendance (but not income) showed an inverse association with stomach cancer and positive association with pancreatic cancer in males. Still other tumor sites showed “suggestive” associations with each of these exposure variables. In the analyses producing these results, age, race, sex, smoking, drinking, education, in- come, parity, foreign birth, marital status, and geographic location were used as stratification variables separately or in combination when appropriate to assess and control for their potentially con- founding affects and to examine results in different strata to assess interaction. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Wilson, R.E.; Donegan, W.L.; Mettlin, C.; et al. The 1982 National Survey of Carcinoma of the Breast in the United States by the American College of Surgeons 340 Surg Gynecol Obstet 159(4):309-318, 1984 Subjects Treatment; Cancer Site: breast; Cancer Measures: survival for patients diagnosed in 1972 and 1977 were compared for indica- Study Characteristics tions of long term changes in patterns. The 1982 survey Design: prospective, time trend; Measures: crude survival rate, documented changes in the management of carcinoma of the breast descriptive statistics, higher level statistics; Time Frame: prior in this country. If one compares the current survey results with to 1981 the results of the 1978 survey, an inescapable conclusion is that Sample Characteristics there is a changing understanding of carcinoma of the breast and 26,551 breast cancer cases (blacks 2,253; whites 23,902; others a willingness to apply new knowledge to clinical practice. These 396) and 19,131 breast cancer cases (blacks 1,523; whites 17,249; results also show the potential of the hospital Tumor Registry others 359): female, hospital patients; United States when records are systematically aggregated and data from different Variables points compared. Future applications of this survey procedure and Demographic: age, race-ethnicity; Cancer: stage, nodal further analyses of these data may be used to document progress involvement in the control of cancer. While useful for these purposes, it should Data Sources be noted that these data are not intended, nor suited, for assessing survey, SEER cancer registry the superiority of given treatments. They may not reflect the out- Abstract come that would result from a clinical trial. Comparisons made The article presents the results of a recent survey of the hospital herein often involve patients who differed in many respects other records of women with carcinoma of the breast with respect to than that on which the comparison was focused. However, within the characteristics of the patient, disease, treatments, and outcome. these constraints, the data may be useful for clinicians and scien- The data came from several hundred hospitals and account for tists interested in the dynamic state of the treatment of carcinoma about one-fifth of the incident cases of carcinoma of the breast in of the breast in large numbers of hospitals in the United States. the United States. Data from similar surveys conducted in 1978 Wogalter, H.; D’Esposito, R.F. Letters to the Editor: Malignant Leydig Cell Tumor in a 94-year-old Man 341 J Urol 122:428, 1979 Subjects Abstract Cancer Site: testis While testicular tumors are rare in blacks, Leydig cell tumors, Study Characteristics primary lymphoma of the testis and granulomatous orchitis, are Design: case study most frequently reported in the black population. Sample Characteristics One testis cancer case: black, male, 94 years old Wolpowitz, A.; Van Heerden, J.; Punt, A.M; et al. Carcinoma of the Oesophagus 342 S Afr Med J 56:1043-1044, 1979 Subjects Abstract Cancer Site: esophagus; Cancer Measures: incidence A survey of carcinoma of the oesophagus at Kalafong Hospital Study Characteristics revealed a greater incidence of the disease in males, being highest Design: cross-sectional; Measures: descriptive statistics; Time in the age group 50-59 years, with the middle third of the Frame: 1972-1977 oesophagus the most prevalent site and with oesophagotracheal Sample Characteristics or oesophagobronchial fistulae a common complication. 530 esophagus cancer cases (males 440; females 90): presumed black, hospital patients; South Africa Variables Demographic: age, sex; Cancer: tumor location Data Sources Hospital medical records 138 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Wyndham, C.H. A Comparison of Mortality Rates From Cancer in White, Indian and Coloured Adults in 1970 and 1980 S Afr Med J 67:709-711, 1985 343 Subjects Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: time trend; Measures: age-adjusted mortality rate Sample Characteristics Cancer mortality data: Indians, coloureds, 25-74 years old; South Africa Variables Demographic: sex, race-ethnicity Data Sources Death registration Abstract In 1980 lung cancer was associated with the highest mortality rate (MR) in white, Indian and coloured men in the RSA. MRs for this type of cancer were higher in 1980 than in 1970 in all three populations, the increase of 47% in that for coloured men being especially marked. On the other hand, MRs for stomach cancer were lower in 1970 than in 1980 in men from all three populations; however, this type of cancer was still associated with the second highest MR in all groups in 1980. Breast and lung cancer had the highest MRs in white women. In this group there was little change in the MR for breast cancer, but a 27% increase in that for lung cancer over the 10-year period. The three types of cancer associated with the highest MRs in coloured women were cancer of the cervix, breast and stomach. In this group the MR for breast cancer dropped by 22% and that for lung cancer increased by 135% over the 10-year period. In Indian women MRs for cancer of the stomach, breast and cervix ranked highest, a marked increase occurring only in the MR for breast cancer. The MR for lung cancer in Indian women showed a decrease of 71% in 1980 compared with 1970. Wyndham, C.H. Comparison and Ranking of Cancer Mortality Rates in the Various Populations of the RSA in 1970 S Afr Med ] 67:584-587, 1985 344 Subjects Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cross-sectional; Measures: age-adjusted mortality rate; Time Frame: 1970 Sample Characteristics Cancer mortality data: blacks, coloureds, Indians, whites, 25-74 years old; South Africa Variables Demographic: age, sex, race-ethnicity Data Sources Death registration Abstract Age-adjusted mortality rates (MRs) in all four population groups in the RSA (age range 25 - 74 years) for different types of cancer were compared and ranked. Lung and stomach cancer had the highest MRs in white, Indian and coloured males. In white males lung cancer ranked 1st (MR more than twice as high as that for stomach cancer), while in Indian and coloured males stomach cancer ranked 1st and lung cancer 2nd. The MR for lung cancer in coloured males was a little higher than that in white males. In black males oesophageal cancer ranked 1st and liver cancer 2nd. In white females breast cancer ranked 1st and lung cancer 2nd. In coloured females cancer of the cervix ranked 1st followed by cancer of the breast and of the stomach. In black females cancer of the oesophagus and of the liver ranked 2nd and 3rd after cancer of the cervix, and in Indian females the rank order was stomach cancer 1st, breast cancer 2nd, and cervical cancer 3rd. 139 EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Yatani, R.; Chigusa, I.; Akazaki, K.; et al. Geographic Pathology of Latent Prostatic Carcinoma Am J Cancer 29:611-616, 1982 345 Subjects Risk Factor Exposure: geographic area; Cancer Site: prostate; Cancer Measures: prevalence Study Characteristics Design: cross-sectional; Measures: age-adjusted incidence rate, descriptive statistics, higher level statistics; Time Frame: 1965-1979 Sample Characteristics 1,606 prostate glands (Louisiana blacks 178; Louisiana whites 253; Columbians 182; Hawaiian Japanese 417; Japanese 576): males, 50+ years old, no prior prostate surgery or diagnosis of prostate carcinoma; New Orleans, LA; Japan; Columbia; Hawaii Variables Demographic: age, race-ethnicity, geographic area; Cancer: histology, tumor size Data Sources Autopsy records, laboratory test Abstract Prostates obtained at autopsy from black and white males in the United States of America, from Colombians, from Japanese migrants in Hawaii, and from Japanese in Japan (all over 50 years old) were serially step-sectioned and examined microscopically using identical techniques and diagnostic criteria. The age-adjusted overall prevalence of latent carcinoma was significantly higher in US blacks (36.9%), in US whites (34.6%) and in Colombians (31.5%) than in Japanese in Japan (20.5%). There was no significant differ- ence in prevalence between Japanese immigrants in Hawaii (25.6%) and Japanese in Japan. When the carcinomas were subdivided into latent infiltrative stage (LIT) and latent non-infiltrative type (LNT), the LIT component reflected upon the overall prevalence of latent carcinoma. There was an increase in the overall prevalence of latent carcinoma and in the prevalence of LIT tumors with age in Japanese migrants to Hawaii and in Japanese of Japan but only a suggestive increase in blacks and whites in the United States. There was no significant relation between age and prevalence in Colombia. In LNT tumors, there was no consistent trend related to race or age. The size of LIT tumors was significantly greater than that of LNT tumors in all races or in each decade of total material. There were many large LIT tumors in blacks and whites in the United States. These results suggest that the prevalence of LIT tumors shows a race variation similar to the incidence of clinical carcinoma of the prostate, and that LNT tumors probably remain latent during an individual's life span. Young, J.L., Jr.; Devesa, S.S.; Cutler, S.J. Incidence of Cancer in United States Blacks Cancer Res 35:3523-3536, 1975 346 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, mortality Study Characteristics Design: cross-sectional, time trend; Measures: crude incidence rate, age-adjusted incidence rate, age-specific incidence rate, crude mortality rate; Time Frame: 1935-1971 Sample Characteristics 15,183 cancer cases: TNCS areas (excluding Minneapolis-St. Paul, MN; Colorado; and Iowa); cancer incidence data: Nigeria, Rhodesia, and South Africa; cancer mortality data: United States Variables Demographic: age, sex, race-ethnicity (black, white), geographic area Data Sources Third National Cancer Survey 1969-1971 140 Abstract Incidence rates for the black population of six Standard Met- ropolitan Statistical Areas in the United States are examined using data collected in the Third National Cancer Survey, 1969 to 1971. For all sites combined, black males had the highest rates among the four major race-sex groups; black females had the lowest rates. For fourteen common sites accounting for 80% of the cancers among blacks, black/white ratios, survival data, trends between 1935 and 1969, and geographic variation are presented. United States black data adjusted to an African Standard are compared with similar data from Nigeria, Rhodesia, and South Africa. EPIDEMIOLOGY: DESCRIPTIVE/ANALYTIC Young, J.L.; Ries, L.G.; Pollack, E.S. Cancer Patient Survival Among Ethnic Groups in the United States J Natl Cancer Inst 73(2):341-352, 1984 347 Subjects Cancer Site: multiple sites; Cancer Measures: survival Study Characteristics Design: cross-sectional; Measures: relative survival rate, crude survival rate, higher level statistics; Time Frame: 1973-1979 Sample Characteristics 402,752 cancer cases (blacks 30,253; Anglos 350,302; Hispanics 8,622; American Indians 1,264; Chinese 3,048; Japanese 5,030; Filipinos 2,355; Hawaiians 1,878): SEER Program areas. Variables Demographic: age, sex, race-ethnicity; Cancer: stage Data Sources SEER cancer registry Abstract Data from the Surveillance, Epidemiology and End Results Program of the National Cancer Institute were used in the evalu- ation of cancer patient survival experience for 8 racial-ethnic groups in the U.S. population — Anglos, Hispanics, blacks, American Indians, Chinese, Japanese, Filipinos, and Hawaiians. This report contains an actuarial (life table) survival analysis on 402,752 pa- tients with first primary cancer diagnosed in 1973-79 and followed through December 31, 1981. Relative and observed survival rates are shown for each sex separately by primary site. Survival rates for females exceeded those for males for each primary site except for gallbladder and urinary bladder. The primary site having the highest survival rate among each group studied was the thyroid gland with a 5-year relative rate of 91% for all races combined. Rates were uniformly low among each group for cancers of the esophagus, liver, and pancreas. Survival rates for Hispanics were almost identical to those for Anglos; the largest differences were found for bladder (73% for Anglos vs. 64% for Hispanics), Hodgkin's disease (70% for Anglos vs. 61% for Hispanics), and ovary (35% for Anglos vs. 42% for Hispanics). For many primary sites Japanese experience the highest survival rates and American Indians the lowest. In comparison to Anglos, Japanese had higher survival for cancers of the stomach, colon, prostate gland, and breast. The higher survival for stomach and breast cancers per- sisted even when controlling for age of patient and stage of disease at diagnosis. Survival rates for blacks were much lower than those for whites for cancers of the corpus uteri and urinary bladder. Survival rates for Chinese and Hawaiians were roughly comparable to those for Anglos, whereas survival rates for Filipinos tended to be similar to those for blacks. Zimmerman, R.R.; Kung'u A. Testicular Neoplasms in Kenyan Africans Cancer 41:2452-2455, 1978 348 Subjects Cancer Site: testis Study Characteristics Design: cross-sectional; Measures: crude incidence rate; Time Frame: 1968-1976 Sample Characteristics 40 testis cancer cases: black, male; Kenya Variables Demographic: age; Cancer: histology Data Sources Hospital medical records 141 Abstract Over a 9-year period 40 testicular and paratesticular neoplasms were seen at the Kenyatta National Hospital, Nairobi, Kenya. Their incidence rate was 0.08 per annum per 100,000 Kenyan males. This low incidence was largely accounted for by a decrease in tumors of germ cell origin. The proportional distribution of the testicular neoplasms, however, was not significantly different from findings in the United States. An inheritable factor apparently controls the decreased susceptability to testicular neoplasms. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Adam, E.; Sanders, E.K.; Melnick, ]J.L.; et al. Antibodies to Herpesvirus Type 2 in Breast Cancer and Cervical Cancer Patients Cancer 330:147-152, 1974 349 Subjects Biologic Marker: herpesvirus type 2 antibodies; Risk Factor Exposure: other diseases, sexual practices, reproductive factors; Cancer Site: breast, cervix uteri Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics Sample Characteristics 279 individuals (breast cancer cases 43; cervix uteri cancer cases 50; controls 186): Negro, female, low income; Harris County, TX Variables Demographic: age, age at first pregnancy, parity, age at first marriage, number of marriages Data Sources Laboratory test, survey, hospital medical records Abstract Antibody titers to herpesvirus Types 1 and 2 were measured in sera from 43 patients with breast cancer, 50 patients with cervical cancer, and 186 women without malignant or gynecologic disease. All were Negroes in the lower socioeconomic class. The three groups of women were compared with respect to age at first inter- course, age at first marriage, age at first pregnancy, number of marriages, number of sex partners, and number of live births. The patients with breast cancer differed significantly from the other women only in being slightly older at first marriage. The distribu- tion of antibodies to the herpesviruses was similar for breast cancer patients and control women; however, the antibody distribution was different for breast cancer patients when compared to cervical cancer patients. The results of the study support the view that the high frequency of herpesvirus Type 2 antibodies among cervical cancer patients is not due solely to increased sexual promiscuous- ness in this group of women. Alfred, L.]J.; Venkatesan, N.; Mandal, A.K; et al. Release of Lymphotoxin by Control and Chemical Carcinogen-Treated Lymphocyte Cultures Derived from Black Healthy Subjects and Cancer Patients J Natl Med Assoc 74(8): 775-781, 1982 350 Subjects Cancer Site: lung, head and neck Study Characteristics Design: experimental; Measures: descriptive statistics, higher level statistics Sample Characteristics 55 individuals (lung, head and neck cancer cases 33; healthy volunteers 22): Los Angeles, CA Data Sources Laboratory test, hospital medical records Abstract In an age-adjusted comparison with white men, black men have a significantly higher increase in esophageal and other types of cancer associated with environmental causes. The basis of this increase in cancer rates in blacks over the last two decades is unknown. Since cancer patients generally show an impairment in cell-mediated immune (CMI) functions, we measured certain CMI reactions in cultured lymphocytes derived from black healthy sub- jects and cancer patients. We also determined the levels of aryl 142 hydrocarbon hydroxylase (AHH) induced in these lymphocytes. AHH catalyzes the activation of polycyclic aromatic hydrocarbons (PAH) to intermediates which might alter CMI functions. Lympho- cytes from 33 black patients with squamous cell carcinoma and 22 healthy volunteers were mitogen-activated with phytohemaggluti- nin and concurrently treated with the environmental carcinogen, 3-methylchol- anthrene (MCA). Measurements were made of the effects of MCA (0.5 to 3.0 micro-moles) on blastogenesis, T-cell growth, lymphotoxin (LT) release, and AHH induction in these lymphocyte cultures. MCA treatment depressed blastogenesis but had no depressive effect on T-cell growth in cultures. Blastogenesis and T-cell growth were mitogen dose-dependent, while LT release was independent of mitogen concentration. There was a signific- antly lower LT release by lymphocytes from lung cancer patients, compared to those from healthy and head/neck cancer subjects. The reduced levels of LT release in lung cancer patients might reflect an impairment in this CMI function. Studies on the role of lymphocyte subpopulations in CMI functions are in progress. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Austin, H.; Cole, P.; Wynder, E. Breast Cancer in Black American Women Int J] Cancer 24:541-544, 1979 351 Subjects Risk Factor Exposure: breastfeeding, reproductive factors; Cancer Site: breast; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1969-1975 Sample Characteristics 444 hospital patients (breast cancer cases 127; controls 317— Premenopausal 203; postmenopausal 241): black, female, 30 + years old; New York City, NY Variables SES: education; Demographic: age at menarche, age at first birth, menopausal status Data Sources Hospital medical records, survey Abstract A case-control study of breast cancer among Black American women was conducted in seven hospitals in New York City from 1969 to 1975. Results are reported for 127 cases and 317 controls. Compared to women with a first birth before age 19, those with a first birth after 25 had a relative incidence rate for breast cancer of 3.8 and 2.2 for the pre- and postmenopausal age-groups, respec- tively. Compared to nulliparous women, parous women had a relative incidence rate of 0.6 for premenopausal and 0.7 for post- menopausal women. The incidence rate of breast cancer for women with a menopause after age 49 was estimated to be 3.1 times that of women with a menopause before age 45. Thus, the known risk factors for breast cancer among Whites are also related to the etiology of the disease among Blacks. The incidence rate of breast cancer has increased among younger Blacks since 1947 and is now similar to that among younger Whites. However, among older women, the incidence rate is still appreciably higher for Whites. The most likely explanation of this pattern is that Black women born since about 1925 are being exposed at the same frequency as White women to the causes of breast cancer. Blot, W.].; Davies, J.E.; Brown, L.M.; et al. Occupation and the High Risk of Lung Cancer in Northeast Florida Cancer 50:364-371, 1982 352 Subjects Risk Factor Exposure: occupation, tobacco; Cancer Site: lung; Cancer Measures: mortality, incidence Study Characteristics Design: cross-sectional, case-control; Measures: relative risk, age-adjusted mortality rate, descriptive statistics, higher level statistics; Time Frame: 1970-1979 Sample Characteristics 428 hospital patients (blacks 90; whites 338—Lung cancer cases 152; controls 276) and 327 deaths (blacks 78; whites 249— Lung cancer deaths 169; controls 158): male; northeast Florida; lung cancer mortality data from Florida and the United States Variables SES: education, industry; Demographic: age, race-ethnicity, geographic area; Cancer: histology Data Sources Hospital records, U.S. Population Census, death registration Abstract A case-control study involving interviews with 321 male pa- tients with lung cancer and 434 controls, or their next of kin, was undertaken to identify reasons for the high lung cancer mortality along the northeast coast of Florida. In Duval county (Jacksonville), the age-adjusted rate for lung cancer, 1970-1975, among white males was the highest of all urban counties in the United States. Increased risks on the order of 40-50% were associated with em- ployment in the shipbuilding, construction, and lumber/wood in- dustries, particularly among workers with reported exposures to asbestos or wood dust. Excess risks were also linked to fishing and forestry occupations, although the numbers of cases involved were small. Occupational factors did not appear to fully account for the area-wide excess of lung cancer, but no evidence was found to implicate smoking habits, migration patterns, or diagnostic and reporting practices as factors responsible for the exceptional mor- tality rates. 143 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Bradshaw, E.; Schonland, M. Oesphageal and Lung Cancers in Natal African Males in Relation to Certain Socio-Economic Factors Br J Cancer 23(2):275-284, 1969 353 Subjects Risk Factor Exposure: diet-nutritional factors, occupation, alcohol, tobacco; Cancer Site: esophagus, lung; Socioeconomic Factors Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1964-1966 Sample Characteristics 484 individuals (esophagus cancer cases 98; lung cancer cases 45; controls 341): black, male; South Africa Variables SES: education, occupation, reads English, age first wore shoes; Demographic: age Data Sources Survey, hospital medical records Abstract In view of the high incidence of oesophageal and lung cancer in African males of Durban, an investigation of certain socio- economic factors has been undertaken. Interviews on 98 male Af- rican oesophageal cancer patients, 45 male African lung cancer patients and 341 male African patients not suffering from malig- nant disease were undertaken at a large Durban hospital. Although the two cancer groups used emetics and purgatives more fre- quently than the control group, it is possible that this finding was connected with the presence of a chronic illness, rather than with cancer causation. Nevertheless this survey indicates that native remedies are still frequently taken. The regular use of emetics represents a type of oesophageal insult. The consumption of al- coholic concoctions was more frequent in both cancer groups than in the control group, with the lung cancer group drinking more concoctions than the oesophageal cancer group. On considering the use of tobacco, it was found that both oesophageal and lung cancer groups had more tobacco users than the control group. Analysis of the type of tobacco used indicated that oesophageal cancer cases smoked pipes more than the other groups, and lung cancer cases smoked cigarettes more than the other groups. Both oesophageal and lung cancer groups had an excess of males who had smoked for over 30 years in comparison with the control group. Both cancer groups had a higher lifetime consumption of tobacco than the control group, and the oesophageal cancer group had the highest lifetime consumption. It was found that both oesophageal and lung cancer groups had been exposed to more possible occupational carcinogens and for longer periods than the control group. An analysis of the possible carcinogens involved indicated that petrol, oil and tar were the commonest carcinogens associated with the lung cancer group, and asbestos and lead were most commonly associated with the the oesophageal cancer group. The authors feel that the connection between smoking and lung cancer in African males has been established, in previous papers, and in this paper. It is suggested by the authors that a variety of oesophageal insults, including those which emerge from the study, taken in conjunction with a general state of malnutrition in African males, may combine to promote the development of oesophageal cancer. Connection with a specific carcinogen has yet to be estab- lished. Browne, R.M.; Camsey, M.C.; Waterhouse, ]J.A.H.; et al. Etiological Factors in Oral Squamous Cell Carcinoma Community Dent Oral Epidemiol 5:301-306, 1977 354 Subjects Risk Factor Exposure: occupation, alcohol, tobacco; Cancer Site: oral cavity Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1957-1971 Sample Characteristics 225 individuals (oral squamous cell carcinoma cases 75; controls 150): 28-90 years old; County Borough of Stoke-on-Trent, England Variables SES: occupation; Demographic: sex; Cancer: histology, tumor location Data Sources non-SEER population-based cancer registry, survey Abstract A retrospective survey of 75 patients with oral squamous cell carcinoma (ICD Nos. 143, 144, 145) resident in the County Borough of Stoke-on-Trent, England, and 150 controls has been carried out by interview. The controls were matched for age, sex, occupation and place of residence. There was no difference in the prevalence of duration of denture wearing, although male controls had their dentures remade more frequently. Male patients practised oral hygiene procedures less frequently when they had their own teeth than controls. Habitual beer-drinking was more common and greater quantities were drunk by male patients than controls. To- bacco chewing, which was restricted to miners, was equally com- mon (45%) in the two groups. Among miners, the combined habit of tabacco chewing and pipe smoking was more common among patients (100%) than among controls (25%). 144 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Dales, L.G.; Friedman, G.D.; Ury, HK; et al. A Case-Control Study of Relationships of Diet and Other Traits to Colorectal Cancer in American Blacks Am ] Epidemiol 109(2):132-144, 1979 355 Subjects Risk Factor Exposure: diet-nutritional factors, tobacco, other diseases; Cancer Site: colon-rectum Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1973-1976 Sample Characteristics 379 individuals (colon-rectum cancer cases 99; hospital controls 189; multiphasic screening controls 91): black; San Francisco, CA Variables Demographic: age, sex Data Sources Survey, hospital medical records Abstract Ninety-nine black colorectal cancer patients and 280 matched controls from hospitals and multiphasic health checkup clinics were interviewed about past dietary habits and other traits. The colon cancer cases tended to report less frequent use of foods with atleast 0.5% fiber content than did their controls. This relationship, though small, showed a consistent dose-response gradient, ap- peared in both case-hospital control and case-multiphasic health checkup control comparisons, and could not be accounted for by the effects of other variables. Colon and rectosigmoid junction cancer patients tended to have eaten foods with at least 5% satu- rated fat somewhat more often than controls. When consumption of these two groups of foods was considered in combination, sig- nificantly more colon cancer patients than controls reported a high saturated fat foods-low fibrous foods eating pattern, as opposed to a low saturated fat foods-high fibrous foods diet. Statistically significant excesses of the following traits were also reported by the colorectal cancer patients: prolonged cigar smoking in men, nulliparity in women, and history of colorectal polyps. Delzell, E.; Monson, R.R. Mortality Among Rubber Workers: X. Reclaim Workers Am J Ind Med 7:307-313, 1985 356 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cohort; Measures: crude mortality rate, descriptive statistics, higher level statistics; Time Frame: 1940-1978 Sample Characteristics 16,827 individuals (nonwhites 1,184; whites 15,643—Reclaim workers 1,790; nonreclaim workers 15,037): male, rubber manufacturing company employees Variables Demographic: race-ethnicity Data Sources Industry records Abstract This study evaluated the mortality experience of 1,352 white and 438 nonwhite men who worked in the rubber-reclaiming divi- 145 sion of a large rubber manufacturing company. In comparisons of mortality of white reclaim workers with that of nonreclaim workers rate ratios were 2.7 for esophageal cancer (six observed deaths among reclaim workers), 2.1 for bladder cancer (seven observed deaths), and 4.5 for multiple myeloma (six observed deaths). The excess of bladder cancer among white reclaim workers may be associated with their employment in other high-risk areas of the plant, whereas no such explanation was found for the excesses of esophageal cancer and multiple myeloma. Overall, the lung cancer mortality rate of white reclaim workers was similar to the rate of U.S. white males and other white rubber workers. There was a 50% excess of lung cancer deaths among nonwhite reclaim workers compared with other nonwhite rubber workers. However, this observation is based on small numbers, and no firm conclusions can be reached about the risk of lung cancer associated with reclaim operations in this group of rubber workers. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Dubin, N.; Hutter, R.V.P.; Strax, P.; et al. Epidemiology of Minimal Breast Cancer Among Women Screened in New York City J Natl Cancer Inst 73(6):1273-1279, 1984 357 Subjects Screening: breast; Risk Factor Exposure: familial factors, breastfeeding, diet-nutritional factors, reproductive factors; Cancer Site: breast Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1968-1979 Sample Characteristics 3,463 individuals (blacks 207; whites 2,691; Puerto Ricans 67; others 43—Breast cancer cases 1,290; controls 2,173): female, screened for breast cancer; New York City, NY Variables SES: religion; Demographic: age, race-ethnicity, age at first birth, parity; Cancer: histology, stage Data Sources Histopathology records, medical records Abstract A case-control study based on a screened population in New York City examined epidemiologic risk factor differences between minimal breast cancer (in situ and small invasive carcinomas) and all other breast carcinomas, referred to as clinical breast cancer. Histopathologic re-review of the original slides identified 113 min- imal and 792 clinical breast cancer among 1,290 eligible cases; 2,173 randomly selected screenees served as controls. Among those who developed cancer, black women were twice as likely to develop minimal, as compared to clinical, breast cancer. Women who were less than 20 years of age at first live birth had more than double the probability of being diagnosed with minimal breast cancer, whereas women with first live birth at age 30 years or greater and nulliparous women were at 1.5 times the risk of clinical breast cancer. The relative proportion of minimal breast cancer increased with increasing number of children breast fed, being twofold among women who nursed 2 children or more. Unlike clinical breast cancer, minimal breast cancer was not associated with either family history of breast cancer or obesity. Meaningful histologic differences were not apparent between the case subgroups. Except possibly for obesity, these results could not be explained by any plausible diagnostic bias. 146 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Dunham, L.]J.; Rabson, A.S.; Stewart, H.L.; et al. Rates, Interview, and Pathology Study of Cancer of the Urinary Bladder in New Orleans, Louisiana J Natl Cancer Inst 41:683-709, 1968 358 Subjects Risk Factor Exposure: diet-nutritional factors, occupation, tobacco, other diseases; Cancer Site: bladder; Cancer Measures: incidence Study Characteristics Design: retrospective, case-control; Measures: crude incidence rate, age-specific incidence rate, age-adjusted incidence rate, descriptive statistics, higher level statistics; Time Frame: 1958-1964 Sample Characteristics 702 bladder cancer cases (Negroes 147; whites 555) and 1,020 hospital patients (Negroes 248; whites 772—Bladder cancer cases 493; controls 527): New Orleans, LA Variables SES: occupation; Demographic: age, sex, race-ethnicity; Cancer: histology, tumor grade Data Sources non-SEER population-based cancer registry, hospital medical records, survey, histopathology records, U.S. Population Census Abstract An incidence, interview, and pathology study of cancer of the urinary bladder in whites and Negroes was carried out in New Orleans, Louisiana, from 1958-64. A total of 702 patients were identified with a presumptive or confirmed diagnosis of tumor of the bladder. The physicians in practice in New Orleans initially diagnosed the cancers of the patients in this study. Insofar as possible, pathologists at the National Cancer Institute reviewed the diagnoses. The reviewers aimed to make as certain as possible the diagnosis of the neoplastic lesion, the site of its origin, and its histologic type. In the pathology study the following patients in- itially reported as having bladder cancer were not included: 2.3% with clinical diagnoses only, 1.1% with the histologic diagnosis made by the New Orleans pathologists but not reviewed by Na- tional Cancer Institute pathologists, and 6.7% with a diagnosis of primary cancer in the bladder that was not confirmed after review. Histologic types of 22 (3.5%) of the 631 cancers in the pathology study were changed after review; 92% were transitional cell car- cinomas (29% Grade 1; 39% Grade 2; and 24% Grade 3); 4.3% were squamous cell carcinomas; 1.1% were adenocarcinomas; 1.7% were undifferentiated cell carcinomas; malignant tumors, unclassified type. Negro females (49) had larger percentages of squamous cell carcinoma, adenocarcinoma, and undifferentiated cell carcinoma than the other groups, and Negro males had a larger percentage of squamous cell carcinoma than the white males. Age-standar- dized incidence rates (new cases of bladder cancer per 100,000 per year) were determined for the three parishes of New Orleans (St. Bernard, Jefferson, and Orleans) combined. There were 564 pa- tients, 97.2% with histologic diagnosis of their cancer. In the 6 years 1958-64, incidence rates ranged from 14.3-25.2 for white males, 7.5-15.9 for Negro males, 5.5-10.2 for white females, and 5.5-8.7 for Negro females, indicating considerable yearly variation. The incidence rate for bladder cancer for the total 6-year period was 11.4, i.e., 19.0 for white males, 11.2 for Negro males, 7.4 for white females, and 7.2 for Negro females. The corresponding rates for 1947, as determined in the study by Dorn and Cutler, were 21.2, i.e,. 34.6, 7.0, 17.6, and 14.6. An interviewer questioned 493 patients with bladder cancer (98.8% histologically confirmed) and 527 control patients about factors that might have influenced their disease. Cystitis, in all males, and prostatic diseases in white males were more frequent for patients with cancer of the bladder than for the controls. The ratios of “Observed” to “Expected” cases in white male patients from the incidence rates study were excessive for sailors and deckhands, tinsmiths and coppersmiths, machinists, civil engineers, plumbers, structural metal workers, barbers, and carpenters. Exposure to dusts, especially chromate dusts, was slightly more frequent in white male cancer patients than in the controls. However, doubts about the validity of some of the methods of study concerned with these factors left us unwil- ling to conclude that the associations observed were related to the causes of cancer of the bladder. There was a significant correlation between bladder cancer and tobacco use, all forms combined, among males only; however, this correlation did not exist when data were analyzed by type of tobacco or amounts used. Bladder cancer in our study was not associated with the history of amounts of total daily fluid ingested, alcoholic or nonalcoholic beverages, or various types of coffees, or with dietary habits, or preparations for the hair and scalp. 147 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Fair, W.R.; Heston, W.D.W., Kadmon, D.; et al. Prostatic Cancer, Acid Phosphatase, Creatine Kinase-BB and Race: A Prospective Study J Urol 128:735-738, 1982 359 Subjects Diagnosis; Treatment; Cancer Site: prostate; Biologic Marker: prostaticacid phosphatase (PAP), prostatic creatine kinase-BB Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1979-1980 Sample Characteristics 451 hospital patients (blacks 94; whites 357—Untreated prostate cancer cases 118; benign prostatic hypertrophy controls 333): male, 41+ years old Variables Cancer: stage Data Sources Laboratory test, hospital medical records Abstract To examine the effectiveness of prostatic acid phosphatase and creatine kinase-BB determinations in detecting prostatic cancer serum from 594 men more than 40 years old was assayed for prostatic acid phosphatase with the thymolphthalein monophos- phate substrate and a radioimmunoassay kit. Creatine kinase-BB levels also were measured with a radioimmunoassay kit. Patients with benign prostatic hyperplasia had higher prostatic acid phos- phatase levels than normal controls. Accordingly, to avoid a high incidence of false positives in patients with benign prostatic hyperplasia the 92.5 percentile level of the patients with benign prostatic hyperplasia (3.9 ng/ml) was chosen as the upper limit of normal. With this critical value elevated prostatic acid phosphate levels were observed in 6 per cent of the patients with clinical stage A disease, 8 per cent with stage B, 35 per cent with stage C and 68 per cent with stage D. The radioimmunoassay was no more effective than the enzymatic assay in detecting prostatic cancer. A correlation between prostatic acid phosphatase levels and patient race was observed, with 80 per cent of the black patients with extracapsular prostatic cancer having elevated prostatic acid phos- phatase levels compared to 34 per cent of the white patients with similar stage disease. Creatine kinase-BB was elevated only in patients with advanced disease and was of little value in the diag- nosis of prostatic cancer. Flanders, W.D.; Rothman, K.]. Occupational Risk for Laryngeal Cancer Am ] Public Health 72(4):369-372, 1982 360 Subjects Risk Factor Exposure: occupation; Cancer Site: larynx; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1969-1971 Sample Characteristics 1,023 cancer cases (nonwhites 138; whites 885—Larynx cancer cases 90; controls 933): male, TNCS areas (Iowa excluded) Variables SES: industry, occupation; Demographic: geographic area Data Sources Third National Cancer Survey 1969-1971 Abstract In a case-control analysis, we studied the effects of type of employment on laryngeal cancer risk using the interview data from the Third National Cancer Survey. Effects were measured relative to the risk for those employed in a group of arbitrarily defined industries and occupations with low risk. We excluded females and controlled for age, tobacco use, alcohol use, and race in the analysis. We found ratio estimates above 3.0 for workers in the railroad industry and the lumber industry; and for sheetmetal workers, grinding wheel operators, and automobile mechanics. 148 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Freedman, R.S.; Joosting, A.C.C.; Ryan, J.T; et al. A Study of Associated Factors, Including Genital Herpes, in Black Women with Cervical Carcinoma in Johannesburg S Afr Med ] 48:1747-1752, 1974 361 Subjects Biologic Marker: herpesvirus type 2 antibodies; Screening: Pap smear; Risk Factor Exposure: other diseases, sexual practices, reproductive factors; Cancer Site: cervix uteri Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1971 Sample Characteristics 96 hospital patients (invasive cervix uteri cancer cases 48; controls 48): black, female; Johannesburg, South Africa Variables SES: education; Demographic: age, race-ethnicity (Nguni, Sotho, Venda, Shangaan); parity, age at menarche; Cancer: histology, stage Data Sources Hospital medical records, laboratory test Abstract Herpesvirus hominis antibodies were measured by a kinetic neutralisation test. Among Black patients in the Johannesburg area there is a high incidence of carcinoma of the cervix and H. hominis type 2 infection. A correlation exists between carcinomas of the cervix on the one hand and antibodies to H. hominis type 2 on the other. In addition, patients with keratinising tumours were more promiscuous (mean of 2,9 consorts each) and had more ven- eral disease (81% positive fluorescent treponemal antibody tests) than those with non-keratinising tumours (mean of 2,1 consorts and 52% positive FTAs). The latter had a poorer antibody response (mean 1,9 times lower than matched controls). It is clear that more specific differentiation between the Herpesvirus types is needed before firm conclusions can be reached. These findings confirm those found by other workers elsewhere but suggest in addition that differentiation into tumour types is needed to separate tumours of different aetiological origin. Gold, E.B.; Gordis, L.; Diener, M.D.; et al. Diet and Other Risk Factors for Cancer of the Pancreas Cancer 55(2):460-467, 1985 362 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, occupation, tobacco; Cancer Site: pancreas Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1978-1980 Sample Characteristics 603 individuals (blacks 138; others 465—Pancreas cancer cases 201; hospital controls 201: random digit dietary controls 201): Baltimore, MD Variables SES: occupation; Demographic: sex Data Sources Hospital medical records, survey Abstract The findings of a case-control study of cancer of the pancreas, which was conducted in the Baltimore metropolitan area, are re- ported. Two hundred one patients with pancreatic cancer were matched on age (+5 years), race, and sex to hospital and non-hos- pital controls, the latter selected by random-digit-dialing (RDD). All subjects were interviewed regarding diet, beverage consump- tion, occupational and environmental exposures, and medical and surgical history. Significantly decreased risks were associated with consumption of raw fruits and vegetables and diet soda, and sig- nificantly increased risks were associated with consumption of white bread when cases were compared with hospital and RDD controls. A significantly reduced risk was associated with con- sumption of wine when cases were compared to RDD controls. Risk ratios for consumption of coffee were not significantly differ- ent from one, although there appeared to be a dose-response re- lationship in women. A moderate but statistically nonsignificant increase in relative odds was found for cigarette smoking, and cessation of smoking was associated with a marked reduction in risk. No significant associations were found with particular occu- pational exposures. Tonsillectomy was associated with a signific- antly reduced risk, a finding that has been observed for other cancers as well. The current evidence indicates that pancreatic cancer is likely to result from a complex interaction of factors and suggests that the study of its etiology requires a multidisciplinary approach involving both laboratory and epidemiologic compo- nents. 149 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Goldsmith, D.F.; Smith, A.H.; McMichael, A.]. A Case-Control Study of Prostate Cancer Within a Cohort of Rubber and Tire Workers J Occup Med 22(8):533-541, 1980 363 Subjects Risk Factor Exposure: occupation; Cancer Site: prostate; Cancer Measures: mortality Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1964-1974 Sample Characteristics 346 individuals (blacks 68; whites 278—Prostate cancer cases 88; controls 258): male, rubber and tire workers Variables SES: education, occupation; Demographic: age, race-ethnicity, birthplace, rural-urban, number of dependants, marital status Data Sources Industry records, death registration Abstract A matched case-control study was conducted to test the hypothesis that prostate cancer is associated with employment in the compounding area of a rubber and tire manufacturing plant. A search of death certificates from the years 1964 to 1975 led to the identification of 88 cases of prostate cancer. These were indi- vidually matched with 258 controls on the factors of age, race, and date of entry into the plant. To eliminate possible confounding by sociodemographic differences, several variables were assessed but none differed significantly between cases and controls. The batch preparation work area showed statistically significant (p< 0.025) risk ratios over three exposure periods - more than one month, more than 24 months, and more than 60 months. The service to batch preparation and shipping and receiving work areas showed significant risks for blacks. The data were used to assess a latent period or etiologic fraction for prostate cancer for those employed in batch preparation. Twenty-nine years was estimated to be the latent period while the period of greatest risk occurred during the years 1940 to 1947. Gottlieb, M.S.; Pickle, L.W.; Blot, W.].; et al. Lung Cancer in Louisiana: Death Certificate Analysis J Natl Cancer Inst 63(5):1131-1137, 1979 364 Subjects Risk Factor Exposure: occupation; Cancer Site: lung; Cancer Measures: mortality Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1960-1975 Sample Characteristics 6,654 deaths (blacks 1,578; whites 5,068; American Indians 8— Lung cancer cases 3,327; controls 3,327): Louisiana Variables SES:industry, occupation; Demographic: sex, age, race-ethnicity Data Sources Death registration Abstract In a search for etiologic clues, a review was made of death certificates of residents in a cluster of Louisiana parishes, mainly in the southern part of the state, where lung cancer mortality was high. A comparison of the statements on occupation for 3,327 patients with lung cancer and those of 3,327 controls (matched by sex, race, age, and parish of residence) during 1960-75 revealed an approximately twofold excess risk associated with transporta- tion equipment manufacture, mainly shipbuilding, and the fishing industry. Smaller elevations of lung cancer risk were found among older men who had been employed in petroleum exploration and production and among male and female residents of towns where the petroleum industry was a major employer. In addition, Aca- dian ancestry was associated with a higher risk of lung cancer among older male and female residents. 150 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Gutensohn, N.M.; Shapiro, D.S. Social Class Risk Factors Among Children with Hodgkin's Disease Int J] Cancer 30:433-435, 1982 365 Subjects Risk Factor Exposure: occupation; Cancer Site: lymphoma; Socioeconomic Factors Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics; Time Frame: 1959-1977 Sample Characteristics 248 families (black families with children under 15 years old with Hodgkin's disease 3; white families with children under 15 years old with Hodgkin's disease 63; control families 182): Boston, MA Variables SES: occupation, religion, census tract income, census tract single unit housing; Demographic: age, race-ethnicity; Cancer: histology Data Sources U.S. Population Census Abstract We compared the social class characteristics of 66 families with children diagnosed with Hodgkin's disease (HD) from 1959 through 1977 in a defined population with that of 182 “control families” identified by a random process from the population base. The 14 youngest cases (<10 years at diagnosis) were from some- what lower social-class backgrounds than their 37 controls as evi- denced by the distribution of median income, single-unit housing, and poverty level of their census tract of residence, as well as by the occupational class of head-of-household. In contrast, the social class characteristics of the 52 older children with HD (10-14 years) were quite similar to that of their 145 controls. This apparent shift from lower to average social class between younger and older children with HD Hodgkin's disease may reflect a shift in their age of exposure to common infections. If true, these findings are consistent with the hypothesis that HD may develop as a rare consequence of a common infection. However, these findings are based on small numbers of cases and on indirect measures of social class. Hartge, P.; Hoover, R.; West, D.W._; et al. Coffee Drinking and Risk of Bladder Cancer J Natl Cancer Inst 70(6):1021-1026, 1983 366 Subjects Risk Factor Exposure: diet-nutritional factors; Cancer Site: bladder Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics Sample Characteristics 8,764 individuals (bladder cancer cases 2,982; controls 5,782): 21-84 years old, SEER Program areas Variables Demographic: age, sex, race-ethnicity, geographic area Data Sources SEER cancer registry, survey Abstract The relationship between coffee drinking and risk of bladder cancer was assessed with the use of data from a case-control study of bladder cancer. Incident cases (2,982) and general population controls (5,782) were interviewed. Overall, the relative risk (RR) of bladder cancer for subjects who had ever drunk coffee was estimated as 1.4 (95% confidence interval=1.1-1.8). There was no consistent relation between the RR estimate and the current con- sumption level. Among men who drank coffee, those who drank more than 49 cupfuls of coffee per week had an apparent excess in risk, but women who drank that much had an apparent deficit in risk. Henderson, E.; Louie, E.; Bogdanoff, E.; et al. Antibodies to Herpes Group Viruses in Patients with Nasopharyngeal and Other Head and Neck Cancers Cancer Res 34:1207-1210, 1974 367 Subjects Risk Factor Exposure: other diseases; Cancer Site: head and neck, nasopharyngeal Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics Sample Characteristics 31 nasopharyngeal cancer cases (Negroes 5; Caucasians 14; Chinese 6; Oriental non-Chinese 3; Mexican-Americans 3) and 31 matched controls: California. 12 head and neck cancer cases: California Variables Demographic: race-ethnicity; Cancer: histology, stage Data Sources Hospital medical records, histopathology records, non-SEER population-based cancer registry 151 Abstract | Elevated Epstein-Barr virus antibody titers were found in cases of nasopharyngeal carcinoma in all racial and ethnic groups tested. Titers were highest in patients with active, advanced disease. Anti- body titers to other herpesviruses were not increased above con- trols. Elevated EBV titers were found also in patients with epider- moid cancer of the nasal cavity and hypopharynx. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Heshmat, M.Y.; Kaul, L.; Kovi, J.; et al. Nutrition and Prostate Cancer: A Case-Control Study Prostate 6:7-17, 1985 368 Subjects Risk Factor Exposure: diet-nutritional factors; Cancer Site: prostate Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1979-1982 Sample Characteristics 362 hospital patients (prostate cancer cases 181; controls 181): black, male; Washington, DC Variables Demographic: age Data Sources Hospital medical records, survey Abstract This one-to-one, age- and race-matched case-control study in- volved 181 histologically confirmed black prostate cancer patients and 181 controls seen at three major hospitals in Washington, D.C., during the period 1979-1982. Personal interviews were con- ducted to obtain the number of times food items of specified serv- ing size were consumed per week by cases and controls during the age periods 30-49 and 50 years and older. Then the average daily consumption of each of 18 nutrients per 1,000 calories was calculated. There was risk enhancement associated with increased intake of proteins, total fat, saturated fat, oleic acid, and vitamin A during the age period 30-49 years. The association was highly significant for vitamin A during the age period 30-49 years. The association was highly significant for vitamin A and approached statistical significance for the other four nutrients. A hypothesis based on disturbance of the zinc-retinol binding protein-vitamin A axis was put forward to explain the relative risk enhancement effect of vitamin A on prostate cancer. Hill, P.; Wynder, E.L.; Garbaczewski, L.; et al. Effect of Diet on Plasma and Urinary Hormones in South African Black Men with Prostatic Cancer Cancer Res 42:3864-3869, 1982 369 Subjects Risk Factor Exposure: diet-nutritional factors, endogenous hormones; Cancer Site: prostate Study Characteristics Design: controlled intervention study; Measures: descriptive statistics, higher level statistics Sample Characteristics 40 individuals (prostate cancer cases 21; BPH cases 6; healthy subjects 13): black, male, 60-73 years old; South Africa Data Sources Laboratory test Abstract Epidemiological evidence suggests that the incidence and death rates from prostatic cancer, an endocrine-associated disease, are related to environmental factors including diet. In this study, a comparison of serum and urinary levels of steroid hormones was carried out in healthy elderly rural vegetarian South African black men, a low-risk population, and a comparable group of men with prostatic cancer. In these prostatic cancer patients, plasma andro- gen levels decreased, while estrogen levels increased. Concomit- antly, the androsterone:etiocholanolone ratio increased, and a greater proportion of estrogens was excreted as estradiol. When transferred to a Western diet, plasma androgens showed a further decrease and a greater increase in estrone in prostatic cancer pa- tients. In prostatic cancer patients, the total urinary androgen and estrogen levels were unaltered. However, in elderly health men, the Western diet decreased the excretion of estrogens and andro- gens. Thus, a Western diet supplemented the decrease in plasma androgens initially present in these patients. Evidence suggests that the decrease in plasma androgens increases the estrogen:an- drogen ratio, which may lead to hyperplasia of the prostatic ductal epithelia, a change enhanced by a Western diet. Changes in urinary steroid hormone levels in South African black patients indicate that hormonal changes must be related to several environmental factors, apart from diet. A simultaneous study of the steroid hor- mone composition of blood and prostatic fluid in this low-risk population is suggested. 152 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Hill, P.; Wynder, E.L.; Garbaczewski, L.; et al. Diet and Urinary Steroids in Black and White North American Men and Black South African Men Cancer Res 39:5101-5105, 1979 370 Subjects Risk Factor Exposure: diet-nutritional factors, endogenous hormones, geographic area; Cancer Site: prostate Study Characteristics Design: controlled intervention study; Measures: descriptive statistics, higher level statistics Sample Characteristics 55 individuals (South African blacks 21; North American blacks 16; North American whites 18): male, 40-73 years old, South Africa, United States Variables Demographic: age, race-ethnicity, geographic area Data Sources Laboratory test Abstract Urinary steroid hormone content was determined in Black and White North American men and in rural Black South African men between 40 and 55 years of age and in Black South African men over 60 years of age when maintained on their customary diets or when transferred to a vegetarian or Western diet, respectively. When eating their customary diets, Black South African men had lower levels of urinary estrogens and androgens than did Black and White North American men. The total androgen and estrogen content decreased significantly in Black North American men on the vegetarian diet and increased in Black South African men fed a Western diet. Urinary excretion of estrogens was higher in older than in younger rural Black South African men. Data indicated that a vegetarian diet modified androgen and estrogen metabolism in North American men and that a Western diet was associated with higher levels of urinary steroid hormones in young Black South African men. Diet-related changes in steroid metabolism in rural Black South African men were age dependent. The relation- ship of the increased urinary excretion of steroid hormones in Black South African men, a low-risk group fed a Western diet, and the decreased excretion in Black and White North American men, high-risk groups fed a vegetarian diet, to the development of prostatic cancer remains to be clarified. Hill, P.; Wynder, E.L.; Garnes, H.; et al. Environmental Factors, Hormone Status, and Prostatic Cancer Prev Med 9:657-666, 1980 371 Subjects Risk Factor Exposure: diet-nutritional factors, endogenous hormones, geographic area; Cancer Site: prostate Study Characteristics Design: controlled intervention study; Measures: descriptive statistics, higher level statistics Sample Characteristics 44 individuals (South African blacks 20; North American blacks 13; North American whites 11): male, 40-73 years old; South Africa, United States Variables Demographic: age, race-ethnicity, geographic area Data Sources Laboratory test Abstract Plasma hormone levels were determined in black and white 153 North American (N.A.) men and black South African (S.A.) men maintained on their customary diets or transferred to a vegetarian or Western diet, respectively. Black S.A. men had higher estradiol, estrone, and androsterone levels than N.A. men. In young black S.A. and N.A. men plasma testosterone was decreased when the men were transferred from their customary diets, while compari- son of the plasma-urinary hormone levels suggested transfer from a Western to a vegetarian diet, or vice versa, produced opposite levels in urinary excretion in young black men. The diet hormone changes in black S.A. men were age dependent. A Western diet increased the excretions of estrogens and androgens in young men and decreased them in elderly men. Evidence indicates that diet modified hormonal activity and that the dietary effect was age dependent. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Jackson, M.A.; Ahluwalia, B.S.; Herson, J.; et al. Characterization of Prostatic Carcinoma Among Blacks: A Continuation Report. Cancer Treat Rep 61(2):167-172, 1977 372 Subjects Risk Factor Exposure: endogenous hormones; Cancer Site: prostate; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: crude incidence rate, age- adjusted incidence rate, descriptive statistics, higher level statistics; Time Frame: 1973-1976 Sample Characteristics 329 hospital patients (Nigeria prostate cancer cases 65; Nigeria controls 33; U.S. prostate cancer cases 132; U.S. controls 99) and 492 necropsy cases (Nigeria 243; United States 249): black, male; Ibadan, Nigeria; Washington, DC Variables Demographic: age, geographic area; Cancer: histology, stage Data Sources Laboratory test, survey Abstract A clinical and pathologic comparison of carcinoma of the pros- tate in a high-risk US (Washington, DC) black population and in a low-risk Nigerian (Ibadan) black population is presented. Fifty- two percent of American patients were in clinical stages I and II, whereas only 10% of Nigerian patients were in the same stages. Testosterone and estradiol blood levels were significantly lower (p< 0.05) in patients with carcinoma of the prostate from Ibadan compared with American black prostate cancer patients from Washington, DC. Age-standardized incidence rates (world stan- dard) for 1000 autopsies for micro (incidental)-carcinoma were ap- proximately equal in American and African black men. The inci- dence rate of invasive carcinoma was, however, even after adjust- ment for age, higher in American black men than in African men. Jackson, M.A.; Kovi, J.; Heshmat, M.Y.; et al. Factors Involved in the High Incidence of Prostatic Cancer Among American Blacks 373 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 111-132 Subjects Risk Factor Exposure: endogenous hormones, reproductive factors, sexual practices; Cancer Site: prostate; Cancer Measures: incidence ° Study Characteristics Design: case-control, cross-sectional; Measures: crude incidence rate, age-adjusted incidence rate, descriptive statistics, higher level statistics; Time Frame: 1973-1978 Sample Characteristics 436 individuals (prostate cancer cases 231; matched controls 205) and 417 necropsy cases (prostate cancers 108; nonprostate cancers 328): black, male; Washington, DC Variables SES:income, education, occupation; Demographic: age, marital status, geographic area; Cancer: stage, tumor grade Data Sources Histopathology records, laboratory test Abstract A total of 231 black patients with histologically confirmed car- cinoma of the prostate and 205 matched control patients were entered into a study in 1973-1978. Blood samples for hormone and phosphatase analyses were obtained from all male patients after admission to the Urologic services of Howard University and D.C. General Hospitals in Washington, D.C. Mean plasma testosterone and estrone levels were significantly higher in cancer patients than in age-matched hospitalized controls. The factor with the highest risk ratio was urinary tract symptoms occurring ten years or longer before the onset of the present complaints which led to the diag- nosis of prostatic carcinoma. Johnson, C.C.; Spitz, M.R. Neuroblastoma: Case-Control Analysis of Birth Characteristics J Natl Cancer Inst 74(4):789-792, 1985 374 Subjects Cancer Site: neuroblastoma; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1964-1978 Sample Characteristics 471 births (blacks 77; whites 286; Hispanics 108— Neuroblastoma cases 157; controls 314): Texas Variables Demographic: sex, race-ethnicity Data Sources Birth registration, death registration Abstract The peak incidence of neuroblastoma during early infancy in- dicates that prenatal factors may play a role in the pathogenesis of this disease. A population-based case-control study was con- ducted comparing birth certificate data of 157 children who later died from neuroblastoma in Texas with 314 controls randomly selected from all Texas live births. Analysis of birth certificate data revealed a protective relative risk estimate for preterm births (<37- wk gestation), with an overall odds ratio of 0.29 (95% confidence limits of 0.10-0.86). This effect was independent of birth weight and ethnic group. A statistically significant odds ratio of 3.22 was detected for term babies whose birth weight was low. The findings suggest that the fetus is susceptible to an in utero oncogenic in- itiator or promoter during the last 4 weeks of gestation. 154 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Kark, J.D.; Smith, A.H.; Switzer, B.R.; et al. Serum Vitamin A (Retinol) and Cancer Incidence in Evans County, Georgia J Natl Cancer Inst 66(1):7-16, 1981 375 Subjects Risk Factor Exposure: diet-nutritional factors; Cancer Site: multiple sites Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1960-1976 Sample Characteristics 259 individuals (blacks 65; whites 194—Cancer cases 85; matched controls 162; unmatched controls 12): Evans County, GA Variables Demographic: age, sex, race-ethnicity; Cancer: histology Data Sources Medical records, death registration, survey, laboratory test Abstract A total community sample of 3,102 individuals from Evans County, Georgia, was followed for 12-14 years. During this period, 129 documented new cases of cancer were ascertained from med- ical records and death certificates. Cases were considered for inclu- sion only if documented at least 12 months after subjects were inducted into the cohort study. Cases were classified as definite, probable, and possible by strict criteria. Blood samples were drawn at the beginning of the study in 1960-62 and sera were frozen. Serum vitamin A (retinol) levels were measured in 1976 on the stored sera of 85 cancer patients and for 174 age-, race-, and sex- matched controls. Retinol estimations were performed by a fluorometric method after alumina column separation. Experi- ments conducted to simulate the exposure to light, thawing, and refreezing that sera may have undergone during the 14-16 years of storage showed retinol to be quite stable in response to these possible insults. As compared to controls, persons that eventually developed cancer had significantly lower mean serum retinol levels at least 12 months before the cancer diagnosis. The association was in the same direction for all 4 race-sex groups, although stronger overall for males than females, and was consistent for the various cancer sites and cell types. Both matched and regres- sion residual analyses were used to control for the confounding variable considered: age, race, sex, obesity, social class, and smok- ing. Keller, A.Z. Selected Factors in the Risks of Upper Alimentary Cancers Prev Med 12:541-553, 1983 376 Subjects Risk Factor Exposure: alcohol, occupation, tobacco, other diseases; Cancer Site: esophagus, oral cavity, larynx, pharynx Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1969-1979 Sample Characteristics 529 cancer cases (with cirrhosis 204; without cirrhosis 325): black, male, discharged from U.S. Veterans Administration hospitals Variables SES: occupation; Demographic: age Data Sources Hospital medical records Abstract This study investigates whether liver cirrhosis, alcoholic bever- ages, smoked tobaccos, age, occupations, and multiple primary cancers are significant risk factors for malignancies of the upper alimentary tract. It utilizes maximum likelihood estimates of logis- 155 tics parameters in multivariate analyses of these factors to assess risks of upper alimentary cancer (UAC) among 529 cancer patients, 204 with liver cirrhosis and 325 without liver cirrhosis. All patients are black males, discharged from U.S. Veterans Administration Hospitals from 1969 to 1975 with newly diagnosed and histologi- cally confirmed primary cancers. In comparison with cirrhotics without UAC, cirrhotics with UAC have similar frequencies of liver cirrhosis by type, severity, and histological confirmation but are notably younger and have liver cirrhosis before cancer at a significantly higher frequency and over a significantly longer period. When there is simultaneous adjustment for all six factors, only liver cirrhosis, age, and multiple primaries (mainly at floor of mouth, soft palate, and esophageal sites) are significantly as- sociated with increased cancer risk at upper alimentary sites where such processes as glycogen storage in some squamous cells, as in the liver, may contribute to further elucidation of the implication of liver cirrhosis as a major pathway in cancer risk at upper alimen- tary sites. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Keller, A.Z. Alcohol and Tobacco Use Among Cirrhotic and Non-Cirrhotic Males with Cancer 377 In: Cancer Among Blacks Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 213-223 Subjects Risk Factor Exposure: alcohol, tobacco, other diseases; Cancer Site: multiple sites Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1969-1979 Sample Characteristics 474 cancer cases (cirrhosis 237; non-cirrhosis controls 237): black, male, veterans hospital patients; United States and 675 non-cirrhosis cancer cases (blacks 305; whites 370): male veterans hospital patients; United States Variables Demographic: age, race-ethnicity Data Sources Hospital medical records Abstract This paper investigates the anatomical site distributions of can- cers among black cirrhotics and non-cirrhotics with cancer in order to assess relative risks of multiple and major primary cancer sites. It also compares 305 blacks and 370 whites, with cancer but without liver cirrhosis, by cancer site distributions, tobacco use and alcohol consumption in an attempt to explicate racial differences in cancer risks within the United States of America. Keller, A.Z. Liver Cirrhosis, Tobacco, Alcohol, and Cancer Among Blacks J Natl Med Assoc 70(8):575-580, 1978 378 Subjects Risk Factor Exposure: alcohol, tobacco; Cancer Site: multiple sites Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1969-1970 Sample Characteristics 479 cancer cases (blacks 105; whites 374—Black cancer patients 73; black cancer patients with cirrhosis 32; white cancer patients 374): male Variables Demographic: age, race-ethnicity Data Sources Hospital medical records Abstract Attributes of age, tobacco use, and alcohol consumption were studied in order to elucidate their roles in the increased risks of blacks for selected neoplasms. Black cancer patients with and with- out liver cirrhosis were compared by cancer sites, age, tobacco usage, and alcohol consumption. Subsequently, non-cirrhotic blacks and whites with cancer were characterized on the same variables. Black males with cancer and liver cirrhosis, when com- pared with similiar males without liver cirrhosis, were significantly younger and had more than triple the frequencies of esophageal and hepatic cancers but less than one fourth the frequencies of gastric and prostatic cancers. Cirrhotic patients were rarely non- drinkers but drank whiskey excessively. Non-cirrhotic blacks, when compared with non-cirrhotic whites, had very high risks of liver, stomach, and prostate cancers and smoked less heavily but drank significantly more whiskey. Hence, factors associated with patterns of smoking cigarettes and drinking, especially whiskey, if not these habits themselves, are probably related to the increased risks of blacks for stomach and liver cancers when compared with non-cirrhotic whites and for esophageal and hepatic cancers when compared with non-cirrhotic blacks. 156 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Keller, A.Z. Histology, Survivorship and Related Factors in the Epidemiology of Eye Cancer Am J Epidemiol 97(6):386-393, 1973 379 Subjects Risk Factor Exposure: geographic area; Cancer Site: eye; Cancer Measures: incidence, survival Study Characteristics Design: case-control, cross-sectional; Measures: age-adjusted survival rate, relative risk, crude incidence rate, age-adjusted incidence rate descriptive statistics, higher level statistics; Time Frame: 1955-1970 Sample Characteristics 542 hospital patients (Negroes 49; whites 493—Eye cancer cases 271; controls 271): male, veterans; United States and 66 eye and orbital cancer cases (Negroes 3; whites 63): TNCS areas Variables Demographic: age, race-ethnicity, geographic area; Cancer: tumor location, histology, stage Data Sources Hospital medical records, Third National Cancer Survey 1969-1971 Abstract This study characterizes 271 cases of eye cancers in males by laterality, histology, race, age, survivorship and residence. These histologically confirmed cases constitute a 7-year 100% patient sam- ple from all veterans’ hospitals in the United States. They are compared with male controls to whom they are matched on age and hospital, and with all veterans living in 1960. Relative risks and incidence rates of eye cancers are determined. The following results are reported. 1) Malignant melanomas are 65% more com- mon in the left than in the right eye, while squamous cell and intra-epithelial carcinomas are comparably frequent in both eyes. 2) The worst survival experience is with intra-orbital melanomas in contrast to that with conjunctival cancers of any type. 3) Cumula- tive 5-year survival rates for all cases with localized cancers are similar to those of the male population counterparts of similar age and race. 4) The relative risk of malignant melanomas is extremely high above age 59 and in regions of the south. These results em- phasize that eye cancers select on factors which strongly relate to race, age, laterality and probably geography as well. They establish further that survivorship depends on the degree of spread, anatomical site and histological type of cancers within the eye. Keller, A.Z. Survivorship With Mouth and Pharynx Cancers and Their Association With Cirrhosis of the Liver, Marital Status, and Residence Am ] Public Health 59(7):1139-1153, 1969 380 Subjects Cancer Site: oral cavity, pharynx; Cancer Measures: survival Study Characteristics Design: case-control; Measures: crude survival rate, descriptive statistics, higher level statistics; Time Frame: 1959-1966 Sample Characteristics 1,412 hospital patients (Negroes 153; whites 1,259—Oral cavity and pharynx cancer cases 706; controls 706): male, veterans; United States Variables SES: occupation, religion; Demographic: age, marital status, birthplace; Cancer: tumor location, stage Data Sources Hospital medical records Abstract A study of 706 males with squamous cell carcinoma of the 157 mouth and pharynx and an equal number of age-matched and hospital-matched controls, who were diagnosed and treated in 101 veterans’ hospitals in the United States from 1959 through 1961, demonstrates that: 1. Cirrhosis of the liver is strongly associated with cancer of the mouth and pharynx. 2. Notably few Jews have this cancer. 3. Survivorship with mouth and pharynx cancers is not significantly altered, either by one’s race, religion, marital status, occupation, residence, or by the clinical diagnoses of liver cirrhosis, syphilis, rheumatoid arthritis, diabetes mellitus, or other body cancers. 4. Case survivorship is significantly better for those with a small, anatomically localized cancer than it is for others. These findings are consistent with the conclusion that, apart from the cancer’s size and its initial location, neither demographic factors nor other clinical diseases appear to influence markedly the survi- val experience of patients with cancer of the mouth and pharynx. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Kew, M.C.; Marcus, R.; Geddes, E.W. Some Characteristics of Mozambican Shangaans with Primary Hepatocellular Cancer S Afr Med J 51:306, 1977 381 Subjects Biologic Marker: alpha-fetoprotein, hepatitis B surface antigen; Risk Factor Exposure: diet-nutritional factors, alcohol, other diseases; Cancer Site: liver; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1965-1975 Sample Characteristics 613 hospital patients (Mozambican Shangaan liver cancer cases 328; non-Shangaan liver cancer cases 122; Mozambican Shangaan controls 163): black, male, miners; Johannesburg, South Africa Variables Demographic: age, race-ethnicity Data Sources Laboratory test, hospital medical records Abstract Certain characteristics of 328 Mozambican male Shangaans with primary hepatocellular cancer (PHC) have been compared with those of 163 Shangaan men with hepatomegaly from causes other than PHC and with those of 122 Black Southern African men with the same tumour but who belonged to tribes other than the Shan- gaan. Shangaans with PHC were significantly younger than non- Shangaans with the tumour (mean age 33,4 cf. 40,0 years; p< 0,001). They also had a significantly higher positivity rate of alpha- fetoprotein by immunodiffusion (71,4%) than the non-Shangaans (61%), although in other respects the tumours appeared to be similar. Cirrhosis of the non-tumorous part of the liver was present at necropsy in 62% of the Shangaans and in 66% of the non-Shan- gaans. The hepatitis B (surface) antigen (HB,Ag) was detected in the serum of 60% of the Shangaans with PHC compared with only 9% of the controls. The antigen was present in 53,4% of the non- Shangaans with PHC (the difference between this figure and that in the Shangaans was not significant). HB, Ag was detected in the serum of 64% of the Shangaans with PHC and cirrhosis, but also in 74% of those with the tumour without associated cirrhosis. The possible role of the hepatitis B virus in the aetiology of PHC is considered in the light both of these findings and of the possibility that the frequency with which the tumour is associated with cir- rhosis may be decreasing in Shangaans. Some of the dietary habits of the Shangaans with PHC were compared with those of the controls. Virtually all the patients with PHC, but also almost all the controls, ate groundnuts in large quantities from an early age, as well as cashew nuts in smaller amounts. Cycad pips, mopani leaves and pods, mopani worms and locusts were not eaten by significantly more of the Shangaans with PHC than the controls. The limitations of this type of dietary analysis are discussed. Krain, L.S. Epidemiologic Variables in Prostatic Cancer Geriatrics, May 1973, p 93-98 382 Subjects Risk Factor Exposure: sexual practices, other diseases; Cancer Site: prostate; Cancer Measures: incidence, survival Study Characteristics Design: case-control; Measures: age-specific incidence rate, age-adjusted incidence rate, relative survival rate, descriptive statistics, higher level statistics; Time Frame: 1942-1972 Sample Characteristics 15,142 prostate cancer cases: male, California and 272 individuals (Negroes 142; whites 116; Japanese 2; Chinese 2; Mexican-Americans 4—Prostate cancer cases 136; controls 136): male; Los Angeles, CA Variables SES: religion, public-private hospital status; Demographic: age, race-ethnicity, urban-rural, marital status; Cancer: stage Data Sources non-SEER population-based cancer registry, survey Abstract Data from the California and Alameda County Tumor Registries on prostatic carcinoma were analyzed, and a special epidemiologic questionnaire was given to 136 patients with prostatic cancer. These were matched by age and race to 136 patients with noncan- cerous, nongenitourinary conditions from various Los Angeles area hospitals. There was a statiscally significant (p<0.05) excess of familially associated cancers, history of veneral disease, coital frequency, number of sexual partners before or after marriage, and use of contraceptive agents in the prostatic cancer cases as compared with the controls. Data on a relationship between socioeconomic class and prostatic cancer were inconsistent; data for all races showed no association; data for the nonwhite races showed a strong negative association with social class. Negroes have the highest incidence of prostatic cancer and Chinese and Japanese have the lowest. The five year relative survival rates for prostatic cancer were 64 percent (1955 to 1969) for localized cases and 52 percent (1955 to 1969) for all stages of disease; although these rates have improved, the mortality for prostatic cancer con- tinues to increase in California, implying an increase in incidence. 158 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Leech, S.H.; Bryan, C.F; Elston, R.C.; et al. Genetic Studies in Multiple Myeloma, 1. Association with HLA-Cw5 Cancer 51:1408-1411, 1983 383 Subjects Biologic Marker: HLA; Cancer Site: multiple myeloma Study Characteristics Design: case-control; Measures: relative risk, higher level statistics Sample Characteristics 160 individuals (multiple myeloma cases 22; controls 138): black; New Orleans and LaFayette, LA Variables Demographic: sex Data Sources Hospital medical records, laboratory test Abstract Human leukocyte antigens (HLA) were identified in 22 black Americans with multiple myeloma. No significant association was observed between antigens at either the A or the B locus. At the C locus, in contrast, HLA-Cw5 was more prevalent in the patient group, four of 22 having it, compared with the control group, in which two of 138 individuals possessed it. All four patients with HLA-Cwb5 were males. Those results suggest that genetic factors, perhaps in conjunction with an environmental change, may be responsible for the recent increase in incidence in myeloma in black Americans, especially in males. Lloyd, J.W. Long-Term Mortality Study of Steelworkers, V. Respiratory Cancer in Coke Plant Workers J Occup Med 13(2):53-68, 1971 384 Subjects Risk Factor Exposure: occupation; Cancer Site: respiratory system; Cancer Measures: mortality Study Characteristics Design: cohort; Measures: standardized mortality ratios (SMR), descriptive statistics, higher level statistics; Time Frame: 1953-1961 Sample Characteristics 3,530 individuals (nonwhites 1,161; whites 2,369—Coke oven workers 2,048; nonoven workers 1,482): male, coke plant employees Variables SES: occupation; Demographic: race-ethnicity Data Sources Industry records Abstract Mortality experience of coke plant workers was examined ac- cording to calendar period and length of employment at several work stations within a coke plant. The cause-specific mortality experience of men employed was contrasted with mortality pre- dicted by the age, race, and nativity-specific rates for the total steelworker population. Findings showed that an excess of re- spiratory cancer was limited to men employed at the coke ovens where the relative mortality for the disease was 2-1/2 times that predicted. The apparent differential in respiratory cancer rates for white and nonwhite coke plant workers was accounted for by differing distribution by work area. Mabuchi, K.; Bross, D.S.; Kessler, I.1. Epidemiology of Cancer of the Vulva: A Case-Control Study Cancer 55:1843-1848, 1985 385 Subjects Risk Factor Exposure: diet-nutritional factors, tobacco, sexual practices, reproductive factors, occupation, other diseases; Cancer Site: vulva Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics, odds ratio; Time Frame: 1972-1975 Sample Characteristics 298 hospital patients (blacks 30; whites 268—Vulvar cancer Cases 149; controls 149): female, 30+ years old; Buffalo-Erie Co., NY; Detroit-Wayne Co., MI; Miami-Dade Co., FL; Minneapolis-St. Paul, MN; New York City, NY Variables SES: education, religion, occupation; Demographic: number of marriages, age at first marriage, age at first intercourse, number of pregancies, age at first pregnancy, age at menarche, menopausal status, age at menopause Data Sources Hospital medical records, survey 159 Abstract Few epidemiologic studies have been undertaken to investigate the etiology of cancer of the vulva. To identify risk factors as- sociated with this cancer, a case-control study was conducted in 149 patients with histologically proven vulvar carcinoma and the same number of control patients matched for age, race, marital status, and hospital from five U.S. metropolitan areas. As previ- ously suspected, we found prior histories of leukoplakia of the vulva, inflammation of the vulva or vagina, and urogenital cancer to be significantly associated with vulvar cancer. In addition, we found moderately high odds ratios associated with occupational histories of private household maids and servants, and work in laundry, cleaning, and other garment services. A slightly increased odds ratio was associated with coffee consumption and a dose-re- sponse effect was demonstrated. The findings suggest that en- vironmental exposures may play a role in vulvar carcinogenesis. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL MacDonald, P.C.; Edman, C.D.; Hemsell, D.L.; et al. Effect of Obesity on Conversion of Plasma Androstenedione to Estrone in Postmenopausal Women With and Without Endometrial Cancer Am |] Obstet Gynecol 130:448-453, 1978 386 Subjects Risk Factor Exposure: diet-nutritional factors, endogenous hormones; Cancer Site: corpus uteri Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics Sample Characteristics 50 individuals (Negroes 22; Caucasians 26; Latins 1; Indians 1— Endometrial cancer cases 25; controls 25): female, 47 + years old; Texas Data Sources Laboratory test Abstract The purpose of this study was to ascertain if a relationship exists between the transfer constant of conversion of plasma and androstenedione to estrone ([p] AE1/BU) and total body weight or excessive body weight in 50 postmenopausal women, of whom 25 had adenocarcinoma of the endometrium and 25 had no en- dometrial disease. The [p]AE1/BU ranged from 0.015 to 0.129 in these 50 women. The [p]AE1/BU in the women with endometrial cancer was 0.051+/-0.006 (mean+/-S.E.), whereas that in the women with no endometrial disease was 0.039+/-0.004. These values are not significantly different (p >0.05). The body weights of these 50 women ranged from 104 to 430 pounds. Weight of the patients with endometrial cancer was 234 +/-16 pounds (mean +/- S.E.), and that for the women with no endometrial cancer was 194+/-12 pounds. A statistically significant correlation (p<0.001) was found between [p]AE1/BU and body weight and between [P]AE1/BU and excessive body weight in both groups of women. Moreover, obesity and aging appear to act in concert to potentiate the conversion of plasma androstenedione to estrone in extraglan- dular sites once the [p]AE1/BU is considerably greater among obese postmenopausal women than among comparably obese pre- menopausal women. Morrison, A.S. Some Social and Medical Characteristics of Army Men with Testicular Cancer Am J Epidemiol 104(5):511-516, 1976 387 Subjects Risk Factor Exposure: other diseases; Cancer Site: testis Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1950-1970 Sample Characteristics 1,326 individuals (blacks 73; whites 1,238; race not given 51— Testis cancer cases 702; controls 624): male, active service in the U.S. Army Variables SES: religion, education; Demographic: age, race-ethnicity, marital status, geographicarea, place of birth; Cancer: histology Data Sources non-SEER hospital cancer registry Abstract Risk of developing cancer of the testis was related to social and medical characteristics in a case-control study of men on active duty in the US Army, 1950-1970. Service records were obtained for 702 newly-diagnosed patients and 624 members of a systematic sample of unaffected servicemen. Risk ratios were computed for seminomas, tumors of other histologic types, and for all tumors combined. Respectively, these risk ratios were 0.3, 0.1, and 0.2 for blacks compared to whites, 1.5, 1.1 and 1.3 for Protestants compared to Catholics, and 1.5, 1.0 and 1.2 for the ever married compared to the never married. Risk increased regularly with du- ration of schooling for seminomas but not for other types of tumors. Risk appeared relatively low for men from large sibships. Men from the North Central or Western parts of the country were at higher risk than those from the South or Northeast, a trend that was stronger for tumors other than seminomas. Observed changes in risk with respect to history of mumps and ABO blood group were small. 160 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Newell, G.R.; Gordon, J.E.; Monlezun, A.P.; et al. ABO Blood Groups and Cancer J Natl Cancer Inst 52(5):1425-1430, 1974 388 Subjects Biologic Marker: ABO blood group; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1966-1970 Sample Characteristics 6,950 blood samples (blacks 5,247; whites 1,703—Cancer cases 1,955; controls 4,995): New Orleans, LA Variables Demographic: sex, race-ethnicity Data Sources non-SEER hospital cancer registry, hospital medical records Abstract Problems noted in the methods used to study differences in the distribution of ABO blood groups in persons with various malignant disease were examined. Attention to these problems might result in more consistent confirmation of such work. Diag- nostically confirmed cases of cancer of the breast, cervix, pancreas, colon, stomach, and leukemia were obtained from the Charity Hospital of Louisiana Tumor Registry, and their ABO blood group distribution was compared to a sample 5000 individuals who vol- untarily donated blood to the same hospital. For the comparison group, there were no differences in ABO distribution between the sexes, whereas blacks had a greater frequency of type B and a lesser frequency of type A than whites. No excess risk for cancer of the breast, cervix, or colon was associated with blood group A. A relationship between blood group A and cancer of the stomach was found among blacks as well as whites. Previous reports of blood group A was found among black males and white females, but its interpretation is open to question. Peleg, I.; Heyden, S.; Knowles, M.; et al. Serum Retinol and Risk of Subsequent Cancer: Extension of the Evans County, Georgia, Study J Natl Cancer Inst 73(6):1455-1458, 1984 389 Subjects Risk Factor Exposure: diet-nutritional factors; Cancer Site: multiple sites; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1967-1981 Sample Characteristics 362 individuals (blacks 84; whites 278—Cancer cases 131; controls 231): Evans County, GA Variables SES: social class; Demographic: sex, race-ethnicity Data Sources Hospital medical records, histopathology records, death registration, laboratory test Abstract A previous report from the Evans County, Ga., cohort corre- lated low base-line retinol levels in 1960-62 to an increased risk of cancer developing in the following 12-14 years. So that this inverse association could be retested, retinol determinations were per- formed on sera in 135 incident cancer cases, identified during a vital status follow-up in 1981, and 237 controls matched by race, sex, age, and date of venipuncture. The serum was drawn during the second cohort examination between 1967 and 1969 and stored at -20 degrees C until analysis by high-pressure liquid chromatog- raphy in January 1982. Eligible for the study were patients whose initial cancer diagnosis was established more than 24 months after venipuncture. Base-line retinol levels of subsequent cancer cases and their matched controls were similar in the 4 race-sex groups. The risk of cancer at all sites associated with the lowest retinol quintile was similar to that associated with the highest quintile. In multivariate analysis, with control for age, cholesterol, body mass, smoking status, and social class, no significant relationship of serum retinol and case status was found. In summary, these data fail to confirm the strong dose-response relationship between baseline retinol levels and subsequent cancer reported in the pre- vious study. Persaud, V. Epidemiology of Cancer of the Uterine Cervix in Jamaica West Indian Med ] 24(171):171-178, 1975 390 Subjects Risk Factor Exposure: reproductive factors, sexual practices; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics Sample Characteristics 690 individuals (Negroes 607; mixed blood 69; East Indians 14— Cervix uteri cancer cases 300; controls 390): female; Jamaica Variables Demographic: age, age at first pregnancy, parity; Cancer: histology Data Sources non-SEER population-based cancer registry, survey Abstract Despite the very high incidence of cervix cancer in developing countries including Jamaica, controlled studies aimed at defining the significant aetiological factors are practically non-existent. Such investigations have been carried out only in metropolitan countries with relatively low incidences, and much of the controversy con- cerning the role of certain environmental influences has been due in part to the improper selection of control cases. The Jamaican scene with a well-defined population in which there is a very high incidence of this malignancy offered an excellent opportunity for conducting such a survey. 161 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Pickle, L.W.; Gottlieb, M.S. Pancreatic Cancer Mortality in Louisiana Am |] Public Health 70:256-259, 1980 391 Subjects Risk Factor Exposure: occupation, geographic area; Cancer Site: pancreas; Cancer Measures: mortality Study Characteristics Design: case-control; Measures: odds ratio; Tine Frame: 1960-1975 Sample Characteristics 1,752 deaths (blacks 436; whites 1,316—Pancreas cancer cases 876; controls 876): Louisiana Variables SES:industry, occupation; Demographic: age, sex, race-ethnicity Data Sources Death registration Abstract As a preliminary step in the investigation of high pancreas cancer mortality among White males in a cluster of Louisiana parishes, we examined 876 pairs of certificates of death which occurred in this area during 1960-75. The pancreas cancer death records were matched to controls by age, race, sex, year of death, and parish of residence. The odds ratios were increased about twofold for workers in the oil refining and paper manufacturing industries, and slight elevations were seen among residents near refineries and food processing plants. Despite the limited residen- tial and occupational information available on death certificates, this study suggests leads to environmental factors that can be further investigated by a case-control interview study in Louisiana. Pickle, L.W.; Brown, L.M.; Blot, W.]. Information Available from Surrogate Respondents in Case-Control Interview Studies Am |] Epidemiol 118(1):99-108, 1983 392 Subjects Cancer Site: lung; Methodology Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1970-1979 Sample Characteristics 2,606 individuals (blacks 595; whites 2,011—Cancer cases 1,281; controls 1,325): Tidewater, VA; Jacksonville, FL; Savannah, GA Variables Demographic: age, sex, race-ethnicity, geographic area Data Sources Death registration Abstract Epidemiologic studies of fatal diseases often require that infor- 162 mation be sought from relatives or friends of deceased or disabled patients. The authors have evaluated the ability of several types of surrogate respondents to provide information on the smoking, occupational, medical history, and demographic characteristics of their next of kin in three recent case-control studies involving interviews with 2606 individuals. The ability of the surrogates to provide this information varied by topic, degree of detail re- quested, race, sex, age, and study area, but was most affected by the type of respondent. Sibs were best able to respond to questions about the subject's immediate family or events that occurred during early life, while spouses and offspring were best able to describe events that occurred during adult life. Several recommendations are made to improve the design of future interview studies. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Pottern, L.M.; Morris, L.E.; Blot, W.].; et al. Esophageal Cancer Among Black Men in Washington, D.C.: I. Alcohol, Tobacco, and Other Risk Factors J Natl Cancer Inst 67(4):777-783, 1981 393 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, tobacco; Cancer Site: esophagus; Cancer Measures: mortality Study Characteristics Design: case-control, cross-sectional; Measures: odds ratio, age-adjusted mortality rate, age-specific mortality rate, higher level statistics; Time Frame: 1970-1977 Sample Characteristics 370 deaths (esophagus cancer cases 120; controls 250): black, male; Washington, DC; esophagus mortality data for nonwhite males from the United States Variables SES: education; Demographic: age, geographic area Data Sources Death registration, survey Abstract A case-control study involving interviews with the next of kin or close friends of 120 black males who recently died of esophageal cancer and 250 similarly aged black males who died of other causes was undertaken to discover reasons for the exceptionally high mortality from this cancer in Washington, D.C. The age-adjusted annual death rate in Washington, D.C., for nonwhite males, 1970- 75, was 28.6/100,000, far higher than the national rate of 12.4/ 100,000 and the rates in other metropolitan areas of the country. The major factor responsible for the excess was alcoholic beverage consumption with an estimated 81% of the esophageal cancers attributed to its use; high use of alcoholic beverages was also found among the controls. The relative risk (RR) of esophageal cancer associated with use of alcoholic beverages was 6.4 (95% confidence interval=2.5, 16.4). The RR increased with amount of ethanol con- sumed and was highest among drinkers of hard liquor, although the risk was also elevated among consumers of wine and/or beer only. The RR associated with cigarette smoking was 1.9 (1.0, 3.5) when controls with smoking-related causes of death were excluded but declined to 1.5 (0.7, 3.0) when adjusted for ethanol consump- tion. Significant differences of approximately twofold were found between low and high levels of a) consumption of fresh or frozen meat and fish, fruits and vegetables, and dairy products and eggs and b) relative weight (wt/ht sqd). The inverse trends with these general measures of nutritional status were not explained by al- coholic beverage consumption or socioeconomic status as mea- sured by educational level. Prince, A.M.; Szmuness, W.; Michon, J.; et al. A Case/Control Study of the Association Between Primary Liver Cancer and Hepatitis B Infection in Senegal Int J] Cancer 16:376-383, 1975 394 Subjects Cancer Site: liver; Biologic Marker: hepatitis B surface antigens (HBsAg), alpha-fetoprotein (AFP) Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics; Time Frame: 1972-1974 Sample Characteristics 647 hospital patients (primary liver cases 165; cancer controls 154; noncancer controls 328): Dakar, Senegal Data Sources Laboratory test, hospital medical records, histopathology records, autopsy records Abstract A case/control study has been carried out to determine by radioimmunoassay and passive hemagglutination techniques the prevalence of hepatitis B surface antigen (HBsAg) and antibody (anti-HBs) in patients with primary liver cancer (PLC) and age/sex- matched hospital controls with cancers of other sites (OCC) and similarly matched controls without cancer (NCC). HBsAg was found in 61.2% of 165 cases of PLC as compared to 11.7% of 154 OCC and 11.3% of 328 NCC. The frequency of HBsAg in PLC patients was significantly higher (72.2%) in those with detectable alpha fetoprotein as compared to those without (40.3%). There was no difference in the frequency of HBsAg in PLC patients with and without accompanying cirrhosis. No significant difference in potential hepatitis exposure history was found in the three study groups. 163 EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Redmond, C.K.; Strobino, B.R.; Cypess, R.H. Cancer Experience Among Coke By-Product Workers Ann NY Acad Sci 271:102-115, 1976 395 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cohort; Measures: relative risk, higher level statistics; Time Frame: 1953-1966 Sample Characteristics 58,828 individuals: male, steelworkers; Allegheny County, PA Variables SES: industry-occupation; Demographic: race-ethnicity (native-born white, foreign-born white, nonwhite) Data Sources Death registration Abstract This paper presents the site-specific cancer mortality, 1953- 1966, for men employed in by-product coke plants in Allegheny County, PA. Approximate relative risks, which take into account race, age, and calendar years of follow-up, have been calculated for various work areas of the coke plant. The major findings are: 1. Men with five or more years at the coke ovens have an excess risk of dying from lung cancer and kidney cancer. 2. Cancers of the digestive system are significantly elevated in nonoven workers. Cancers of two sites, the colon and pancreas, account for the total excess in cancers of the digestive system. 3. Cancers of the buccal cavity and pharynx appear high in nonoven workers, although the number of deaths involved is small. These observations clearly indicate the need to consider nonoven as well as oven workers when evaluating cancer hazards in the coke plant. Rockette, H.E.; Redmond, C.K. Long-Term Mortality Study of Steelworkers J Occup Med 18(8):541-545, 1976 396 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites; Cancer Measures: mortality Study Characteristics Design: cohort; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1953-1966 Sample Characteristics 58,828 individuals (nonwhites 7,349; whites 51,479—Mason department workers 1,743; controls 57,058): male, steelworkers; Allegheny County, PA Variables SES: occupation; Demographic: age, race-ethnicity, birthplace Data Sources Industry records Abstract This study examines the cause-specific mortality of males em- ployed as masons in the steel industry. Their mortality experience is compared with the mortality predicted by the age - and race - specific death rates for a control group consisting of those steelwor- kers whose first job in 1953 was not in the mason department. The relative risks of dying from selected causes were highly depen- dent on race. The most important observations are: (1) The non- whites have an overall excess risk when compared to the nonmason group. This excess is not confined to any particular category and appears to be due to the cumulative effect of an excess in many different categories. (2) The white masons have an excess risk for nonmalignant respiratory disease. Furthermore, this risk increases with length of exposure. (3) The whites have an excess risk for respiratory cancer when compared to the nonmason group. This excess is not present in the later years of the follow-up period. (4) The interpretation of the findings for cardiovascular disease remain unclear. Results for cerebral vascular disease and hypertensive disease in nonwhites are suggestive, but need further investigation before any strong statements can be made. Rogers, E.L.; Goldkind, L.; Goldkind, S.F. Increasing Frequency of Esophageal Cancer Among Black Male Veterans Cancer 49(3):610-617, 1982 397 Subjects Risk Factor Exposure: alcohol, tobacco; Cancer Site: esophagus; Cancer Measures: incidence Study Characteristics Design: case-control, time trend; Measures: descriptive statistics, higher level statistics; Time Frame: 1975-1980 Sample Characteristics 182 hospital patients (esophagus cancer cases 62; 1980 controls 60; presume 1975 controls 60): male, 43-87 years old, veterans; Baltimore, MD Variables SES: education, occupation; Demographic: age, race-ethnicity (black, white), urban-rural Data Sources Hospital medical records 164 Abstract Between the years of 1975 and 1979, the frequency of diagnosis of esophageal cancer has doubled at the Baltimore VA Medical Center due to a rapid increase of esophageal cancer among black males. This increase was not related to increased yearly hospital admission rates, percentage of black patients admitted yearly, or increased use of the hospital for chronic disease processes. Detailed chart review and comparison with consecutive medical admissions as controls revealed heavy alcohol use and urbanization to be risk factors experienced more frequently by black than white male vet- erans. A serious question needs to be quickly answered: Does the rise of esophageal cancer at the Baltimore VAMC reflect a rise among black males only in Baltimore or does it reflect a rise nation- wide among black males with a history of previous employment in the armed forces? EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Rothschild, H.; Mulvey, J.J. An Increased Risk for Lung Cancer Mortality Associated with Sugarcane Farming J Natl Cancer Inst 68(5):755-760, 1982 398 Subjects Risk Factor Exposure: occupation, tobacco; Cancer Site: lung; Cancer Measures: mortality Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1971-1977 Sample Characteristics 568 deaths (nonwhites 88; whites 480—Lung cancer cases 284; controls 284): Louisiana Variables SES: occupation; Demographic: age, sex, race-ethnicity, geographic area; Cancer: histology Data Sources Death registration, survey Abstract In a study to determine putative occupation-related factors re- sponsible for the excessive mortality due to lung cancer in southern Louisiana, the next-of-kin were interviewed of 284 of 400 persons (71%) randomly sampled from a total of 815 persons who died of lung cancer during 1971 through 1977 and had lived in any of 10 parishes (counties) of southern Louisiana. Of the decendents, 108 (38%) had been employed for at least 6 months as sugarcane farm workers at some time during their lives. Logistic regression analysis indicated this industrial involvement differed signifi- cantly (p<0.0001) from that of a control group, consisting of per- sons who had died of any cause other than lung cancer and who were matched for year of death, age, sex, race, and parish of residence; only 58 (20%) matched controls had had sugarcane farm employment (odds ratio=2.4; 95% confidence interval, 2.0-2.9). Employment in other industries or tobacco consumption could not account for the elevated risk of lung cancer mortality associated with sugarcane farming. After adjustment for smoking, the relative risk estimate of lung cancer mortality for sugarcane farm workers was 2.4 (95% confidence limits, 1.7-3.6). The sugarcane farmers who died with lung cancer had worked for longer periods in the sugarcane farm industry than did those sugarcane farmers in whom lung cancer did not develop (P=0.006). No specific his- topathologic cell type was noted to be increased in persons who had been employed in sugarcane farming; however, 2 sugarcane farmers had had mesotheliomas. Stagner, L.T.; Wegman, D.H. Smoking, Occupation, and Histopathology of Lung Cancer: A Case-Control Study with the Use of the Third National Cancer Survey J Natl Cancer Inst 70(3):421-426, 1983 399 Subjects Risk Factor Exposure: occupation, tobacco; Cancer Site: lung and bronchus Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1969-1971 Sample Characteristics 1,320 cancer cases (blacks 141; whites 1,159; others 20— Respiratory cancer cases 420; controls 900): male, 30-84 years old; TNCS areas (Iowa excluded) Variables SES: occupation; Cancer: histology Data Sources Third National Cancer Survey 1969-1971 Abstract The relationships between occupation, smoking, and the three most common histologic types of lung cancer (squamous cell car- cinoma, small-cell carcinoma, and adenocarcinoma) were explored in a case-control study with the use of data collected during the Third National Cancer Survey. The largest histologic group was 165 squamous cell carcinoma (152 cases), followed by adenocarcinoma (50 cases), and small-cell carcinoma (45 cases). The control series was comprised of cancers at all anatomic sites except those believed to be associated with either smoking or occupational exposures. Cigarette smoking was significantly associated with all three his- tologic types of lung cancer. Overall, the relationship with small- cell carcinoma was strongest (odds ratio = 5.1), whereas those with squamous and adenocarcinoma were approximately equiva- lent (odds ratio = 3.1). Dose-response relationships were evident for all three histologic types; however, the linear relationship was found to be statistically significant (p<<0.05) with “blue collar” professions (odds ratio = 2.1). No occupational categories were significantly associated with adenocarcinoma. In addition, no oc- cupational categories were associated with all histologic types of lung cancer combined. This last observation suggested that the sensitivity of epidemiologic studies might not only be increased by use of improved occupational histories but more specifically by consideration of histology in examination of associations between occupation and respiratory cancer. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Terris, M.; Wilson, F.; Nelson, J.H., Jr. Comparative Epidemiology of Invasive Carcinoma of the Cervix, Carcinoma In Situ, and Cervical Dysplasia Am ] Epidemiol 112(2):253-257, 1980 400 Subjects Diagnosis; Risk Factor Exposure: reproductive factors, sexual practices; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: descriptive statistics, higher level statistics Sample Characteristics 2,296 hospital patients (invasive cases 454; in situ cases 411; dysplasia cases 283; controls 1,148): female; New York Variables SES: education; Demographic: age, age at first marriage, number of pregnancies; Cancer: histology Data Sources Hospital medical records, survey Abstract Findings of this study support the concept that carcinoma in situ and invasive cancer of the cervix are different aspects of the same process. The risk factors for cervical dysplasia, however, are not the same as for invasive cancer of the cervix. The study also concludes that women who present with dysplasia are different than those who present with invasive cancer. Tollerud, D.].; Brinton, L.A.; Stone, B.].; et al. Mortality from Multiple Myeloma Among North Carolina Furniture Workers J Natl Cancer Inst 74(4):799-801, 1985 401 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple myeloma Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1956-1980 Sample Characteristics 1,159 deaths (blacks 220; whites 939—Multiple myeloma cases 301; controls 858): North Carolina Variables SES: occupation; Demographic: age, race-ethnicity Data Sources Death registration Abstract The risk of multiple myeloma among furniture workers was investigated in a case-control study with the use of death certificate statements on occupation in North Carolina counties with heavy employment in furniture-manufacturing industries. From com- puterized mortality listings, 301 male deaths from multiple myeloma were identified, and 858 controls were selected from deaths due to other causes; controls were matched by sex, race, county of usual residence, age at death, and year of death. Employ- ment in the furniture industry was associated with a nonsignificant excess risk of multiple myeloma [relative risk (RR) = 1.3], particu- larly among subjects who died before age 65 (RR = 1.7) and among those born before 1905 (RR = 1.5). A significantly elevated risk (RR = 5.4) was seen for furniture workers who were born before 1905 and died prior to age 65. Unexpectedly, farmers were found to have a significantly decreased risk of multiple myeloma (RR = 0.6). The finding of an elevated risk of myeloma in earlier cohorts of furniture workers may be a clue to the nature of the environmen- tal exposure. Trevathan, E.; Layde, P.; Webster, L.A.; et al. Cigarette Smoking and Dysplasia and Carcinoma In Situ of the Uterine Cervix JAMA 250:499-502, 1983 402 Subjects Risk Factor Exposure: tobacco; Cancer Site: cervix uteri Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1980-1981 Sample Characteristics 662 individuals (carcinoma in situ cases 99; severe dysplasia cases 81; mild-moderate dysplasia cases 194; controls 288): blacks, female, 17-55 years old, hospital clinic attenders; Atlanta, GA Variables SES: socioeconomic status (not defined); Demographic: age Data Sources Survey, histopathology, records Abstract We conducted a case-control study of cigarette smoking and dysplasia and carcinoma in situ of the uterine cervix. Cases were black women 17 to 55 years of age who were attending a dysplasia 166 clinic and had biopsy-confirmed cervical pathologic conditions. Controls were women who were attending the family planning clinic at the same hospital and who had at least two normal Papanicolaou smears. Results were adjusted for age, number of sexual partners, age at first intercourse, socioeconomic status, and oral contraceptive use. Cigarette smoking was significantly as- sociated with carcinoma in situ, severe dysplasia, and mild-mod- erate dysplasia (relative risks, 3.6, 3.3, and 2.4, respectively). Cumulative exposure to cigarette smoking (as measured by pack- years smoked) was strongly related to the risks of these conditions; women with 12 or more pack-years of exposure had relative risks of 12.7, 10.2, and 4.3, respectively, for the three conditions. There was some evidence that the risk was greatest in women who began smoking in their early teenage years. A reduction in the risk of cervical cancer appears to be another inducement for young women not to smoke. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Van Rensburg, S.].; Bradshaw, E.S.; Bradshaw, D.; et al. Oesophageal Cancer in Zulu Men, South Africa: A Case-Control Study Br J Cancer 51:399-405, 1985 403 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, tobacco, treatment modalities; Cancer Site: esophagus Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1978-1981 Sample Characteristics 422 hospital patients (esophagus cancer cases 211; controls 211): Zulu, presumed male; Durban, South Africa Variables SES: education, quality of house Data Sources Hospital medical records, survey Abstract The high rate of oesophageal cancer amongst southern African blacks has also been recorded amongst the Zulus. Data embracing a wide spectrum of factors pertaining to socio-economic status, nutrition, exposure to carcinogens, tobacco and alcohol usage and traditional health practices were obtained from 211 hospitalized oesophageal cancer patients and compared with hospital popula- tion controls matched forage and urban-rural background. Step- wise logistic regression analysis with adjustment for age effects showed that four of the many factors could adequately model the odds of being a cancer case. They were the daily consumption of purchased maize meal (relative risk (RR) 5.7) currently smoking commercial cigarettes (RR 2.6), pipe smoking (RR 2.1), and a reduc- tion of risk in those using butter or margarine daily (RR 0.51). Further significant differences (p<<0.05) in 12 other factors suggest that those with rural assets but an ability to earn a modest income external to the subsistence economy are at highest risk. They rep- resent a transitional state of Westernization which is characterised by excessive smoking habits and a diet having a low vitamin and mineral density. These results provide further evidence for the need to combat smoking and for a program of nutrient enrichment of maize meal. Wassertheil-Smoller, S.; Romney, S.L., Wylie-Rosett, J.; et al. Dietary Vitamin C and Uterine Cervical Dysplasia Am | Epidemiol 114(5):714-724, 1981 404 Subjects Screening: Pap smear; Risk Factor Exposure: reproductive factors, sexual practices, diet-nutritional factors; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics Sample Characteristics 169 individuals (blacks 75; whites 63; others 31—Cervical abnormality cases 87; controls 87): female, 15+ years old, screening clinic attenders; Bronx, NY Variables SES: income; Demographic: age, number of pregnancies Data Sources Health clinic medical records, survey Abstract A case-control study of women with cervical abnormalities identified through Pap smears, was conducted in the Bronx, New York, to explore the relationship between nutritional intake and cervical dysplasia. Nutrient intake was estimated from computer 167 analysis of three-day food records and 24-hour recall for 169 study participants (87 cases, 82 controls), including a subset of 49 pairs matched for age, race and parity. Mean vitamin C intake per day from three-day food record for controls was 107 mg, compared to 80 mg for cases (p<<0.01). Analysis of matched pairs showed similar results; 29% of cases compared to 3% of controls in matched subset had vitamin C intake less than 50% of the recommended daily allowance, yielding a ten-fold increase in risk of cervical dysplasia as estimated by odds ratio (p<0.05). Younger age, greater fre- quency of sexual intercourse and younger age at first intercourse were associated with higher risk of cervical dysplasia. Multiple logistic analyses indicated that low vitamin C intake is an indepen- dent contributor to risk of severe cervical dysplasia when age and sexual activity variables are controlled. Approximately 35% of US women in their reproductive years have daily vitamin C intake below 30 mg, and 68% have vitamin C intake below 88 mg. If other studies confirm these findings, it may be important to explore a possible protective role of supplementary vitamin C for women at high risk of cervical cancer. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Willet, W.C.; Polk, B.F.; Morris, J.S.; et al. Prediagnostic Serum Selenium and Risk of Cancer Lancet 11(8324):130-134, 1983 405 Subjects Risk Factor Exposure: diet-nutritional factors, tobacco; Cancer Site: multiple sites Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1973-1977 Sample Characteristics 321 individuals (blacks 144; whites 177—Cancer cases 111; controls 210): 30-69 years old Variables Demographic: age, sex, race-ethnicity Data Sources Laboratory test, hospital medical records, histopathology records, death registration Abstract Selenium levels in serum samples collected in 1973 from 111 subjects in whom cancer developed during the subsequent 5 years were compared with those in serum samples from 210 cancer-free subjects matched for age, race, sex, and smoking history. The mean selenium level of cases (0.129+/-SEM 0.002 micro-gm/ml was significantly lower than that of controls (0.136 +/-.002 micro- gm/ml). The risk of cancer for subjects in the lowest quintile of serum selenium was twice that of subjects in the highest. Mul- tivariate adjustment for geographical area and serum levels of lipids, vitamins A and E, and carotene, did not alter this relation. The association between low selenium level and cancer was strongest for gastrointestinal and prostatic cancers. Serum levels of vitamins A and E compounded the effect of low selenium; relative risks for the lowest tertiles of selenium were 2.4 and 3.9 in the lowest tertiles of vitamins E and A, respectively. Winn, D.M.; Blot, W.].; Shy, C.M.; et al. Snuff Dipping and Oral Cancer Among Women in the Southern United States N Engl J Med 304(13):745-749, 1981 406 Subjects Risk Factor Exposure: tobacco, alcohol; Cancer Site: oral cavity, pharynx Study Characteristics Design: case-control; Measures: odds ratio, descriptive statistics, higher level statistics; Time Frame: 1975-1978 Sample Characteristics 642 individuals (blacks 98; whites 544—Oral and pharynx cancer cases 232; controls 410): female; North Carolina Variables SES: education; Demographic: age, race-ethnicity, urban-rural, birthplace; Cancer: tumor location Data Sources Hospital medical records, death registration, survey Abstract A case-control study in North Carolina involving 255 women with oral and pharyngeal cancer and 502 controls revealed that the exceptionally high mortality from this cancer among white women in the South is primarily related to chronic use of snuff. The relative risk associated with snuff dipping among white nonsmokers was 4.2 (95 per cent confidence limits, 2.6 to 6.7), and among chronic users the risk approached 50-fold for cancers of the gum and buccal mucosa - tissues that come in direct contact with the tobacco powder. In the absence of snuff dipping, oral and pharyngeal cancer resulted mainly from the combined effects of cigarette smoking and alcohol consumption. The carcinogenic hazard of oral snuff is of special concern in view of the recent upswing in consumption of smokeless tobacco in the United States. Wynder, E.L.; Goldsmith, R. The Epidemiology of Bladder Cancer: A Second Look Cancer 40:1246-1268, 1977 407 Subjects Risk Factor Exposure: diet-nutritional factors, occupation, tobacco; Cancer Site: bladder; Cancer Measures: incidence Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1969-1974 Sample Characteristics 1,464 hospital patients (bladder cancer cases 732; controls 732): 40-80 years old; New York City, NY; Houston, TX; Los Angeles, CA; Miami, FL; Birmingham, AL; New Orleans, LA Variables SES: education, religion, Demographic: age, sex Data Sources Survey, hospital medical records Abstract A case-control study among 574 males and 158 female bladder 168 cancer patients and equal numbers of matched controls was con- ducted between 1969 and 1974 in 17 hospitals in six United States cities. We determined that cigarette smokers of both sexes were at higher relative risk than nonsmokers. Cigarette smoking was responsible for about one-half of male and one-third of female bladder cancer. There was an excess of bladder cancer patients with some previous occupational exposure, such as rubber, chem- icals, and textiles. A weak association with coffee drinking, which appeared to be independent of smoking, was found for males. Users of artificial sweeteners were not over-represented among the cases. The authors conclude that the epidemiologic pattern of bladder cancer cannot be fully accounted for by cigarette smoking and occupational exposure and suggest a series of metabolic studies to assess the role of additional factors, such as nutrition. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Wynder, E.L.; Mabuchi, K.; Whitmore, W.F., Jr. Epidemiology of Adenocarcinoma of the Kidney J Natl Cancer Inst 53:1619-1634, 1974 408 Subjects Risk Factor Exposure: diet-nutritional factors, tobacco, alcohol, endogenous hormones, occupation, geographic area, other diseases; Cancer Site: kidney; Cancer Measures: incidence, mortality, survival Study Characteristics Design: case-control, retrospective; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1950-1973 Sample Characteristics 596 hospital patients (blacks 36; whites 560—Adenocarcinoma of the kidney cases 202; controls 394): Houston, TX; New York City, NY; Los Angeles, CA; and 196 renal adenocarcinoma cases (blacks 7; whites 189): hospital patients; New York City, NY Variables SES: religion, education, occupation; Demographic: sex, age, marital status, birthplace, parity Data Sources Hospital medical records, survey Abstract A case-control study was conducted in 202 patients with renal adenocarcinoma and 394 controls matched for sex, age, race, and hospital. The results were evaluated with other epidemiologic and laboratory findings in the literature. The case-control study showed a significant but moderate association between cigarette smoking and renal adenocarcinoma, though the specific nature of the associ- ation was not determined. Male cases included a significantly larger proportion than did controls of persons with prior myocar- dial infarction. Female cases contained a significantly larger prop- ortion of persons >19% overweight than did controls. A relation- ship to marital status was suggested in females. No significant relationship was found with coffee drinking, alcohol consumption, occupational exposure, or previous disease history. From the cur- rent epidemiologic and pathophysiologic data, a working hypothesis was derived which suggested that dietary factors, pos- sibly related to a fat and/or cholesterol intake, were involved in the pathogenesis of renal adenocarcinoma. Wynder, E.L.; Stellman, S.D. Comparative Epidemiology of Tobacco-related Cancers Cancer Res 37:4608-4622, 1977 409 Subjects Risk Factor Exposure: alcohol, tobacco; Cancer Site: multiple sites Study Characteristics Design: case-control; Measures: relative risk, descriptive statistics, higher level statistics; Time Frame: 1969-1975 Sample Characteristics 7,068 hospital patients (blacks 1,196; whites 5,872—Cancer cases 3,718; matched controls 3,350): New York City, NY; Birmingham, AL; Miami, FL; New Orleans, LA Variables SES: occupation, religion, education; Demographic: age, sex, race-ethnicity Data Sources Survey, hospital medical records Abstract In a retrospective study, interviews were obtained with 3,716 patients with histologically proven cancer of the lung (Kreyberg 169 types I and II), mouth, larynx, esophagus, or bladder and with over 18,000 controls. For each of these cancers, the relative risk of both male and female present smokers increased with the quantity smoked and the duration of the habit. The strongest increase oc- curred for cancer of the lung and larynx, and the least increase occurred for cancer of the esophagus and bladder. For ex-smokers the risk decreased with years of cessation. The risk for mouth cancer of pipe and cigar smokers who inhaled much less than cigarette smokers was less than that of the latter and increased with the quantity smoked. The risk of mouth, larynx, and esophagus cancer among smokers increased with the quantity of alcohol consumed. Greater smoking habits and lesser cessation rates were noted among lower socioeconomic groups, suggesting that these groups will bear an ever increasing proportion of the burden of tobacco-related cancer. EPIDEMIOLOGY: CASE-CONTROL/COHORT/EXPERIMENTAL Ziegler, R.G.; Morris, L.E.; Blot, W.].; et al. Esophageal Cancer Among Black Men in Washington, D.C.: II. Role of Nurtition J Natl Cancer Inst 67(6):1199-1206, 1981 410 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol; Cancer Site: esophagus Study Characteristics Design: case-control; Measures: odds ratio, higher level statistics; Time Frame: 1975-1979 Sample Characteristics 490 deaths (120 esophagus cancer cases; 250 controls): black, male; Washington, DC Data Sources Death registration, survey Abstract A case-control study of esophageal cancer was conducted among the black male residents of Washington, D.C., to find reasons for the exceptionally high risk in this population. The next of kin of 120 esophageal cancer cases who died during 1975-77 and of 250 D.C. black males who died of other causes were inter- viewed. Five indicators of general nutritional status - fresh or fro- zen meat and fish consumption, dairy product and egg consump- tion, fruit and vegetable consumption, relative weight (wt/ht sqd), 170 and number of meals eaten per day - were each significantly and inversely correlated with the relative risk of esophageal cancer. Associations with other food groups were not apparent. The least nourished third of the study population, defined by any of these measures, was at twice the risk of the most nourished third. None of these associations was markedly reduced by controlling for ethanol consumption, the other major risk factor in this population; smoking; socioeconomic status; or the other nutrition measures. When the three food group consumption measures were combined into a single overall index of general nutritional status, the relative risk of esophageal cancer between extremes was 14. Estimates of the intake of vitamin A, carotene, vitamin C, thiamin, and ribofla- vin were inversely associated with relative risk; but each micronut- rient index was less strongly associated with risk than were the broad food groups that provide most of the micronutrient. Thus no specific micronutrient deficiency was identified. Instead, gen- erally poor nutrition was the major dietary predictor of risk and may partially explain the susceptibility of urban black men to esophageal cancer. EPIDEMIOLOGY: POLICY/REVIEW Adeloye, A. Neoplasms of the Brain in the African Surg Neurol 11:247-255, 1979 411 Subjects Treatment; Risk Factor Exposure: geographic area; Cancer Site: brain; Cancer Measures: incidence, prevalence Abstract At the Second Pan African Congress of Neurological Sciences held at Ibadan, Nigeria, in January, 1970, it became clear that, in spite of the active neurological and neurosurgical centers existing in Africa, the pattern of neurological disease in general, and of CNS tumours in particular, lacked definition. Hence, in January, 1972 in Nairobi, Kenya, a Pan African Symposium on “Tumours of the Nervous System in the African” was organized where neuroscientists from various parts of the African continent, and a few experts from outside Africa, met to relate their experiences on neoplasms of the central nervous system in the African and those of African descent. Much of what was presented at that Symposium forms the basis of this paper. Patterns of neoplasms of the brain are reviewed for Egypt, Eritrea, Sudan, Uganda, Kenya, Senegal, Ivory Coast, Ghana, Nigeria, Rhodesia, Zambia, Transvaal, and South Africa. Adelstein, A.M. Current Vital Statistics: Methods and Interpretation Br Med J 2:983-987, 1978 412 Subjects Methodology Abstract The paper discusses some general principles related to the man- agement of national medical statistics in the United Kingdom and presents recent statistics. Mortality rates in England and Wales are favorable for children and young adults compared to other countries but lag behind for infants and after middle age. Analysis of occupational mortality for 1970-1972 highlights the continuing effects of social factors on health at every age in the life cycle. Since 1969 country of birth has been noted on death certificates, and with the addition of similar information from the census, an analysis of mortality of immigrant for 1970-1972 is in progress. Because country of birth does not identify race, names have been used to decide between Asian and British backgrounds for those born in the Indian subcontinent and between Africans, Europeans, and Asians for those born in the African commonwealth. Since there are no population statistics of ethnic origin (apart from coun- try of birth) corresponding to the figures derived from death cer- tificates, death rates for specific causes have to be based on prop- ortions of all deaths age-standardized. Mortality rates from hyper- tension are very high among Africans and West Indians, both men and women. For ischaemic heart disease Asians have fairly high rates, while African and West Indians have low rates. American Cancer Society Cancer Statistics, 1986 CA 36(1):9-25, 1986 413 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, mortality, survival Abstract The article presents U.S. cancer incidence, mortality and survi- val data. Five-year cancer survival rates (1977-1982) from the Na- tional Cancer Institute are presented by race. Trends in cancer survival for cases diagnosed in 1960-63, 1970-73, 1973-76, and 1977- 82 are presented by race for 24 sites. Blacks had lower five-year survival rates for each of these four time periods. Anteby, 5.0.; Mor Yosef, S.; Schenker, J.G. Ovarian Cancer Geographical, Host and Environmental Factors Arch Gynecol 234:137-148, 1983 414 Subjects Prevention; Risk Factor Exposure: endogenous hormones, familial factors, reproductive factors, other diseases, chemical, geographicarea; Cancer Site: ovary; Cancer Measures: incidence Abstract The incidence of ovarian cancer is reviewed among different races and ethnic groups. We present the effect of socio-economic class, age, endocrinological factors, menstrual cycle, reproductive history, familial, and genetic factors, viral infections, chemical car- cinogens, and previous exposure to radiation on the incidence of ovarian cancer. There is evidence that environmental factors affect the distribution of various types of ovarian cancer. Migration may also change the susceptibility to this disease and the mean age was varied between countries being lowest where incidence was highest. The disease appeared to have a familial incidence. The protective effect of oral contraceptive steroids and the role of asbes- tos as a chemical carcinogen are discussed. 171 EPIDEMIOLOGY: POLICY/REVIEW Bader, M. Letter to the Editor Arch Environ Health 34(3):189, 1979 415 Subjects Risk Factor Exposure: occupation, tobacco; Cancer Site: lung Abstract In response to the article “Smoking Patterns by Occupation, Industry, Sex and Race” (Arch Environ Health, Nov/Dec 1978, 313-317), the author states that the data presented do not warrant the conclusion that the increased incidence of lung cancer among blacks is due to occupational exposure. He suggests that the in- crease may be due to black males smoking different brands of cigarettes than white males and smoking them more completely due to economic considerations. In addition, he notes that blacks have a 12-14 percent smaller vital lung capacity and that there may be significant differences in the age distributions of blacks and whites. Baquet, C.R.; Ringen, K. Cancer Control in Blacks: Epidemiology and NCI Program Plans 416 In: Advances in Cancer Control: Health Care Financing and Research (eds) Mortenson, L.E.; Engstrom, P.F.; Anderson, P.N. New York: Alan R. Liss, 1986, p 215-227 Subjects Access and Utilization of Health Services; Diagnosis; Treatment; Risk Factor Exposure: tobacco, alcohol, diet-nutritional factors, occupation; Cancer Site: multiple sites; Cancer Measures: incidence, mortality, survival Abstract Cancer data derived from the Surveillance, Epidemiology and End Results (SEER) program of the National Cancer Institute and the National Center for Health Statistics indicate that U.S. blacks are disproportionately affected by cancer and have excesses in cancer incidence and mortality as well as poorer survival overall and for specific cancer sites when compared to whites: 1) Cancer Incidence: Breast (under age 40), cervix, esophagus, lung (males), multiple myeloma, pancreas, prostate, stomach; 2) Cancer Mortal- ity: Cervix, esophagus, larynx, lung, multiple myeloma, pancreas, prostate; and 3) Cancer Survival: Bladder, breast, corpus uterus, prostate, rectum. Factors which may explain the differences in black and white cancer experience include: 1) Risk Factors/Expo- sure: Tobacco use patterns, combined effects of tobacco and al- cohol, dietary and nutritional patterns, and greater exposures to hazardous substances in the workplace; 2) Socioeconomic Status (SES): This affects general health status, health and medical care resources as well as risk factors and exposure. Berg, ].W.; Connelly, R.R. Updating the Epidemiologic Data on Pancreatic Cancer Semin Oncol 6(3):275-283, 1979 417 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, occupation, tobacco; Cancer Site: pancreas; Cancer Measures: incidence, mortality Abstract The purpose of this article is to supplement and update cited reviews of the topic, to provide more actual estimates of risk than other reviews have included, and to cite references not mentioned therein. The demography of pancreatic cancer is described includ- ing racial, ethnic, and socioeconomic differences in cancer risk. U.S. blacks as a whole have the world’s highest pancreatic cancer rates. Evidence linking specific risk factors to pancreatic cancer are reviewed including tobacco, occupation, diabetes, and alcoholism. The authors conclude that the epidemiology of pancreatic cancer does not yet point to specific carcinogens except cigarette smoke and that we cannot yet estimate how much of pancreatic cancer is due to special environmental-cultural factors and how much of the observed differences between populations are due to differ- ences in case ascertainment. Blot, W.].; Fraumeni, J.F., Jr.; Morris, L.E. Patterns of Laryngeal Cancer in the United States Lancet 11:674-675, 1978 418 Subjects Risk Factor Exposure: occupation, alcohol, tobacco, geographic area; Cancer Site: larynx; Cancer Measures: mortality Abstract This study examined factors that may be responsible for the geographical distribution of laryngeal cancer in the United States. The association of the age-adjusted laryngeal cancer rates with demographic and industrial variables for 3,056 counties of the contiguous U.S. were reviewed. The U.S. death rates for laryngeal cancer in white males have been more or less constant over the period 1950-1969, but analysis of the county data showed a steady increase of 20 - 30 percent in rural areas throughout the U.S. Rates among non-white males during this period nearly doubled in the South in both urban and rural areas, while increasing more slowly elsewhere. 172 EPIDEMIOLOGY: POLICY/REVIEW Boxer, R.]. Adenocarcinoma of the Prostate Gland Urological Survey 27:75-79, 1977 419 Subjects Treatment; Diagnosis; Risk Factor Exposure: endogenous hormones; Cancer Site: prostate; Cancer Measures: incidence, mortality Abstract The author reviews the history, etiology, and epidemiology of adenocarcinoma of the prostate gland. Attention is given to tumor staging, diagnosis, laboratory studies, immunology, and treat- ment. The author states that adenocarcinoma of the prostate gland is the second most common type of cancer and the third leading cause of death by a cancer in the male population. The estimated number of new cases for nonwhite male patients in 1976 is 8,500, 21.8 percent of the total cancers diagnosed. In white male patients there will be 47,500 new cases, 15.8 percent of the total new cases of cancer. The number of deaths is estimated at 19,300 in 1976. Prostatic cancer is a major problem, especially in black subjects for whom the incidence is much higher (78.8 cases per 100,000 population versus 44.7 for white subjects) and this margin is in- creasing. Bracken, M.B.; Brinton, L.A.; Hayashi, K. Epidemiology of Hydatidiform Mole and Choriocarcinoma 420 Epidemiol Rev 6:52-75, 1984 Subjects Cancer Site: gestational trophoblastic disease; Risk Factor Exposure: reproductive factors; Cancer Measures: incidence Abstract The article focuses on 1) the incidence and distribution of hydatidiform mole and choriocarcinoma and 2) the major etiologic hypotheses for each disease. The authors assert that the study of hydatidiform mole and choriocarcinoma offers the possibility for insight into other important questions, such as the origin of some congenital malformations, the mechanism for maintaining normal pregnancies, and the etiology of other cancers. Bradshaw, E.; Harington, J.S. The Cancer Problem Among Blacks: South Africa 421 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: AlanR. Liss, 1981, p 17-33 Subjects Risk Factor Exposure: geographic area; Cancer Site: multiple sites; Cancer Measures: mortality Abstract The overall mortality rate for cancer in South African urban blacks is lower than that for South African whites or coloreds, both males and females. In addition, there is a different pattern for cancer among South African blacks. When compared to U.S. blacks, South African blacks still show lower mortality rates. How- ever, the cancer pattern in white South Africans of both sexes is similar to that of U.S. whites. Brink, A.J. Medical Research in the Republic of South Africa S Afr Med J 51:493-494, 1977 422 Subjects Cancer Site: colon-rectum, esophagus, liver Abstract A brief overview of the policies and research priorities of the South African Medical Research Council is given. Some of the unique medical problems of South Africa on which research is being done are indicated. 173 EPIDEMIOLOGY: POLICY/REVIEW Buckley, J.D. The Epidemiology of Molar Pregnancy and Choriocarcinoma Clin Obstet Gynecol 27(1):153-159, 1984 423 Subjects Biologic Marker: HLA, ABO blood group; Risk Factor Exposure: reproductive factors, familial factors, exogenous hormones; Cancer Site: gestational trophoblastic disease; Cancer Measures: incidence Abstract The available evidence suggests quite strongly that a hydatidiform mole arises as a consequence of production of a defective ovum. That the defect is maternal is implied by the risk associated with increasing maternal (but not paternal) age, by the recurrence in some women of multiple moles (even in one instance by different husbands), and by reports of familial aggregation of hydatidiform moles among sisters. The absence of maternal genetic material in most complete hydatidiform moles and the dispermic fertilization of some complete and most partial moles further suggests that the ovum is defective. The causes of defective or “blighted” ova are not known. The dramatic increase in risk with age, the smaller increase in young women, and the high incidence in women in parts of Asia must all be important clues. Consanguin- ity, malnutrition, racial (genetic) predisposition, oral contracep- tives, and viral infections have been advanced as causative or contributory factors. Disorders of ovulation, with prolongation of the follicular phase or an inadequate luteal phase, may produce a suboptimum ovum. At some time after ovulation, when the viabil- ity of the ovum has become borderline, abnormal (e.g., dispermic) fertilization may become more likely. These and other speculations are being investigated in the detailed case-control studies of trophoblastic disease epidemiology, currently in progress. Centers for Disease Control Cancer Patient Survival by Racial/Ethnic Group-United States, 1973-1979 MMWR 34(18):248-255, 1985 424 Subjects Cancer Site: multiple sites; Cancer Measures: survival Abstract The article reports on a National Cancer Institute study of SEER 5-year survival rates from 1973 to 1979 for eight racial-ethnic groups in the United States - Anglos, Hispanics, blacks, American Indians, Chinese, Japanese, Filipinos, and Hawaiians. Black males experi- enced poorer survival than Anglo males for cancers of the rectum, prostate, bladder, and thyroid; black females had poorer survival than Anglo females for cancers of the bladder, corpus uteri, and breast. In an editorial note, MMWR reports on two studies which investigated the role of socioeconomic status and treatment factors in the survival differentials of U.S. racial-ethnic groups. Chiedozi, L.C.; Lippman, M.E. Management of Advanced Breast Cancer in Africa: Strategies in the Absence of Hormone Receptor Assay Data Int Surg 70:57-61, 1985 425 Subjects Treatment; Risk Factor Exposure: reproductive factors; Cancer Site: breast Abstract The oncologist in Africa and many other third world countries is hampered in the management of breast cancer by the absence of hormone receptor assays in most teaching hospitals. The little information that is available in the literature on hormone receptors in black, breast cancer patients as well as other available hormone receptor information can serve as a data base for the development of a rational approach to this management dilemma. Using the available information, an approach, in the form of a flow chart, has been developed for the management of breast cancer in Africa. It is further suggested that breast cancer should be managed using protocols designed to answer various questions about breast cancer in Africa. Cook, P.].; Burkitt, D.P. Cancer in Africa Br Med Bull 27(1):14-20, 1971 426 Subjects Methodology; Risk Factor Exposure: geographic area; Cancer Site: multiple sites; Cancer Measures: prevalence Abstract This review article discusses the distribution in select popula- tions in Africa of the following cancer sites: liver, esophagus, 174 stomach, penis, cervix, Kaposi's sarcoma, bladder, scar epithelium, Burkitt's lymphoma, and large bowel. Discusses prob- lems and significance of epidemiological investigations of cancer in Africa. EPIDEMIOLOGY: POLICY/REVIEW Cook-Mozaffari, P. Symposium on Tumours in the Tropics Trans R Soc Trop Med Hyg 76(2):157-163, 1982 427 Subjects Risk Factor Exposure: diet-nutritional factors, reproductive factors, tobacco, other diseases, sexual practices, geographic area; Cancer Site: esophagus, bladder, cervix uteri, penis; Cancer Measures: incidence Abstract The article reviews the relative frequencies and incidence levels within Africa of cancers of the esophagus, bladder, cervix, and penis. Correa, P. Gastrointestinal Cancer Among Black Populations 428 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 197-211 Subjects Risk Factor Exposure: alcohol, tobacco, diet-nutritional factors, geographic area; Cancer Site: esophagus, stomach, colon- rectum; Cancer Measures: incidence Abstract This paper supports the idea that cancer in blacks reflects the interaction between man and the environment. It purports that some conclusions concerning blacks and cancer can be explained on the basis of postmigration conditions in the United States as well as the interaction of social factors and disease patterns. Correa, P. Epidemiology of Polyps and Cancer Major Probl Pathol 10:126-152, 1978 429 Subjects Risk Factor Exposure: other diseases, diet-nutritional factors, geographic area, bacterial flora; Cancer Site: colon-rectum; Cancer Measures: incidence Abstract This paper examines the polyp-cancer sequence from the epidemiologic point of view. It also briefly reviews the epidemiol- ogy of large bowel cancer and then compares it to what is known of the epidemiology of the different types of polyps of the large bowel. It concludes that adenomatous polyps are a good epidemiologic indicator of colon cancer risk, and that their presence should be helpful in advancing from studies of the epidemiology of colon cancer to the epidemiology of precursor lesions. Cutler, S.J. Some Epidemiologic Observations on Cancer of the Female Breast 430 Int J] Radiation Oncol Biol Phy (2):755-761, 1977 Subjects Cancer Site: breast; Cancer Measures: incidence increasing more rapidly and has not yet levelled off. The author Abstract concludes that it is likely that the development of breast cancer is Using data from published sources, variation in female breast cancer incidence by race, age, and geographic areas is described. Data from three morbidity surveys carried out in 1937-39, 1947-48, and 1969-71 indicate that the incidence rate in black women is still lower than in white women but the rate in black women has been influenced by a complex of factors including genetically deter- mined characteristics, such as endocrine levels, as well as be- havioral and environmental factors such as age at first childbearing and diet. Davies, J.N.P. Kaposi's Sarcoma 431 Presented to the Consultants Seminar at the Institute of Dermatology, 19th April, 1963 Subjects Cancer Site: sarcoma Abstract This is an early review of the biology of Kaposi's sarcoma in Africa. EPIDEMIOLOGY: POLICY/REVIEW Davis, D.L.; Bridbord, K.; Schneiderman, M. Estimating Cancer Causes: Problems in Methodology, Production, and Trends 432 In: Quantification of Occupational Cancer, Banbury Report 9 Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. (eds) Peto, R.; Schneiderman, M. p 285-316, 1981 Subjects Methodology; Risk Factor Exposure: diet-nutritional factors, alcohol, occupation, tobacco; Cancer Site: multiple sites; Cancer Measures: incidence, mortality Abstract This three-part paper reviews difficulties in allocating the prop- ortion of cancers attributable to specific causes: The first part briefly notes problems associated with allocating causes of cancer, includ- ing a discussion of previous efforts to make such estimates based on epidemiological studies. The second part discusses recent trends in production, consumption, and use of materials known or suspected to be carcinogens. The third part examines trends in the 1970s of cancer incidence and mortality within the United States and speculates about explanations for the decline in cancer in people under age 45 and the increases in those over 45. The possible contributions of cigarette smoking and industrial expo- sures also are examined. Davis, M.E. Occupational Hazards and Black Workers Urban Health, Aug 1977, p 16-18 433 Subjects Risk Factor Exposure: occupation; Cancer Site: multiple sites Abstract This article presents an overview of the occupational health hazards faced by black workers. Particular attention is given to the steel, rubber, laundry and dry cleaning and sanitation indus- tries. The author discusses some of the general carcinogens to which black workers are exposed. Devesa, S.S. Descriptive Epidemiology of Cancer of the Uterine Cervix Obstet Gynecol 63(5):605-612, 1984 dad Subjects Cancer Site: cervix uteri; Cancer Measures: incidence, mortality, survival Abstract Substantial decreases in both incidence of and mortality from cancer of the uterine cervix occurred during the past 30 years. Decreases are apparent among both whites and nonwhites, al- though rates among black women remain considerably higher than those among white women. In spite of geographic differences in rates, decreases have been observed in all areas of the United States for which data are available. The decline in mortality and in invasive cancer incidence has been occurring within virtually all age groups. Diagnosed infrequently in the past, carcinoma in situ now has an incidence rate two to three times higher than that of invasive cancer and also has a much younger age distribution. Carcinoma of the cervix occurs much more frequently among women of lower income and education level; this accounts for much of the observed racial difference in incidence. Doll, R. Letters to the Editor: Flouridation and Cancer Lancet 11(8032):296, 1977 435 Subjects Risk Factor Exposure: fluoridation; Cancer Site: multiple sites Abstract This letter to the editor is critical of an article which concerns fluoridation and cancer. The author states that there has been a 176 tremendous urbanization of nonwhites. He feels that urbanization has caused a substantially higher increase in cancer mortality among nonwhites. EPIDEMIOLOGY: POLICY/REVIEW Dunn, J.E., Jr. Geographic Considerations of Endometrial Cancer Gynecol Oncol 2:114-121, 1974 436 Subjects Methodology; Risk Factor Exposure: geographic area; Cancer Site: breast, cervix uteri, corpus uteri, ovary; Cancer Measures: incidence Abstract Investigating the geographic variation in the occurrence of cancer of the uterine corpus presents certain problems. Mortality rates are particularly unsuitable for this because of the large prop- ortion of deaths from uterine cancer not specified as to whether they originated in the cervix or the corpus. Incidence rates, on the other hand, may be somewhat inflated if cytology screening is being extensively used in a population of women, but the designa- tion between corpus and cervix is nearly complete. Incidence rates compiled by the International Union Against Cancer indicate con- siderable variation in the frequency of uterine corpus cancer among female populations throughout the world. There is a correlation between the frequency of this cancer and that of the ovary and breast, but not with cervical cancer. In the United States southern white women tend to have cancers of the uterine corpus, breast, and ovary less frequently and of the cervix more frequently than women of the north and west. American-born Japanese women are developing breast cancer nearly as frequently in California as white women. This may be true for ovarian cancer as well. Uterine corpus cancer may be increasing but the population is still too young to be sure, since this cancer reaches its highest rates at older ages. All these cancers are less frequent in Japan. The Amer- ican-born Japanese will be followed with considerable epidemiological interest. Dunn, J.E., Jr.; Austin, D.F. Cancer Epidemiology in the San Francisco Bay Area Natl Cancer Inst Monogr 47:93-98, 1977 437 Subjects Methodology; Risk Factor Exposure: diet-nutritional factors, occupation; Cancer Site: multiple sites, colon, colon-rectum; Cancer Measures: incidence Abstract The Third National Cancer Survey of 1969-71 included the five counties of the San Francisco-Oakland Standard Metropolitan Statistical Area. The complete cancer reporting for this area, begun by the Third National Cancer Survey, was continued by the Califor- nia Tumor Registry as part of the San Francisco Bay Area Resource for Cancer Epidemiology. The population-based cancer-reporting system provided an excellent data base for epidemiologic studies, a number of which (planned or in progress) were described briefly. Those in progress include: cancer of the ovary, corpus uteri, and breast as related to child bearing, fertility, exogenous hormones, etc.; the relationship of diet to breast cancer occurrence among Japanese; diet and colorectal cancer among blacks; and the relation- ship of cervical cancer to cytology in Alameda County. Other study proposals are under consideration. Earles, L.C., III Cancer Epidemic in the Black Community J Natl Med Assoc 76(2):93-94, 1984 438 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, prevalence, survival Abstract The article reviews cancer rates in black Americans and presents information from a national survey of urban black attitudes, knowl- edge and practices regarding cancer. The author concludes that socioeconomic factors rather than inherent biological characteris- tics have caused the prevailing differences observed both between races and within the black racial group itself. Enstrom, J.E.; Austin, D.F. Interpreting Cancer Survival Rates, The Available Data on Survival Are Not a Sensitive Measure of Progress Science 195:847-851, 1977 439 Subjects Cancer Site: multiple sites; Cancer Measures: survival; Methodology Abstract The purpose of this article is to put the discussion of the “prog- ress or lack of progress” in cancer research and control in the United States in perspective by pointing out limitations and qual- ifications in the interpretation of survival rates and trends. Factors influencing survival rates are reviewed. Data from the California Tumor Registry and the End Results Evaluation Program are pre- sented. In summary, the author states that survival rates should not be used as the sole or primary measure of progress in cancer control because factors unrelated to the efficacy of treatment play an important role in their determination. Instead, it is suggested that progress can better be measured by the use of incidence and mortality rates. 177 EPIDEMIOLOGY: POLICY/REVIEW Flanders, W.D. Review: Prostate Cancer Epidemiology Prostate 5:621-629, 1984 440 Subjects Rick Factor Exposure: geographic area, sexual practices, familial factors, diet-nutritional factors, endogenous hormones, occupation; Cancer Site: prostate; Cancer Measures: incidence, mortality, survival Abstract Prostate cancer is common among men in the United States. Factors of possible importance in the etiology of prostate cancer include diet, primarily implicated by ecologic studies of national, regional, and ethnic variation in rates; endocrine function, impli- cated by the importance of endocrine function in normal prostatic growth and in the treatment of prostate cancer; genetic susceptibil- ity, supported by familial aggregation; some aspect of sexual, suggested by case-control differences in sexual behavior; and occu- pational exposure, particularly cadmium exposure. Despite the public health importance of prostate cancer, it has received only moderate epidemiologic study; thus the etiologic importance of these and other possible determinants of prostate cancer risk is uncertain. Fraser, P.; Adelstein, A.M. Recent Trends Recent Results Cancer Res 83:1-10, 1982 441 Subjects Cancer Site: colon, rectum; Cancer Measures: incidence, mortality Abstract Recent trends in vital statistics relating to colorectal cancer re- veal geographic, ethnic, and regional differences in the distribution of these cancers. They show changes in mortality over time and after migration, a higher risk in urban populations and the lower social classes, and generally higher mortality among males. The low rates of survival serve to emphasize that while clinicians are striving towards earlier diagnosis and more effective treatment, clues to aetiology, which may lead to prevention, must be vigor- ously pursued. Fraumeni, J.F., Jr. Epidemiological Opportunities in Alcohol-related Cancer Cancer Res 39:2851-2852, 1979 442 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, occupation; Cancer Site: oral cavity, pharynx, esophagus, larynx Abstract Alcohol combined with tobacco smoking is an established risk factor for cancers of the oropharynx, esophagus, and larynx. It should be possible to clarify further the role of alcohol itself, the modifying effects of tobacco, dose-response relationships, and nut- ritional cofactors. Studies are also needed to delineate the steps by which alcohol consumption leads to liver cancer and to resolve the suggestion that certain beverages may predispose to other 178 cancers including those of the pancreas and rectum. Epidemiolog- ical investigations should be combined with experimental work to identify hazardous fractions in alcoholic beverages and to delineate the mechanisms by which alcohol promotes carcinogenesis. Epidemiologists and biometricians may also contribute toward the development of programs aimed at primary prevention and early detection of cancers related to alcohol and tobacco. Incorporation of research questions into data collection systems deserves serious consideration as a means of obtaining additional valuable informa- tion for etiological studies. EPIDEMIOLOGY: POLICY/REVIEW Fraumeni, J.F., Jr. Cancers of the Pancreas and Biliary Tract: Epidemiological Considerations Cancer Res 35:3437-3446, 1975 443 Subjects Risk Factor Exposure: diet-nutritional factors; alcohol, occupation, tobacco, geographic area, other diseases, familial factors; Cancer Site: pancreas, biliary tract; Cancer Measures: incidence Abstract The epidemiological patterns for pancreatic and biliary cancers reveal more differences than similarities. Pancreatic carcinoma is common in western countries, although 2 Polynesian groups (New Zealand Maoris and native Hawaiians) have the highest rates in- ternationally. In the United States the disease is rising in frequency, predominating in males and in blacks. The rates are elevated in urban areas, but geographic analysis uncovered no clustering of contiguous counties except in southern Louisiana. The origin of pancreatic cancer is obscure, but a twofold increased risk has been documented for cigarette smokers and diabetic patients. Alcohol, occupational agents, and dietary fat have been suspected, but not proven to be risk factors. Except for the rare hereditary form of pancreatitis, there are few clues to genetic predisposition. In con- trast, the reported incidence of biliary tract cancer is highest in Latin American populations and American Indians. The tumor predominates in females around the world, except for Chinese and Japanese who show a male excess. In the United States the rates are higher in whites than blacks, and clusters of high-risk counties have been found in the north central region, the south- west, and Appalachia. The distribution of biliary tumors parallels that of cholesterol gallstones, the major risk factor for biliary cancer. Insights into biliary carcinogenesis depend upon clarification of lithogenic influences, such as pregnancy, obesity, and hyperlipop- roteinemia, exogenous estrogens, familial tendencies, and ethnic- geographic factors that may reflect dietary habits. Noncalculous risk factors for biliary cancer include ulcerative colitis, clonor- chiasis, Gardner's syndrome, and probably certain industrial expo- sures. Within the biliary tract, tumors of the gallbladder and bile duct show epidemiological distinctions. In contrast to gallbladder cancer, bile duct neoplasms predomiate in males: they are less often associated with stones and more often with other risk factors. In some respects, bile duct and pancreatic tumors are alike. The male predominance of both tumors, an association between cholecystectomy and pancreatic cancer, and other considerations have prompted the notion that the same biliary carcinogens may affect the bile duct, ampulla of Vater, or, by reflux, the pancreatic duct. Various epidemiological and interdisciplinary approaches are needed to further clarify the origins of biliary tract and pancrea- tic cancers, but nutritional studies hold special promise in laying the groundwork for prevention of these tumors. Garfinkel, L.; Sarokhan, B. Trends in Brain Cancer Tumor Mortality and Morbidity in the United States Ann NY Acad Sci 381:1-5, 1982 dad Subjects Cancer Site: brain; Cancer Measures: incidence, mortality Abstract This paper presents statistical and epidemiological information on the patterns of brain cancer mortality and morbidity in the United States over the past 40 years. Mortality data were obtained from Annual Vital Statistics of the United States (National Center for Health Statistics) volumes; incidence data were from the Na- tional Cancer Institute's SEER (Surveillance, Epidemiology, and End Results) program, 1973-1977. Since 1950, whites have higher mortality rates for malignant tumors of the brain and central nerv- ous system than nonwhites. Gilder, S.S.B. Carcinoma of the Esophagus Ann Intern Med 87(4):494, 1977 445 Subjects Cancer Site: esophagus; Cancer Measures: incidence, survival Abstract In South Africa, blacks have a higher incidence of cancer of the esophagus than whites or Indians, especially in the eastern section of the country, Transkei. The cancer also appears at least 10 years earlier in blacks than whites and is much more aggressive in the black man (not necessarily the woman). It is important to find the cause of esophageal cancer so that the disease may be prevented. The vast majority of black patients is South Africa are 179 first seen when the lesion is already inoperable. This may be due to the traditional attitude of the black man toward cancer and surgery or communication problems when the patient presents himself at a hospital. The tragic contrast between the races was pointed up by a series reported by Marchand of Johannesburg. With exactly the same management policy (operation and roentgenographic therapy), one third of the white patients sur- vived 5 years, although they were on average almost 20 years older. No black patient survived nearly that long. EPIDEMIOLOGY: POLICY/REVIEW Gori, G.B. Role of Diet and Nutrition in Cancer Cause, Prevention and Treatment 446 Bull Cancer (Paris) 65:115-126, 1978 Subjects Treatment; Risk Factor Exposure: diet-nutritional factors, geographic area; Cancer Site: multiple sites; Cancer Measures: incidence, mortality Abstract The roles of diet and nutrition as etiologic factors in cancer incidence is discussed. Diet and nutrition have recently been rec- ognized as causative agents in other chronic diseases. Epidemiologic evidence supportive of the hypothesis that different diets rather than other environmental contaminants are important in the causation of certain forms of cancer is reviewed, including studies on several migrant populations and ethnic groups. Specific dietary components have been correlated with certain forms of cancer, such as bowel cancer and fat consumption, and a theory of the mechanism is set forth. The animal studies discussed strongly confirm the results of epidemiologic studies in man. Ad- ditionally, nutrition’s important role in cancer therapy and rehabili- tation is described. A variety of intervention techniques are avail- able to maintain the nutrition status of the cancer patient. Finally, the research priorities and the Diet, Nutrition and Cancer Program (DNCP) of the National Cancer Institute are explained. Graham, S.; Mettlin, C. Diet and Colon Cancer Am ] Epidemiol 109(1):1-29, 1979 447 Subjects Prevention; Methodology; Risk Factor Exposure: diet-nutritional factors; Cancer Site: colon Abstract The article reviews two dietary hypotheses about the etiology of cancer of the colon. One hypothesis suggests that our allegedly over-refined diet, which supposedly reduces the amount of fiber in foods may figure in the etiology. The other suggests that a high animal-fats diet is related to an increased risk of colon cancer. The reliability of various methods of obtaining diet information is re- viewed. The article concludes that inconsistencies in findings and the paucity of data linking ingestion of meat to bowel cancer suggest that there is not enough evidence to relate a high-meat or beef diet to cancer of the colon at this time. Neither is their enough evidence to conclude that decreased risk is associated with inges- tion of vegetables. But the consistency of both epidemiologic and laboratory findings regarding vegetables and the logic and variety of the evidence implicating animal fats suggests the need for further inquiry regarding both factors as well as others reviewed in the paper. Laboratory studies are needed as well as studies on the past diets of humans. The interview approach, despite its crudeness, has some promise of shedding light on the epidemiol- ogy of colon cancer. Grant, M.C.G. Carcinoma of the Cervix—A Tragic Disease in South Africa S Afr Med J 61:819-822, 1982 448 Subjects Screening: Pap smear; Prevention; Cancer Site: cervix uteri; Cancer Measures: incidence, mortality, prevalence Abstract Reports from several large gynaecological oncology units in South Africa reveal that most patients with carcinoma of the cervix present in an advanced stage of the disease when it is incurable. Mass Papanicolaou smear campaigns in other areas in the world have resulted in almost complete eradication of the disease, and it is suggested that invasive carcinoma of the cervix is preventable in most cases. It is considered that it is the responsibility of the State Health Department to institute mass Papanicolaou smear campaigns in South Africa for certain sections of the population who are at risk, and these have been defined. This would reduce the enormous cost of management of most of these cases in their advanced stages. Grufferman, S.; Delzell, E. Epidemiology of Hodgkin's Disease Epidemiol Rev 6:76-106, 1984 449 Subjects Risk Factor Exposure: occupation, familial factors; Cancer Site: lymphoma; Cancer Measures: incidence, prevalence, survival Abstract This paper reviews the clinical picture and pathology of Hodgkin's disease, particularly as they relate to nosology and the possibly infectious etiology of Hodgkin's disease. The descriptive epidemiology of Hodgkin's disease and the major etiologic hypoth- eses that have emerged are discussed, as well as specific epidemiologic studies related to the disease’s putative infectious- ness and to occupational exposures as risk factors for the disease. Finally, several areas are identified in which additional epidemiologic research is needed. EPIDEMIOLOGY: POLICY/REVIEW Haddy, T.B. Cancer in Black Children Am | Pediatr Hematol Oncol 4(3):285-292, 1982 450 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, survival Abstract A review of the literature indicates that black children in the United States have a lower overall incidence of cancer and are less prone to leukemia and certain solid tumors, including neuroblas- toma, rhabdomyosarcoma, Ewing's sarcoma, testicular tumors, liver tumors, and malignant melanoma, than are white children. Black children with acute lymphoblastic leukemia and retinoblas- toma, but not with neuroblastoma, Wilms’ tumor, and rhab- domyosarcoma, have poorer survival rates than white children. Socioeconomic status appears to be an important reason for the discrepant outlook, but genetic differences may also play a role. Consideration of these issues will assist in planning appropriate treatment regimens. Higgins, L.T.T. Epidemiology of Lung Cancer in the United States IARC Sci Pub 16:191-203, 1977 451 Subjects Risk Factor Exposure: tobacco, air pollution, geographic area; Cancer Site: lung; Cancer Measures: incidence, mortality Abstract This paper has reviewed the temporal and spatial distribution of lung cancer in the United States. Some of the personal charac- teristics which are important determinants of this type of cancer and which need to be considered in assessing the role of environ- mental air pollution have also been considered. Recent studies which have attempted to relate lung cancer mortality to possible arsenic exposure or to levels of benzo[a]pyrene in different places are noted. While an urban effect is undoubted, it is still not certain that it is due to carcinogenic pollutants in the air. Although there may be exceptional areas, environmental air pollution probably contributes only a very small fraction to the effect of cigarette smoking on this type of cancer. Hill, P.; Garbaczewski, L.; Helman, P.; et al. Environmental Factors and Breast and Prostatic Cancer Cancer Res 41:3817-3818, 1981 452 Subjects Risk Factor Exposure: diet-nutritional factors, reproductive factors, endogenous hormones, geographic area; Cancer Site: breast, prostate Abstract Despite epidemiological evidence supporting a causal relation- ship between environmental factors and the development of breast and prostatic cancers, direct confirmation is unproven. Differences in hormone profiles in cancer patients and in patients with an increasing number of potential risk factors together with differ- ences in life style and diet, suggest a relationship between diet, hormonal metabolism, and these endocrine-associated cancers. Modification of hormonal status by diet in men and women at different risk suggest that specific dietary components play a major role in these diseases. Hutchison, G.B. Incidence and Etiology of Prostate Cancer Sup to Urology 17(3):4-10, 1981 453 Subjects Methodology; Risk Factor Exposure: diet-nutritional factors, occupation, reproductive factors, endogenous hormones, geographic area; Cancer Site: prostate; Cancer Measures: incidence, mortality Abstract The author reviews prostate cancer incidence and mortality trends in the United States and examines age and race differentials. The author states, after reviewing prostate mortality data from 46 181 countries and migration data, that U.S. blacks may have the highest rate in the world and that this must reflect some factor to which they alone are exposed. Information about a variety of risk factors are reviewed. The author concludes that causal theories involving mechanisms of viral infection, hormonal imbalance, sexual activity, and occupational exposure are not supported sufficiently to lead to preventive programs. EPIDEMIOLOGY: POLICY/REVIEW Hutchison, G.B. Epidemiology of Prostatic Cancer Semin Oncol 3(2):151-159, 1976 454 Subjects Prevention; Risk Factor Exposure: occupation, geographic area, sexual practices, reproductive factors, air pollution; Cancer Site: prostate; Cancer Measures: incidence, mortality Abstract The article reviews time trends in prostate cancer incidence and mortality for both nonwhite and white U.S. males. Interna- tional variations, etiology, risk factors, and possible preventive actions are examined. Jones, G.W. Diagnosis and Management of Prostate Cancer Cancer 51(12):2456-2459, 1983 455 Subjects Diagnosis; Treatment; Cancer Site: prostate; Cancer Measures: incidence Abstract Representing 18% of male cancers, carcinoma of the prostate gland is the second most common cancer and the third leading cause of cancer deaths in U.S. males. In 1982, it is estimated 73,000 new cases and 23,000 deaths will occur from the most common cancer in men older than 65 years of age. U.S. blacks have the highest incidence in the world. Annual rectal digital examination of the prostate gland discovering a suspicious nodule or indurated area of the gland remains the earliest means of detection of the prostatic cancer in the asymptomatic man. However, more com- monly the earliest unsuspected detection occurs in the surgical specimen removed following operations to relieve obstructive symptoms of benign prostatism. Survival of patients with this disease correlates with the stage and grade of disease at time of diagnosis, consequently it is imperative to ascertain the extent of disease in order to select the appropriate therapy for the patient. Radical prostatectomy, interstitial radiation implantation, and ex- ternal radiation are the current means of treatment for patients with low-stage disease, while hormonal manipulation alone or in combination with external radiation and chemotherapy are the primary means of treatment for symptomatic patients with car- cinoma of the prostate. The expected five-year survival for low- stage disease is 70-80%. King, H.; Diamond, E.; Lilienfeld, A.M. Some Epidemiological Aspects of Cancer of the Prostate J Chron Dis 16:117-153, 1963 456 Subjects Risk Factor Exposure: geographic area, occupation, familial factors; Cancer Site: prostate; Cancer Measures: incidence, mortality, survival Abstract The study of the distribution of an illness in a population in terms of selected demographic and genetic characteristics repre- sents one method for discovering disease patterns and for develop- ing etiological hypotheses upon which more definitive research can be based. A cancer site which has not been the subject of systematic epidemiological investigation is the prostate, despite the fact that it is one of the most common sites of cancer in the male. It is the purpose of this paper to assemble and review pre- vious significant findings on certain epidemiological aspects of this site of malignancy. Hopefully, this endeavor will serve as a stimulus for further investigation into this much neglected area. In this presentation, statistical data from divergent sources pertain- ing to prostatic cancer mortality and morbidity are reviewed. First are considered the trends of the frequency of this disease, after which the distribution by such characteristics as age, race, geog- raphical area, nativity and religion, marital status, socio-economic standing, and familial aggregation is discussed. An attempt is made to indicate potentially fruitful areas of epidemiological re- search which could possibly provide answers to some of the ques- tions which are raised by this review. Age-specific incidence rates for nonwhites were found to be much higher than for whites for all ages up to 80. The age-adjusted rate per 100,000 male population for nonwhites exceeded that for whites by 44 percent. Furthermore, the percentage increase in incidence between 1937 and 1947 is much greater among nonwhites than whites, 63 and 17 per cent respectively. Kleinstein, R.N.; Lehman, H.F. Optometric Public Health, Incidence and Prevalence of Eye Cancer Am J Optom Physiol Opt 54(1):49-51, 1977 457 Subjects Cancer Site: eye; Cancer Measures: incidence, mortality, survival, prevalence Abstract Cancer of the eye is the only ocular disease which directly threatens life. The average annual incidence rate (new cases) in the U.S. is about 1/100,000 population and the estimated preva- lence rate (total cases) is about 12/100,000 population. With few exceptions, eye cancer occurs more frequently in whites than in blacks and more often in older than in younger people. The survival rate for persons with localized ocular tumors receiving early treat- ment is almost 80%. This statistic indicates the importance of op- tometric detection of this uncommon disease. 182 EPIDEMIOLOGY: POLICY/REVIEW Larson, E. The Epidemiology of Primary Brain Tumors 458 J Neurosurg Nurs 12(3):121-127, 1980 Subjects Cancer Site: brain; Cancer Measures: incidence, mortality, survival as in lung, hepatic, and bladder cancer for which specific risk Abstract factors and high risk groups can be identified, brain tumors elude The article reviews the classification and severity (stage and our predictive skill and that unlike tumors such as in cervical and grade) of brain tumors. Incidence, mortality and survival data are breast cancer, there are no means of successfully screening for presented for the United States and other countries. The following brain tumors. Unlike leukemia and endometrial cancer which have risk factors are briefly discussed: chemical carcinogens, familial decreasing mortality and improving prognosis, brain tumor treat- factors, drugs and immunosuppression, infectious agents, radia- ment remains only minimally successful. However, diagnostic tion, and trauma. The author concludes that unlike tumors such technology for brain tumors is advancing rapidly. Leffall, L.D., Jr. Program Priorities for the Future 459 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 253-263 Subjects Screening: inultiple cancer screening tests; Prevention; Education; Grants for Cancer Cause and Prevention Research, and 2) the Cancer Site: multiple sites Cancer Prevention Study. Key initiatives that are identified are: 1) Abstract the modification of the Society's recommendations to the public This article describes programs which have been initiated by about checkups for the early detection of cancer; and 2) the Public the American Cancer Society, including: 1) the Special Institutional Education program “Priority Activities in Cancer Education.” Leffall, L.D., Jr. Breast Cancer in Black Women 460 CA 31(4):208-211, 1981 Subjects Biologic Marker: steroid receptors; Diagnosis; Risk Factor Exposure: discussed. The primary conclusion is that poor results in black diet-nutritional factors, Cancer Site: breast; Cancer Measures: women may be attributed to later stage of disease at the time of incidence, mortality diagnosis and fewer well-differeniated tumors. Increased em- Abstract phasis on monthly breast self-examination for black women is A general overview of studies on breast cancer and black warranted, as is more frequent use of low dose mammography, women is presented. Data from various sources on the mortality where appropriate, if one is to achieve earlier diagnosis and greater rates, genetic predisposition, diagnostic differences between black possibility of long-term survival. and white women, and the use of estrogen receptor status are Levin, D.L.; Connelly, R.R. Cancer of the Pancreas, Available Epidemiologic Information and Its Implications 461 Cancer 31:1231-1236, 1973 Subjects Risk Factor Exposure: geographic area; Cancer Site: pancreas; and mortality are examined to determine epidemiologic patterns. Cancer Measures: incidence, mortality Racial differences in incidence/mortality data and their relation to Abstract industrial hazards, quality of medical care, and geographic distri- Cancer of the pancreas is the fourth leading cause of death bution of cancer are discussed. The need for more definitive studies among all sites of cancer in the United States. Very little is known is noted, and suggestions for areas of future work are made. about the epidemiology of the disease. Available data on incidence Levin, M.L.; Thomas, D.B. The Epidemiology of Breast Cancer 462 In: Breast Cancer. New York: Alan R. Liss, 1977, p 9-35 Subjects Abstract Risk Factor Exposure: diet-nutritional factors, reproductive The article reviews the epidemiology of breast cancer among factors, exogenous hormones, breastfeeding, other diseases, females and males and suggests possible cancer control activities familial factors; Cancer Site: breast; Cancer Measures: incidence in two areas: primary and secondary prevention. 183 EPIDEMIOLOGY: POLICY/REVIEW Macdonald, E.J. Epidemiological Aspects JAMA 228(7):884-886, 1974 463 Subjects Risk Factor Exposure: diet-nutritional factors; Cancer Site: stomach; Cancer Measures: mortality Abstract The article suggests that a geographic pattern of extreme high and low incidence rates for gastric cancer predicate environmental factors as causative agents. As gastric cancer patients give a familial history of gastric cancer more often that do patients with other primary sites, except skin, nutrition is a prime suspect. Diet- nutritional factors which appear to be linked to gastric cancer in- clude: an antivitamin a factor in reheated fats and other foods; smoked foods; talc-treated rice; salted fish, vegetables, and meat products; and nitrosamines. Mack, T.M. Uterine Cancer and Estrogen Therapy Front Hormone Res 5:101-116, 1978 464 Subjects Methodology; Risk Factor Exposure: exogenous hormones; Cancer Site: corpus uteri; Cancer Measures: incidence Abstract Recent studies concerning the relationship of exogenous estro- gens and endometrial cancer have been reviewed, and the objec- tions to their conclusions discussed. Even considering the most serious of these reservations, it is difficult to avoid the tentative conclusion that estrogens are causally linked to endometrial cancer, and that the attributable risk is numerically sizable. Whether the medical costs will be ultimately shown to outweigh the benefits of estrogen replacement therapy will depend largely on whether or not a relationship to breast cancer is shown, and on the mag- nitude of the effect of estrogens in preventing serious complica- tions of ovarian failure. For the time being, estrogenic treatment of menopausal symptoms should be as brief as possible, should use the lowest dose possible, and should probably be given in cyclic fashion, particularly to women with intact uteruses. Every attempt should be made to give patients a quantitative understand- ing of both potential risks and benefits before they decide for or against estrogen therapy. MacMahon, B. Risk Factors for Cancer of the Pancreas Cancer 50(11):2676-2680, 1982 465 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, occupation, tobacco; Cancer Site: pancreas Abstract The descriptive epidemiology of cancer of the pancreas is ham- pered by the difficulty of differential diagnosis and the variation in diagnostic expertise and resources around the world. The con- dition is a disease of the elderly and there are few patients under 40 years of age. The disease is somewhat more common in Amer- ican blacks than in whites. Two populations of Polynesian origin - Maoris and Hawaiians - show high incidence rates, especially in males. Geographic variation is not striking and is difficult to inter- pret. The most consistently observed exogenous risk factor is cigarette smoking, the disease being approximately twice as com- mon in heavy smokers as in nonsmokers. In sharp contrast to chronic pancreatitis, pancreatic cancer does not appear to be related to use of alcoholic beverages. Recent observations of association of risk with coffee consumption and with use of decaffeinated coffee require further evaluation. Mancuso, T.F. Epidemiological Study of Tumors of the Central Nervous System in Ohio Ann NY Acad Sci 381:17-39, 1982 466 Subjects Risk Factor Exposure: occupation; Cancer Site: central nervous system Abstract Historically, the first epidemiological lead of an excess risk of death due to malignancies of the brain and central nervous system among the former employees of the rubber industry was in 1949, in a study correlating causes of death with prior employment de- signated on the death certificates for the state of Ohio in 1947. The geographical distribution of deaths (1955) due to histologically con- firmed tumors of the CNS identified one county, Summit, of the 88 counties of Ohio, as having significantly more deaths than expected for ages 25-64. Summit County represents the principal rubber manufacture center in the U.S. A series of analyses were carried out which identified a higher death risk among rubber workers compared to nonrubber worker residents of Summit County according to histologically confirmed diagnoses of tumors of the CNS in 1963. Longitudinal employee cohort studies of four different companies engaged in the manufacture of rubber prod- ucts, in the same state, Ohio, indicated a higher death rate due to tumors of the central nervous system. Epidemiological leads were cited in a broad industry survey comparison in which higher death rates due to neoplasms of the central nervous system were observed among workers in electrical machinery and paper and allied products as well as the rubber industry. Potential chemical linkage of the excess of CNS neoplasm deaths among rubber work- ers and workers in the rayon industry was explored. 184 EPIDEMIOLOGY: POLICY/REVIEW McBride, W.; Richter, G.C. Gastrointestinal Malignancies in Black Americans NY State J Med 85(4):157-159, 1985 467 Subjects Diagnosis; Prevention; Cancer Site: digestive system; Cancer Measures: incidence, prevalence Abstract In 1982, the last year for which figures are obtainable approx- imately 81,000 black Americans learned that they had cancer: 27% had cancers of the oral cavity or digestive organs. Studies of the incidence of cancer during the past several decades show that the rates for black Americans have been higher than for white Amer- icans. In reference to gastrointestinal malignancies, death rates have increased for cancers of the oropharynx, esophagus, pan- creas, and colorectal region during a 25-year period ending in 1978. Although death rates have declined for gastric carcinoma for both black and white Americans, black Americans still have higher in- cidence rates for gastric carcinoma. This article will discuss the major gastrointestinal malignancies that occur in black Americans and, when possible, make specific references to environmental or socioeconomic factors as they relate to cancer incidence rates, sur- vival trends, and cancer prevention. McConnell, R.B. Epidemiology of Gastrointestinal Tumors: A Review. J R Soc Med 71:278-281, 1978 468 Subjects Risk Factor Exposure: diet-nutritional factors, occupation, tobacco, alcohol, other diseases, geographic area; Cancer Site: colon-rectum, esophagus; Cancer Measures: incidence, mortality Abstract The article examines the distribution of gastrointestinal cancers in various countries. The author states that variation in death rates from country to country are greater for esophageal cancer than for other gastrointestinal cancers. International evidence linking vari- ous risk factors with esophageal cancer are reviewed. Mettlin, C. Diet and the Epidemiology of Human Breast Cancer Cancer 53:605-611, 1984 469 Subjects Risk Factor Exposure: diet-nutritional factors; Cancer Site: breast Abstract There are substantial data on breast tumorigenesis in animals that suggest that diet may be an important factor in human breast cancer etiology. The promotional effects of dietary fat, and, in particular, unsaturated fats, on mammary tumors in rodents is well established. The geographic distribution of breast cancer in humans correlates with international differences in average fat intake. Differences in dietary habits among populations in the United States and their breast cancer risk also have been observed. In the United States, the trend has been toward increased total fat consumption and increased use of polyunsaturated fats. However, breast cancer incidence among white women in the United States has changed very little. Case-control studies of dietary intake and breast cancer risk have shown inconsistent results, and prospective studies of breast cancer mortality and serum cholesterol and serum lipids show no differences in risk between women with high levels of cholesterol and serum lipids compared with women with low levels. Laboratory studies also suggest the possibility that natural inhibitors of breast cancer may occur in the diet as well. Antioxid- ants, inducers of microsomal enzyme activity, and retinoids, all have been implicated in the metabolic epidemiology of breast cancer. Research results at Roswell Park Memorial Institute have associated lower levels of intake of dietary vitamin A with a slightly elevated risk of breast cancer. To date, the epidemiologic data do not indicate with confidence that any specific dietary risk factor may be associated with breast cancer risk in the United States population. Additional epidemiologic studies on inhibition or promotion of breast cancer following the leads of previous labora- tory research may clarify the nature and practical significance of the relationship between diet and breast cancer. Michaels, D. Occupational Cancer in the Black Population: The Health Effects of Job Discrimination J Natl Med Assoc 75(10):1014-1018, 1983 470 Subjects Risk Factor Exposure: occupation; Cancer Site: lung; Cancer Measures: mortality Abstract Cancer mortality rates among black workers employed in sev- eral United States industries are higher than those of their white co-workers and of the national black population. As a result of discriminatory employment patterns, blacks are disproportion- ately employed in “high-hazard” jobs where they are exposed to carcinogens and other disease-producing substances. The elevated cancer mortality rates in black workers in these industries are a direct consequence of discriminatory employment patterns. The implications of these findings for physician practice, research, and national regulatory policy are discussed. 185 EPIDEMIOLOGY: POLICY/REVIEW Miller, A.B. Risk Factors from Geographic Epidemiology for Gastrointestinal Cancer Cancer 50:2533-2540, 1982 471 Subjects Risk Factor Exposure: diet-nutritional factors, tobacco, other diseases, geographic area; Cancer Site: colon, rectum; Cancer Measures: incidence, mortality Abstract The incidence of gastrointestinal cancers differ greatly interna- tionally. Blacks have higher rates than whites for esophagus, stomach, liver, and pancreatic cancers in the U.S. Differences also occur between other racial groups. Studies on migrants suggest environmental causes for these differences. Stomach cancer rates have fallen, intestine cancer is stable, pancreatic cancer rates rose, but are now falling in males. The incidence of colon, rectal, breast, corpus-uterus, ovary, and prostate cancer are positively correlated, but colon or rectal and stomach cancer are negatively correlated. Colon and rectal cancer mortality are positively and stomach cancer negatively associated with social class. Esophageal, liver, colon, and pancreatic cancer in males are higher in urban areas. Seventh Day Adventists and Mormons show a low risk of colon cancer. Internationally intake of fat is correlated with colon, rectal, and pancreatic cancer, and starch with stomach cancer. Miller, A.B. An Overview of Hormone-associated Cancers Cancer Res 38:3985-3990, 1978 472 Subjects Risk Factor Exposure: diet-nutritional factors, endogenous hormones; Cancer Site: multiple sites; Cancer Measures: incidence Abstract Data on the etiology of hormone-associated cancers are re- viewed. Although for breast cancer many risk factors point to the relevance of hormonal factors, findings are not uniform. Evidence points to the importance of dietary factors, and one study appears to confirm an association with high total fat intake. Dietary factors, although possibly mediated through a hormonal mechanism, may eventually be found to play a major role in the etiology of breast cancer. Studies of both endometrial and ovarian cancer also show the importance of hormonally associated factors, although not al- ways in the same way as for breast cancer. In addition, both for these sites and for cancer of the prostate, dietary factors may also be relevant in their etiology. As yet, no direct assessment of the importance of diet has been attempted for these sites, but this could be rewarding. Miller, W.]., Jr.; Cooper R. Rising Lung Cancer Death Rates Among Black Men: The Importance of Occupation and Social Class J Natl Med Assoc 74(3):253-258, 1982 473 Subjects Risk Factor Exposure: occupation, tobacco; Cancer Site: lung; Cancer Measures: mortality; Socioeconomic Factors Abstract From 1950 to 1977 the age-adjusted cancer death rates for non- white men in the United States rose an astonishing 63.2 percent, while rates for white men increased 22.2 percent and fell slightly for women of both races. The bulk of this increase can be accounted for by cancer of the lung. As a serious health problem that is increasing in severity, cancer in black men deserves close attention and definitive action. This discussion focuses on basic epidemiological relationships in the origins of this epidemic, par- ticularly in regard to the relative importance of occupation, cigarette smoking, and social class. Murray, M.].; Murray, A.B.; Murray, N.J. The Ecological Interdependence of Diet and Disease in Tribal Societies Yal ] Biol Med 53:295-306, 1980 474 Subjects Biologic Marker: hepatitis B virus associated antigens; Risk Factor Exposure: diet-nutritional factors; Cancer Measures: prevalence Abstract Observations among nomads suggest there is a strong ecolog- ical interdependence of diet and disease in tribal societies which favors survival of man. This relationship may be disrupted by changes in diet to conform to the highly productive technology of the West. Such changes may result in intensification of indigenous disease and in the transfer of disease characteristics of Western societies. To prevent these consequences, relief feeding and long- term attempts to upgrade nutrition should be carried out with traditional foods wherever possible. 186 EPIDEMIOLOGY: POLICY/REVIEW Myers, J. The Social Context of Occupational Disease: Asbestos and South Africa 475 Int J] Health Serv 11(2):227-245, 1981 Subjects Risk Factor Exposure: occupation; Cancer Site: lung weaknesses — the lack of statutory limits on exposure, capital's Abstract responsibility for monitoring exposure and health, the inefficiency General issues of industrial health are raised in relation to the of the state inspection, and the meagerness and racial disparities production of asbestos and asbestos-related diseases in South Af- in compensation — are related to the weakness of organized labor. rica. A historical analysis of occupational diseases with long incu- These weaknesses are linked to the movement of certain industrial bation periods and their implications for capital and labor. In order processes, finally acknowledged as unsafe by most academic re- to understand the role of the research establishment, an attempt search, away from the developed countries. In these countries, is made to situate the state in the conflict between capital and the strength of labor and environmental organizations has caused labor. The terms and weapons of this ideological arena are inves- a decline in capitalist productivity. tigated. The South African situation is then discussed. Its evident Newell, G.R. Multiple Primary Cancers: Suggested Etiologic Implications 476 Cancer Bull 32(4):160-164, 1980 Subjects Risk Factor Exposure: diet-nutritional factors, tobacco, States for multiple primary breast cancers are shown. The lowest endogenous hormones, familial factors, treatment modalities rate was found to be among blacks. The author feels that this Cancer Site: multiple sites; Cancer Measures: incidence observation is worthy of confirmation and should be followed up Abstract by studies on environmental or other etiologic factors. In this article, differences in racial groups within the United Oettle, A.G. Cancer in Africa, Especially in Regions South of the Sahara 477 J Natl Cancer Inst 33:383-439, 1964 Subjects Risk Factor Exposure: geographic area; Cancer Site: multiple sites; of the liver, Kaposi's sarcoma, Burkitt's tumor, and, in some reg- Cancer Measures: incidence, mortality ions, esophageal cancer. The mixed race (Colored) has a high rate Abstract of stomach and liver cancer. Asians have a lower mortality. South Striking differences are noted in cancer incidence, situation, African whites resemble U.S. whites, except for higher rates for and histopathological type between and within races of Africa. lip, tongue, stomach, prostate, skin, and bone cancers. Etiological These suggest that 80 percent or more of the cancers affecting aspects of individual tumors are discussed with regard to lip (to- western races are environmentally induced and potentially pre- bacco and sunlight), mouth, postnasal space, esophagus, liver ventable. Negriform races generally have a much lower incidence (fungal toxins?), sinuses (snuff), lung (smoking, arsenic), than western races or U.S. nonwhites, which disposes of the mesothelium (asbestos), uterus, penis (circumcision, ablution), hypothesis of the uniform liability of mankind to cancer. The bladder (bilharziasis), prostate, skin (sunlight, tropical ulceration, gradual assumption of a western manner of life by African races other ulcers), eye, conjunctiva (sunlight), brain, and lymphomas. has resulted in the availability for study of material from these Since most cases of cancer in Africa are out of reach of methods vast human experiments. Among the Negriform races, cancers of of treatment other than chemotherapy, great need as well as oppor- stomach, large bowel, breast, endometrium, ovary, brain, and tunities exist for the study of carcinostatics. leukemia are generally rare. Common to them are primary cancer Oettle, A.G. Kaposi's Sarcoma in Africa 478 Natl Cancer Inst Monogr 10:281-285, 1963 Subjects Cancer Site: sarcoma; Cancer Measures: incidence, prevalence ern Congo it reaches 12.8 percent of all cancers examined histolog- Abstract ’ ically. No cases associated with leukemia or lymphoma have been Kaposi's sarcoma is undoubtedly common among Negroid recorded in Negroids; in other respects, the disease resembles that races of Africa, and the incidence varies from 1.6 to 3.1 per 100,000 seen in Western races. It is concluded that unknown environmental (standardized to the U.S. population of 1950). Ratio studies show factors must be responsible for the frequency of this disease in a profound variation from between regions, and in the Northeast- Africa. 187 EPIDEMIOLOGY: POLICY/REVIEW Owen, W.L. Cancer of the Prostate: A Literature Review J Chron Dis 29:89-114, 1976 479 Subjects Diagnosis; Treatment; Risk Factor Exposure: geographic area, reproductive factors, sexual practices, familial factors, diet nutritional factors, other diseases; Cancer Site: prostate; Cancer Measures: incidence, mortality, survival Abstract The paper reviews the prostate cancer literature with respect to both the epidemiological and clinical aspects of the disease. The author notes that despite the fact that cancer of the prostate is the second or third leading cause of cancer deaths among U.S. males and, according the NCI's Third National Cancer Survey, itaccounts for 15.2% of all cancers among U.S. white males and 21.8% of all cancers among U.S. Negroes, few well-designed studies of this disease have been made. Persaud, V. Geographical Pathology of Cancer of the Uterine Cervix Trop Geogr Med 29:335-345, 1977 480 Subjects Risk Factor Exposure: geographic area, sexual practices, reproductive factors; Cancer Site: cervix uteri; Cancer Measures: incidence, mortality Abstract The availability of age-standardized cancer incidences for differ- ent parts of the world has enabled a thorough and meaningful analysis of the geographical distribution of cancer of the uterine cervix. The high morbidity and mortality from cervical cancer re- ported for most European countries and North America at the beginning of this century has, in recent years, been superseded by those for Asian, African, Latin American, and Caribbean coun- tries with the provision of additional data. Striking differences in cervical cancer incidences have been observed among various ethnic groups in Africa which seem to reflect variations in the intensity of certain environmental influences. Cervix cancer is un- common in the white population of North America and Europe with the exception of West Germany, Denmark, Sweden, and Yugoslavia. In the United States, the highest incidence occurs in Puerto Rican women of New York City and the Latin population of the South. The very low incidence in Jewish women is virtually the same in New York City as in Israel. This world-wide survey has shown that poor sexual hygiene rather than lack of male cir- cumcision per se is a more important aetiological factor in cervical cancer. Pollak, H. Feedback: Breast Cancer in Black Women CA 32(1):62-63, 1982 481 Subjects Risk Factor Exposure: breastfeeding; Cancer Site: breast Abstract In response to Dr. LaSalle D. Leffall Jr.’s article “Breast Cancer in Black Women” (CA 31:208-211, 1981), the author suggests that the recent rise in breast cancer observed among black women, especially younger women, may be due to changing patterns of breast-feeding. Pratt, C.B. Some Aspects of Childhood Cancer Epidemiology Pediatr Clin North Am 32(3):541-556, 1985 482 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, mortality Abstract Within the first 15 years of life, cancer, although rare, accounts for more deaths than any other disease. Deaths within the first year of life most often result from complications of birth or congen- ital malformations, and in later years accidents account for more 188 deaths. Between 6000 and 7000 persons under the age of 16 will develop cancer in 1985. This review is not intended to be all-inclu- sive but representative of many of the aspects of childhood cancer epidemiology. The reader is referred to review articles for addi- tional information related to these malignancies that develop in children and adolescents. EPIDEMIOLOGY: POLICY/REVIEW Reid, B.L. The Causation of Cervical Cancer 483 Clin Obstet Gynecol 12(1):1-18, 1985 Subjects Abstract Screening: Pap smear; Risk Factor Exposure: sexual practices; This is a review of available literature on epidemiology and Cancer Site: cervix uteri biology of cervical cancer. Roach, W.A. An Attempt on a Cause and Prevention of Oseophageal Cancer in the Transki Bantu 484 S Afr Med | 47:347-348, 1973 Subjects Prevention; Risk Factor Exposure: diet-nutritional factors; Cancer lates that presence of these substances in food might be a cause Site: esophagus of esophageal cancer in the Bantu. Analyses of food stuffs for Abstract nitrosamines is proceeding, but no definitive data are available. The author reviews the production of nitrosamines and specu- Rothman, K.]. The Proportion of Cancer Attributable to Alcohol Consumption 485 Prev Med 9:174-179, 1980 Subjects Risk Factor Exposure: alcohol; Cancer Site: oral cavity, pharynx, contain it, is unclear. The alcohol effect, at least for cancers of the larynx, esophagus, liver; Cancer Measures: incidence, mortality upper aerodigestive tract, depends on tobacco consumption. Abstract About 3% of all U.S. cancers in 1974 are attributable to alcohol; Existing evidence leaves little room for doubt that alcoholic the small proportion and the dependence on tobacco mitigate beverages are carcinogenic. Alcohol is a risk factor for cancer of against antidrinking campaigns for the primary prevention of the oral cavity, larynx, esophagus, and possibly the liver. The cancer. It is likely that antismoking campaigns would be more extent to which carcinogenicity is attributable to the alcohol per effective than antidrinking campaigns in preventing alcohol-re- se, or to other components of some or all of the beverages that lated cancers. Rothman, K.]J.; Cann, C.I.; Flanders, D.; et al. Epidemiology of Laryngeal Cancer 486 Epidemiol Rev 2:195-209, 1980 Subjects Risk Factor Exposure: alcohol, occupation, tobacco; Cancer Site: partial or total laryngectomy. More than 90 per cent of the epithelial larynx; Cancer Measures: incidence, mortality, survival malignancies of the larynx are squamous cell carcinoma; most are Abstract moderately or well differentiated. Glottic carcinomas, the most One person in 74 who is diagnosed with cancer in the United common, have the best prognosis, followed in both frequency and States has cancer of the larynx. The symptoms include hoarseness, favorability of prognosis by supraglottic carcinomas. Subglottic throat or neck discomfort, progressive dyspnea, hemoptysis and carcinomas are the least frequent and have the worst prognosis. weight loss. The diagnosis usually rests on direct inspection of The overall 20-year cumulative relative survival rate for cancer of the larynx and biopsy of suspicious lesions. Therapy depends on the larynx reported by the National Cancer Institute is 34 per cent the extent and location of disease; it usually involves radiation and for white patients. 189 EPIDEMIOLOGY: POLICY/REVIEW Safai, B. Kaposi's Sarcoma: A Review of the Classical and Epidemic Forms Ann NY Acad Sci 437:373-382, 1984 487 Subjects Diagnosis; Cancer Site: sarcoma; Cancer Measures: incidence, survival Abstract Kaposi's sarcoma (KS) represents a potential model for a virally associated human tumor. Prior to 1981, KS was considered a rare tumor occurring in cluster distribution. A sudden increase in the number of KS cases along with pneumonia caused by Pneumocys- tis carinii (PCP) and other opportunistic infections (OI) was ob- served among the victims of the epidemic of acquired immune deficiency syndrome (AIDS). Initially, the cases were seen in New York and California, but now they have been reported from most large cities in the USA and many other countries. The first reports were exclusively among homosexual men and intravenous drug abusers; the disease, however, is now being observed among Hai- tians, hemophiliacs, and even in some children with AIDS. In this new epidemic form KS presents with a more aggressive course involving skin, gastrointestinal tract, and lymph nodes. The epidemic form of KS has a similar course to lymphadenopathic KS seen in African children. Whereas OI have been the major cause of death in this epidemic, cases have been seen in whom KS has spread rapidly to vital organs such as lung, liver, and spleen. Much interest has been generated in the study of KS since the epidemic of AIDS. It is hoped that investigative work will provide a better insight into this interesting tumor. Safai, B.; Good, R.A. Basal Cell Carcinoma With Metastasis Arch Pathol Lab Med 101:327-331, 1977 488 Subjects Treatment; Cancer Site: skin; Cancer Measures: incidence, mortality, survival Abstract Basal cell carcinoma (BCC) is the most common form of skin cancer, accounting for 65% to 75% of all malignant skin tumors. Metastatic BCC is very rare—only 109 documented cases have been reported to date. The mean time of survival after metastasis has been reported to be ten months. No effective therapeutic mod- ality is known for its treatment. We are reporting a case of metas- tatic BCC in which we have found deficiency of cell-mediated immunity. Postmortem studies showed a co-existing squamous cell carcinoma in the lungs, brain, liver, and spleen. The most unusual finding in the postmortem study was evidence of juxtapos- ition of BCC and epidermoid carcinoma with distinct histologic dimorphism. We speculate that a combination of immunodefi- ciency and stromal independence is needed for metastases in BCC. Schottenfeld, D. Epidemiology of Cancer of the Esophagus Semin Oncol 11(2):92-100, 1984 489 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, tobacco, familial factors; Cancer Site: esophagus; Cancer Measures: incidence, mortality Abstract The article reviews the different patterns of occurrence of esophageal cancer in different parts of the world. Its epidemiologic behavior is marked by sharp variations in incidence within discrete geographic areas, among different ethnic groups and between males and females. The authors note that esophageal cancer is a disease of the poor in most areas of the world. In North America and Western Europe the major risk factors are alcohol and tobacco which account for about 80 to 90 percent of the incident cases each year. In many other high esophageal cancer incidence areas of the world etiologic factors other than tobacco and alcohol are under investigation. These factors include nutritional deficiences and the ingestion of food or water that may contain nitrosamine precursors or be contaminated with fungi. Schottenfeld, D. Alcohol as a Co-Factor in the Etiology of Cancer Cancer 43:1962-1966, 1979 490 Subjects Risk Factor Exposure: alcohol, tobacco; Cancer Site: oral cavity, pharynx, larynx, esophagus; Cancer Measures: mortality Abstract Alcohol and tobacco appear to act synergistically in the pathogenesis of epithelial cancers of the oropharynx (excluding lip), larynx, and esophagus. For the subsites within the upper aerodigestive tract, over 10,000 deaths in United States men during 1978 may be attributed to tobacco and alcohol consumption. The cancer sites for which tobacco and alcohol are major determinants occur with greater frequency in men, blacks, lower socioeconomic groups, and with increasing urbanization and increasing age (35-70 years). Because primary hepatocellular carcinoma occurs more commonly in patients with cirrhosis, chronic alcohol abuse is an important risk mechanism for carcinoma of the liver parenchyma. Although experimental animal studies have failed to demonstrate whether ethanol can independently initiate tumorigenesis, various alternative or associated biochemical and immunological mechanisms of action have been proposed. 190 EPIDEMIOLOGY: POLICY/REVIEW Schottenfeld, D. Epidemiology of Thyroid Cancer Laryngoscope 88(1pt2):55-57, 1978 491 Subjects Risk Factor Exposure: diet-nutritional factors, endogenous hormones, familial factors, treatment modalities; Cancer Site: thyroid; Cancer Measures: incidence, mortality Abstract The article briefly reviews data on the incidence and mortality of thyroid cancer in black and white U.S. populations. The distri- bution of cell type and the role of ionizing radiation and iodine deficiency in the development of the disease are discussed. Seeler, R.A. Letters to the Editor: Prognosis for Black Infants with Leukemia 492 Pediatrics 73(4):568, 1984 Subjects Treatment; Cancer Site: leukemia of chemotherapeutic drugs because of their lower absolute granulo- Abstract cyte count. It is also suggested that poly drug chemotherapy pro- This letter suggests the possibility that the poorer prognosis of black children with acute leukemia may be due to “underdosing” tocols should have different absolute standards for dose adjust- ments for black children. Seidman, H.; Silverberg, E.; Holleb, A.I Cancer Statistics, 1976: A Comparison of White and Black Populations 493 CA 26(1):2-29, 1976 Subjects Cancer Site: multiple sites; Cancer Measures: incidence, mortality, survival Abstract The article compares white and black U.S. populations in regard to: 1) the probability of developing and dying from cancer, 2) trends in cancer incidence and mortality over the last several dec- ades, 3) stage of disease at diagnosis, and 4) survival following diagnosis. The authors conclude that blacks are developing more cancers and their prospects for diagnosis in a localized stage are poorer than for whites. Survival rates are poorer, and death rates are higher for blacks. Shoemaker, E.S.; Forney, ].P.; MacDonald, P.C. Estrogen Treatment of Postmenopausal Women: Benefit and Risks 494 JAMA 238(14)1524-1530, 1977 Subjects Risk Factor Exposure: exogenous hormones; Cancer Site: breast, corpus uteri Abstract Estrogen treatment of postmenopausal women is effective in relieving the symptoms of vasomotor instability and urogenital atrophy; estrogen treatment is effective in preventing accelerated bone loss and osteoporosis in young women following castration, but in postmenopausal women aging is a more important deter- minant of accelerated bone loss than is decreased estrogen secre- tion. Low-dose estrogen treatment of postmenopausal women neither prevents nor increases the risk of arteriosclerotic cardiovas- cular disease or cerebral vascular disease. It cannot be definitively established that estrogen treatment of postmenopausal women causes an increased incidence of breast tumors, but it is clear that such treatment does not prevent these tumors. It is established that estrogen treatment of postmenopausal women increases the risk ratio of endometrial carcinoma. Silber, W. Carcinoma of the Oesophagus: Aspects of Epidemiology and Aetiology 495 Proc Nutr Soc 44:101-110, 1985 Subjects Prevention; Risk Factor Exposure: geographic area, diet-nutritional factors, alcohol, tobacco; Cancer Site: esophagus; Cancer Measures: incidence Abstract The article reviews international patterns in the incidence of esophageal cancer. Research concerning a variety of host and en- vironmental factors are examined. In addition, information con- cerning diet and the role of vitamin deficiencies is reviewed. The author concludes that the aetiology of esophageal cancer is mul- tifactorial and that while there have been advances, these are not always consistent. Suggested measures of primary prevention in- clude fortifying maize and the administration of B-carotene. EPIDEMIOLOGY: POLICY/REVIEW Silvers, D.N. Letters to the Editor: Prognosis of Melanoma JAMA 249(11): 1437-1438, 1983 496 Subjects Diagnosis; Cancer Site: malignant melanoma; Cancer Measures: survival Abstract This letter to the editor refers to an article entitled “Malignant Melanoma in Black American and White American Populations.” The letter questions whether or not the authors provide specific data that demonstrates the supposition that volar and subungual melanoma in blacks has a worse prognosis when compared spec- ifically with volar and subungual melanoma in whites (controlled by tumor thickness). Smally, A.J. Letters to the Editor: Screening for Human T-Cell Leukemia-Lymphoma Virus 497 JAMA 251(12):1555, 1984 Subjects Risk Factor Exposure: occupation; Cancer Site: leukemia Abstract The author suggests that the high prevalence of the human T-cell leukemia - lymphoma virus (HTLV) noted in southeastern blacks may be due to large numbers of blacks working in hospitals. He further asks the question, “should populations of all races working in hospitals be screened for HTLV“? Smith, E.M. Epidemiology of Oral and Pharyngeal Cancers in the United States: Review of Recent Literature 498 J Natl Cancer Inst 63(5):1189-1198, 1979 Subjects Risk Factor Exposure: tobacco, alcohol, other diseases, geographic area; Cancer Site: oral cavity, pharynx; Cancer Measures: survival, mortality, incidence Abstract Recent literature on morbidity and mortality patterns, known risk factors, and related sociobehavioral characteristics of oral cancer have been examined. Despite variation in populations and methodologies, alcohol and tobacco were seen in the major inde- pendent etiologic agents; these effects were associated with age, sex, and religion-ethnicity. Other factors were suggested, but their correlation was less consistent: geographic location, race, socioeconomic status, nutrition, dental conditions, and concurrent diseases. Social and behavioral components may alter risk, stage of disease at diagnosis, treatment, or survival from oral cancer. The impact of sociobehavioral elements on the reduction of the incidence of and the mortality from the disease is an important area needing further investigation. Tulinius, H. Epidemiology of Prostate Carcinoma Recent Results Cancer Res 60:3-13, 1977 499 Subjects Risk Factor Exposure: sexual practices, endogenous hormones, geographic area; Cancer Site: prostate; Cancer Measures: incidence, mortality Abstract Thorough and comprehensive reviews exist on the epidemiol- ogy of carcinoma of the prostate. In reviewing descriptive epidemiology of the disease, this chapter will give mortality and incidence rates for carcinoma of the prostate in Iceland. It will also mention some suggestions as to etiology. Van Peenen, P.F.D.; Nelson, N.A.; Wolkonsky, P.M. Letters to the Editor: Re: Occupational Risk Factors and Liver Cancer 500 Am ] Epidemiol 119(1):144-145, 1984 Subjects Risk Factor Exposure: occupation, alcohol; Cancer Site: liver; Cancer Measures: incidence, mortality Abstract This article emphasizes that occupations for which associations are found in a study of liver cancer in New Jersey are generally those of low socioeconomic status. It is critical of another article that concerns occupational risk factors and liver cancer. EPIDEMIOLOGY: POLICY/REVIEW Vianna, N.J. Epidemiology of Hodgkin's Disease: Review And Etiologic Leads 501 CRC Crit Rev Clin Lab Sci, Jan 1975, p 245-287 Subjects Cancer Site: lymphoma; Cancer Measures: incidence, mortality, survival Abstract This review article is a comprehensive overview of Hodgkin's disease, including incidence, mortality and survival. It covers risk factors, including environmental and genetic factors, and presents extensive data in support of an infectious etiologic factor hypothesis. Descriptive data from major cancer registries and vital statistics in the U.S. and other countries presents time trends and international comparisons. Analytic data are drawn from a number of cohort and case control studies. Numerous problems in epidemiologic research are discussed, including classification dif- ferences and the lack of population-based (as opposed to institu- tion-based) research. Walker, A.R.P. Prostate Cancer—Some Aspects of Epidemiology, Risk Factors, Treatment and Survival 502 S Afr Med ] 69:44-47, 1986 Subjects Treatment; Risk Factor Exposure: diet-nutritional factors, endogenous hormones; Cancer Site: prostate; Cancer Measures: incidence, mortality Abstract In Western populations prostate cancer was probably rare in the past, yet in many populations the mortality rate has risen 3 - 5-fold since 1918. The disease now affects 1 man in 20. While the incidence in Third World populations living traditionally remains low, it increases with urbanization and properity, as in South African blacks. While the disease is age-related, more common in married than in single men and family-orientated, knowledge of specific aetiological factors is meagre. Abnormal hormonal status may well be involved, and possibly diet (Western v. Third World diet). There is a strong hormonal influence on tumour develop- ment. Treatment, according to the stage of the disease, includes prostatectomy, hormonal manipulation, external irradiation and chemotherapy; 70% of patients are surgically incurable at the time of presentation. Survival is greatly affected by the stage at diag- nosis, the percentage surviving 5 years is 3 times higher for patients at stage A than for those at stage D. Little can be done to avoid prostate cancer because of inadequate understanding of risk fac- tors. Annual rectal examination between 40 and 65 years is urged, since surgical cure is possible when metastases are absent. Walker, A.R.P. Cancer of the Cervix—Some Aspects of Epidemiology, Screening, Risk Factors and Survival 503 S Afr Med ] 68:316-320, 1985 Subjects Screening: Pap smear; Risk Factor Exposure: sexual practices, reproductive factors, tobacco, exogenous hormones, other diseases; Cancer Site: cervix uteri; Cancer Measures: incidence, mortality, survival Abstract Early statistics on the occurrence of cancer of the cervix are unreliable. What is certain is that between 1950 and 1965 incidence rates (IRs) and mortality rates (MRs) fell considerably in most Western populations. Recently, however, rates in the 15 - 35-year age group of some populations have risen two- to threefold. In some developing populations, especially in South America and Africa, rates are very high — up to eight times those in most Western populations. Risk factors include early age at first inter- course, having multiple partners, smoking, being widowed or di- vorced, and having a husband who has had multiple partners. The modus operandi of carcinogenesis is not known. Screening, using the Papnicolaou test, has markedly reduced the IRs and MRs, particularly in certain Scandinavian populations. In Western populations 40 - 60% of patients attain 5-year survival. Understand- ably, in individual patients the crucial regulating factor is the stage of disease when diagnosed. Among those at stage, 70 - 90% survive for 5 years or more, whereas only 0 - 25% of patients diagnosed at stage IV do. In Soweto, Johannesburg, length of survival in a series of black patients whose staging pattern did not markedly differ from that in Western series was less than half that in the latter. Undoubtedly, as repeatedly urged, appropriate, selective screening and health education could effectively reduce the toll of mortality, especially in high-risk developing populations. Walker, A.R.P. The Relationship Between Bowel Cancer and Fiber Content in the Diet 504 Am J Clin Nutr 31:5248-5251, 1978 Subjects Risk Factor Exposure: diet-nutritional factors, geographic area, bowel characteristics; Cancer Site: colon; Cancer Measures: incidence Abstract Colon cancer, very rare in rural South African blacks, is also rare in urban dwellers despite considerable rise in prosperity. The disease has scarcely increased during the last quarter of a century. The same situation applies to appendicitis. Endeavors to charac- terize different black populations in transition (including subjects who have had appendicitis) are being made respecting 1) diet, especially dietary fiber intake; 2) bowel physiology (e.g., transit time); 3) concentrations of fecal bile acids and other metabolites; and 4) the activity of certain fecal enzymes. EPIDEMIOLOGY: POLICY/REVIEW Walker, A.R.P.; Burkitt, D.P. Colon Cancer: Epidemiology Semin Oncol 3(4):341-350, 1976 505 Subjects Prevention; Risk Factor Exposure: diet-nutritional factors, geographicarea, bowel characteristics; Cancer Site: colon; Cancer Measures: incidence, mortality Abstract The article discusses the epidemiology of colon cancer as it prevails in: 1) western populations, past and present, 2) primitive and developing populations, and 3) immigrant populations. Walker, A.R.P.; Segal, I. Epidemiology of Noninfective Intestinal Diseases in Various Ethnic Groups in South Africa Israel ] Med Sci 15(4):309-313, 1979 506 Subjects Risk Factor Exposure: diet-nutrition factors; Cancer Site: colon; Cancer Measures: prevalence Abstract The prevalences of bowel diseases (hemorrhoids, appendicitis, polyps, ulcerative colitis, irritable bowel syndrome, diverticular disease, and colon cancer) are similar in South African whites and in populations of prosperous western countries. Among rural South African blacks with a traditional life style, these diseases are very uncommon or almost unknown. Among the urban South African blacks with a partially westernized life style, the diseases remain uncommon. Frequencies of appendectomies and colon cancer in urban blacks have increased little during the last three decades, although an increase undoubtedly has occurred in the frequency of diverticular disease. Frequencies of bowel diseases in South African Indian and colored (Eurafrican, Malay) popula- tions are intermediate. Because the diseases are almost entirely of environmental causation (due principally to changes in diet), pre- valences almost certainly will increase in blacks, Indians and col- oreds, as their way of life becomes further westernized. Walton, R.J.; Blanchet, M.; Boyes, D.A_; et al. Cervical Cancer Screening Programs: The Walton Report Can Med Assoc J 114(11):2-32, 1976 507 Subjects Screening: Pap smear; Risk Factor Exposure: reproductive factors, sexual practices; Cancer Site: cervix uteri; Cancer Measures: incidence, mortality Abstract The Walton report puts into focus the situation concerning screening for cervical cancer in Canada. This report discusses the value of cervical screening in the prevention of death among Cana- dian women from cancer of the cervix and provides an authoritative review of current knowledge about the disease. Recommendations of the Walton report are based on careful reviews of many aspects of cancer of the cervix and deals with such matters as frequency of screening, quality control and follow-up mechanisms. It is ac- cepted as scientifically important to the medical profession and the public. White, J.E.; Enterline, ]J.P.; Alam, Z.; et al. Cancer Among Blacks in the United States—Recognizing the Problem 508 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 35-53 Subjects Risk Factor Exposure: geographic area; Cancer Site: multiple sites; Cancer Measures: incidence, mortality, survival Abstract In the mid-1950’s, both the cancer incidence and mortality rates for black males and females surpassed those of their white coun- terparts. In 1970, the overall cancer mortality rate among blacks was nearly 20 percent higher than that of whites. The highest cancer death rates among both blacks and whites are found to occur in urbanized areas. An analysis of 15 major metropolitan areas in the U.S. found that three cities have extremely high cancer rates—Washington, D.C., Baltimore, MD, and New Orleans, LA. Blacks not only have a higher incidence of cancer when compared to whites, but have a higher frequency of cancers with dire prog- noses. Survival after the diagnosis of cancer is estimated to account for 70 percent of the black-white differential in mortality, with cancer incidence accounting for the other 30 percent. The relatively late stage-of-disease at which cancer is diagnosed among blacks accounts for approximately 50 percent of the black-white differen- tial in survival. Specific recommendations concerning future direc- tions for cancer research and control activities among blacks should include: (1) concentrating a greater effort on primary and secondary cancer prevention among high-risk black populations; 2) the fund- ing of epidemiologic studies in three high-risk black populations, i.e., Washington, D.C., Baltimore, MD, and New Orleans, LA; and 3) the funding of studies to determine effective methods for reaching blacks with appropriate health information. It is hoped that the overview of the cancer problem among U.S. blacks pre- sented in this paper will assist in motivating both the public and private sector to carefully consider implementing these recommen- dations. 194 EPIDEMIOLOGY: POLICY/REVIEW White, J.K. Increasing the Odds for Cancer Survival Health Aff 3(3):121-127, 1984 509 Subjects Cancer Site: multiple sites; Cancer Measures: mortality, survival Abstract The article reviews cancer statistics and relates them to the probability of increasing the odds for cancer survival. It is stated that a combination of increased funding and improved research efforts, stemming in part from enactment of the National Cancer Act of 1971, and some key discoveries about cancer genes through recombinant DNA technologies, have led to the improvements in cancer survival rates. The survival rates for black cancer patients lag considerably behind those of whites. Data indicate that total cancer death rates have increased slightly over the past three dec- ades, up 16 percent for white males and 39 percent for nonwhite males. For females, mortality rates fell slightly for whites and changed negligibly for nonwhites. Mortality rates for lung cancer have shown the most sizeable increases, 116 percent for white males, 199 percent for white females, and 185 and 188 percent for nonwhite males and females, respectively. By the end of 1984, lung cancer is expected to surpass breast cancer as the leader of cancer deaths in women. Wynder, E.L. Opportunities for Prevention of Cancer in Blacks 510 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 237-252 Subjects Prevention; Risk Factor Exposure: tobacco, alcohol, occupation, diet-nutritional factors; Cancer Site: multiple sites; Cancer Measures: incidence, mortality Abstract This paper briefly focuses on those sites for which blacks and whites have different rates and to see to what extent these differ- ences can be explained by our current knowledge about environ- mental contributions to cancer. It also proposes preventive strategies which can be implemented now. It is asserted that for certain risk factors and for certain cancers, preventive approaches are possible today and need to be applied with great vigor. Wynder, E.L.; Gori, G.B. Contribution of the Environment to Cancer Incidence: An Epidemiologic Exercise J Natl Cancer Inst 58(4):825-832, 1977 511 Subjects Risk Factor Exposure: diet-nutritional factors, occupation, tobacco, air pollution, water pollution, drugs; Cancer Site: multiple sites; Cancer Measures: incidence, mortality Abstract This article attempts to estimate the extent that cancer incidence and mortality may be related to certain suspect causative factors. To determine more precisely the influence of individual causative factors, the evidence related to the following environmental factors is examined: tobacco use, nutrition, air and water pollution, occu- pational exposure and drugs. Differences between population groups, sex ratios, time trends, and demographic and socioeconomic variables are investigated. Wynder, E.L.; Laakso, K.; Sotarauta, M.; et al. Metabolic Epidemiology of Prostatic Cancer Prostate 5:47-53, 1984 512 Subjects Risk Factor Exposure: diet-nutritional factors, sexual practices, endogenous hormones; Cancer Site: prostate Abstract A review of the epidemiological evidence indicates that dietary fat very likely has an etiologic role in the development of prostatic carcinoma. While this effect may be mediated by way of altered hormonal action of the prostate, there is little supporting evidence from assays of plasma of urinary hormones in case-control studies or the investigation of high-risk and low-risk groups. The applica- tions of metabolic epidemiology to this problem is most likely to succeed by direct studies of the prostate gland, and the perfor- mance of relevant assays on prostatic fluid. Estradiol and estrone levels were found to be higher in prostatic fluid than in serum, whereas for prolactin the reverse was true. Testosterone concentra- tions were very low in prostatic fluid, perhaps because of the high degree of plasma protein binding. Preliminary data indicated that prostatic fluid estradiol and prolactin levels are elevated in some prostate cancer patients; estrone levels appear to be normal. 195 EPIDEMIOLOGY: POLICY/REVIEW Young, J.L.; Ries, L.G.; Pollack, E.S.; et al. Cancer Patient Survival by Racial-Ethnic Group—United States, 1973 to 1979 JAMA 253(21):3069, 1985 513 Subjects Treatment; Cancer Site: multiple sites; Cancer Measures: survival Abstract The authors present the results of an National Cancer Institute (NCI) study to examine survival differences of patients with a first primary cancer by race and ethnic group. The study included data from NCI's Surveillance, Epidemiology and End Results Program (SEER) on cancer cases diagnosed from 1973 to 1979 and followed through December 31, 1981. Survival rates for Hispanics were almost identical to those for Anglos. Black males experienced poorer survival than Anglo males for cancer of the rectum, pros- tate, bladder, and thyroid; black females had poorer survival than Anglo females for cancers of the bladder, corpus uteri, and breast. For many primary sites, Japanese experienced the highest survival rates, and American Indians, the lowest. In an Editorial note fol- lowing the presentation of study results, it is noted that survival time after diagnosis is related to the extent of disease at diagnosis (stage), the effectiveness of treatment, and biologic and behavioral differences. Zaridze, D.G.; Boyle, P.; Smans, M. International Trends In Prostatic Cancer Int J] Cancer 33:223-230, 1984 514 Subjects Diagnosis; Cancer Site: prostate; Cancer Measures: incidence, mortality Abstract The most recent data available demonstrates a 120-fold differ- ence between the lowest and highest incidence rates of prostatic cancer, the disease being very common in North America, particu- larly among Blacks, and in Scandinavia, while it is rare in Japan and other oriental countries. The highest mortality for prostatic cancer is reported from St. Vincent and Grenadines, Martinique and Bermuda, from countries where the morbidity statistics are not available; the mortality rates reported from the United States and Canada are considerably lower. The incidence of the mortality from prostatic cancer have increased in most countries, in particu- lar in areas with an initially low frequency of this disease. The ratio of mortality to incidence for prostatic cancer varies rather widely, being low in North America, Hawaii and Scandinavia. It is suggested that the observed variation in the mortality to inci- 196 dence ratio for prostatic cancer could be due to differences in diagnostic practices between countries. This could explain, at least in part, the fact that the increasing trends of prostatic cancer inci- dence in North America are not accompanied by an increase in mortality from this tumour. This notion, however, does not exclude advances in treatment as possible determinants of the improved survival rate from prostatic cancer in this part of the world. The available statistics on prostatic cancer are based on the sum of clinically diagnosed carcinomas and those latent tumours found unexpectedly at prostatectomy and autopsy. The proportion of latent carcinomas among all prostatic cancer depends on the detection rate and varies from country to country, thus casting uncertainty on the comparability of prostatic cancer statistics from different areas. To avoid confusion in the statistics of prostatic cancer, it would be useful to consider introducing latent prostatic cancer as a separate entity in the next revision of the International Classification of Diseases (ICD). SECTION II Knowledge, Attitudes, and Practices (KAP) be bk r ' ; E 5 REN a cia A bet fon ie ae ne merase Bah ee Sip Section II Knowledge, Attitudes, and Practices (KAP) Information concerning the knowledge, attitudes, and practices (KAP) of indi- viduals and population groups is generally derived from survey research and the behavioral sciences. Knowledge and attitudes in this report refer to an indi- vidual’s or population's knowledge base and attitudes toward various aspects of cancer such as cancer incidence, detection, treatment, curability, and risk factors. Practices refer to cancer-related behaviors such as screening, delay be- havior, and exposure to cancer risk factors. Articles in this section are divided into three parts. The first is labeled KAP: Descriptive/ Analytic and contains 35 articles which present descriptive or analytic studies on the cancer knowledge, attitudes, and practices of blacks. The second, labeled KAP: Experimental, contains three articles with experimental study de- signs. The third part, KAP: Policy/Review, includes six articles which examine policy issues or review the literature. Data on the cancer-related knowledge, attitudes, and practices of blacks are limited. For example, no articles containing prospective study designs which relate KAP to cancer outcome were identified. Longitudinal studies are therefore needed to document specific relationships between KAPs and cancer incidence, mortality, and survival within black populations. 199 KAP: DESCRIPTIVE/ANALYTIC American Cancer Society Black Americans’ Attitudes Toward Cancer and Cancer Tests: Highlights of a Study CA 31(4):212-218, 1981 515 Subjects Screening: mammography, breast self-exam (BSE), Pap smear, proctoscopic exam; Treatment; Risk Factor Exposure: tobacco; Cancer Site: multiple sites; Communications Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1980 Sample Characteristics 750 individuals: black, 18+ years old; United States Variables SES: income Data Sources Survey Abstract The article presents a series of tables highlighting the results of a 1980 American Cancer Society-sponsored study of black Amer- ican attitudes toward cancer and cancer tests. The study was con- ducted by EVAXX, Inc., a black-owned evaluation organization. It was based on personal, in-home interviews conducted by black interviewers with a nationwide sample of 750 black American men and women, 18 years old and older. The study findings are com- pared to a similar American Cancer Society-sponsored study con- ducted in 1978 among a sample of the general population. Baker, M.D.; Johnston, J.R.; Turner, D.M. Smoking Habits of Blacks in Industry S Afr Med J 54:67-69, 1978 516 Subjects Risk Factor Exposure: occupation, tobacco Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 676 individuals: black, male, factory workers; Johannesburg, South Africa Variables Demographic: age, urban-rural Data Sources Survey Abstract A number of Black workers from a factory near Johannesburg were questioned about their smoking habits. Compared with an earlier study, a significantly greater number of younger workers now smoke — and mainly cigarettes. The roles played by the strong tobacco lobby and that of the Department of Health are considered. Breslow, L.; Hochstim, J.R. Sociocultural Aspects of Cervical Cytology in Alameda County, Calif. Public Health Rep 79(2):107-112, 1964 517 Subjects Screening: Pap smear; Diagnosis; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: cross-sectional; Measures: age-specific incidence rate, descriptive statistics; Time Frame: 1960-1962 Sample Characteristics 946 individuals: female, 20+ years old; Alameda, CA. 266 cervix uteri cancer cases (Negroes 57; whites 205; others 4): female, 20+ years old; Alameda County, CA Variables SES: occupation, education, public-private hospital status; Demographic: age, race-ethnicity; Cancer: stage Data Sources Survey, non-SEER population-based cancer registry Abstract Use of the cervical cytologic test in Alameda County has been profoundly influenced by sociocultural factors. Women in minority racial and ethnic groups, those of low social status as defined by their husbands’ occupations, and those with low educational at- tainment have used the test to a far lesser extent than women in more favorable social circumstances. Of particular importance to cancer control, the very groups of women with least utilization of the test have been the ones with the highest rate of cervical cancer, as measured by morbidity and mortality data. By far the greatest influence on women in obtaining the test was the advice or action of their physician. Nine of ten women who took the test stated that they obtained their first test in the course of a routine physical examination or at the specific suggestion of their physician. Only 1 of 10 reported that she independently asked to have the test. Participation in a health insurance plan was positively related to obtaining the Papanicolaou test, particularly a health plan em- phasizing screening examinations. Finally, knowledge of the test has an important bearing on utilization. Among those women who knew of the test, social class made little difference in whether or not they took the test. A major conclusion of this study is that further efforts to promote the use of the cytologic test for cancer of the cervix should be concentrated in the minority ethnic-racial- socioeconomic complex identified as being at greatest risk of cer- vical cancer yet with least use of the test up to the present time. The importance of using cervical cytology at an early age (20-30 years) also is emphasized. 201 KAP: DESCRIPTIVE/ANALYTIC Butler, L.; King, G.; White, ].E. Communications Strategies, Cancer Information and Black Populations: An Analysis of Longitudinal Data 518 In: Progress in Cancer Control IV: Research in the Cancer Center. New York: Alan R. Liss, 1983, p 171-182 Subjects Abstract Communications The authors discuss the inherent difficulties in reaching blacks Study Characteristics (males and females) due to the limitations in the use of mass media Design: time trend; Measures: descriptive statistics; Time Frame: and sociological factors. Also discussed are the communication 1978-1982 strategies which are presently employed that are beginning to Sample Characteristics prove effective. Sociological factors which may inhibit dissemina- 2,934 telephone calls: black, Washington, DC, SMSA tion to blacks are explored as well as the positive role of the health Variables professional in the community outreach segment of cancer control. Demographic: age, sex Data Sources Cancer Information Service Cantor, A.B.; Miller, M.C., III; Murphy, E.; et al. Comparison of Media for Dissemination of Information About Cancer in Rural Communities 519 South Med J 72(10):1232-1235, 1979 Subjects Abstract Communications; Education A study was done in Orangeburg County, South Carolina, to Study Characteristics compare the effectiveness of the various mass media-television, Design: cross-sectional; Measures: descriptive statistics radio, and newspaper-for the dissemination of educational mater- Sample Characteristics ial about cancer. Effectiveness of these media in reaching certain 80 individuals (urban 40; rural 40): Orangeburg County, SC subpopulations defined by race, sex, income, and population de- Variables nsity was also examined. The newspaper was found to be generally SES: income; Demographic: age, sex, race-ethnicity (black, more effective than other media. This superiority held for almost white), urban-rural all subpopulations. Furthermore, it was found that certain sub- Data Sources populations differed from others in the extent to which they were Survey reached by a particular medium. Cardwell, ].J.; Collier, W.V. Racial Differences in Cancer Awareness, What Black Americans Know and Need to Know About Cancer 520 Urban Health, Oct 1981, p 29-32 Subjects Screening: Pap smear, mammography, proctoscopic exam; of blacks will survive cancer. In 1980, the American Cancer Society Diagnosis; Treatment; Communications; Education commissioned a study of black American attitudes toward cancer Study Characteristics and cancer tests. Using face-to-face interviews, EVAXX, Inc., a Design: cross-sectional; Measures: descriptive statistics; Time black-owned evaluation research firm, conducted a survey among Frame: 1980 a scientific sample of 750 blacks representing adult urban blacks. Sample Characteristics Results from this study were compared with a similar 1978 study 750 individuals: black, urban; United States of the general population. Findings from the 1980 study indicate Abstract that black Americans are terrified of cancer; they underestimate Black Americans are slightly more likely than whites to get its incidence, are pessimistic about its cure and are not aware of cancer and much more likely to die from it. Estimates are that diagnostic tests. According to the survey information, blacks can one-quarter of the U.S. population will contract cancer sometime be reached with cancer information through local community or- in their lifetime. Forty-one percent of whites, but only 30 percent ganizations, doctors, the black media and schools. 202 KAP: DESCRIPTIVE/ANALYTIC Celentano, D.D.; Holtzman, D. Breast Self-Examination Competency: An Analysis of Self-Reported Practice and Associated Characteristics Am J Public Health 73(11):1321-1323, 1983 521 Subjects Screening: breast self-exam (BSE); Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 308 individuals (blacks 132; whites 176): female, 18 + years old; Baltimore, MD Variables SES: education, income, employment status; Demographic: age, race-ethnicity, marital status Data Sources Survey Abstract Of 308 Baltimore, Maryland women surveyed by telephone, 76.3 per cent reported having peformed breast self-examination (BSE) during the last year, with only 35 per cent reporting monthly BSE. Four BSE ability scores showed that most women had little knowledge of the proper BSE technique. Utilization factors, socioeconomic status, and knowledge/attitude about cancer were not related to BSE competency. Higher BSE competency scores were related to performing BSE at the recommended interval, hav- ing been taught the procedure by a health professional, and per- ceived confidence in BSE practice were also related to proper per- formance. Corel, J. Ethnicity and Cancer Prevention in a Tri-Ethnic Urban Community 522 J Natl Med Assoc 76(10):1013-1019, 1984 Subjects Screening: proctoscopic exam, mouth exam; Prevention; Risk Factor Exposure: diet-nutritional factors, tobacco, solar radiation; Cancer Site: multiple sites; Communications; Education Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: Winter, 1980 Sample Characteristics 64 individuals (blacks 20; Mexican-Americans 20; whites 24): Galveston, TX Data Sources Survey, U.S. Population Census Abstract A pilot study of knowledge and behavior regarding primary cancer prevention was conducted in a tri-ethnic urban community. Knowlege of smoking and dietary risk factors was substantial, but awareness of cancer safeguards involving reduced sun exposure and mouth and proctological examinations was low. One of five respondents had taken measures to prevent cancer, and these persons tended to rate their own risk higher than respondents who made no life-style changes. Self-motivated behavior change focused on avoidance of cancer-promoting food. Blacks differed from whites and Mexican-Americans in awareness of cancer cause and prevention, particularly regarding dietary factors and behavior modification. The findings have implications for the design of cancer control measures in multiethnic communities. Correa, P.; Johnson, W.D. Cancer and Lifestyle in Louisiana J La State Med Soc 135(3):4-6, 1983 523 Subjects Risk Factor Exposure: diet-nutritional factors, alcohol, occupation, tobacco; Cancer Site: multiple sites Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 1,686 individuals (blacks 229; whites 1,457—North Louisiana 817; South Louisiana 869): male, 30+ years old; Louisiana Variables SES: occupation Data Sources Survey Abstract The purpose of this study was to investigate the link between cancer risk and lifestyle in Louisiana. A random sample survey was conducted of male holders of driver's licenses from four south- ern and five northern Louisiana parishes with contrasting lung cancer risks. A questionnaire was mailed to them inquiring about their lifestyle. Lifestyle features related to race rather than geog- raphy include education, spraying of home for insects, work in farming, construction, white collar occupations, and consumption of coffee. The smoking habits of parents appear to influence the smoking habits of their children, especially among blacks. KAP: DESCRIPTIVE/ANALYTIC Covey L.S.; Mushinski, M.H.; Wynder E.L. Smoking Habits in a Hospitalized Population: 1970-1980 Am J Public Health 73(11):1293-1297, 1983 524 Subjects Risk Factor Exposure: tobacco Study Characteristics Design: time trend; Measures: descriptive statistics, higher level statistics; Time Frame: 1970-1980 Sample Characteristics 23,953 hospital patients (blacks 3,628; whites 20,325—Period 1: 16,484; Period 2: 7,469): New York, NY; Los Angeles, CA; Houston, TX; Birmingham, AL; Philadelphia, PA; Miami, FL; San Francisco, CA; Chicago, IL; New Orleans, LA Variables SES: education; Demographic: age, sex, race-ethnicity Abstract The smoking habits of 23,953 hospitalized men and women aged 20 to 80, interviewed in nine United States cities between 1970 and 1980, were examined. Comparisons were made of age-ad- justed smoking rates in 1970-1975 vs. 1976-1980, stratified by sex, race, and educational level. An overall decrease in smoking expo- sure between the time periods studied was observed; a decline in rates of current cigarette smokers and an increase in use of cigaret- tes containing 12 mg tar or less. Nevertheless, among current smokers, there was no reduction in the proportions of those who smoke 31+ cigarettes daily. An enhancing effect of higher educa- tion on rates of smoking cessation and use of cigarettes containing 12 mg tar or less was observed. In both time periods, proportion- ately more Black than White men were smokers, although the difference was smaller in college educated groups. These findings point up the limited effectiveness of publc health education about smoking in lower socioecomimic groups, and suggest that in the future, lung cancer and other diseases for which smoking is a major factor will be increasingly social class-related diseases. Denniston, R.W. Cancer Knowledge, Attitudes, and Practices Among Black Americans 525 In: Cancer Among Black Populations (eds) Mettlin C., Murphy G. New York: AlanR. Liss, 1981, p 225-235 Subjects Screening: Pap smear, breast self-exam (BSE), mammography, Education; Communications; Cancer Site: breast Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1979 Sample Characteristics 1,995 individuals (blacks 415; whites 1,580): female; United States Variables Demographic: age, race-ethnicity, age at first birth Data Sources Survey Abstract This article states that it usually is individual human actions that increase or decrease risk, that detect cancer early or late, that seek treatment promptly, or refuse treatment altogether. It is re- ported that: 1) the level of knowledge about cancer tends to be lower among Blacks than among Whites, 2) many blacks believe that Whites get more cancers, 3) myths and misconceptions about cancer tend to be strong among Blacks compared to Whites, and 4) Blacks tend to report health problems later in the disease process. Emdon, S.; Gerard, U.; Jones, R. Knowledge About and Utilization of Facilities for Cervical Smears Among Black Women in Johannesburg S Afr Med J 65:289-290, 1984 526 Subjects Screening: Pap smear; Cancer Site: cervix uteri Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 89 hospital patients: black, female, childbearing age, urban; Johannesburg, South Africa Data Sources Survey Abstract Cancer of the cervix is the most common cause of death from 204 carcinoma among Black South African women. Mortality from this disease can be reduced by screening using the cervical smear. To determine the knowledge about and utilizaton of facilities for cer- vical smears among Black women in Johannesburg, 89 women were interviewed at Baragwanath Hospital. Only 4 of these had heard of or had had a cervical smear. This indicates that education about cervical carcinoma and cervical smears needs to be more extensive. KAP: DESCRIPTIVE/ANALYTIC Friedman, G.D.; Seltzer, C.C.; Siegelaub, A.B.; et al. Smoking Among White, Black and Yellow Men and Women 527 Am J Epidemiol 96(1):23-35, 1972 Subjects Risk Factor Exposure: tobacco Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1964-1968 Sample Characteristics 111,024 individuals (blacks 12,735; whites 88,570; yellows 4,322; others and unknown 5,397): health maintenance organization members; San Francisco-Oakland, CA Variables Demographic: age, sex, race-ethnicity Data Sources Survey Abstract Substantial differences in smoking habits were noted in relation to age, sex, and race among 89,066 white, black, and yellow men and women. The study subjects reported their smoking habits at the Kaiser-Permanente multiphasic health checkup, given in San Francisco and Oakland, California during the years 1964-1968. Cigarette smoking was more prevalent among men than women. The age decades 20-29 and 30-39 contained the largest proportion of smokers. On the average the yellows contained the lowest per- centage of cigarette smokers; and in most age-sex groups of smok- ers, yellows were the least apt to inhale. Current cigarette smoking was more prevalent in black men than in white men. Among cigarette smokers whites were most apt to inhale and to smoke at least one pack per day. A smaller proportion of cigarette smokers was noted among Chinese than among Japanese, particularly in women and younger men. Friedman, G.D.; Siegelaub, A.B.; Seltzer, C.C. Cigarette Smoking and Exposure to Occupational Hazards Am J Epidemiol 98(3):175-183, 1973 528 Subjects Risk Factor Exposure: occupation, tobacco Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1964-1968 Sample Characteristics 70,289 individuals (blacks 9,812; whites 57,352; yellows 3,125— Smokers 37,122; nonsmokers 33,167): 15-79 years old Variables SES: occupation; Demographic: age, sex, race-ethnicity Data Sources Health maintenance organization records Abstract The frequencies of reported exposures to work-related environ- mental hazards were compared in 70,289 white, black and oriental male and female cigarette smokers and nonsmokers who received Kaiser-Permanente Multiphasic Health Checkups. In almost all race-sex groups, smokers more often reported occupational expo- sures to a variety of chemicals, fumes, sprays and dusts, and to extreme heat and loud noise. No consistent smoker-nonsmoker differences were noted for exposures to insect and plant sprays, x-rays, radioactivity or ultraviolet radiation. Men were more often exposed than women, and black men more often than white men, to several of these hazards. It is suggested that differences in these exposures be taken into account in studies of smoking and certain diseases, just as smoking should be taken into account in studying the consequences of certain occupational exposures. Hendershot, G.E. Coitus-related Cervical Cancer Risk Factors: Trends and Differentials in Racial and Religious Groups Am ] Public Health 73(3):299-301, 1983 529 Subjects Risk Factor Exposure: sexual practices, contraceptive devices; Cancer Site: cervix uteri Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1965, 1976 Sample Characteristics 13,708 individuals (blacks 3,471; whites 10,237—Period 1: 5,884; Period 2: 7,824): female, 15-44 years old; United States Variables SES: religion; Demographic: race-ethnicity Data Sources National Center for Health Statistics survey Abstract Data from nationally representative surveys conducted in 1965 and 1976 are presented which describe differences between women of different religious and racial groups with respect to the preva- lence of primary marriage and use of barrier methods of contracep- tion. These practices were significantly less likely among Black women and significantly more likely among Jewish women than among comparison groups, which is consistent with the hypothesis that cervical cancer risk is related to coital behaviors. KAP: DESCRIPTIVE/ANALYTIC Hendershot, G.E. Screening for Cervical Cancer, 1973-1976 Public Health Briefs 71(8):851-852, 1981 530 Subjects Screening: Pap smear; Cancer Site: cervix uteri; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1973-1976 Sample Characteristics Screening information: females, ever married and single mothers, 25-44 years old; United States Variables SES: income; Demographic: race-ethnicity (black, white), urban-rural Data Sources National Center for Health Statistics survey Abstract Using data from the National Survey of Family Growth, Cycle II, the per cent of women without a Pap test 1973-76 is shown according to race, residence, and income, updating an earlier re- port based on 1973 data. Poor metropolitan and nonmetropolitan women continued to be significantly less likely than other women to have had a Pap test. Hunter, S.M.; Webber, L.S.; Berenson, G.S. Cigarette Smoking and Tobacco Usage Behavior in Children and Adolescents: Bogalusa Heart Study Prev Med 9:701-712, 1980 531 Subjects Risk Factor Exposure: tobacco Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1976-1977 Sample Characteristics 3,147 individuals (blacks 1,092; whites 1,922): 8-17 years old; Bogalusa, Louisiana Variables Demographic: age, sex, race-ethnicity Data Sources Survey Abstract As part of a comprehensive program to assess cardiovascular disease risk factor variables in an entire biracial pediatric popula- tion, a study of cigarette smoking was included in the Bogalusa Heart Study. The purpose of the study was to obtain reliable and valid data on attitudes, beliefs, and cigarette-smoking behavior. The questionnaire demonstrates good internal consistency, test-re- test reliability, and external validation with plasma thiocyanate. For all sex and race groups the percentage of smokers increases with age. White boys begin smoking exposure and adopt cigarette smoking earlier (32% of 14- to 15-year-olds) while white girls in- crease and surpass the boys by ages 16 to 17 (41%). Black boys and girls lag behind white children in early experience and adop- tion of smoking behavior. The largest percentage of experimental nonadopters (those who said they tried cigarettes to see what they were like but do not smoke now) occurs among the 12- to 13-year- olds for each race-sex group except black girls. Kleinman, J.C.; Kopstein, A. Who Is Being Screened for Cervical Cancer? Am ] Public Health 71(1):73-76, 1981 532 Subjects Screening: Pap smear Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1973 Sample Characteristics 64,815 individuals (blacks 6,265; whites 58,550): female, 25+ years old; United States Variables SES: income; Demographic: age, race-ethnicity, geographic area Data Sources National Center for Health Statistics Survey Abstract Data from the 1973 National Health Interview Survey, a prob- 206 ability sample of the United States population, are used to examine the relationship between Pap testing and four socioeconomic vari- ables. It was found that women at highest risk of cervical cancer are least likely to have had Pap tests. The proportion of women who report never having had a Pap test is greater among Blacks, the poor, the elderly and nonmetropolitan residents. In particular, poor Black women in nonmetropolitan areas have extremely high proportions reporting non Pap test. However, high risk women are only slightly less likely to have visited a doctor in the two years preceding interview. These results suggest that improvement in Pap test coverage among high risk women could be attained by encouraging the use of the Pap test in regular ambulatory medical care. KAP: DESCRIPTIVE/ANALYTIC Maclean, U.; Sinfield, D.; Klein, S.; et al. Women Who Decline Breast Screening J Epidemiol Community Health 38:278-283, 1984 533 Subjects Screening: clinical breast exam, breast self-exam (BSE); Cancer Site: breast Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 146 individuals (breast screening clinic attenders 21; non- attenders 125): female, 45-64 years old; Edinburgh, United Kingdom Variables SES: social class Data Sources Survey Abstract The health related behaviours and attitudes of a random sample of 125 women aged 45-64 who declined to accept an invitation to attend a breast screening clinic are described. The women differed from attenders in their socioeconomic class and in their customary use of measures to promote health. They were basically unsym- pathetic towards the idea of screening and the invitation had caused them considerable anxiety. Policy implications are raised. Manfredi, C.; Warnecke, R.B.; Graham, S.; et al. Social Psychological Correlates of Health Behavior: Knowledge of Breast Self-Examination Techniques Among Black Women Soc Sci Med 11:433-440, 1977 534 Subjects Screening: breast self-exam (BSE); Cancer Site: breast Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 696 cancer cases: black, female, inner city; Buffalo, NY Data Sources Survey Abstract This paper examines the relationship between fear, perceived susceptibility and belief in the efficacy of early detection of breast cancer as correlates of the likelihood that a procedure for early detection, self-examination (BSE), will actually be known. The data were obtained via personal interviews as part of a study of practices and beliefs related to cancer detection and prevention among black females in Buffalo, New York. Belief in the efficacy of early detec- tion of the disease to reduce the danger from the disease was found to be the strongest correlate of the ability to perform BSE. However, independent effects of fear as reflected in perceived threat and feelings of personal susceptibility were also apparent. Michielutte, R.; Diseker, R.A. Racial Differences in Knowledge of Cancer: A Pilot Study Soc Sci Med 16:245-252, 1982 535 Subjects Communications; Cancer Knowledge Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 140 individuals (blacks 42; whites 98): 21+ years old; Winston-Salem, NC Variables SES: education; Demographic: age, sex, race-ethnicity Data Sources Survey Abstract The present study examined the reported sources of informa- tion on cancer and the level of cancer knowledge for a sample of black and white adults. Black respondents had significantly less knowledge and the relationship between race and cancer knowl- edge persisted even when controlling for education, sex, and age. Possible reasons for the observed difference include (a) the ten- dency for blacks to obtain information on cancer from television and radio, while whites rely more on printed materials, (b) differ- ences in the quality of education received by black and white adults, and (c) a possible lack of motivation on the part of black respondents to acquire knowledge of cancer due to lower access to medical care. Intervention programs designed to provide all blacks with information about cancer should take into account the preferred sources of information, and should be oriented toward reducing the barriers to taking action related to prevention and early detection as well as increasing perceptions of the benefits of taking such action. KAP: DESCRIPTIVE/ANALYTIC Michielutte, R.; Diseker, R.A. Children’s Perceptions of Cancer in Comparison To Other Chronic Illnesses J Chron Dis 35:843-852, 1982 536 Subjects Education; Cancer Site: all sites combined Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1978 Sample Characteristics 295 individuals (blacks 51; whites 244): 11-14 years old, 7th grade; Forsyth County, NC Variables SES: parent's education and occupation; Demographic: race-ethnicity Data Sources Survey Abstract Research which deals with beliefs and feelings about cancer suggests that cancer is somewhat unique among chronic illnesses with respect to the intensity of negative feelings and stigma at- tached to this disease. The research results to date fit well in the context of the health belief model and indicate that cancer is ex- tremely high in perceived severity, moderate in perceived suscep- tibility, and extremely low in perceived benefits of preventive and treatment programs. This paper reports the results of a study which used the semantic differential technique to determine a sample of seventh-grade students’ perceptions of cancer in com- parison to heart disease, diabetes, and mental illness. The results indicated that children generally view cancer as higher in severity (except for heart disease), higher in susceptibility, and lower in benefits of treatment than the other illnesses. Perceptions did not differ by sex, socio-economic background, or knowledge of cancer. However, in addition to perceiving cancer as being very high in severity and feeling pessimistic about the chances of recovering, black children believe they are personally more susceptible than do white children. In general, this combination of perceptions provides a unique stigma to cancer as a fearsome chronic disease with little hope of cure. The low perceived benefits of treatment presents a barrier to action which has important implications for health education programs. Parker, D.F.; Enterline, J.P.; White, ].E. Differences Between Black and White Responses to Cancer Information Service Promotion Mechanisms In: Issues in Cancer Screening and Communication. New York: 537 Alan R. Liss, 1982, p 429-436 Subjects Communications Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1978-1981 Sample Characteristics 7,493 telephone calls (blacks 2,178; whites 5,315): Washington, DC, SMSA Variables Demographic: race-ethnicity, sex Data Sources Cancer Information Service Abstract The purpose of this paper is to present the findings of a study of the response to promotion of the Cancer Information Service (CIS) and how this response varies between black and white audi- ences. The study provides an analysis of the distribution of sources by which people learned about the CIS and an understanding of which promotional mechanisms might prove to be effective and deserving of further study. Blacks were found to be more likely than whites to learn of CIS through the electronic media (radio and TV) and less likely than whites through printed sources. Blacks and whites appear to respond differently to CIS promotion. Tele- vision promotion appears to be particularly effective in reaching blacks while targeted informational campaigns using the print media were much less effective. 208 KAP: DESCRIPTIVE/ANALYTIC Roberson, N.L. A Cancer Control Intervention for Black Americans in Buffalo, New York: A Case Study 538 Buffalo, New York: State Univ. of New York at Buffalo, 1986. Available from: University Microfilms, Ann Arbor, Mich.; Publication No. 86-09,147. p 509. Dissertation. Subjects Screening: breast self-exam (BSE); Education; Communications; study sought to evaluate whether community organization Risk Factor Exposure: tobacco; Cancer Site: breast, lung through networking could effectively improve access to cancer Study Characteristics education programs and determine the impact of education prog- Design: cross-sectional; Measures: descriptive statistics, higher rams on health practices about breast self-examination (BSE) and level statistics; Time Frame: 1982-1983 smoking cessation. Program intervention involved implementa- Sample Characteristics tion of cancer education programs through a network where com- 542 individuals (BSE education group 130; smoking education munity health guides, community volunteers, and mass media group 23; cancer information group 389): black, 18+ years, were the vehicles for recruiting and educating neighborhood resi- moderate to low income; Buffalo, NY dents. This case study was conducted in the Black communities Variables of Buffalo, New York, where six geographically defined census SES: education; Demographic: age, sex tracts were targeted study areas. The study population consisted Data Sources of five hundred forty-two (542) persons, eighteen years of age and Survey older, who attended programs. Of these subjects, one hundred Providers fifty-three (153) subjects agreed to a three-month follow-up inter- Nurse view. Results of this study indicated that community organization Abstract through networking was a major strategy for improving access to Cancer mortality is particularly high among Blacks for two cancer education programs and was useful for recruiting particip- major sites: breast and lung. Factors believed to contribute to the ants to attend training sessions. In addition, results showed that problem of control of these diseases include underutilization of cancer education programs had an impact on participants’ cancer- the limited cancer education programs available, limited knowl- related knowledge and health practices about breast and lung edge, certain beliefs about breast and lung cancer and minimal cancer, but demonstrated a lesser impact on cancer-related beliefs. disease prevention practices. Epidemiological studies suggest that Moreover, a large proportion of women showed improvement in mortality for these diseases can be reduced by increasing cancer reported BSE technique and frequency. Also, a majority of subjects control activities focused on cancer education programs. However, either quit or reduced smoking after intervention. This research there is little evidence that interventions to improve accessibility is significant because it contributes to establishing a data base to cancer education and utilization of cancer control services have about practical program strategies and responses to cancer control been designed to meet these needs for this population. This case intervention. Rose, C.L. Social Correlates of Smoking in a Healthy Male Population 539 Aging Hum Dev 3:111-124, 1972 Subjects Abstract Risk Factor Exposure: tobacco This is a preliminary exploration of smoking behaviour relative Study Characteristics to social characteristics in the Normative Aging population. There Design: cross-sectional; Measures: descriptive statistics, higher were definitive findings with respect to the non-smoker and the level statistics ex-smoker, but none with regard to the amount of cigarette smok- Sample Characteristics ing. 1,984 individuals (nonwhites 40; whites 1,943), male, veterans Variables SES: religion, education, occupation; Demographic: age, race-ethnicity, marital status 209 KAP: DESCRIPTIVE/ANALYTIC Sterling, T.D.; Weinkam, J.J. Smoking Patterns by Occupation, Industry, Sex and Race Arch Environ Health 33(6):313-317, 1978 540 Subjects Risk Factor Exposure: tobacco; Socioeconomic Factors Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1970 Sample Characteristics 76,675 individuals: 17+ years old; United States Variables SES: industry, occupation; Demographic: sex, race-ethnicity (black, white) Data Sources National Center for Health Statistics survey Abstract Patterns of prevalence, amount, and cessation of smoking are computed for occupations by socioeconomic class, sex, and race, based on a probability sample of 39,011 households collected by the 1970 Health Interview Survey. Smoking is most prevalent in blue-collar occupations, while a high proportion of professionals and managers who smoke, stop smoking. Within industries, sub- stantially higher percentages of individuals smoke in lower pre- stige paying jobs, while more smokers quit in the higher prestige paying jobs. Smoking is most prevalent among women managers and professionals, and least among those employed in traditional work. One surprising and possibly very important finding is that white smokers smoke about 20% more cigarettes per day than black smokers. Not only would it seem unreasonable to ascribe the larger rate of lung disease among blacks than whites (especially cancer), to smoking when blacks smoke significantly fewer cigaret- tes than whites, but this same negative relationship points to oc- cupational exposure as the possible major cause for lung cancer. Sterling, T.D.; Weinkam, J.J. Smoking Characteristics by Type of Employment J Occup Med 18(11):743-754, 1976 541 Subjects Risk Factor Exposure: occupation, tobacco Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1970 Sample Characteristics 75,827 individuals: 17+ years old; United States Variables SES: occupation, industry; Demographic: age, sex, race-ethnicity (black, white) Data Sources National Center for Health Statistics survey Abstract Smoking habits are examined in a probability sample of 75,827 American men and women. The data was obtained from The 1970 Household Interview Survey conducted by the National Center for Health Statistics. There are pronounced differences in preva- lence intensity, kind, cessation of smoking related to type of em- ployment. Occupations tend to be more of a factor in white males, where occupational choice is least encumbered, than in black males or in females. While blacks uniformly smoke with greater preva- lence, they are much lighter smokers than whites. Employed women tend to smoke with prevalence approaching males for many occupations. Prevalence of smoking almost uniformly is highest among blue collar occupations and lowest among profes- sionals, managers, and proprietors. Stern, E.; Clark, V.A.; Coffelt, C.F. Contraceptive Methods: Selective Factors in a Study of Dysplasia of the Cervix Am ] Public Health 61(3):553-558, 1971 542 Subjects Risk Factor Exposure: exogenous hormones, reproductive factors; Cancer Site: cervix uteri Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics Sample Characteristics 204 individuals (dysplasia cases 87; normals 117): female, 15-35 years old, low income, inner-city; Los Angeles County, CA Variables SES: income, religion, employment status, education; Demographic: race-ethnicity (Negro, not Negro), age at menarche, age at first marriage, age at first birth, number of pregnancies, parity Data Sources Family planning clinic medical records 210 Abstract In a homogeneous population from a large urban ghetto, there are pre-existing differences among women who choose the IUD, the contraceptive pill, or other methods. Others have reported a difference in demographic variables, while, in addition, the present authors find differences in the prevalence of dysplasia according to contraceptive choice, prior to the use of the contraceptive. KAP: DESCRIPTIVE/ANALYTIC Stern, E.; Misczynski, M.; Greenland, S.; et al. “Pap” Testing and Hysterectomy Prevalence: A Survey of Communities With High and Low Cervical Cancer Rates Am ] Epidemiol 106(4):296-305, 1977 543 Subjects Screening: Pap smear; Cancer Site: cervix uteri; Cancer Measures:incidence Study Characteristics Design: cross-sectional; Measures: descriptive statistics, higher level statistics; Time Frame: 1974-1975 Sample Characteristics 673 individuals (blacks 291; whites 218; Spanish surnamed 164—Low income area 361; high income area 312): female, 35+ years old; Los Angeles County, CA Variables SES: income; Demographic: age, race-ethnicity, hysterectomy Data Sources Death registration, non-SEER population-based cancer registry, survey Abstract There is an inverse relation between cervical cancer rates and income in Los Angeles County communities, and there is a positive association between level of Papanicolaou (Pap) testing and in- come. Ethnic differences in cervical cancer rates and Pap testing are dependent on income. The relatively high rate of cervical cancer and low level of systematic screening in low income communities suggest that a community trial to assess the value of cytologic screening in reducing cervical cancer rates is feasible. Information on hysterectomy prevalence by type of procedure supports the idea that the long-observed decline in cervical cancer rates is in part attributable to a concomitant decrease in the ratio of subtotal to total hysterectomy. Tietz, W.; McSherry, L.; Britt, B. Family Sequelae After a Child's Death Due to Cancer Am ] Psychother 31:417-425, 1977 544 Subjects Psychosocial Factors: reactions to dying/grief; Cancer Site: multiple sites Study Characteristics Design: case-study Sample Characteristics Nine households (blacks 2; whites 2; Mexican-Americans 5): lower socioeconomic status, history of child dying within 2 years prior to study; California Variables Demographic: age, sex, race-ethnicity Data Sources Survey Abstract In a small study assessing the psychologic sequelae on disad- vantaged families after a child's death from cancer, a high incidence of psychologic problems was found. The implication is that the coping mechanisms failed partially as a result of inadequate anti- cipatory griefing. Van Der Burgh, C. Smoking Behavior of White, Black, Coloured and Indian South Africans: Some Statistical Data on a Major Public Health Hazard S Afr Med J 55:975-978, 1979 545 Subjects Risk Factor Exposure: tobacco Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1975-1977 Sample Characteristics 14,000 cancer cases (blacks 4,000; whites 5,000; coloureds 2,500; Indians 2,500): 20+ years old; South Africa Variables Demographic: age, sex, race-ethnicity Data Sources Survey 211 Abstract An analysis of the smoking habits of the population groups in the Republic of South Africa indicates an alarmingly high incidence of smoking, particularly among Coloureds. Significant numbers of men and women in all the population groups, especially Whites, are former smokers. KAP: DESCRIPTIVE/ANALYTIC Warnecke, R.B.; Graham, S. Characteristics of Blacks Obtaining Papanicolaou Smears Cancer 37:2015-2025, 1976 546 Subjects Screening: Pap smear Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1972 Sample Characteristics 700 individuals: black, female; Buffalo, NY Variables SES: education; Demographic: age Data Sources Survey, physician medical records Abstract An understanding of the characteristics of women who do not obtain cervical cytology may shed light on procedures which can be instituted by physicians and public health workers to increase use of screening. Of particular interest are women at high risk of cervical cancer. We interviewed a random sample of about 700 blacks living in central Buffalo census tracts in regard to their past use of pelvic examinations as well as circumstances surrounding such use or non-use. We employed only data on screening which was verified in records. We found that there was decreasing use with increasing age and lack of contact with physicians, clinics, or hospitals. The relationship between increasing use and increas- ing education was found to be a function of the fact that women with more education are likely to be younger. There was clear evidence that making such examinations part of the contact women have with physicians or hospitals for any purpose, including childbirth, would increase the proportions screened. Warnecke, R.B.; Graham, S., Rosenthal S.; et al. Social and Psychological Correlates of Smoking Behavior Among Black Women J Health Soc Behav 19:397-410, 1978 547 Subjects Risk Factor Exposure: tobacco Study Characteristics Design: cross-sectional; Measures: descriptive statistics; Time Frame: 1972 Sample Characteristics 696 individuals: black, female, inner city; New York Data Sources Survey Abstract This paper reports a study of factors associated with smoking behavior among black females. One purpose of the study was to test the applicability of a conceptualization of the process of cessa- tion that combined two of the most commonly applied theoretical approaches used by other researchers in the field. A second goal was to ascertain the factors that best differentiated among four patterns of smoking behavior: continuous smoking, continuous smoking with some modification of amount, successful cessation, and lifetime abstention. The patterns of association in the data that were analyzed appeared to be consistent with the hypotheses derived from the expanded model, and the combined conceptuali- zation seemed helpful in dealing with difficulties inherent in each of the conceptualizations used by other investigators. It also ap- peared that the subjects in the study did not differ substantially from the white males who had been the subjects of most earlier studies. The results indicated that personal relevance of the threat to the individual, the use of communications, and the influence of others close to the respondent differentiated among the four behavior groups studied. 212 KAP: DESCRIPTIVE/ANALYTIC Warnecke, R.B.; Havlicek, P.L.; Manfredi, C. Awareness and Use of Screening by Older-Aged Persons In: Perspectives on Prevention and Treatment of Cancer in the Elderly (eds) Yancik R., et al. New York: Raven Press, 1983, p 275-287 548 Subjects Screening: breast self-exam (BSE), Pap smear, proctoscopic exam, rectal exam, prostate palpation Study Characteristics Design: panel study; Measures: descriptive statistics, higher level statistics Sample Characteristics 1,607 individuals: 36+ years old, low income; Illinois Variables SES: education; Demographic: age, sex Data Sources Survey Abstract Three health behavior models can be used to explain the use of early detection procedures by the elderly: economic or sociodemographic, behavioral, and structural. This article presents a new more general model with age and education indirectly affect- ing the receipt for medical procedures. The data presented are from a panel study of a sample of Illinois residents. The study population included those with limited access to the health care system because of income or isolated residence, who were ap- proaching the age of risk for many of the more prevalent cancers. The baseline data were obtained in telephone interviews. The model was applied to the data in two phases to investigate what elements most influence the receipt of early detection procedures. In phase one demographic, economic, and structural factors related to receiving a regular check-up were examined and in phase two behavioral factors were considered. Factors predicting receipt of specific cancer detection procedures were examined. Behavior re- lated to detection and diagnosis of cancer does not appear to be directly related to age except for cervical cancer where the receipt of a Pap smear decreases with age. The three procedures that required the intervention of a health care professional (prostate palpatin, Pap smear, proctoscopic examination) were most influ- enced by receipt of a regular check-up. However, the behavioral variable, knowledge of symptoms and health beliefs, most influ- enced breast self-examination. Wilkinson, G.S.; Wilson, J. An Evaluation of Demographic Differences in the Utilization of a Cancer Information Service Soc Sci Med 17:169-175, 1983 549 Subjects Communications Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 9,593 telephone calls: New York Variables SES: combined SES rank, occupation; Demographic: age, sex, urban-rural, race-ethnicity (nonwhite, white) Abstract A detailed empirical evaluation of the utilization of a public 213 telephone cancer information program was conducted. This paper reports demographic differences between utilizers and the general population. It was found that utilization varied according to socioeconomic status, area of residence, age and sex. The greatest proportion of users were females, from the higher social classes, city residents and 20-39 years of age. Attempts to look at inter- relationships among these variables did not produce any consistent findings. KAP: EXPERIMENTAL Cassileth, B.R.; Heiberger, R.M.; March, V._; et al. Effect of Audiovisual Cancer Programs on Patients and Families 550 J Med Educ 57:54-59, 1982 Subjects Treatment; Education; Cancer Site: multiple sites Study Characteristics Design: quasi-experimental; Intervention Study; Measures: descriptive statistics, higher level statistics Sample Characteristics 240 individuals (blacks 47; whites 193—Cancer patients 106; patient's relatives 120; patient's friends 14): Pennsylvania Variables SES: education; Demographic: race-ethnicity Data Sources Survey Providers Hospital Abstract Four new audiovisual programs about cancer and cancer treat- ment were evaluated. A total of 240 cancer patients, their families, and their friends were asked to complete self-report questionnaires before and after watching a program to determine the effects of the program on their knowledge of cancer, anxiety levels, and perceived ability to communicate with the staff. Education and race were the only demographic or medical variables that contri- buted to knowledge scores. Patients’ anxiety levels were signific- antly lower than those of families. Neither anxiety nor knowledge was a function of time from diagnosis. This study demonstrates that audiovisual programs can effectively increase patients’ under- standing of their disease, decrease anxiety, and facilitate communi- cation with staff. Kane-Williams, E.; White, ].E. Community-Based Cancer Education for the Elderly 551 In: Progress in Cancer Control IV: Research in the Cancer Center. New York: AlanR. Liss, 1983, p 113-122 Subjects Education; Cancer Site: breast, colon-rectum, prostate Study Characteristics Design: controlled intervention study; Measures: descriptive statistics Sample Characteristics 267 individuals (blacks 227; whites 17; others 23—Pre-test group 204; post-test group 63), 60+ years old; Washington, DC Providers Howard University Cancer Center Abstract This paper describes the experiences of the Howard University Cancer Center in developing and implementing cancer education programs for the elderly in the District of Columbia. Approximately 600 senior citizens participated in this project. A pre-and post-test instrument was developed to assist in the evaluation of this project. The preliminary results indicate that while the sampled elderly appear to have adequate knowledge about basic cancer concepts, future programs should focus on early warning signals and dispel- ling cultural myths about cancer. Special targeted efforts should be directed at older men, who have poorer knowledge about cancer in general and specifically about prostate cancer. Lynch, P.D.; Ronis, D.L. Cancer Information for Blacks: A Radio Program Evaluation 552 In: Issues in Cancer Screening and Communication. New York: Alan R. Liss, 1982, p 399-408 Subjects Communications; Cancer Site: breast, cervix uteri, colon-rectum, larynx, oral cavity Study Characteristics Design: controlled intervention study; Measures: descriptive statistics, higher level statistics Sample Characteristics 152 individuals (blacks 114; whites 38—Ten groups of 10-20 persons): urban; Detroit, MI Variables Demographic: age, sex, race-ethnicity Abstract This paper reports on a study designed to evaluate a series of five radio programs for an urban black audience. The purposes of the study were 1) to determine if the radio programs were approp- riate for the target audience and 2) to examine the effects of target- ing messages to a particular audience. The research indicated that the programs were successfully designed for a black audience. But, the research also indicated a potential problem of targeted messages: people outside of the target groups may have negative reactions to such messages. KAP: POLICY/REVIEW Askey, D.G.; Parker, D.; Alexander, D.; et al. Clergy As Intermediary—An Approach To Cancer Control 553 In: Progress in Cancer Control IV: Research in the Cancer Center. New York: AlanR. Liss, 1983, p 417-424 Subjects Prevention; Communications; Education; Cancer Site: multiple sites Abstract The article reviews data collected on the role of the clergy in the black community and data on the cancer experience of blacks in the United States and the District of Columbia. The Howard University Cancer Center's “Clergy as Intermediary Project” is described. This project used the PRECEDE model as a framework to help plan an effective program which integrates scientific theories and educational technologies. Initial reactions to the pro- ject have been favorable. Clausen, J.P. Cancer Diagnoses in Children: Cultural Factors Influencing Parent/Child Reactions 554 Cancer Nurs 1(5):395-401, 1978 Subjects Diagnosis Abstract. This paper focuses on the importance of a nurse’s ability to assess the impact of childhood cancer on the patient and family, taking into consideration the relationship of cultural background and reaction to diagnosis. The author has prepared this paper as a result of participation on a program for professional nurses spon- sored by the Association of Pediatric Oncology Nurses and the Nursing Department, St. Jude Children’s Research Hospital, Mem- phis, Tennessee, titled “A Child Living with Cancer.” DeVita, V.T., Jr. Cancer Prevention Awareness Program: Targeting Black Americans 555 Public Health Rep 100(3):253-254, 1985 Subjects Prevention; Education; Communications Abstract Data collected over the past decades show that cancer incidence rates have increased 27 percent among blacks and only 12 percent among whites. National Cancer Institute (NCI) researchers believe that differences in lifestyle contribute to these different cancer rates. Access to and use of medical care for screening tests, early diagnosis, and prompt treatment may contribute to differences in cancer death rates. In 1984 NCI announced its national goal: to reduce the 1980 U.S. cancer death rate by one-half by the year 2000. In that same year NCI launched the Cancer Awareness Prog- ram designed to inform and educate the public about cancer pre- vention. To inform and educate blacks and other high-risk popu- lations about cancer and cancer prevention, NCI plans to bring together health professionals, media, churches, and businesses representing black interests in a “Joint Health Venture.” NCI en- visions this group as a coalition for organizations whose unifying objective is to improve the overall health and well-being of the American population. Hall, H.; Bell, X. Increases in Cancer Rates Among Blacks J Black Psychology 12(1):1-14, 1985 556 Subjects Screening: multiple cancer screening tests; Psychosocial Factors: coping mechanisms; Risk Factor Exposure: diet-nutritional factors, alcohol, occupation, tobacco, stress; Cancer Site: multiple sites Abstract The recent increases in cancer rates among Blacks in the United States and in other countries is examined. Black males have the highest rates, but at one time had lower rates than White males. Levy's (1983) model is employed to examine direct and indirect psychological and behavioral risk factors that might explain racial differences in the initiation and progression of cancer. Differences are noted and suggestions for needed research, including a Black Studies perspective of this issue, are offered. KAP: POLICY/REVIEW Holtzman, D.; Celentano, D.D. The Practice and Efficacy of Breast Self-Examination: A Critical Review 557 Am ] Public Health 73(11):1324-1326, 1983 Subjects Screening: breast self-exam (BSE); Cancer Site: breast Abstract Evaluation of the results of breast self-examination (BSE) is inconclusive. Studies which address the question of proper method of BSE find that most women do not know how to carry out the procedure correctly. Socio-demographic characteristics most con- sistently related to BSE practice are age and education. New re- search on BSE must not only take into account such correlates but also provide some verification of the procedure to determine the importance of BSE in the detection and control of breast cancer. Mettlin, C.; Cummings, K.M. Communication and Behavior Change for Cancer Control 558 In: Issues in Cancer Screening and Communication. New York: Alan R. Liss, 1982, p 135-148 Subjects Communications Abstract This article discusses the impact of cancer communications on behavior and examines ways to improve the effectiveness of cancer communications. Conditions identified as being conducive to effec- tive attitude and behavior change through communciations in- clude: 1) novelty in the way the message is presented to attract attention, 2) use of multiple channels of communication to increase the probability of reaching the target audience, 3) pretesting of health messages to ensure that the information is understood by the target audience, 4) repetition of the message by multiple sources of high credibility, 5) targeting of specific behaviors to be changed rather than a general orientation, and 6) incorporation of specific behavioral instructions (e.g. tips on how to quit smoking) in the communications. The authors note that a recent study of breast cancer patients from the surveys of the American College of Surgeons Commission on Cancer shows a significant improve- ment in the stage at which diagnosis were made, especially among black patients. Public education campaigns may have contributed to the recent improvement observed. Other studies suggest that subgroups in the population may rely on different channels of communication for their information about health matters. For example, data from a survey at Roswell Park Memorial Institute indicate that nonwhites are more likely than whites to hear about special hospital clinics from personal sources such as relatives and friends, while whites tend to get their information from impersonal sources such as T.V., radio, or newspapers. SECTION III Health Services Research ET en ary - « 2 Te TATE Al a Ee mc. og ATT #* Section III Health Services Research The delivery, utilization, and quality of health and medical services appears to be strongly related to cancer mortality and survival rates. Although health ser- vices research has emerged as a major activity in public health in the last 30 years, it is not a clearly defined field of research. The classification of health services research articles in this bibliography was made according to the following framework: 1. Health Services Research concerns the use and delivery of health services in the population and typically considers the effectiveness, efficacy, or efficiency of particular health services and the delivery of such services. 2. Health Services Research is by its nature intervention-oriented, and usually includes research questions related to a) the delivery, availability, cost, quality, and continuity of health services; and b) health services access and compliance. 3. Health Services Research may be descriptive or analytical, retrospective, or prospective. This section includes 13 articles which deal with health services research and cancer within black populations. There is a paucity of literature on the topic which directly relates health services to cancer outcomes. The few articles iden- tified are strongly suggestive but in general do not provide direct evidence of an association between cancer and health and medical resources. This indicates a serious deficiency in the literature and points out the need for more research in this area. 219 HEALTH SERVICES RESEARCH: DESCRIPTIVE/ANALYTIC Fruchter, R.G.; Boyce, ].; Hunt, M. Invasive Cancer of Cervix: Failures in Prevention New York State ] Med 80(5):740-745, 1980 559 Subjects Access and Utilization of Health Services; Health Care Provider Knowledge, Attitudes, and Practices (KAP); Screening: Pap smear; Prevention; Cancer Site: cervix uteri Study Characteristics Design: retrospective; Measures: descriptive statistics, higher level statistics; Time Frame: 1976-1978 Sample Characteristics 97 cervix uteri cancer cases (blacks 74; whites 17; Hispanics 8; Orientals 1): female, hospital patients; New York Variables SES: education, income; Demographic: age, geographic area Data Sources Hospital medical records, physician medical records, survey Abstract Ninety-seven low-income urban women who developed inva- sive cancer of the cervix were interviewed about their previous Pap smears and medical history; 52 percent had no previous smear, and 62 percent had no smear in the previous five years. Lack of previous smears was most frequent among older women. Twenty- five women had been hospitalized 44 times for nongynecologic reasons in the ten years before the first sign of disease, but only 18 percent had a Pap smear. One half the women had received regular medical care in the previous five years, but 59 percent had no Pap smear. In low-income communities, new cervical cancer screening programs should be located in hospitals and ambulatory facilities that women use for acute health care. Greene, A.B.; Salber, E.]. Pro Bono Publico—A Century Later Am ] Public Health 67:735-739, 1977 560 Subjects Access and Utilization of Health Services; Risk Factor Exposure: tobacco Study Characteristics Design: cross-sectional; Measures: bed days, utilization of health services, descriptive statistics; Time Frame: 1973 Sample Characteristics 1,179 individuals (blacks 380; whites 799—Smokers 465; non-smokers 573; ex-smokers 141): 17 + years old; Durham, NC Variables SES: education, income; Demographic: sex, age, race-ethnicity Data Sources Survey Abstract From the introduction of Pro Bono Publico smoking tobacco in Durham, NC a century ago, the production of tobacco products has become a vital part of the state’s economy. How this may relate to the smoking behavior, and consequently to the health of its residents is assessed from smoking patterns of adult residents of a rural area of Durham County. Male smoking rates are consid- erably higher than U.S. estimates whereas female rates are lower. The national trend of increasing incidence of lung cancer can be expected to continue in North Carolina and may be even greater due to the high rate of smoking. McCusker, J. Where Cancer Patients Die: An Epidemiologic Study Public Health Rep 98(2):170-180, 1983 561 Subjects Access and Utilization of Health Services; Cancer Site: multiple sites Study Characteristics Design: cross-sectional; Measures: utilization of health services, odds ratio, descriptive statistics, higher level statistics; Time Frame: 1976-1978 Sample Characteristics 2,989 cancer deaths (blacks 128; whites 2,840; other or unknown 21): 15+ years old; Monroe County, NY Variables SES: census tract SES rank; Demographic: age, sex, race-ethnicity Data Sources non-SEER population-based cancer registry Abstract In a sample of deaths among cancer patients, the relationship of place of death to age, sex, length of time between diagnosis and death, cancer site, and patient's socioeconomic status was investigated. The Rochester (N.Y.) Regional Tumor Registry pro- vided these data for all cancer patients who died in Monroe County, N.Y., during 1976, 1977, and 1978. Patients who had not been residents of the county were excluded from the sample, as were patients under 15 years of age at death and those whose cancers had been diagnosed only at autopsy. Analysis with a logit model was used to estimate odds ratios that compared the prob- abilities of death in an acute care hospital and in a chronic care facility with the probability of death at home. Patients whose can- cers had been diagnosed less than 1 month before their deaths were significantly more likely to die in a hospital than were patients whose cancers had been diagnosed earlier. Cancer sites, too, were significantly related to place of death: persons with leukemia or lymphoma were most likely to die in a hospital, followed by pa- tients with lung, breast, and upper gastrointestinal tract cancers; persons with colorectal, genitourinary, and miscellaneous cancers were most likely to die at home. The patients whose deaths were studied were classified by socioeconomic area (SEA) ranking. Pa- tients who had resided in higher level SEAs were more likely to die at home than those from lower level SEAs; however, this trend was reversed among patients from the lowest level SEAs, who had a relatively high rate of death at home and a low rate of death in chronic care facilities. 221 HEALTH SERVICES RESEARCH: DESCRIPTIVE/ANALYTIC Misczynski, M.; Stern, E. Detection of Cervical and Breast Cancer: A Community-Based Pilot Study Med Care 17(3):304-313, 1979 562 Subjects Access and Utilization of Health Services; Screening: clinical breast exam, breast self-exam (BSE), Pap smear; Cancer Site: breast, cervix uteri Study Characteristics Design: prospective; Measures: descriptive statistics Sample Characteristics 499 individuals (kept appointment 72; broke appointment 28; clinic nonparticipants 399): female, 25-55 years old, low income, age-eligible for clinic; Compton and El Monte census tracts, Los Angeles County, CA Variables Demographic: geographic area Data Sources U.S. Population Census, health clinic medical records Providers Health clinic Abstract Pilot screening clinics were conducted in two Los Angeles County health districts where cervical cancer incidence and mor- tality rates were high. Results are of interest because they provide information on planning and operating community-based clinics in low-income areas where there are women who are unlikely to seek or are ineligible for available health services. Owen, W.L.; Anderson, P.S.; Parry, W.L. Health Services Delivery to Prostatic Cancer Patients Oklahoma State Med Assoc J 70:436-442, 1977 563 Subjects Quality of Care; Health Care Provider: physician; Cancer Site: prostate Study Characteristics Design: restrospective; Measures: descriptive statistics, higher level statistics; Time Frame: 1969-1971 Sample Characteristics 2,244 prostate cancer cases (blacks 227; whites 1,785; Indians 29): male, hospital patients; Oklahoma Variables Demographic: age, race-ethnicity; Cancer: stage, symptoms Data Sources Hospital medical records Abstract In attempting to assess differences in health care delivery to prostatic cancer patients, the relations among the variables of age, race, degree and specialty of primary physician, distance to care, stage and symptoms at diagnosis, and number of tests done have been examined. Age does not seem to be much of a factor in patient care. Race, on the other hand, does appear to be a deter- minant of care. On a percentage basis, more blacks are diagnosed in later stages; fewer blacks than whites see specialists, particularly urologists; and more blacks go to larger hospitals and undergo more diagnostic tests. It seems likely that these observations are interrelated and possibly due to the fact that many blacks suffer from low socioeconomic status. A racial difference was observed, related particularly to the number of men seeing urologists. This is possibly socioeconomic in origin. Men who travel more than 100 miles for care tend to consult urologists more than physicians in other specialties. Moreover, men who are under care of urologists tend to be diagnosed in earlier stages. To some extent this is expected, since stage I disease must be diagnosed during treatment for another condition; however, why urologists should treat a high percentage of stage II patients is unclear. A partial explanation may lie in those symptoms which lead men to consult a urologist. Finally, the number of tests administered to a patient is related to the degree and specialty of the physician. This differ- ence may be partially related to size of hospital, as well as stage and symptoms at diagnosis. 222 HEALTH SERVICES RESEARCH: DESCRIPTIVE/ANALYTIC Walton, L.A.; Kernodle, W., Jr. Profiles and Perspectives in Patients With Advanced Carcinoma of the Cervix in Eastern and Piedmont North Carolina NCM] 40(12):751-754, 1979 564 Subjects Access and Utilization of Health Services; Screening: Pap smear; Diagnosis; Cancer Site: cervix uteri; Cancer Measures: incidence Study Characteristics Design: retrospective; Measures: descriptive statistics; Time Frame: 1975-1977 Sample Characteristics 170 advanced cervix uteri cancer cases (non-Caucasians 100; Caucasians 70): female; North Carolina Variables SES:income; Demographic: age, race-ethnicity, geographic area; Cancer: symptoms, stage, histology Data Sources Hospital medical records Abstract Invasive carcinoma of the cervix is not declining in the South. A review of patients with advanced stages of this disease was undertaken. The majority were from Eastern North Carolina. They were poor and did not seek medical attention until symptomatic. An opportunity for early diagnosis was missed in some patients exposed to the health care system. Factors influencing the lack of routine Pap screening are discussed. Delivery of medical care to this population needs re-evaluation. Warnecke, R.B. Intervention in Black Populations 565 In: Cancer Among Black Populations (eds) Mettlin, C.; Murphy, G. New York: Alan R. Liss, 1981, p 167-183 Subjects Screening: clinical breast exam, Pap smear, proctoscopic exam, rectal exam, prostate palpation; Diagnosis; Health Care Provider: physician; Access and Utilization of Health Services; Health Care Provider Knowledge, Attitudes, and Practices (KAP) Study Characteristics Design: cross-sectional; Measures: descriptive statistics Sample Characteristics 1,160 individuals: low SES; Illinois Variables Demographic: race-ethnicity (black, white) Data Sources Survey, physician medical records Abstract This paper explores some issues involved in the delivery and active acceptance of those modes of intervention which are not likely to permit detection of asymptomatic or early symptomatic cancers. The experience of mass screening programs and the effects of similar contacts between the patient and the health care deliverer as reported in the literature are briefly reviewed. Some data col- lected as part of a telephone survey which pertain to the current experience of blacks and whites in the lowest socioeconomic stratum where risk is believed to be highest and prognosis to be poorests are examined. Blacks were substantially less likely to have received breast exams and Pap smears than whites. Variations were less obvious for the other examinations perhaps because so few of the respondents of either race received these examinations. The positive relationship between having received a general phys- ical and any of the specific examinations was considerably stronger for whites than for blacks. Unlike whites, regular contact with the provider appeared to have little if any affect on the awareness of specific examinations for blacks. 223 HEALTH SERVICES RESEARCH: EXPERIMENTAL Shapiro, S.; Venet, W.; Strax, P.; et al. Prospects for Eliminating Racial Differences in Breast Cancer Survival Rates Am J Public Health 72(10):1142-1145, 1982 566 Subjects Screening: mammography, clinical breast exam; Cancer Site: breast; Cancer Measures: mortality, survival Study Characteristics Design: controlled intervention study; Measures: adjusted survival rate; Time Frame: 1963-198 Sample Characteristics 62,000 individuals (group offered screening 31,000; controls 31,000— Nonwhite breast cancers 81; white breast cancers 514; noncancers 61,405): female, 40-46 years old, Health Insurance Plan of Greater New York members; New York Variables Demographic: race-ethnicity; Cancer: stage Data Sources Health maintenance association medical records Abstract Results from the randomized trial underway in the Health In- surance Plan of Greater New York to determine the efficacy of periodic screening with mammography and palpation of the breast have been examined to determine the effect of screening on racial differences in breast cancer survival rates. Consistent with experi- ence in general populations, the control group showed a lower five-year survival rate among non-White women with breast cancer than among White women. In the study group, 65 per cent of whom participated in the screening program, there was no diffe- rential in the survival rates of the two racial groups. The elimination of the unfavorable status among non-Whites through screening does not appear to be explained by various artifacts explored. Secondary prevention measures may offer the possibility of reduc- ing or closing the gap in breast cancer survival rates between White and non-White women. 224 HEALTH SERVICES RESEARCH: POLICY/REVIEW Cornely, P.B. Health Status of the Negro Today and in the Future 567 Am ] Public Health 58(4):647-654, 1968 Subjects Access and Utilization of Health Services; Health Care Provider workers. Recommended actions include: 1) a massive, coordi- Knowledge, Attitudes, and Practices (KAP) nated, and comprehensive attack on the nation’s social, economic, Abstract and health problems sponsored by the federal government; 2) The author summarizes what occurred at the first nationwide research into the factors which affect the full utilization of health conference on the health status of the Negro, held on March 13 services, 3) greater involvement of Negro health professionals in and 14, 1967, during the centennial observance of Howard Univer- the correction of factors which promote behavior problems among sity. Three major health problems were identified: 1) the widening Negro youth, and 4) regional conferences financed by the federal gaps between black and white morbidity and mortality rates, 2) government to address the health problems of the Negro. In addi- the need to alleviate conditions leading to mental retardation, and tion, shortages in minority health manpower and health behaviors 3) the need for improved health care for unskilled and semiskilled of both the donors and recipients of health care are addressed. Haynes, A.; Densen, P.M. Implications for Research About the Health of the Negro 568 In: Proceedings of the Centennial Conference on the Health Status of the Negro Today and in the Future. Howard University, Washington, D.C., March 13-14, 1967, p 36-41 Subjects Access and Utilization of Health Services; Cancer Measures: mortality ernmental and voluntary agencies. It considers present and future Abstract trends in the morbidity and mortality of Blacks in the United States. The purpose of this conference was to consider selected health It was concluded that there is ample justification for an accelerated problems in the Black population so that some suggested national program designed to resolve the problems confronting guidelines may be developed which will be of service to gov- the Black population of the United States. Howe, H.L. Increasing Efficiency in Evaluation Research: The Use of Sequential Analysis 569 Am ] Public Health 72(7):690-697, 1982 Subjects Screening: Pap smear; Methodology; Access and Utilization of Health were being reviewed. In the second case, prompt feedback of data Services to a breast screening program was facilitated by the sequential Abstract method allowing program administrators to refocus their efforts This paper describes Wald's sequential analysis and briefly re- on problem areas before the end of the funding period. On the views the history of its applied use. Two public health applications average, Wald's sequential method results in a savings of 50 per are presented as examples of how the method helped overcome cent in observations as compared to classical sampling procedures. two common problems associated with evaluation research. In one The sample sizes will always terminate with a finite number and case, the sequential technique reduced dramatically the workload will rarely exceed three times the average sample size for any in an evaluation project of Pap smear testing where hospital records single sample. Muller, C. A Window on the Past: The Position of the Client in Twentieth Century Public Health Thought and Practice 570 Am ] Public Health 75(5):470-476, 1985 Subjects Health Care Provider: physician; Access and Utilization of Health century. This position is viewed in relation to two elements: 1) Services inheritance versus environment (nature vs nurture) particularly as Abstract related to racial characteristics, and 2) poverty as a cause of ill This paper draws from materials published in the American health and a determinant of personal interaction with health sys- Journal of Public Health supplemented by other selected manu- tems. Each decade is reviewed so that historical changes in public scripts to look at professional thought and practice concerning the health ideas can be identified. position of clients in the public health system during the twentieth 225 HEALTH SERVICES RESEARCH: POLICY/REVIEW Panek, D.; Brown, N.C. Conflicts in Practice J Assoc Pediatr Oncol Nurses 2(1):37-40, 1985 571 Subjects Treatment; Cost-financing; Access and Utilization of Health Services; Cancer Site: leukemia Abstract With this article a new section is introduced in JAPON which depicts moral, ethical, personal or professional dilemmas or con- flicts in practice. Two people or groups are asked to respond with a pro or con stand on the issues. In this, the first situation, a case report of a 10 year old black male with acute lymphocytic leukemia (ALL) is presented. This patient was doing well until three months ago at which time he had an isolated bone marrow relapse. Bone marrow transplantation is the treatment of choice for ALL in first relapse. However, this patient has limited financial resources. This article presents arguments for and against the provision of bone marrow transplantation to this patient. Special emphasis is given to the legal, medical, ethical and financial aspects of the issue. 226 APPENDIX METHODOLOGY ~ So - a - = vr name wean gry ’ ib / ‘ Appendix Methodology The annotated bibliography presented here was designed to be both a com- puterized database of the literature on cancer in black populations for NCI staff use and a reference book for health researchers interested in the topic. To identify the literature, a broad search strategy was designed for computerized bibliog- raphic databases to identify articles published since 1965. In December 1985, the following six databases were searched yielding a total of 17,504 citations with abstracts: MEDLINE 15,591, CANCERLINE 753, HEALTHLINE 433, ERIC 445, PSYCHINFO 273, and Index to Nursing and Health 9. Each article abstract was reviewed for its relevance to cancer and race-ethnicity. Approximately 800 articles were initially selected for inclusion in the bibliography. Copies of the original articles were then obtained from the National Library of Medicine and the Na- tional Institutes of Health Library. Each article was reviewed for specific reference to black populations; articles which dealt with socioeconomic status and cancer but did not directly address blacks were deleted. Five hundred and seventy-one articles were ultimately selected. Before designing the full classification scheme, a decision was made to classify all articles in one and only one category in the classification structure with no articles in an ambiguous or “left over” position." After reviewing the literature, three broad categories of health research were identified: epidemiology; health services research; and health knowledge, attitudes, and practices (KAP). These categories or topic areas were used as the basic categories within which all articles were classified. The rest of the classification scheme developed over a period of many months through trial and error. The literature classification scheme was initially designed around the following eight basic article characteristics: subjects, type of article: policy/review or original research, geographic areas of study or interest, study characteristics, sample characteristics, study variables, data sources, and the kinds of health care providers giving the health services under study. PND The first three characteristics are generally applicable to all articles, while the remaining five relate only to original research articles where the results of primary or secondary data analysis are presented. Within the “study characteristics” category four aspects of research studies are examined: 1) study design, 2) study measures, 3) years of data collection, and 4) whether the research qualifies as a cancer intervention study. Likewise, within the “sample characteristics” category 229 four aspects of research samples or populations are examined: 1) the defining overall characteristics of the sample/population, 2) the name and size of each racial/ethnic group studied or contained within the sample/population, 3) the name and size of the major groups of comparison, and 4) the unit of analysis. Because the literature contained such a wide variety of articles the initial class- ification scheme needed to be flexible enough to capture important article sub- groupings. As the scheme developed it became apparent that there were impor- tant subgroupings of articles which were unique to each of the three categories. Therefore, to simplify the classification process and to keep the classification scheme adaptable, slightly different summary forms were used for articles as- signed to each topic area. The forms had spaces to insert “other” responses which were judged important by the coder but were not among the given re- sponse categories. Exhibit 1 presents the three article summary forms. Throughout the classification process, a set of classification instructions with category definitions was updated and revised (Exhibit 2). A list of standard terms was also developed and regularly updated (Exhibit 3). This list was eventually used for the database code book. After all articles were classified, a single-page coding sheet was used to prepare the material for data entry (Exhibit 4). This sheet grouped the different items into a standard format. An edit sheet outlining how to fill-in the coding sheet is presented in Exhibit 5. Rather than write original abstracts for each article, the decision was made to use existing, published article abstracts and summaries where available. Nearly 520 articles had some kind of abstract or summary which could be directly used. For those articles which had none, a brief abstract was written describing the article’s purpose and major findings or conclusions. 1. Bailar, J.C. III; Louis T.A.; Lavori, P.W.; et al. Statistics in Practice: A Classification for Biomedical Research Reports. N Engl ] Med 311(23):1482-7, 1984 230 Coder EXHIBIT 1A ANNOTATED BIBLIOGRAPHY OF CANCER-RELATED LITERATURE ON BLACK POPULATIONS ARTICLE SUMMARY ARTICLE NUMBER: AUTHOR : TITLE: SOURCE : PAGE NUMBER(S): PUBLICATION DATE: TOPIC AREA: Epidemiology GENERAL SUBJECTS (check all that apply): Screening general mammography clinical breast exam breast self exam (BSE) pap smear Diagnosis Treatment Prevention Risk Factor Exposure occult blood testing proctoscopic exam rectal exam prostate palpation other (specify) diet-nutritional factors alcohol occupation tobacco other (specify) 11] Cancer Site all sites combined multiple sites breast cervix colon oral rectum larynx lung pharynx prostate pancreas esophagus other (specify) 231 Article: Cancer Measures incidence prevalence mortality survival Biologic Markers (epidemiology only) ABO blood type estro & progesterone receptors breast fluid G6PD CEA gamma-feo protein cerumen type HLA other Methodology Other (Specify) 9. TYPE OF ARTICLE/STUDY: policy/review descriptive/analytic epidemiology experimental 10. INTERVENTION STUDY: yes no 11. STUDY DESIGN (check all that apply): case study retrospective (descriptive studies) cross-sectional case-control (retrospective time trend analytic studies) prospective controlled intervention study (CIS) (descriptive studies) clinical trial nonconcurrent other (soecify) prospective cohort (prospective analytic studies) 12. PROVIDER OF SERVICE: public health physician department nurse private voluntary other (specify) organization (PVO) media 232 Article: 13. MEASUREMENT INDICES (check all that apply): Cancer Incidence Cancer Survival crude incidence rate crude survival rate age-specific incidence rate relative survival rate age-adjusted incidence rate Cancer Mortalit prevalance rate crude mortality rate Relative Risk age-specific mortality rate 0dds Ratio (relative odds) age-adjusted mortality rate Standardized Mortality Ratio (SMR) descriptive statistics higher level statistics 14. CHARACTERISTICS OF POPULATION/SAMPLE STUDIED (specify): income age racial/ethnic group sex geographic location education other 15. STUDY VARIABLES (check all that apply): Socioeconomic and Demographic Indicators age income sex racial/ethnic group (specify) education industry/occupation geographic location other (specify) Cancer Characteristics histologic type of cancer stage of cancer at diagnosis other (specify) 16. YEAR(s) OF DATA COLLECTION (specify): 17. POPULATION/SAMPLE SIZE (specify): 18. SIZE OF EACH RACIAL/ETHNIC GROUP STUDIED: (specify): 233 19. Article: Cope OF SONTROL AND EXPERIMENTAL GROUPS OR MAJOR GROUPS OF COMPARISON specify): 20. SOURCE OF DATA (make one check for each data source used in the study): 21. Death Registration ationa State Local Cancer Registration First National Cancer Survey, 1937-1939 Second National Cancer Survey, 1947-1948 Third National Cancer Survey, 1969-1971 End Results Program SEER non-SEER state or county (pop. based) non-SEER hospital Patholo Records autopsy histopathology Medical Records hospita health maintenance organization (HMO) other U.S. Population Census Cancer Information Service (CIS) Survey National Center for Health Statistics (NCHS) Other Other STUDY ABSTRACT: 234 3. 4. Coder EXHIBIT 1B | ANNOTATED BIBLIOGRAPHY OF CANCER-RELATED LITERATURE ON BLACK POPULATIONS ARTICLE Bony ARTICLE NUMBER: AUTHOR : TITLE: SOURCE : PAGE NUMBER(S): PUBLICATION DATE: TOPIC AREA: Knowledge, Attitudes, and Practices (KAP) | GENERAL SUBJECTS (check all that apply): Screening | general occult blood testing mammography proctoscopic exam clinical breast exam rectal exam breast self exam (BSE) prostate palpation pap smear other (specify) Diagnosis Treatment Prevention Risk Factor Exposure diet/nutritional factors alcohol occupation tobacco other (specify) | Cancer Site multiple sites rectum oral breast lung larynx cervix prostate | pharynx colon esophagus pancreas | other (specify) 235 9. 10. 11. 12. 13. 14. 15. Article Number: Communications Education Compliance Other (specify) TYPE OF ARTICLE/STUDY: policy/review descriptive experimental INTERVENTION STUDY: yes no STUDY DESIGN: case study cross-sectional panel study controlled intervention study (CIS) other (specify) 111 PROVIDER OF SERVICE: public health physician department nurse private voluntary other (specify) organization (PVO) media MEASUREMENT INDICES (check all that apply): simple descriptive statistics higher level statistics CHARACTERISTICS OF POPULATION/SAMPLE STUDIED (check and specify all that apply): general U.S. population income age racial/ethnic group sex geographic location education other STUDY VARIABLES (check all that apply): Socioeconomic and Demographic Indicators age income sex racial/ethnic groups (specify) education industry/occupation geographic location other (specify) 236 16. 17. 18. 19. 20. 21. YEAR(s) OF DATA COLLECTION (specify): POPULATION/SAMPLE SIZE (specify): SIZE OF EACH RACIAL/ETHNIC GROUP STUDIED (specify): Article Number: SIZE OF CONTROL AND EXPERIMENTAL GROUPS OR MAJOR GROUPS OF COMPARISON (specify): SOURCE OF DATA (make one check for each data source used in the study): Death Registration National (NCHS) State Local Cancer Registration First National Cancer Survey, 1937-1939 Second National Cancer Survey, 1947-1948 Third National Cancer Survey, 1969-1971 End Results Program SEER non-SEER hospital Pathology Records autopsy histopathology Medical Records ospita health maintenance organization (HMO) other non-SEER state or county (pop. based) U.S. Population Census Cancer Information Service (CIS) Survey National Center for Health Statistics (NCHS) Other Other STUDY ABSTRACT: 237 3. 4. Coder EXHIBIT 1C ANNOTATED BIBLIOGRAPHY OF CANCER-RELATED LITERATURE ON BLACK POPULATIONS ARTICLE SUMMARY ARTICLE NUMBER: AUTHOR : TITLE: SOURCE : PAGE NUMBER(S): PUBLICATION DATE: TOPIC AREA: Health Services GENERAL SUBJECTS (check all that apply): Screening general occult blood testing mammography proctoscopic exam rectal exam prostate palpation other (specify) clinical breast exam breast self exam (BSE) pap smear Diagnosis Treatment Prevention Risk Factor Exposure diet/nutritional factors alcohol occupation tobacco other (specify) Cancer Site multiple sites rectum oral breast Tung larynx cervix prostate pharynx colon esophagus pancreas other (specify) 239 9. 10. 11. 12. 13. Article Number: Health Care Provider hospital physician nurse other (specify) Cost/Financing Access and Utilization of Health Services Quality of Care Health Care Provider Education Health Care Provider Knowledge, Attitudes, and Behavior (KAP) Public/Patient Education Programs Other TYPE OF ARTICLE/STUDY: policy/review descriptive experimental INTERVENTION STUDY: yes no STUDY DESIGN: case study cross-sectional panel study controlled intervention study (CIS) other (specify) PROVIDER OF SERVICE: public health physician department nurse private voluntary other (specify) organization (PVO) media MEASUREMENT INDICES (check all that apply): Health Services Measures average length of stay (ALOS) hospital admissions rate disability days utilization of health services other simple descriptive statistics higher level statistics 240 14. 15. 16. 17. 18. 19. 20. Article Number: CHARACTERISTICS OF POPULATION/SAMPLE STUDIED (check and specify all that apply): general U.S. population ___ income age racial/ethnic group sex ___ geographic location ____ education ___ other STUDY VARIABLES (check all that apply): Socioeconomic and Demographic Indicators age income sex racial/ethnic groups (specify) education industry/occupation geographic location other (specify) YEAR(s) OF DATA COLLECTION (specify): POPULATION/ SAMPLE SIZE (specify): SIZE OF EACH RACIAL/ETHNIC GROUP STUDIED (specify): SIZE OF CONTROL AND EXPERIMENTAL GROUPS OR MAJOR GROUPS OF COMPARISON (specify): SOURCE OF DATA (make one check for each data source used in the study): Death Registration =e on aT (NCHS) State Local 241 Article number: Cancer Registration First National Cancer Survey, 1937-1939 Second National Cancer Survey, 1947-1948 Third National Cancer Survey, 1969-1971 End Results Program SEER non-SEER state or county (pop. based) non-SEER hospital Pathology Records autopsy histopathology Medical Records hospital health maintenance organization (HMO) other U.S. Population Census Cancer Information Service (CIS) Survey National Center for Health Statistics (NCHS) ____ Other Other 21. STUDY ABSTRACT: 242 EXHIBIT 2 ANNOTATED BIBLIOGRAPHY OF CANCER-RELATED LITERATURE ON BLACK POPULATIONS General Instructions 1. DO NOT SKIP ANY ITEM NUMBERS. Write N/A next to an item number which: a. does not apply to the article or b. where the information is not specifically supplied within the article. 2. If there are questions about an item, paper clip the article to the back of the form and put a note explaining the question on a yellow "stickum". 3. Put your name in the upper right hand corner. Classification Definitions Item #7 Topic Area: * 0 ** * kk 0 Epidemiology - Cancer epidemiology deals with the distribution and determinants of cancer. Epidemiology articles focus on who gets cancer and why. Epidemiology deals with outcome measures which relate to cancer rates. KAP - These articles deal with the health knowledge, attitudes, and practices of black populations especially as they relate to cancer. The study group will generally be "well" individuals. The focus of these studies is on the population as opposed to clinicians. Health Services - Health services research is concerned with the quality and delivery of health services. Health services articles will focus on the characteristics of health services and the providers and provider institutions. These articles generally use process measures as proxy measures for outcomes. Item #8 General Subject (check all that apply): Most articles will cover more than one subject, therefore, check all descriptors that apply. 0 Risk factor exposure/behavior - Note the kind of cancer risk factor examined in the article. Examples of the kinds of risk factors that could be coded as "other" are: sexual practices, diabetes, hypertension, and hormones. Factors which are more indirectly associated with cancer such as socioeconomic and geographic variables are not to be coded here. ** For KAP articles, this subject will include surveys and studies related to diet, tobacco, and alcohol consumption patterns of black populations, as well as occupational patterns. 243 o Cancer Site: - Check the cancer sites of interest to the article. If more than five, check "multiple sites". When data for cancer as a whole are presented check, "all sites combined". ** An example of this subject for KAP articles would be an article about an educational program or media campaign concerned with cancer--the cancer site of interest would be checked here. * o Cancer Measures - Cancer epidemiology articles are frequently concerned *kk *kk *kk *kk * kk * kk with one or more of the following aspects of cancer (measures of cancer). These may or may not be expressed as rates. - cancer incidence - The frequency with which cancer occurs in a population. This is generally expressed as a rate--the number of cancer cases newly diagnosed during a calendar year per 100,000 population - cancer prevalence - The number of cancer cases present in the population. - cancer mortality - Deaths from cancer. This is often expressed as a rate--the number of cancer deaths occuring during a calendar year per 100,000. - cancer survival - The period of survival for a cancer patient. Biologic Marker--A "marker" measured on a biologic specimen, i.e. blood, cells, cerumen, that may be associated in some way with cancer. The association of cancer to biologic markers may include prognosis after therapy, heightened cancer risk, or early diagnosis. Health Care Provider--Note the types of health care providers of interest to the article or study. Cost/Financing - This involves the economic costs associated with the delivery of health services and the methods used to finance these services (health insurance coverage, Medicare, Medicaid, etc.) Access and Utilization of Health Services - This deals with the social and geographic availability of health services to individuals or groups and the patterns of their use. Quality of Care - This involves examinimg the performance of physicians, nurses, and allied health personnel, as well as hospitals and other health organizations. It includes "patterns of care" studies which look at the management practices of physicians and health care facilities. Provider Education - Articles which deal with the education of physicians, nurses and other health care providers. This includes professional and in-service training. Provider Knowledge, Attitudes, and Behavior (KAP) - Articles which deal with various aspects of health providers' KAP as they relate to the health care of black populations. 244 *** o Patient Education Programs - Articles dealing with education programs which are implemented by a health facility or provider and are directed toward patients, their families and friends, or in some cases, the general public. ** 0 Compliance - Articles which are concerned with patients' behavior following instructions by health care providers. ** 0 Communication - This subject, in general, deals with "non-provider" oriented communication strategies related to cancer, for example, articles about NCI's Cancer Information Service (CIS). ** 0 Education - Articles which describe cancer education activities directed toward the public - not providers. This can include school health programs, community health fairs, etc. Item #9 Type of Article/Study: The "type of article/study" defines the overall kind of information being presented in the article or the type of study being described. 0 Policy/review - This category includes articles which describe the author's views on a subject without involving primary or secondary data collection or analysis. Articles which review other studies are included here. Items #10 through #19 do not apply to articles classified on policy/review; therefore, write N/A beside each of these items. 0 Descriptive - Articles which present descriptive information or in- formation from observational studies such as surveys. These articles simply describe what is being observed. This category includes analytic epidemiology studies. o Experimental - Articles which describe studies where the investigator has manipulated in some way a study variable in order to affect the study outcome. Item #10 Intervention: A cancer control intervention is an action taken to affect a positive change in cancer rates. Cancer intervention studies are concerned with the process of developing (Phases I and II) and evaluating and applying (Phases III - V) cancer control interventions. In general, articles describing cancer control Phases III-V intervention studies will be experimental studies where the researcher manipulates some variable to influence an outcome. Certain quasi- experimental and epidemiologic cohort or cross-sectional studies may be classified as intervention research (Phase III) if they attempt to assess the impact of a cancer intervention. Studies classified as Phases III-V intervention research will always use a comparison group of some sort to show an influence. Intervention studies are frequently unable to demonstrate changes in cancer rates because of latent periods (time between exposure and cancer outcome), therefore, they often use proxy measures of success such as screening and smoking behavior, health services use, and care seeking practices and knowledge. 245 Item #11 Study Design: In general, the type of study design used for collecting and analyzing the data will be mentioned in the article's methodology section. Identify the type of study design described in the article; fill in the type of study if needed. * o Case Study - Information on one to several persons. For example, 0 1) description of patient(s)' natural history of disease, 2) clinical description of patient(s). Cross-Sectional - These studies focus on describing the state of a phenomenon during a fixed period of time; they provide a "picture" of what the phenomenon looks like. Change is not the object of these studies. Cross-sectional studies differ from case studies in that they involve a larger group of cases. Cross-sectional studies can be used to define prevelence. An example of a cross-sectional study is a descriptive study of the demographic characteristics of breast cancer patients diagnosed from 1980-1982. Time Trend - This is where data which has been collected cross-sectionally from 2 or more time periods (not following the same people) is observed over a period of time. For example, a descriptive study of the pattern of change in the annual incidence rate of esophageal cancer in U.S. blacks during the last 20 years. Prospective (descriptive studies) - In prospective studies, sampling is done on the basis of input variables (the presumed determinants of an outcome). Prospective studies focus on change -- selected subjects are followed forward in time. Unlike cohort studies, no comparison or control groups are used in descriptive prospective studies. For example: 1) study subjects selected for their exposures to a carcinogen are followed to see what kind of cancer develops and 2) cancer patients are selected and followed to see the progression of their disease and their period of survival. A major problem with prospective studies is follow-up, therefore, many articles with prospective designs will use words like "lost to follow-up" and "patient tracing". Nonconcurrent Prospective - As with prospective studies, these studies sample on input variables and follow subjects forward in time. However, the samples are drawn from the past and traced forward. For purposes of this bibliography, nonconcurrent prospective studies include studies of cancer cases selected from hospital records and followed forward in time to examine survival patterns. Cohort (analytic studies) - As with descriptive prospective designs, cohort studies follow individuals forward in time. A defined population is selected and followed. For example, individuals working in an industrial plant from 1950-60 form a cohort. This cohort is then disagregated by worksite. The entire cohort is followed forward to determine which workers develop cancer. This enables relative risk to be calculated. 246 * 0 Retrospective (descriptive studies) - In retrospective studies, sampling is done on the basis of output variables (the presumed effect of inputs). Retrospective studies focus on change -- cases are selected and tracked backwards in time in order to observe prior exposures. An example of a retrospective study is a study where the researcher selects breast cancer patients and reviews their records to see how the cancer was initially detected -- through BSE, clinical exam or mammography. Unlike case-control studies, no comparison or control groups are used. * 0 Case-control/retrospective (analytic studies) - As with retrospective designs, the focus in case-control studies is on change -- cases are selected and tracked backwards in time. However, in these studies, a comparison group of cases are selected who do not have the same output variabTe which was used to select the cases. Both groups are tracked backwards and their prior exposures to certain variables are compared. For example, researchers select persons with and without cancer and then determine what their past history of exposure was to a possible carcinogen. This is an analytic study because the researchers have no control over the exposure. o Panel-Study - Information from the same people at various time intervals. o Controlled Intervention Study (CIS) - These studies are Phase III cancer control studies where a study group is compared with a group that does not receive an intervention, or where different interventions are compared with one another or with a control group. Controlled Intervention Studies are experimental studies. * 0 Clinical trial - In clinical trials the efficacy of a therapeutic agent or procedure is tested in individual subjects. This is an experi- mental method where individuals are randomly allocated to two groups (experimental and control) and the experimental group is exposed to the agent or procedure. For purposes of this review, the "clinical trial" description will be narrowly applied to articles which describe therapeutic agents or procedures which have been given in an attempt to relieve symptoms and/or improve the survivorship of those with the disease. It will not include "intervention" or "preventive" clinical trials as defined by Lilienfeld--these kinds of clinical trials will instead be coded as controlled intervention studies (CIS). Item #12 Provider of Service: Identify the agent providing the population/sample group with the service or intervention being studied. This will usually apply to articles describing experimental studies. 247 Item #13 Measurement Indices (check all that apply): Identify the kinds of statistics presented in the article. * 0 Crude cancer incidence rate - Calculated by taking the number of cancer cases newly diagnosed during a calendar year, dividing it by the mid-year population, and multiplying by 100,000 (these rates are not adjusted for age or made specific for a specified age group). * 0 Age-specific cancer incidence rate - Calculated by taking the number of cancer cases within a stated age group newly diagnosed during a calendar year, dividing it by the mid-year population of that same age group, and multiplying by 100,000. * 0 Age-adjusted cancer incidence rate - Incidence rates are usually age- adjusted when 2 or more populations are being compared. The adjustment (or standardization) for age serves the purpose of statistically eliminating differences in the age compositions of the populations studied which could be an "extraneous" source of variation. Articles should state either in the text or in the tables and figures that the rates have been age-adjusted or age-standardized. * 0 Crude cancer mortality rate - Calculated by taking the number of deaths from cancer that occur during a calendar year, dividing by the mid-year population, and mulitplying by 100,000 (these rates are not adjusted for age or made specific for a specified age group). * 0 Age-specific cancer mortality rate - Calculated by taking the number of deaths from cancer within a specific age grou that occur during a calendar year, dividing by the mid-year population of that same age group, and multiplying by 100,000. * 0 Age-adjusted cancer mortality rate - (see definition of age-adjusted incidence rate). * 0 Crude cancer survival rate - The proportion of newly diagnosed cancer patients surviving for a specified period of time after diagnosis (usually 5 years). * 0 Relative cancer survival rate - The ratio of the observed survival rate for the patient population to the expected survival rate for persons in the general population who are similar to the patient group with respect to age, sex, race, and calendar year of observation. * 0 Prevalence rate - The number of cancer cases present in the population at a specified time divided by the number of persons in the population at that specified time. * 0 Relative Risk/0Odds Ratio (relative odds) - A measure of association frequently used in epidemiology studies where the incidence rate of a disease in an exposed group is divided by the incidence rate of the disease in a nonexposed group. It can be directly calculated for cohort/prospective studies and estimated for case-control/ retrospective studies. 248 * o Standardized Mortality Ratio (SMR) - The ratio of the number of deaths expected for a specified group divided by the number of deaths observed for the total population. The SMR is age-adjusted and expressed as a percentage. o Simple descriptive statistics - These measures describe what is observed. They include percentage distributions, measures of central tendency (mean, median, and mode), and measures of dispersion (range, variance, and standard deviation). Simple retrospective rates are also included. Not included here are the more complex measures - of multivariate analysis and measures of significance. 0 Higher level statistics - These measures include tests of association (measures which signify how strongly two or more variables are related to each other--they include chi square and correlation coefficients) and tests of significance (tests which help assess whether or not observed differences between 2 or more study groups are real or due to random variation--results from tests of significance are generally reported as a P value). Item #14 Characteristics of Population/Sample Studied (check and specify all that apply): Fill in, where possible, the overall characteristics of the group or sample described or studied in the article. For example, if the sample contains women over age 50 from California with family incomes of less than $10,000 a year, 3 indicators are checked and filled in: 1) age, >50; 2) geographic area, Calif.; and 3) income, <$10,000. The criteria used to select the study sample should be identified. If the sample contains cancer patients or deceased cancer patients, it should be noted. Item #15 Study Variables (check all that apply): Identify the socioeconomic and demographic variables which are measured in the study and presented in the article. If "racial/ethnic group" is used as a variable, specify the specific groups identified using the same terminology, i.e. nonwhite, anglo. For purposes of this bibliography, "demographic variables" includes common menstrual and reproductive variables (i.e. age at menopause and parity). * For expidemiology studies, indicate the cancer characteristics used as variables. Item #16 Year(s) of Data Collection (specify): Specify the year or years that the data were collected. Item #17 Population/Sample Size (specify): Fill in the total size of the population or sample group studied. 249 Item #18 Size of Racial/Ethnic Groups Studied (specify): Label and state the size of each racial/ethnic group identified in the study. Item #19 Size of Control and Experimental Groups or Major Groups of Comparison (specify): This item generally applies to experimental and quasi-experimental studies. These studies will most likely have one of the following study designs identified in Item #11: CIS, cohort/prospective, case-control/retrospective, or clinical. Label and state the size of each experimental and control group or major group of comparison used in the study. If the comparison groups are the same as those ‘listed in #18, don't code & put "same as #19". Item #20 Source of Data (make one check for each data source used in the study): Identify the major sources of data used in the study. Each data source should be given only one check. For example, if both census data and hospital records are used to create a data base for a study of SES and cancer, two sources of data would be checked, U.S. Population Census and hospital medical records. Item #21 Study Abstract: Wherever possible, use the abstract presented at the beginning of the article or from the literature search print-out. To save time and reduce errors, xerox the article's abstract and tape it to the form in the space allotted. If the abstract does not specifically mention blacks, use the existing abstract but add a couple of sentences about the article's major findings or ideas concerning blacks. If the article does not have an abstract, write a short paragraph which includes the purpose and results of the article or study. 250 EXHIBIT 3 ANNOTATED BIBLIOGRAPHY OF CANCER-RELATED LITERATURE ON BLACK POPULATIONS LIST OF STANDARD TERMS 1. General Subjects a. Screening: bladder cytology breast breast self exam (BSE) clinical breast exam mammography mouth exam multiple cancer screening tests Pap smear (cervical smear) proctoscopic exam prostate palpation rectal exam b. Diagnosis c. Treatment d. Prevention e. Risk Factor Exposure: air pollution alcohol bacterial flora bowel characteristics breastfeeding treatment modalities (chemotherapeutic agents, radiation, etc.) chemical (talc, asbestos, DMBA, etc.) circumcision contraceptive devices diet-nutritional factors drugs endogenous hormones exogenous hormones familial factors fluoridation geographic area occupation industry other diseases (includes other pathological conditions, i.e. polyps) reproductive factors sexual practices (sexual activity, includes age at 1st coitus and number of partners) 251 solar radiation stress tobacco treatment modalities ultraviolet light (i.e. light bulbs and heat lamps) water pollution f. Cancer Site: anus malignant melanoma all sites combined multiple myeloma biliary tract (gall bladder) multiple sites bladder muscle brain nasopharyngeal bone neuroblastoma breast oral cavity (includes, mouth, Burkitt's tumor (Burkitt's tongue, buccal mucosa, gums, lymphoma) lip, and gingiva) central nervous system ovary cervix uteri pancreas colon penis colon-rectum pharynx corpus uteri (includes rectum endometrium and myometrium) respiratory system digestive system salivary glands esophagus scrotum eye skin (basal cell carcinoma, gestational trophoblastic squamous cell carcinoma, disease (choriocarcinoma, Bowen's disease) hydatidiform mole) sarcoma (soft tissue, Kaposi's head and neck sarcoma, bone or osteogenic jaw sarcoma) larynx small intestine kidney spinal cord leukemia (blood, hematopoietic) stomach liver (hepatocellular) testis lung & bronchus thyroid lymphoma (Hodgkin's and vagina non-Hodgkin's disease) vulva g. Cancer Measures (Epidemiology Studies Only): incidence mortality survival prevalence 252 h. Biologic Markers: ABD blood group ABO blood group alpha-fetoprotein anti-HTLV antibodies beta 2-microglobulin breast fluid carcinoembryonic antigen catecholamines cerumen type Duffy gene Epstein-Barr virus FAB antibodies gamma-glutamyl transferase G6PD HLA hepatitis B antigen hepatitis B surface antigens (HBsAg) hepatitis B virus associated antigens herpesvirus type 2 antibodies human T-cell leukemia/lymphoma virus (HTLV) HTLV-1 immunoglobulin insulin antibody karyotype monoclonal antibodies prostatic acid phosphatase (PAP) prostatic creatine kinase-BB serum ferritin serum lipoprotein concentrations steroid receptors tyrosine hydroxylase unsaturated B12 binding capacity Methodology Socioeconomic Factors Communications (KAP Studies Only) Psychosocial Factors (KAP Studies Only) o Dying/Grief 0 Coping mechanisms Education (KAP Studies Only) Psychosocial factors Compliance (KAP Studies Only) Health Care Provider (Health Services studies only): clinic health maintenance organization (HMO) hospital nurse physician 253 p. Cost Financing (Health Services study only) q. Access and Utilization of Health Services (Health Services study only) r. Quality of Care (Health Services study only) s. Health Care Provider Education (Health Services study only) t. Health Care Provider Knowledge, Attitudes and Practices (KAP) (Health Services study only) u. Public-Patient Education Program (Health Services study only) Study Characteristics a. Design: case study cross-sectional controlled intervention study (CIS) (Epidemiology Only): time trend prospective prospective (nonconcurrent) cohort retrospective case-control clinical trial (KAP and Health Services): panel study b. Intervention Study Cc. Measures: descriptive statistics higher level statistics (Epidemiology Only): Incidence crude incidence rate age-specific incidence rate age-adjusted incidence rate proportional incidence ratio (PIR) Mortality crude mortality rate age-specific mortality rate age-adjusted mortality rate 254 Survival crude survival rate age-adjusted survival rate relative survival rate prevalence rate relative risk odds ratio (relative odds) standardized mortality ratio (SMR) (Health Services Only): average length of stay (ALOS) bed days hospital admissions rate disability days utilization of health services d. Time Frame: (Example) 1945-1962 3. Sample Characteristics - Follow instructions provided in the “Coding Sheet Guidelines" a. Units of analysis (following the sample total): autopsy cases blood samples breast leisons breast tumor biopsies cancer cases cancer deaths cell lines counties deaths households hospital cancer admissions hospital patients household heads individuals necropsy cases neighborhoods prostate specimens surgical specimens telephone calls b. Major Groups of Comparison (use these labels): cases controls experimental group 255 c. racial-ethnic groups: American Indians Anglos blacks Caucasians coloreds Filipino Hispanics Hispano Latin Americans Latins Negroes nonwhites Orientals other whites others whites yellows unknown d. sex - use male or female; if both are used in study leave this blank e. age group - present the age or age range of the sample in the following manner: 45 - 60 years old 45 + years old under 61 years old f. education: g. income: income Tow income h. other: inner-city railroad workers mastectomy i. geographic area: For U.S. studies, use city name and U.S. Post Office abbreviation for state. 256 4. a. b. Variables SES: ability to pay age first wore shoes census census census census census census census census census combined SES county income block tract tract tract tract tract tract tract tract education/income education emp loyment income rental value number of school years completed by adult population SES rank single unit housing ZIP code education rank county number of school years completed by adult population education employment status income industry industry-occupation medical insurance status occupation parent's education/occupation public-private patient status public-private hospital patients quality of house reads English religion residency social class socioeconomic background socioeconomic status westernization Demographic: age age at first birth age at marriage age at menarche age at menopause age at first pregnancy birthdate birthplace county percent urban county population density geographic area home territory hysterectomy marital status menstrural status menopausal status number of dependents number of pregnancies number of marriages 257 parity population size pregnancy status race-ethnity sex urban-rural c. Cancer: age at presentation of first primary tumor biologic characteristics clinical diagnosis clinical presentation histology hydrocortisone production immunophenotype metastases nodal involvement number of primary cancers organ involvement primary site stage (extent of disease) surface phenotype symptoms tumor grade tumor location tumor size Data Sources: birth registration death registration First National Cancer Survey 1937-1939 Second National Cancer Survey 1947-1948 Third National Cancer Survey 1969-1971 End Results Program SEER cancer registry non-SEER population-based cancer registry non-SEER hospital cancer registry non-SEER cancer registry autopsy records histopathology records industry records laboratory test family planning clinic medical records health clinic medical records health maintenance organization medical records hospital clinic records hospital medical records medical records physician medical records 258 country census U.S. Population Census National Center for Health Statistics survey Cancer Information Service survey (includes information collected through interviews and questionnaires) Provider: university cancer center Notes: For indexing purposes, list the names of U.S. states and territories and countries which are of major interest to the article or from which data are obtained. 259 pov res rem EXHIBIT 4 ANNOTATED BIBLIOGRAPHY OF CANCER-RELATED LITERATURE ON BLACK POPULATIONS CODING SHEET* ARTICLE NO. CATEGORY (7) (9) SUBJECTS (8) Screening Risk Factor Exposure Cancer Site Cancer Measures STUDY CHARACTERISTICS Design (11) Intervention (10) Measures (13) Time Frame (16) SAMPLE CHARACTERISTICS (No of cases 17) (14) (cancer site or unit of analysis) (racial gps 18) (cntrl/exp gps 19) (characteristics 14) (race) (sex) years old, (age) (ed.) (income) (other) (geo. Tocation) VARIABLES (15) SES Demographic Cancer DATA SOURCES (20) PROVIDER (12) Note: STATE OR COUNTRY (14) * Numbers in parentheses refer to the Article Summary items to be coded. 261 EXHIBIT 5 ANNOTATED BIBLIOGRAPHY OF CANCER-RELATED LITERATURE ON BLACK POPULATIONS CODING SHEET EDIT PROCEDURES FOR DATABASE ENTRY Use (;) semi-colons between major italicized elements within the following catagories: Subjects, Study Characteristics, and Variables. For example, - Cancer Site: lung; Treatment; Prevention - SES: income; Demographic: age, sex Use (,) commas between Data Source elements. For example, - autopsy records, survey Use (,) commas between minor elements in all 5 catagories. For example, - Screening: pap smear, rectal exam; Diagnosis For Sample Characteristics, insert the unit of analysis after the total sample number. For example, - 25 hospital patients This should be followed by the racial-ethnic breakdown of the sample in parenthesis and, where applicable, the sizes of the major groups of comparison used in the study. The formats to use are as follow: - 50,000 breast cancer cases (blacks 25,000; whites 25,000) - 50,000 individuals (blacks 800; whites 200 -- Cases 500; controls 500) For groups of comparison which are time periods, use the term "period 1." and "period 2": etc. For example, (Period 1: 50,000; period 2: 20,000) When the sample is made up of just one racial/ethnic group use the singular form. For example, - 1700 cancer cases, black, female For articles which use 2 or more different sets of data (i.e. 1 set for mortality analysis and 1 set for incidence analysis) use the following format where appropriate: Study Characteristics - Design: cross-sectional, time trend; - Time Frame: 1965-1980 Sample Characteristics 1,460 cervical cancer deaths (blacks 1,000; whites 460) and 60 cervical cancer cases (blacks 40; whites 20) Buffalo, NY 263 AREA SUBJECT INDEX Numbers following subject refer to article numbers. ACCESS AND UTILIZATION OF HEALTH SERVICES 416, 559, 560, 561, 562, 564, 565, 567, 568, 569, 570, 571 ALCOHOL (see Risk Factor Exposure) ANUS 92 BILIARY TRACT 193, 443 BIOLOGIC MARKER ABD blood group 325 ABO blood group 68, 209, 388, 423 alpha-fetoprotein 4, 151, 279, 286, 293, 312, 381, 394 anti-HTLV antibodies 276 beta 2-microglobulin 50 breast fluid 240 carinoembryonic antigen 180, 277 catecholamines 142 cerumen type 240 Duffy gene 239 Epstein-Barr virus 84 FAB antibodies 324 gamma-glutamyl transferase 151, 153 G6PD 232, 290 HLA 115, 116, 117, 166, 383, 423 hepatitis B antigen 312 hepatitis B surface antigens (HBsAg) 45, 152, 237, 293, 381, 394 hepatitis B virus associated antigens 45, 151, 474 herpesvirus type 2 antibodies 252, 349, 361 human T-cell leukemia/lymphoma virus (HTLV) 21 HTLV-1 271 immunoglobulin 325 insulin antibody 146 karyotype 16, 17 monoclonal antibodies 84 prostatic acid phosphatase (PAP) 359 prostatic creatine kinase-BB 359 serum ferritin 46 serum lipoprotein concentrations 198 steroid receptors 138, 200, 211, 234, 277, 460 tyrosine hydroxylase 142 unsaturated B12 binding capacity 320 BLADDER 23, 275, 358, 366, 407, 427 BONE 129, 244 BRAIN 81, 190, 248, 411, 444, 458 BREAST access and utilization of health services 562 biologic marker: ABD blood group 325 breast fluid 240 carcinoembryonic antigen 277 cerumen 240 herpesvirus type 2 antibodies 349 immunoglobulin 325 steroid receptors 138, 200, 211, 234, 277, 460 communications 515, 525, 538, 552 diagnosis 55, 85,98, 109, 124, 136, 220, 225, 234, 267, 277, 296, 338, 460 education 525, 538, 551 incidence 26, 55, 70, 98, 105, 109, 112, 121, 136, 179, 200, 225, 243, 246, 267, 351, 430, 436, 460, 462 methodology 169, 436 mortality 105, 109, 460, 566 prevalence 70, 220, 235 risk factor exposure: alcohol 150 breastfeeding 351, 357, 462, 481 diet-nutritional factors 133, 246, 327, 357, 452, 460, 462, 469 endogenous hormones 56, 133, 138, 149, 179, 211, 235, 246, 452 exogenous hormones 138, 168, 225, 462, 494 familial factors 26, 124, 211, 225, 246, 357, 462 geographic area 436, 452 other diseases 56, 349, 462 reproductive factors 105, 138, 179, 200, 211, 219, 225, 240, 277, 327, 338, 349, 351, 357, 425, 452, 462 sexual practices 349 tobacco 150, 538 screening: breast 357 breast self exam (BSE) - 85, 213, 521, 525, 533, 534, 538, 557, 562 clinical breast exam 213, 533, 562, 566 mammography 213, 525, 566 Pap smear 525, 562 socioeconomic factors 62, 70, 150, 169, 243 survival 55, 62, 109, 127, 150, 169, 200, 211, 219, 220, 322, 327, 338, 340, 566 treatment 47, 55, 109, 127, 200, 219, 220, 322, 327, 340, 425 BREAST SELF EXAM (see Screening) BURKITT'S TUMOR 50, 63, 73, 84, 115 CANCER KNOWLEDGE 535 CANCER SITE/CATEGORY see specific site: anus, biliary tract, bladder, brain, bone, breast, Burkitt's tumor, central nervous system, cervix uteri, choriocarcinoma, colon, colon-rectum, corpus uteri, digestive system, esophagus, eye, gestational trophoblastic disease, head and neck, jaw, larynx, kidney, leukemia, liver, lung & bronchus, lymphoma, malignant melanoma, multiple myeloma, muscle, nasopharyngeal, neuroblastoma, oral cavity, ovary, pancreas, penis, pharynx, prostate, rectum, respiratory system, salivary glands, scrotum, skin, sarcoma, small intestine, spinal cord, stomach, testis, thyroid, vagina, vulva all sites combined 202, 230, 273, 315, 319, 324, 536 multiple sites: access and utilization of health services 416, 561 biologic marker ABO blood group 388 Duffy gene 239 HLA 117 communications 515, 522, 553 diagnosis 10, 39, 117, 156, 263, 311, 416 education 459, 522, 550, 553 incidence 6, 12, 30, 39, 48, 71, 75, 107, 117, 156, 189, 210, 217, 238, 239, 245, 257, 263, 273, 301, 309, 311, 346, 388, 389, 413, 416, 432, 437, 438, 446, 450, 472, 476, 477, 482, 493, 508, 510, 511 methodology 169, 173, 185, 207, 263, 426, 432, 437, 439 mortality 12, 29, 32, 34, 59, 64, 65, 71, 88, 93, 125, 173, 174, 183, 185, 188, 199, 202, 245, 273, 280, 284, 309, 314, 321, 328, 343, 344, 346, 356, 395, 396, 413, 416, 421, 432, 446,477, 482, 493, 508, 509, 510, 511 prevention 314 risk factor exposure: air pollution 59, 88, 511 alcohol 39, 93,339, 377, 378, 409, 416, 432, 510, 523, 556 contraceptive devices 130 267 SUBJECT INDEX (Cancer Site/Category-risk factor exposure continued) diet-nutritional factors 39, 93, 375,389, 405, 416, 432, 437, 446, 472,476,510, 511, 522, 523, 556 drugs 511 endogenous hormones 472, 476 exogenous hormones 130 familial factors 476 fluoridation 435 geographic area 30, 88, 93, 107, 156, 189, 210, 223, 421, 426, 446, 477, 508 occupation 64, 65, 93, 173, 174, 189, 202, 257, 284, 321, 328, 356, 395, 396, 416, 432, 433, 437, 510, 511, 523, 556 other diseases 377 solar radiation 522 stress 556 tobacco 88, 93, 238, 339, 377, 378, 405, 409, 416, 432, 476, 510, 511, 515, 522, 523, 556 treatment modalities 32, 476 water pollution 511 screening 459, 515, 522, 556 socioeconomic factors 15, 169, 172, 231, 280, 339 survival 9, 10, 12, 15, 169, 170, 172, 207, 231, 245, 347, 413, 416, 424, 438, 439, 450, 493, 508, 509, 513 treatment 9, 10, 32, 416, 446, 476, 513, 515, 550 CENTRAL NERVOUS SYSTEM 82, 250, 466 CERVIX UTERI access and utilization of health services 559, 562, 564 biologic marker: ABD blood group 325 G6PD 232, 290 herpesvirus type 2 antibodies 252, 349, 361 immunoglobulin 325 communications 552 diagnosis 31, 43, 139, 147, 208, 400, 517, 564 health care provider KAP 559 incidence 27, 41, 43, 44, 53, 70, 76, 90, 104, 126, 155, 192, 205, 208, 232, 288, 294, 306, 310, 390, 400, 404, 427, 434, 436, 448, 480, 503, 507, 517, 543, 564 methodology 436 mortality 43,44, 155, 310, 434, 448, 480, 503, 507 prevalence 70, 448 prevention 448, 559 risk factor exposure: contraceptive devices 529 diet-nutritional factors 404, 427 exogenous hormones 503, 542 geographic area 427, 436, 480 other diseases 104, 252, 349, 361, 427, 503 reproductive factors 1, 27, 76, 155, 192, 208, 262, 349, 361, 390, 400, 404, 427, 480, 503, 507, 542 sexual practices 27,76, 208, 288, 349, 361, 390, 400, 404, 427, 480, 483, 503, 507, 529 tobacco 402, 427, 503 screening: Pap smear 41, 43, 44, 72, 90, 139, 155, 268, 306, 310, 331, 361, 404, 448, 483, 503, 507, 517, 526, 530, 543, 559, 562, 564 socioeconomic factors 70, 205, 530 survival 1, 147, 262, 434, 503 treatment 1, 31, 72, 139, 147, 262, 290 CHEMICAL (see Risk Factor Exposure) CHORIOCARCINOMA 283 COLON diagnosis 74, 92, 97, 110 incidence 37, 110, 176, 181, 429, 437, 441, 471, 504, 505 methodology 201, 437, 447 mortality 201, 441, 471 prevalence 176, 506 prevention 447, 505 risk factor exposure: alcohol 181 bacterial flora 429 bowel characteristics 504, 505 diet-nutritional factors 181, 264, 429, 437, 447, 471, 504, 505, 506 familial factors 97, 176 geographic area 201, 429, 471, 504, 505 occupation 176, 437 other diseases 97, 161, 429, 471 tobacco 181, 471 survival 74, 92, 176, 181 treatment 74, 92, 110 COLON-RECTUM/COLORECTAL 422 communications 552 diagnosis 14, 242 education 551 incidence 242, 278, 428, 429, 437, 468 methodology 437 mortality 468 prevalence 14 risk factor exposure: alcohol 150, 428, 468 diet-nutritional factors 278, 355, 428, 429, 437, 468 geographic area 428, 429, 468 occupation 99, 278, 437, 468 other diseases 52,242, 355, 429, 468 survival 150, 241, 242 tobacco 150, 355, 428, 468 treatment 241, 242 COMMUNICATIONS 515, 518, 519, 520, 522,525, 535, 537, 538, 549, 552, 553, 555, 558 CONTRACEPTIVE DEVICES (see Risk Factor Exposure) CORPUS UTERI diagnosis 43 incidence 42,43, 53,104, 155, 192, 205, 436, 464 methodology 436, 464 mortality 43, 155 risk factor exposure: diet-nutritional factors 386 endogenous hormones 386 exogenous hormones 168, 464, 494 geographic areas 436 other diseases 104 reproductive factors 49, 155, 192 screening: Pap smear 42, 43, 155 socioeconomic factors 205 survival 40, 49 treatment 40, 49 COST FINANCING 571 DIAGNOSIS 554 see Cancer Site/Category access and utilization of health services 565 communications 520 health care provider KAP 565 health care provider: physician 565 screening 520 treatment 520 DIET-NUTRITIONAL FACTORS (see Risk Factor Exposure) DIGESTIVE SYSTEM 67, 214, 467 268 SUBJECT INDEX EDUCATION 459, 519, 520, 522, 525, 536, 538, 550, 551, 553, 555 ENDOGENOUS HORMONES (see Risk Factor Exposure) ESOPHAGUS 422 biologic marker: ABD blood group 325 HLA 116 immunogloblin 325 diagnosis 289 incidence 187, 221, 258, 289, 318, 326, 342, 397, 427, 428, 445, 468, 485, 489, 495 mortality 94, 187, 274, 393, 468, 485, 489, 490 prevention 484, 495 risk factor exposure: alcohol 187, 265, 274, 353, 376, 393, 397, 403, 410, 428, 442, 468, 485, 489, 490, 495 diet-nutritional factors 353, 393, 403, 410, 427, 428, 442, 468, 484, 489, 495 familial factors 489 geographic area 94, 221, 274, 427, 428, 468, 495 occupation 94, 353, 376, 442, 468 other diseases 376, 427, 468 reproductive factors 427 sexual practices 427 tobacco 187, 265, 274, 353, 376, 393, 397,403, 427, 428, 468, 489, 490, 495 treatment 403 socioeconomic factors 353 survival 5, 184, 251, 289, 318, 326, 445 treatment 5, 184, 251, 289, 318 EYE 102, 145, 196, 218, 379, 457 GESTATIONAL TROPHOBLASTIC DISEASE 11, 22, 122, 420, 423 HEAD AND NECK 50, 350, 367 HEALTH CARE PROVIDER: PHYSICIAN 563, 565, 570 HEALTH CARE PROVIDER KAP 559, 567 INCIDENCE (see Cancer Site/Category) JAW 83 KIDNEY 66, 175, 408 LARYNX 86, 171, 360, 376, 418, 442, 485,486, 490, 552 LEUKEMIA 87 access and utilization of health services 571 biologic marker: anti-HTLV antibodies 276 human T-cell leukemia/lymphoma virus (HTLV) 21 HTLV-1 271 karyotype 16, 17 cost financing 571 diagnosis 17, 148, 329 incidence 33, 103, 144, 178, 191, 215, 276, 329 mortality 33 prevalence 332 risk factor exposure: familial factors 33 occupation 497 other diseases 21, 36 survival 8, 16, 36, 118, 144, 148, 178, 285, 302, 329 treatment 8, 144, 148, 285, 302, 329, 332, 492, 571 LIVER 422 biologic marker: alpha-fetaprotein 4, 151, 279, 286, 293, 312, 381, 394 carcinoembryonic antigen 180 gamma-glutamyl transferase 151, 153 hepatitis B antigen 312 hepatitis B surface antigens (HBsAg) 45, 151, 152, 237, 293, 381, 394 hepatitis B virus associated antigens 45 insulin antibody 146 serum ferritin 46 unsaturated vitamin B12 binding capacity 320 diagnosis 146, 151, 320 incidence 106, 151, 229, 237, 313, 381, 485, 500 mortality 96, 266, 485, 500 prevalence 151 risk factor exposure: alcohol 143, 312, 381, 485, 500 diet-nutritional factors 381 familial factors 96 geographic area 96, 312 occupation 500 other diseases 45, 46, 96, 152, 229, 237, 266, 312, 313, 381 survival 151 LUNG AND BRONCHUS communications 538 diagnosis 120 education 538 incidence 35,69, 120, 123, 134, 135, 254, 292, 352, 451 methodology 392 mortality 135, 182, 292, 352, 364, 398, 451, 470, 473 risk factor exposure: air pollution 451 alcohol 150, 353 diet-nutritional factors 353 geographic area 182, 451 occupation 123, 182, 352, 353, 364, 398, 399, 415, 470, 473, 475 tobacco 123,150, 254, 352, 353, 398, 399, 415, 451, 473, 538 socioeconomic factors 69, 150, 353, 473 survival 120, 123, 150 treatment 120, 123 LYMPHOMA biologic marker: anti-HTLV antibodies 276 HLA 166 human T-cell leukemia/lymphoma virus (HTLV) 21 HTLV-1 271 incidence 215, 276, 295, 304, 449, 501 mortality 95, 323, 501 prevalance 332, 449 risk factor exposure: familial factors 95, 166, 449 geographic area 95 occupation 323, 365, 449 other diseases 21, 36, 95 socioeconomic factors 365 survival 36, 166, 304, 449, 501 treatment 332 MALIGNANT MELANOMA 300 diagnosis 163, 224, 496 incidence 38, 222, 224, 233, 253, 255 mortality 233 risk factor exposure: other diseases 256 solar radiation 233, 255 survival 154, 160, 222, 253, 255, 256, 496 treatment 154, 160, 163, 224, 247 MAMMOGRAPHY (see Screening) 269 SUBJECT INDEX METHODOLOGY 169, 412, 447, 569 access and utilization of health services 569 cancer site: breast 169, 436 cervix uteri 436 colon 201, 437, 447 colon-rectum 437 corpus uteri 436, 464 lung & bronchus 392 multiple sites 169, 173, 185, 207, 263, 426, 432, 437, 439 ovary 436 prostate 453 stomach 201 diagnosis 263 incidence 263, 432, 436, 437, 453, 464 mortality 173, 185, 201, 432, 453 prevalence 426 risk factor exposure: alcohol 432 diet-nutritional factors 432, 437, 447, 453 endogenous hormones 453 exogenous hormones 464 geographic area 201, 426, 436, 453 occupation 173, 432, 437, 453 reproductive factors 453 tobacco 432 survival 169, 207, 439 MORTALITY (see Cancer Site/Category) access and utilization of health services 568 MULTIPLE MYELOMA 18, 20, 215, 269, 299, 305, 383, 401 MUSCLE 167 NASOPHARYNGEAL 50, 63, 108, 115, 317, 367 NEUROBLASTOMA 142, 374 OCCULT BLOOD TEST (see Screening) OCCUPATION (see Risk Factor Exposure) ORAL CAVITY 216 communications 552 diagnosis 171, 270 incidence 3,162, 226,228,270, 287, 303, 485, 498 mortality 485, 490, 498 prevention 226 risk factor exposure: alcohol 24, 162, 171, 212, 354, 376, 406, 442, 485, 490, 498 diet-nutritional factors 442 geographic area 24, 498 occupation 354, 376, 442 other diseases 212, 376, 498 solar radiation 226, 303 tobacco 24, 162, 171, 212, 354, 376, 406, 490, 498 survival 162, 171, 380, 498 treatment 3, 80, 162, 171 OVARY incidence 53, 104, 192, 291, 335, 336, 337, 414, 436 methodology 436 mortality 291 prevention 414 risk factor exposure: chemical 414 endogenous hormones 414 exogenous hormones 168 familial factors 414 geographic area 414, 436 other diseases 104, 414 reproductive factors 192, 337, 414 PANCREAS diagnosis 204 incidence 158, 159, 165, 204, 417, 443, 461 mortality 13, 25, 158, 159, 165, 391, 417, 461 risk factor exposure: air pollution 159 alcohol 25, 362, 417, 443, 465 diet-nutritional factors 362, 417, 443, 465 endogenous hormones 159 familial factors 443 geographic area 25, 159, 391, 443, 461 occupation 25,362, 391,417, 443, 465 other diseases 25, 204, 443 tobacco 159, 204, 362, 417, 443, 465 survival 165, 298 treatment 298 PAP SMEAR (see Screening) PENIS 58, 104, 272, 294, 427 PHARYNX 216 diagnosis 86, 171 incidence 86, 485, 498 mortality 485, 490, 498 risk factor exposure: alcohol 24, 171, 376, 406, 442, 485, 490, 498 diet-nutritional factors 442 geographic area 24, 498 occupation 376, 442 other diseases 376, 498 tobacco 24, 171, 376, 406, 490, 498 survival 171, 380, 498 treatment 171 PREVALENCE 474 cancer site: bladder 275 brain 248, 411 bone 129 breast 70, 220, 235 cervix uteri 70, 448 colon 14, 176, 506 digestive system 467 eye 457 leukemia 332 liver 151 lymphoma 332, 449 multiple sites 101, 177, 426, 438 prostate 345 rectum 14 sarcoma 478 testis 227 thyroid 119 PREVENTION access and utilization of health services 559 cancer site: cervix uteri 448, 559 colon 447, 505 digestive system 467 esophagus 484, 495 multiplesites 314, 459, 510, 522, 553 oral cavity 226 ovary 414 penis 272 prostate 128, 454 communications 522, 553, 555 diagnosis 467 education 459, 522, 553, 555 health care provider KAP 559 incidence 128, 226, 272, 414, 448, 454, 467, 495, 505, 510 270 SUBJECT INDEX (Prevention continued) mortality 314, 448, 454, 505, 510 prevalence 448, 467 risk factor exposure: air pollution 454 alcohol 495, 510 bowel characteristics 505 chemical 414 circumcision 272 diet-nutritional factors 447, 484, 495, 505, 510, 522 endogenous hormones 414 familial factors 414 geographic area 272, 414, 454, 495, 505 occupation 454, 510 other diseases 414 reproductive factors 414, 454 sexual practices 454 solar radiation 226, 522 tobacco 495, 510, 522 screening: mouth exam 522 multiple cancer screening tests 459 Pap smear 448, 559 proctoscopic exam 522 PROCTOSCOPIC EXAM (see Screening) PROSTATE biologic marker: prostatic acid phosphatase 359 prostatic creatine kinase 359 diagnosis 111, 137, 194, 359, 419, 455, 479, 514 education 551 health care provider: physician 563 incidence 77,79,111,128,132,137,157, 194, 259, 260, 372, 373, 382, 419, 440, 453, 454, 455, 456, 479, 499, 502, 514 methodology 453 mortality 77, 78, 79, 194, 260, 363, 419, 440, 453, 454, 456, 479, 499, 502, 514 prevalence 345 prevention 128, 454 quality of care 563 risk factor exposure: air pollution 454 diet-nutritional factors 19, 131, 132, 260, 368, 369, 370, 371, 440, 452, 453, 479, 502, 512 endogenous hormones 2, 131, 132, 259, 260, 369, 370, 371, 372, 373, 419, 440, 452, 453, 499, 502, 512 familial factors 440, 456, 479 geographic area 131, 132, 345, 370, 371, 440, 452, 453, 454, 456, 479, 499 occupation 19, 260, 363, 440, 453, 454, 456 other diseases 260, 382, 479 reproductive factors 131, 373, 452, 453, 454, 479 sexual practices 260, 373, 382, 440, 454, 479, 499, 512 tobacco 260 socioeconomic factors 60, 61,77,79, 89 survival 60, 61, 89, 111, 137, 194, 382, 440, 456, 479 treatment 89, 111, 137, 359, 419, 455, 479, 502 PROSTATE PALPATION (see Screening) PSYCHOSOCIAL FACTORS reactions to dying/grief 544 coping mechanisms 556 QUALITY OF CARE 563 RECTAL EXAM (see Screening) RECTUM diagnosis 14, 74, 92 incidence 37, 181, 236, 429, 441, 471 mortality 441, 471 prevalence 14 risk factor exposure: alcohol 181 bacterial flora 429 diet-nutritional factors 181, 429, 471 familial factors 195 geographic area 429, 471 other diseases 161, 195, 236,429, 471 sexual practices 236 tobacco 181, 471 survival 74, 92, 181 treatment 74, 92, 181 REPRODUCTIVE FACTORS (see Risk Factor Exposure) RESPIRATORY SYSTEM 261, 384 RISK FACTOR EXPOSURE air pollution 59, 88, 159, 451, 454, 511 alcohol: biliary tract 443 breast 150 colon 181 colon-rectum 150, 428, 468 esophagus 187, 265, 274, 353, 376, 393, 397, 403, 410, 428, 442, 468, 485, 489, 490, 495 kidney 408 larynx 171, 376, 418, 442, 485, 486, 490 liver 143, 312, 381, 485, 500 lung 150 multiple sites 39, 93, 339, 377, 378, 409, 416, 432, 510, 523, 556 oral cavity 24, 162, 171, 212, 354, 376, 406, 442, 485, 490, 498 pancreas 25, 362, 417, 443, 465 pharynx 24, 171, 376, 406, 442, 485, 490, 498 rectum 181 stomach 428 bacterial flora 429 bowel characteristics 504, 505 breastfeeding 351, 357, 462, 481 chemical 414 circumcision 272 contraceptive devices 130, 230, 529 diet-nutritional factors: 474 biliary tract 443 bladder 358, 366, 407, 427 breast 133, 246, 327, 357, 452, 460, 462, 469 cervix uteri 404, 427 colon 181, 264, 429, 437, 447, 471, 504, 505, 506 colon-rectum 278, 355, 428, 437, 468 corpus uteri 386 esophagus 353, 393, 403, 410, 427, 428, 442, 468, 484, 489, 495 larynx 442 liver 381 multiple sites 39, 93, 375, 389, 405, 416, 432, 437, 446, 472, 476, 510, 511, 522, 523, 556 oral cavity 442 pancreas 362, 417, 443, 465 pharynx 442 prostate 19, 131, 132, 260, 368, 369, 370,371, 440, 452, 453, 479, 502, 512 rectum 181, 429, 471 stomach 68, 209, 428, 463 thyroid 491 vulva 385 drugs 511 endogenous hormones: breast 56, 133, 138, 149, 179, 211, 235, 246, 452 271 (Risk Factor Exposure continued) corpus uteri 386 kidney 408 multiple sites 472, 476 ovary 414 pancreas 159 prostate 2, 131, 132, 259, 260, 370, 371, 372, 373, 419, 440, 452, 453, 499, 502, 512 testis 259 thyroid 491 exogenous hormones: all sites combined 230 breast 138, 168, 225, 462, 494 cervix uteri 503, 542 corpus uteri 168, 464, 494 gestational trophoblastic disease 423 multiple sites 130 ovary 168 familial factors: biliary tract 443 breast 26,124,211, 225, 246, 357, 462 colon 97, 176 esophagus 489 gestational trophoblastic disease leukemia 33 liver 96 lymphoma 95, 166, 449 multiple sites 476 muscle 167 ovary 414 pancreas 443 prostate 440, 456, 479 rectum 195 sarcoma 97, 197 thyroid 491 fluoridation 435 geographic area: biliary tract 443 bladder 23, 427 brain 411 breast 436, 452 Burkitt's tumor 63, 73 cervix uteri 427, 436, 480 colon 201, 429, 471, 504, 505 colon-rectum 428, 468 corpus uteri 436 esophagus 94, 221, 274, 427, 428, 468, 495 SUBJECT INDEX eye 379 larynx 418 kidney 408 liver 96, 312 lung and bronchus 182, 451 lymphoma 95 multiple sites 30, 88, 93, 107, 156, 189, 210, 223, 421, 426, 446, 477, 508 nasopharyngeal 63, 317 oral cavity 24, 498 ovary 414, 436 pancreas 25, 159, 391, 443, 461 penis 272, 427 pharynx 24, 498 prostate 131,132, 345, 370, 371, 440, 452, 453, 454, 456, 479, 499 rectum 429, 471 stomach 201, 428 industry and occupation: 516, 528, 541 all sites combined 202 biliary tract 443 bladder 23, 275, 358, 407 central nervous system 466 colon 99, 176, 437 colon-rectum 278, 437, 468 esophagus 94, 353, 376, 442, 468 kidney 408 larynx 360, 376, 418, 442, 486 leukemia 497 liver 500 lung and bronchus 123, 182, 352, 353, 364, 398, 399, 415, 470, 473, 475 lymphoma 323, 365, 449 multiple myeloma 401 multiple sites 64, 65, 93, 173, 174, 189, 202, 257, 284, 321, 328, 356, 395, 396, 416, 432, 433, 437, 510, 511, 523, 556 oral cavity 354, 376, 442 pancreas 25, 362, 391, 417, 443, 465 pharynx 376, 442 prostate 19, 260, 263, 363, 440, 453, 454, 456 : rectum 99 respiratory system 384 vulva 385 other diseases: biliary tract 443 bladder 358, 427 bone 129 breast 56, 349, 462 Burkitt's tumor 50, 63, 115 cervix uteri 104, 252, 349, 361, 427, 503 colon 97, 161, 429, 471 colon-rectum 52, 242, 355, 429, 468 corpus uteri 104 esophagus 376, 427, 468 head and neck 50, 367 kidney 408 larynx 376 leukemia 21, 36 liver 45, 46, 96, 152, 229, 237, 266, 312, 313, 381 lymphoma 21, 36, 95 malignant melanoma 256 multiple myeloma 18 multiple sites 377 nasopharyngeal 50, 63, 115, 367 oral cavity 212, 376, 498 ovary 104, 414 pancreas 25, 204, 443 penis 58, 104, 427 pharynx 376, 498 prostate 260, 382, 479 rectum 236, 471 sarcoma 97 scrotum 58 skin 333 stomach 68, 113 testis 387 vagina 104 vulva 91, 282, 385 reproductive factors: breast 105, 138, 179, 200, 211, 219, 225, 240, 277, 327, 338, 349, 351, 357, 425, 452, 462 cervix uteri 1, 27, 76, 155, 192, 208, 262, 349, 361, 390, 400, 404, 427, 480, 503, 507, 542 corpus uteri 49, 155, 192 gestational trophoblastic disease 11, 22, 122, 420, 423 ovary 192, 337, 414 penis 427 prostate 131, 373, 452, 453, 454, 479 vagina 27, 192 vulva 91, 192, 282, 385 sexual practices: bladder 427 272 (Risk Factor Exposure continued) breast 349 cervixuteri 27,76,208, 288, 349, 361, 390, 400, 404, 427, 480, 483, 503, 507, 529 esophagus 427 penis 427 prostate 260, 373, 382, 440, 454, 479, 499, 512 rectum 236 vagina 27 vulva 385 solar radiation 57, 203, 226, 233, 255, 303, 333, 522 stress 556 SUBJECT INDEX familial factors 97, 197 other diseases 97 survival 54, 140, 141, 334, 487 treatment 54, 100, 334 SCREENING bladder cytology 275 breast 357 breast self exam (BSE) 85, 213, 515, 521, 525, 533, 534, 538, 548, 557, 562 clinical breast exam 213, 533, 562, 565, 566 mammography 213, 515, 520, 525, 566 mouth exam 522 Pap smear 530 survival 15,60, 61, 62, 89,150, 169, 172, 231, 269 treatment 89 SPINAL CORD 114 STOMACH biologic marker: ABO blood group 68, 209 diagnosis 68, 113, 307 incidence 28, 113, 164, 209, 428 methodology 201 mortality 201, 209, 307, 463 risk factor exposure: tobacco: 516, 524, 527, 528, 531, 539, 540, 541, 545, 547, 560 biliary tract 443 bladder 358, 407, 427 breast 150, 538 cervix uteri 402, 427, 503 multiple cancer screening tests 459, 556 Pap smear 41, 42, 43, 44, 72, 90, 139, 155, 268, 306, 310, 331, 361, 404, 448, 483, 503, 507, 515, 517, 520, 525, 526, 530, 532, 543, 546, 548, 559, 562, 564, 565, 569 proctoscopic exam 515, 520, 522, 548, alcohol 428 diet-nutritional factors 68, 209, 428, 463 geographic area 201, 428 other diseases 68, 113 tobacco 428 colon 181, 471 colon-rectum 150, 355, 468 esophagus 187, 265, 274, 353, 376, 393, 397, 403, 427, 428, 468, 489, 490, 495 kidney 408 larynx 171, 376, 418, 486, 490 lung and bronchus 123, 150, 254, 352,353, 398, 399, 415, 451, 473, 538 multiple sites 88, 93, 238, 339, 377, 378, 405, 409, 416, 432, 476, 510, 511, 515, 522, 523, 556 oral cavity 24, 162, 171, 212, 354, 376, 406, 490, 498 pancreas 159,204,362, 417, 443, 465 penis 427 pharynx 24, 171, 376, 406, 490, 498 prostate 260 rectum 181, 428, 471 stomach 428 vulva 385 treatment modalities 32, 330, 403, 476, 491 ultraviolet light 333 water pollution 511 SALIVARY GLANDS 308 SARCOMA 431 diagnosis 54, 97, 487 incidence 140, 197, 249, 478, 487 prevalence 478 risk factor exposure: 565 prostate palpation 548, 565 rectal exam 548, 565 SCROTUM 58, 316 SEXUAL PRACTICES (see Risk Factor Exposure) SKIN 57, 203, 271, 303, 330, 333, 488 SMALL INTESTINE 7 SOCIOECONOMIC FACTORS cancer site: breast 62, 70, 150, 169, 243 cervix uteri 70, 205, 530 colon-rectum 99, 150 corpus uteri 205 esophagus 353 lung and bronchus 69, 150, 353, 473 lymphoma 365 multiple myeloma 269 multiplesites 15, 169, 172, 231, 280, 339 } prostate 60, 61, 77, 79, 89 incidence 69, 70, 77, 79, 205, 243 mortality 77, 79, 280, 473 prevalence 70 risk factor exposure: alcohol 150, 339, 353 diet-nutritional factors 353 occupation 99, 353, 365, 473 tobacco 150, 339, 353, 473, 540 screening: breast self exam (BSE) 521 survival 68 treatment 68 SURVIVAL (see Cancer Site/Category) TESTIS 51, 186, 227, 259, 341, 348, 387 THYROID 119, 206, 281, 491 TOBACCO (see Risk Factor Exposure) TREATMENT 520 (see Cancer Site/Category) VAGINA 27, 104, 126, 147, 192 VULVA 53, 91, 126, 192, 282, 297, 385 273 PUBLICATIONS BASED ON NCI DATA FOR U.S. INDEX Numbers following NCI data for Third National Cancer End Results Program - 1950- US. refer to article numbers. Survey (TNCS) - 1969- 1973 9, 71, 108, 207, 218 1971 53, 69, 70, 71, 75, SEER - 1973-Present 10, . . 77,105, 108, 126, 134, 164, 12, 15, 26, 35, 107, 134 First National Cancer Survey 165, 214, 245, 295, 301, or 166 100 233 245. (FNCS) - 1937-1939 71, 303, 309, 310, 335, 337, 23, ey 205. ol 209, 75, 301 339, 346, 360, 379, 399 a oy Se Second National Cancer Survey (SNCS) - 1947- 1948 53, 71, 75, 245, 301 275 Numbers following geographic area refer to article numbers. - Africa 81, 82,425,426,427,431, 477, 478, 502 Belgium 158 Canada 179,507 Caribbean 21, 276 Columbia 345 Denmark 146, 179 Egypt 271,411 England 354 Finland 179 France 286 Gambia 50 Ghana 157, 271, 411 Hong Kong 179 Israel 317 Italy 158 Ivory Coast 286, 411 Jamaica 30, 121, 249, 282, 283, 304, 305, 390 Japan 159, 179, 345 Kenya 50, 115, 286, 348, 411 Netherlands 179 Nigeria 2, 3, 11, 73, 76, 100, 110, 156, 157, 177, 192, 222, 223, 224, 226, 227, 228, 229, 239, 270, 271, 300, 346, 372, 411 North America 132 Norway 179 Rhodesia 68, 156, 279, 312, 346, 411 Senegal 394, 411 Singapore 286 South Africa 4, 16, 17, 18, 20, 22,27, 29, 45, 51, 66, 67, 74, 83, 86, 90, 102, 106, 112, 116, 117, 119, 129, 131, 132, 143, 144, 146, 151, 152, 153, 156, 178, 180, 181, 187, 188, 189, 196, 221, 234, 235, 242, 248, 255, 256, 257, 258, 264, 271, 278, 281, 292, 293, 306, 308, 313, 320, 323, 324, 325, 326, 327, 342, 343, 344, 346, 353, 361, 369, 370, 371, 381, 403, 411, 421, 422, 445, 448, 475, 484, 504, 506, 516, 526, 545 Sudan 411 Sweden 158, 179 Tanzania 115 Transvaal 411 GEOGRAPHIC AREA INDEX Trinidad 198, 209, 250 Tunisia 271 Uganda 50, 57, 271, 272, 411 United Kingdom 246, 412, 533 United States 2, 12, 13, 23, 24, 25, 32, 34, 71, 78, 82, 88, 93, 94, 95, 96, 105, 108, 110, 122, 125, 135, 145, 147, 165, 166, 167, 179, 185, 191, 194, 195, 211, 213, 218, 220, 231, 239, 266, 270, 273, 274, 295, 315, 319, 332, 335, 340, 346, 352, 370, 371, 376, 377, 379, 380, 385, 387, 393, 407, 408, 409, 413, 418, 424, 432, 434, 444, 469, 470, 490, 498, 508, 509, 515, 520, 524, 525, 529, 530, 532, 540, 541 Alabama 53, 69, 126, 334, 346 California 6, 14, 35, 53, 69, 77, 79,99, 105, 126, 136, 150, 156, 158, 159, 169, 170, 184, 193, 207, 236, 237, 238, 239, 240, 259, 260, 296, 328, 333, 346, 350, 355, 366, 367, 382, 437, 517, 527, 542, 543, 544, 562 Colorado 53, 69, 126, 165 Connecticut 35, 165, 172, 207, 366 District of Columbia 54, 56, 81, 98, 156, 157, 161, 162, 200, 206,.223, 254, 265, 298, 368, 372, 373, 393, 410, 518, 537, 551, 553 Florida 352, 392 New Jersey 284, 309, 310, 311, 323, 366 New Mexico 35, 233, 366 New York 1, 37, 85, 104, 109, 123, 127, 131, 140, 141, 149, 154, 201, 210, 225, 243, 244, 269, 280, 291, 307, 316, 318, 323, 330, 338, 351, 357, 400, 404, 534, 538, 546, 547, 549, 559, 561, 566 North Carolina 65, 138, 253, 331, 401, 406, 535, 536, 560, 564 Ohio 33, 48, 88, 155, 182, 183, 202, 466 Oklahoma 299, 563 Oregon 205 Pennsylvania 28, 53, 55, 59, 69, 75, 118, 126, 139, 164, 173, 174, 182, 212, 289, 321, 346, 395, 396, 550 South Carolina 251, 519 Tennessee 8, 92, 101, 148, 285, 329 Texas 7,53,69,80,91,126, 163, 171, 252, 262, 277, 303, 314, 322, 346, 349, 374, 386, 522 Utah 35, 366 Virginia 5, 47, 60, 62, 89, 197, 264, 392 Washington 35, 366 Washington, DC (see District of Colombia) Georgia 35, 53, 69, 107, 126, 130, 168, 190, 191, 230, 239, 241, 275, 288, 346, 366, 375, 389, 392, 402 Hawaii 35, 345 Illinois 72, 111, 175, 548, 565 Towa 15, 35, 53, 69, 126, 165, 207, 366 Kentucky 40, 41, 42, 43, 44, 49 Louisiana 35, 52, 160, 203, 207, 214, 215, 216, 217, 218, 261, 287, 345, 358, 364, 366, 383, 388, 391, 398, 523, 531 Maryland 31, 103, 124, 199, 302, 362, 397, 521 Massachusetts 172, 365 Michigan 35, 53, 69, 126, 239, 263, 267, 268, 301, 346, 366, 552 Minnesota 53, 126 Mississippi 113, 114, 120 Nebraska 176 Nevada 297 West Indies 36 Yugoslavia 179 Zambia 208, 411 Zanzibar 39 This index represents articles that focus on or provide data for specific geographic areas. 277 ry FIRST AUTHOR INDEX Numbers following author names refer to article numbers. Abad, R.S. 1 Adam, E. 349 Adeloye, A. 411 Adelstein, A.M. 412 Ahluwalia, B. 2 Ajagbe, HA. 3 Alexander, M.G. 4 Alfred, L.J. 350 Amendola, B. 5 American Cancer Society 413, 515 Anteby, S.0. 414 Arellano, M.G. 6 Arthaud, J.B. 7 Askey, D.G. 553 Aur, R.J.A. 8 Austin, H. 351 Axtell, LM. 9,10 Ayangade, O. 11 Bader, M. 415 Baker, M.D. 516 Baquet, C.R. 12, 416 Benarde, M.A. 13 Berg, J.W. 14, 15, 417 Bernstein, R. 16, 17 Bisceglie, A.M. 18 Blair, A. 19 Blattner, W.A. 20, 21 Bloch, B. 22 Blot, W.J. 23, 24, 25, 26, 352, 418 Blumenthal, N.J. 27 Boles, R.S. 28 Boxer, R.J. 419 Bracken, M.B. 420 Bradshaw, E. 29, 353, 421 Bras, G. 30 Brenner, D.E. 31 Breslow, L. 517 Brink, A.J. 422 Brinton, L.A. 32 Browne, RM. 354 Browning, D. 33 Buckley, J.D. 423 Burbank, F. 34 Butler, L. 518 Byers, T.E. 35 Cantor, A.B. 519 Cardwell, J.J. 520 Cassileth, B.R. 550 Catovsky, D. 36 Cedermark, B.J. 37 Celentano, D.D. 521 Centers for Disease Control 424 Charles, N.C. 38 Chiedozi, L.C. 425 Chopra, S.A. 39 Christopherson, W.M. 40, 41, 42, 43, 44 Clausen, J.P. 554 Cohen, C. 45, 46 Cohen, LLK. 47 Cole, S. 48 Connelly, P.J. 49 Cook, P.J. 426 Cook-Mozaffari, P. 427 Cooper, EH. 50 Coovadia, Y.M. 51 Coreil, J. 522 Cornely, P.B. 567 Correa, P. 52, 428, 429, 523 Covey, L.S. 524 Cramer, D.W. 53 Crawford, M. 54 Crichlow, RW. 55 Cuenca, C.R. 56 Cutler, S.J. 430 Dales, L.G. 355 Davies, ].N.P. 57, 431 Davis, D.L. 432 Davis, M.E. 433 Dawson, D.F. 58 Dayal, H. 59, 60, 61, 62 de-The, G. 63 Decoufle, P. 64 Delzell, E. 65, 356 DeMoor, N.G. 66 Denniston, RW. 525 Dent, D.M. 67 Dent, R.I. 68 Devesa,S.S. 69,70,71,434 DeVita, V.T., Jr. 555 Dini, M. 72 Doll, R. 435 Dubin, N. 357 Dunham, L.J. 358 Dunn, J.E., Jr. 436, 437 Earles, L.C., III 438 Edington, G.M. 73 Elliot, M.S. 74 Emdon, S. 526 Emes, J.J. 75 Emovon, A.C. 76 Enstrom, J.E. 439 Ernster, V.L. 77,78, 79 Evans, HL. 80 Fair, W.R. 359 Fan, K.J. 81, 82 Farman, A.G. 83 Favrot, M.C. 84 Feldman, J.G. 85 Fernandes, C.M.C. 86 Festa, R.S. 87 Flanders, W.D. 360, 440 Ford, A.B. 88 Forrest, J.B. 89 Fragoyannis, S. 90 Franklin, EW., III 91 Franklin, R. 92 Fraser, P. 441 Fraumeni, J.F., Jr. 93, 94, 95, 96, 97, 442, 443 Freedman, R.S. 361 Freidman, G.D. 527, 528 Fruchter, R.G. 559 Funderburk, W.W. 98 Garabrant, D.H. 99 Garfinkel, L. 444 Garg, S.K. 100 Gaskin, H.S. 101 Gilder, S.S.B. 445 Gold, E.B. 362 Goldberg, L. 102 Goldsmith, D.F. 363 Gordis, L. 103 Gori, G.B. 446 Gottlieb, M.S. 364 Graham, S. 104, 447 Grant, M.C.G. 448 Gray, G.E. 105 Greenberg, M. 106 Greenberg, R.S. 107 Greene, A.B. 560 Greene, M.H. 108 Gregorio, D.I. 109 Grillo, ILA. 110 Grufferman, S. 449 Guinan, P.D. 111 Gutensohn, N.M. 365 Hacking, E.A. 112 Haddy, T.B. 450 Haerer, A.F. 113, 114 Hall, H. 556 Hall, P.J. 115 Hammond, M.G. 116, 117 Hann, H.L. 118 Hansen, O.E. 119 Hardy, J.D. 120 Harris, M. 121 Hartge, P. 366 Hayashi, K. 122 Haynes, A. 568 Heath, F.R. 123 Heizer, W.D. 124 Hendershot, G.E. 529, 530 Henderson, E. 367 Henschke, U.K. 125 Henson, D." 126 Herbsman, H. 127 Herring, B.D. 128 Heshmat, M.Y. 368 Heydt, D. 129 Higgins, L.T.T. 451 Higgins, J.E. 130 Hill, P. 131, 132, 133, 369, 370, 371, 452 Holtzman, D. 557 Horm, JW. 134, 135 Howard, J. 136 Howe, H.L. 569 Huben, R. 137 279 FIRST AUTHOR INDEX Hulka, B.S. 138, 139 Hunter, S.M. 531 Hutchison, G.B. 453, 454 Huvos, A.G. 140, 141 Imashuku, S. 142 Isaacson, C. 143 Jackson, M.A. 372, 373 Jacobs, P. 144 Jensen, R.D. 145 Joffe, B.I. 146 Johnson, C.C. 374 Johnston, G.A., Jr. 147 Jones, G.W. 455 Kalwinsky, D.K. 148 Kane-Williams, E. 551 Kark, J.D. 375 Katz, J.L. 149 Keirn, W. 150 Keller, A.Z. 376, 377, 378, 379, 380 Kew, M.C. 151, 152, 153, 381 Keyhani, A. 154 Kim, K. 155 King, H. 456 Kleinman, J.C. 532 Kleinstein, R.N. 457 Kovi, J. 156, 157 Krain, L.S. 158, 159, 382 Krementz, E.T. 160 Larson, E. 458 Leech, S.H. 383 Leffall, L.D., Jr. 161, 162, 459, 460 Lepow, G.M. 163 Lerer, T.]. 164 Levin, D.L. 165, 461 Levin, M.L. 462 Levine, P.H. 166 Li, F.P. 167 Liang, A.P. 168 Linden, G. 169, 170 Lipkin, A. 171 Lipworth, L. 172 Lloyd, J.W. 173, 174, 384 Love, L. 175 Lynch, HT. 176 Lynch, P.D. 552 Mabogunje, O.A. 177 Mabuchi, K. 385 Macdonald, E.J. 463 MacDonald, P.C. 386 MacDougall, L.G. 178 Mack, T.M. 464 Maclean, U. 533 MacMahon, B. 179, 465 Macnab, G.M. 180 Maisto, O.E. 181 Mancuso, T.F. 182, 183, 466 Mandal, A.K. 184 Manfredi, C. 534 Manton, K.G. 185 McBride, W. 467 McConnell, R.B. 468 McCusker, J. 561 McDonald, M.W. 186 McGlashan, N.D. 187, 188, 189 McLendon, R.E. 190 McPhedran, P. 191 Megafu, U. 192 Menck, H.R. 193 Mettlin, C. 194, 195, 469, 558 Michaels, D. 470 Michielutte, R. 535, 536 Miller, A.B. 471, 472 Miller, B. 196 Miller CW. 197 Miller, G.J. 198 Miller, JM. 199 Miller, W.]., Jr. 473 Misczynski, M. 562 Mohla, S. 200 Monk, M. 201 Monson, R.R. 202 Mora, R.G. 203 Morgan, J.E. 204 Morrison, A.S. 387 Morton, W.E. 205 Mosley, E.L. 206 Muller, C. 570 Murray, M.J. 474 Myers, J. 475 Myers, M.H. 207 Naik, K.G. 208 Naraynsingh, V. 209 Nasca, P.C. 210 Natarajan, N. 211 Nelson, J.F. 212 Nemoto, T. 213 Newell, G.R. 214,215, 216, 217, 218, 388, 476 Norris, H.J. 219, 220 Oettle, A.G. 221, 477, 478 Ogunbiyi, T.A.J. 222 Olisa, E.G. 223 Oluwasanmi, J.O. 224 Oluwole, S.F. 225 Onuigbo, W.I.LB. 226, 227, 228 Onuigbo, W.J.B. 229 Ory, HW. 230 Owen, W.L. 479, 563 Page, W.F. 231 Panek, D. 571 Park, I-]. 232 Parker, D.F. 537 Pathak, D.R. 233 Pegoraro, R.J. 234, 235 Peleg, I. 389 Persaud, V. 390, 480 Peters, R.K. 236 Peters, R.L. 237 Petitti, D.B. 238 Petrakis, N.L. 239, 240 Petrek, J.A. 241 Pickle, L.W. 391, 392 Pillay, S.P. 242 Polednak, A.P. 243, 244 Pollack, E.S. 245 Pollak, H. 481 Potter, J.F. 246 Pottern, L.M. 393 Pratt, C.B. 482 Prince, AM. 3% Quinn, D. 247 Quinn, R.J.M. 248 Ragbeer, M.M.S. 249 Raju, G.C. 250 Rambo, V.B. 251 Rawls, W.E. 252 Redmond, C.K. 395 Reid, B.L. 483 Reintgen, D.S. 253 Rickard, V.D. 254 Rippey, J.J. 255, 256 Roach, W.A. 484 Roberson, N.L. 538 Robertson, M.A. 257 Rockette, H.E. 396 Rogers, E.L. 397 Rose, C.L. 539 Rose, E.F. 258 Ross, R.K. 259, 260 Rothman, K.J. 485, 486 Rothschild, H. 261, 398 Rutledge, F.N. 262 Safai, B. 487, 488 Saftlas, A.F. 263 Salyers, A.A. 264 Sampson, C.C. 265 Sandler, D.P. 266 Satariano, W.A. 267, 268 Savage, D. 269 Sawyer, D.R. 270 Saxinger, W. 271 Schmauz, R. 272 Schneiderman, M.A. 273 Schoenberg, B.S. 274 Schottenfeld, D. 489, 490, 491 Schulte, P.A. 275 Schupbach, J. 276 Schwartz, M.R. 277 Seeler, R.A. 492 Segal, I. 278 Seggie, J. 279 Seidman, H. 280, 493 Selzer, G. 281 Sengupta, B.S. 282, 283 Shapiro, 5. 566 FIRST AUTHOR INDEX Sheffet, A. 284 Shoemaker, E.S. 494 Silber, W. 495 Silvers, D.N. 496 Simone, J.V. 285 Sizaret, P. 286 Skinner, R.L. 287 Smally, A.J. 497 Smith, D.B. 288 Smith, E.B. 289 Smith, EM. 498 Smith, JW. 290 Soichet, S. 291 Solomon, A. 292 Song, E. 293 Sorahan, T. 294 Spitz, M.R. 295 Spivey, G.H. 296 Stagner, L.T. 399 Stapleton, J.J. 297 Sterling, T.D. 540, 541 Stern, E. 542, 543 Stevens, J. 298 Stober, J. 299 Susseelan, A.V. 300 Swanson, G.M. 301 Szklo, M. 302 Szpak, C.A. 303 Talerman, A. 304, 305 Te Groen, L.H.T. 306 Terris, M. 307, 400 Theron, E.J. 308 Thind, I.S. 309, 310, 311 Thomas, G.E. 312 Tietz, W. 544 Tollerud, D.J. 401 Torres, F.O. 313 Trevathan, E. 402 Tsai, S.P. 314, 315 Tucci, P. 316 Tulinius, H. 499 Turgman, J. 317 Turnbull, A.D.M. 318 Ueshima, H. 319 Van Der Burgh, C. 545 Van Peenen, P.F.D. 500 Van Rensburg, S.J. 403 Van Tonder, S. 320 Vena, J.E. 321 Vernon, S.W. 322 Vianna, N.J. 323, 501 Vos, G.H. 324, 325 Walker, A.R.P. 326, 327, 502, 503, 504, 505, 506 Walrath, J. 328 Walters, T.R. 329 Walther, R.R. 330 Walton, L.A. 331, 564 Walton, R.J. 507 Walzer, P.D. 332 Warnecke, R.B. 546, 547, 548, 565 Wassertheil-Smoller, S. 404 Weaver, S.M. 333 Weiss, A.B. 334 Weiss, N.S. 335, 336, 337 White, J.E. 508 White, J.K. 509 Wilkinson, G.S. 338, 549 Willet, W.C. 405 Williams, R.R. 339 Wilson, R.E. 340 Winn, D.M. 406 Wogalter, H. 341 Wolpowitz, A. 342 Wynder, E.L. 407, 408, 409, 510, 511, 512 Wyndham, C.H. 343, 344 Yatani, R. 345 Young, J.L. 346, 347, 513 Zaridze, D.G. 514 Ziegler, R.G. 410 Zimmerman, R.R. 348 Yr U.S. GOVERNMENT PRINTING OFFICE: 1989— 2 32-180/ 00099 281 453 u. C. BERKELEY LIBRARIES il C0bL94ua098