|This page in the original is blank.|
Reports on Women's Participation in Clinical Studies, 1977-1993
Sex as Reported in a Recent Sample of Psychological Research
P. Reardon and S. Prescott (1977) Psychology of Women Quarterly 2(2):157-160.
The authors reviewed all of the articles that appeared in the Journal of Personality and Social Psychology, volume 30, 1974, for sex of subjects, type of conclusions drawn, and whether sex was mentioned in the abstract, introduction, or methods section. These results were compared with those from a similar study done in 1972 in order to determine whether changes had occurred during the two-year period in scientific sampling and reporting procedures. The authors found that the percentage of all-male studies had dropped 15 percent while all-female studies had risen 22 percent. In addition, there was an increase in the number of both-sex studies that included some analysis of gender differences.
Why Researchers Don't Study Women: The Responses of 62 Researchers
S. Prescott (1978) Sex Roles 4(6):899-905.
The author interviewed 62 researchers who had authored 64 single-sex studies appearing in the Journal of Personality and Social Psychology in 1970 and 1971 about their reasons for limiting their study population to males or females only. Replies were analyzed for thematic content and formed three major types: ''scientific" (e.g., "desire to reduce the
variation in subjects' responses": 56 percent), "practical" (e.g., "limited money and resources for dealing with an increased sample size": 28 percent), and ''extra-scientific" (e.g., "lack of interest in questions related to sex differences": 15 percent). The author concludes that there has been an imbalance in psychology which has led to the study of men rather than women or both sexes, and offers suggestions for counteracting gender-biased trends in psychological research and reporting.
Underrepresentation of Women in New Drug Trials
E. L. Kinney, J. Trautmann, J. A. Gold, E. S. Vesell, and R. Zelis (1981) Annals of Internal Medicine 95(4):495-499.
This article presents findings of a survey of age and gender distribution of subjects participating in 50 clinical trials reported in 1979. The data indicate that young women served less frequently than young men as subjects in premarketing clinical drug trials. Moral, legal, and medical implications of this "underrepresentation" of women are considered; remedies to increase participation of young, nonpregnant women in clinical trials are discussed.
National Institutes of Health: Problems in Implementing Policy on Women in Study Populations
General Accounting Office, Statement of M. V. Nadel (18 June 1990)
In this report, the General Accounting Office (GAO) reported that NIH had not successfully implemented its 1986 policy regarding the inclusion of women in clinical studies. As part of a necessarily informal study (because of the absence of organized data), the GAO reviewed 50 recent NIH grant applications, most of which proposed studies of conditions that affect both men and women. The GAO found that approximately 20 percent of the proposals provided no information on the sex of the study population. Over one-third of the proposals indicated that both men and women would be included, but did not specify in what proportions. In addition, a number of proposals for studies involving only male subjects provided no rationale for the single-sex design.
Some Drug Trials Show Gender Bias
C. Hooper (1990) Journal of NIH Research 2:47-48.
This brief article describes data presented by researcher Dinah Reitman at the 1989 annual meeting of the American Public Health Association. Reitman compared the percentage of women enrolled in five categories of randomized control trials with the percentage of women affected by the corresponding medical conditions. In trials for AIDS drugs, Reitman
found that the percentages were fairly close. In trials of nicotine gum for smoking cessation, she found that women were slightly "overrepresented". In three other therapeutic areas, however, Reitman found that women were "underrepresented" in clinical studies: antiplatelet drugs for preventing stroke, drugs for mild hypertension, and drugs for myocardial infarction.
Women's Health Issues
National Heart, Lung, and Blood Institute (1990)
This publication outlines the past, present, and future of the NHLBI's acquisition and use of new knowledge about how cardiovascular disease affects women. It describes the Institute's efforts to recruit women as investigators in its research and training programs and as subjects in its clinical trials. Of the 18 NHLBI-initiated epidemiologic studies and primary prevention clinical trials active in 1990:
- 2 included exclusively women;
- 3 included between 30 and 45 percent women;
- 10 included between 50 and 58 percent women;
- 3 included exclusively men.
Is There Still Too Much Extrapolation from Data on Middle-Aged White Men?
P. Cotton (1990) Journal of the American Medical Association 263(8): 1049-1050.
The author describes how "efforts to streamline studies by using the most homogeneous population possible have filled medical libraries with data on middle-aged white men." He cites as examples the Multiple Risk-Factor Intervention Trial in 15,000 men, the Physician's Health Study (of aspirin's prophylactic effect in cardiovascular disease) of 22,071 men, and "all the large trials of cholesterol-lowering drugs," which include only men. Important gaps remain, the author notes, despite mounting documentation of important differences in drug responses and risk profiles among women, the elderly, and nonwhite persons.
Most Major Companies Test Medicines in Women, Monitor Data for Gender Differences
L. E. Edwards (1991) In: In Development: New Medicines for Women, Pharmaceutical Manufacturers Association (December)
The author sent a survey to vice presidents of regulatory affairs at 46 pharmaceutical companies; 33 (including almost every major company)
responded. All 33 companies reported that they collect data on gender of trial participants (94 percent always; 6 percent usually). Seventy-six percent of the companies reported that they deliberately recruit "representative" numbers of women for clinical trials.
Sex, Trials, and Datatapes
R. S. Ungerleider and M. A. Friedman (1991) Journal of the National Cancer Institute 83(1):16-17.
The authors examined the representation of women in federally funded clinical studies conducted by the National Cancer Institute's Clinical Trials Cooperative Group Program. In the repertoire of 444 treatment protocols active in January 1991, the only studies that specified gender were those for cancers that occur exclusively in one sex (with the exception of breast cancer protocols, which excluded men). Data from 1989 showed that 57% of all Phase II and Phase III study participants were female. The authors also noted that more women than men entered NCI clinical trials in 1989. During that year, 1.7 percent of all patients with newly diagnosed cancer entered Cooperative Group clinical trials; this total represents 2 percent of women and 1.5 percent of men with newly diagnosed cancer. The authors conclude that women are not "underrepresented" as subjects in federally funded studies conducted by NCI's Clinical Trials Cooperative Group Program.
Assessing Future Research Needs: Mental and Addictive Disorders in Women
Summary of an Institute of Medicine Conference (October 1991)
Excerpt includes discussion of inclusion of women in ADAMHA-supported research. The internal ADAMHA working group on women's health reviewed 907 grants and found that women represented 53 percent of the population of subjects participating in these 907 studies.
In Development: New Medicines for Women
Pharmaceutical Manufacturers Association (December 1991)
This 1991 survey found that 263 drugs were (at that time) being developed for use in women. The top three areas of drug development were cancer (58 medicines), gynecologic diseases (51 medicines), and cardiovascular/cerebrovascular disease (48 medicines). With 79 companies involved in these research projects, the PMA states that "virtually every major pharmaceutical company" is addressing the need to develop medicines that take into account the special medical needs of women. More than
50 percent of the research projects listed in this report were (at the time) in their final stages of development.
Bridging the Gender Gap in Research
B. A. Levey (1991) Clinical Pharmacology & Therapeutics 50:641-646.
The author conducted a "spot comparison" of clinical trials reported in the January 1981 and 1991 issues of Clinical Pharmacology & Therapeutics (CP&T) and found a decline in the number of trials restricted to male subjects and a more than doubling of the number of trials that included both men and women. When the author extended the survey to clinical trials reported in all January issues of CP&T between 1981 and 1991, no consistent pattern (i.e., increase or decrease) emerged for trials that included both men and women. The author concluded that further investigation was required to conclude definitely that there had been a true progression toward an appropriate balance of men and women as research participants in trials published in CP&T in the past decade.
Wanted: Single, White Male for Medical Research
R. Dresser (1992) Hastings Center Report (Jan/Feb):24-29.
The author claims that the failure to include women in research populations is "ubiquitous." She cites large-scale NIH-sponsored studies of heart attacks and aspirin, aging and health, caffeine and heart disease, and AIDS drugs, and even on obesity and breast and uterine cancer that have completely excluded women. The author also notes the "underrepresentation" of racial and ethnic minorities in clinical studies.
Exclusion of Certain Groups from Clinical Research
E. Larson (In press) Image: Journal of Nursing Scholarship
This paper is a descriptive, retrospective review of research protocols approved by an IRB at a major tertiary care center during a two-year (1989, 1990) period. The review focuses on demographic characteristics of subjects and was conducted using exclusions specified in the written protocols. The author found in this review that women were not "underrepresented" in clinical drug trials or other types of research. She concluded that age, race, and socioeconomic status were more likely than gender to be associated with an unjustified exclusion from research protocols—the elderly, poor, and minority groups are excluded unjustifiably from research protocols.
Sex Bias in Psychological Research: Progress or Complacency?
L. Gannon, T. Luchetta, K. Rhodes, L. Pardie, and D. Segrist (1992) American Psychologist (March):389-396.
A total of 4,952 articles published in sample years between 1970 and 1990 in the areas of developmental, clinical, physiological, and social psychology were reviewed for the purpose of assessing various indicators of sexism in human psychological research. Significant changes in sex of first author, sex of participants, sexist language, and inappropriate generalization indicated to the authors that sexism has diminished in the past two decades. Despite these improvements, however, the authors conclude that the data revealed continued evidence of discriminatory practices.
The Exclusion of the Elderly and Women from Clinical Trials in Acute Myocardial Infarction =
J. H. Gurwitz, N. F. Col, and J. Avorn (1992) Journal of the American Medical Association 268(11):1417-1422.
The authors conducted a systematic search of the English-language literature from January 1960 through September 1991 to identify all relevant studies of specific pharmacotherapies employed in the treatment of acute myocardial infarction. They searched MEDLINE, major cardiology textbooks, meta-analyses, reviews, editorials, and the bibliographies of all identified articles. They conclude that age-based exclusions are frequently used in clinical trials of medications used in the treatment of acute myocardial infarction and that such exclusions limit the ability to generalize study findings to the patient population that experiences the most morbidity and mortality from acute myocardial infarction: persons over age 75 (60 percent of deaths from acute myocardial infarction occur in persons over this age). And, since women outlive men by an average of 7.5 years, they are disproportionately represented in the elderly population. Furthermore, the authors found that studies with age-based exclusions had a smaller percentage of women compared with those without such exclusions.
In Development: AIDS Medicines, Drugs, and Vaccines
Pharmaceutical Manufacturers Association (October 1992)
With the dramatic increase in cases of AIDS in women, as well as its continued incidence in children, this PMA survey has identified clinical trials that focus on meeting the specific requirements of women and children with AIDS. The survey results show that there are 50 medicines in development that include women in the human clinical trials
and 13 medicines that include children. The survey also indicates whether products will be specially labeled for use by women and/or children.
Women's Health: FDA Needs to Ensure More Study of Gender Differences in Prescription Drug Testing
General Accounting Office (19 October 1992)
This publication reports the results of a GAO survey of pharmaceutical industry practices regarding inclusion of women in clinical trials. It concludes that:
- A quarter of drug manufacturers in the survey reported that they do not deliberately recruit "representative" numbers of women as participants in drug trials.
- Women were included in clinical trials for all drugs in the survey but were generally "underrepresented" in those trials.
- There were not enough women involved in the surveyed trials to detect gender related differences in response.
- Even when enough women are included in drug testing, often trial data are not analyzed to determine if women's responses to a drug differed from those of men.
Women in AIDS and Orphan Drug Trials
Pharmaceutical Manufacturers Association (October 1992)
Another short newsbrief relating FDA testimony at the October 5th FDLI conference. Marlene Haffner, director of FDA's Office of Orphan Products Development noted that 26 products designated as orphans are in development for diseases that exclusively or predominantly affect women. And because orphan drugs are governed by the same guidelines and regulations as other drugs, Haffner noted, "enrollment of women is pretty much along the lines of the prevalence of the disease in women." Dianne Murphy of the FDA's Center for Drug Evaluation & Research noted that more women are being actively recruited into HIV trials. Women are in some Phase I trials, she added, and some studies even include pregnant women.
Sexism in a Leading Medical Journal—A Quantitative Measure
K. Williams and E. F. Manace Borins (1993) Journal of the American Women's Medical Association 48(5):160-162.
The authors examined 160 randomly selected articles from the 1989 New England Journal of Medicine. Each article was coded using a differentiated analysis of gender bias. Sixteen components of the research process were examined. Significant (present in greater than 60
percent of the articles) gender bias was found in 12 of the 16 components. The authors conclude that medical research in 1989 was still seriously gender-biased, and therefore scientifically flawed.
Understanding the Second Epidemic: The Status of Research on Women and AIDS in the United States
H. Pham, P. Freeman, and N. Kohn (1992) Center for Women Policy Studies, Washington, D.C.
A report from the Center for Women Policy Studies that makes recommendations for woman-focused HIV/AIDS research. The report states that women, intravenous drug users, people of color, and people of low income have been "grossly underrepresented" among study subjects in ACTG clinical trials. It cites two barriers to women's participation in clinical trials: (1) many HIV-positive women had not developed a clinical diagnosis of CDC-defined AIDS and (2) federal regulations placing special emphasis on the avoidance of potential risks to fetuses.
Memo to Human Subjects Committee re: Frequency of Exclusion of Fertile Women from Drug Studies
R. A. Charo (23 March 1992) Unpublished memo.
See also: Restrictions on the Participation of Women in Drug Studies: A Retrospective Analysis, R. A. Charo (1990), unpublished paper.
Based on an evaluation of all drug study protocols submitted for review to the Human Subjects Committee at the University of Wisconsin in 1989 and 1990, this memo provides data on the frequency of exclusion of women of reproductive age from these protocols. One hundred and sixty-nine new drug studies were reviewed (this number excluded Phase I studies, studies of male-only or postmenopause-only disorders, and studies of diseases in infants). Twenty-eight of the 169 studies (16.5 percent/1 in 6) excluded fertile women. Twenty-two of these 28 studies were sponsored by drug companies. The author also analyzed the data after separating out protocols involving cancer agents and other known fetotoxic drugs.
Gender Bias in Medical Research
M. Eichler, A. L. Reisman, and E. Borins (1992) Women and Therapy—A Feminist Quarterly 12(4):61-71.
The authors apply a system of identifying gender bias in research to selected 1988 issues of medical journals representing various subspecialties of medical research: the New England Journal of Medicine, the Canadian Journal of Surgery, the American Journal of Trauma, and the
American Journal of Psychiatry. The authors state that no particular method was used in selecting the journal issues; their intent was not to compare gender bias in different subspecialties or publication outlets, but to provide recent examples of gender bias in medical research. They conclude that gender bias in medical research is pervasive.
Citizen's Petition to the Food and Drug Administration: Statement of Grounds
NOW Legal Defense and Education Fund (February 1993)
This statement describes the level of participation of women in AIDS Clinical Trial Group (ACTG) clinical trials. It states that although the percentage of female enrollment in these trials increased from 6.5 percent to 7.8 percent from 1990 to 1992, the numbers of women enrolled are inadequate to provide sound, meaningful data on the effects of a given drug on women. It cites the even greater exclusion of pregnant women from clinical trials: In a review of 74 AIDS clinical trials open for enrollment in New York and New Jersey in October 1990, 57 (80 percent) excluded pregnant women.
Women's Access to Government-Sponsored AIDS/HIV Clinical Trials: Status Report, Critique, and Recommendations
I. L. Long (March 1993) Prepared for the NIH/ORWH Public Hearing on the Recruitment and Retention of Women in Clinical Trials
The author argues that women with AIDS/HIV disease have been explicitly or implicitly excluded from most clinical trials, whether sponsored by the government or by pharmaceutical manufacturers. She provides data indicating that women are not represented in ACTU (AIDS Clinical Unit) trials proportionately with their reported incidence of AIDS. She also argues that people of color are "underrepresented" in most trials.
Determinants of Accrual of Women to a Large, Multicenter HIV/AIDS Clinical Trials Program in the United States
D. J. Cotton, W. He, J. Feinberg, and D. M. Finkelstein (1993) Journal of Acquired Immune Deficiency 6:1322-1328.
The authors describe their efforts to determine factors influencing enrollment of women in a large, multicenter clinical trials program in the United States, including attributes of participants, sites, and the trials themselves. They found that women accounted for 6.7 percent of the 11,909 AIDS Clinical Trials Group (ACTG) participants from 1987 through 1990. Women entering ACTG trials were significantly more likely to be white and less likely to have ever used intravenous drugs
than U.S. women reported with AIDS. The authors conclude that the low enrollment of women in general in these trials was influenced by demographic and geographic factors rather than attributes of specific trials. They note that an apparent positive influence of female research unit leadership on increasing enrollment of women merits further study.
Memo to Co-chairs of the NAS/IOM Committee on Legal and Ethical Issues Relating to the Inclusion of Women in Clinical Studies re: Johns Hopkins University IRB Counts
C. L. Meinert (14 April 1993)
Dr. Meinert reports on his survey of proposals approved by IRBs at the Johns Hopkins University. Dr. Meinert's survey included all active proposals, proposals pending further action by the IRB, and proposals reviewed and approved and completed or terminated within the last two years. Of a total of 2,801 proposals, 181 involved males only (6.5 percent), 265 involved females only (9.5 percent), and 2,355 (84 percent) involved both males and females.
Memo to Co-chairs of the NAS/IOM Committee on the Legal and Ethical Issues Relating to the Inclusion of Women in Clinical Studies re: NIH Inventory of Clinical Trials
C. L. Meinert (5 May 1993)
Of 293 trials listed in the 1979 NIH inventory of clinical trials, all but 25 involved both males and females. Of the 25 involving only males or females, 13 involved females and 12 involved males. All 12 of the exclusively female trials involved uniquely or primarily female conditions, whereas only 4 of the 12 exclusively male trials involved uniquely or primarily male conditions. [Note: see Dickersin and Min (1993) NIH clinical trials and publication bias. Online Journal of Current Clinical Trials doc. 50 (28 April), in which the authors examined the same data and came to similar conclusions.]
Memo to Co-chairs of the NAS/IOM Committee on the Legal and Ethical Issues Relating to the Inclusion of Women in Clinical Studies re: Clinical Trials in Controlled Clinical Trials
C. L. Meinert (5 May 1993)
Meinert describes the findings of his survey of the reported gender and ethnic mix of trials published in a journal, Controlled Clinical Trials, from its inception to the present. Of a total of 38 papers describing actual clinical trials, only 28 provided explicit statements regarding gender inclusion criteria, and among those 28, only 21 provided exact
counts of males and females. The 21 trials involved a total sample size of 78,840, of whom 72,951 were males and 5,890 were females. The results provide some evidence for a predilection for male-based trials, at least among those authors electing to publish in this journal (or perhaps among the journal's editors). There was very little information about the ethnic mix of the trial populations.
Letter to Curtis Meinert Regarding Gender Representation in NEI Clinical Trials
National Eye Institute; National Institutes of Health (20 May 1993)
The letter states that significant numbers of women are included in a variety of NEI-supported trials. In almost all NEI trials, women comprise at least 40 percent of the participants. Women represent 77 percent of the subject population in trials for treatments of optical neuritis, a condition known to be more prevalent in women than in men.
Memo to Co-chairs of the NAS/IOM Committee on Legal and Ethical Issues Relating to the Inclusion of Women in Clinical Studies re: On Gender Coverage at Johns Hopkins University
C. L. Meinert (21 May 1993)
This memo describes an analysis of gender-specific studies at Johns Hopkins University that were identified in a previous memo (14 April). Gender-specific studies were classified as to whether or not the focus on one gender has a biological or disease basis, as determined by the title of the project. A higher percentage of male-only studies were found to be arbitrarily male-only, as opposed to female-only studies that were arbitrarily female-only (75.2 percent and 21.2 percent, respectively). Also found was a greater propensity to study females than males overall (3:1), as well as reproductive or sex-specific diseases (47:1 and 6:1, respectively).
Letter to Curtis Meinert Regarding Gender Representation in Clinical Trials Performed by the National Heart, Lung, and Blood Institute
National Heart, Lung, and Blood Institute (26 May 1993)
This memo describes data that Dr. Meinert received from the director of the National Heart, Lung, and Blood Institute in response to his request for information about clinical trials under way at the Institute. The figures are as of May 1993. They include a total of 49 trials, one of which is being done in males only and eight of which are being done in females only. The gender mix in the remaining trials ranges from a low of about 10 percent female to about 75 percent female.
Women's Representation as Subjects in Clinical Studies: A Pilot Study of Research Published in JAMA in 1990 and 1992 (in Volume 2 of this report)
C.E. Bird (May 1993) Paper prepared for the IOM Committee on the Legal and Ethical Issues Relating to the Inclusion of Women in Clinical Studies
In this study, all original articles reporting results of clinical studies in two recent years (1990 and 1992) of the Journal of the American Medical Association (JAMA) were examined (see commissioned paper by Bird for full details). Articles were classified by the percent of women included as subjects (grouped into five categories: 0 percent, I percent<x<33 percent, 34 percent<x<66 percent, 67 percent<x<99 percent, 100 percent), by major disease/treatment category, and by major design methodology. Two definitions of gender underrepresentation were used: a "strict" definition: 0 percent (either gender) and a "lenient" definition: one-third or fewer (either gender). There were 243 articles that met the definition of a clinical study and had enough information to classify gender of their subjects. Of these, 207 related to gender-neutral diseases (disease can occur in both genders). The results reported here pertain to these 207 articles studying gender-neutral diseases. (Note: "gender-neutral" disease as used in this study does not imply that the incidence of disease is equal in both genders, but merely that it is reasonably common in both genders. Breast cancer, for example, is not considered to be gender-neutral because it occurs very rarely among men.)
Among clinical studies studying gender-neutral diseases, 49 percent had samples with women representing between one-third and two-thirds of their subjects. Among the remaining 51 percent, there was some fairly strong evidence to suggest that women were more likely to be underrepresented as research subjects compared with underrepresentation of men. Specifically, 17 percent of the studies had no women compared with 6 percent with no men (ratio of 4.3); 38 percent had one-third or fewer women subjects compared with 14 percent with one-third or fewer men (ratio of 2.7).
There were three types of methodological designs with at least 40 articles concerning gender-neutral diseases: cross-sectional (45), longitudinal (100), and RCTs (40). For all three types of designs, about 50 percent had samples with women representing between one-third and two-thirds of their subjects. For simplicity's sake, the remaining articles are presented as the ratio of those which underrepresent women to those which underrepresent men. Using the strict definition of
underrepresentation, this ratio was 0.8 for cross-sectional studies, 3.6 for longitudinal studies, and 5.0 for RCTs. Using the lenient definition, this ratio was 2.5 for cross-sectional, 2.8 for longitudinal, and 3.4 for RCT studies. Thus, there was some evidence to suggest that there are some systematic differences among design methodologies if the strict definition is used (with RCTs being most likely to exclude women in comparison to excluding men), a tendency that was reduced when the lenient definition was used.
One reason for underrepresentation of women in gender-neutral diseases may be that the disease is less common among women than men. Using the strict definition of underrepresentation, the 46 single-gender studies examining gender-neutral diseases were categorized by the primary basis for excluding one gender: prevalence (for example, the disease occurs disproportionately in one gender or the particular vector or risk factor of interest was gender-specific), convenience (for example, the population was veterans or prisoners, data gathering was easier in one sex, secondary analysis of data from sex-specific study), or no discernible rationale.
Several findings emerged. First, overall, the choice of single-gender populations could be rationalized by either the prevalence of the disease or sampling convenience in most instances (over 85 percent). The remaining 15 percent of studies had no apparent rationale, either offered by the authors or inferred on the basis of the disease or site of study. This percentage remained about the same for both subsets of male-only and female-only studies as well. Female-only and male-only studies, however, appeared to differ systematically by whether the basis of the choice was disease prevalence (75 percent of female-only studies compared with 41 percent of male-only studies) or convenience (8 percent of female-only studies compared with 47 percent of male-only studies). Partial explanations for this imbalance were that 53 percent of all studies for which convenience was the primary basis examined the (almost exclusively male) veteran population and another 24 percent of these studies consisted of secondary analyses of single-gender studies, which tended to be all-male.
These findings suggest that one important reason for a tendency for male-only studies to predominate is the differential opportunity for men to be in positions where clinical studies are likely to be funded and carried out (for example, receiving treatment in a Veterans Administration Medical Center or as a member of the armed services or as a prisoner) which in turn can create further imbalance as researchers seek
to take advantage of databases already collected. Perhaps one way to redress this source of gender imbalance in clinical studies is to fund studies in institutional settings where women predominate, e.g., nursing homes, hospital employees, or primary grade teachers.
This study suggests that, among those studies examining gender-neutral diseases, women were more likely than men to be underrepresented. The reasons why women tend to be underrepresented—such as incidence of the disease by gender or convenience of samples by site, such as the Veteran Administration Medical Center population or men's prisons—may help explain the scientific rationale on a study-by-study basis. Nonetheless, viewed as a whole, there is evidence to suggest that the end-result of these individual decisions is to have fewer women in clinical studies of diseases common in both genders.
Letter to Anna Mastroianni from Iris J. Schneider Regarding the Inclusion of Females and Males in Clinical Trials Supported by the National Cancer Institute
National Cancer Institute (23 June 1993)
This memo provides subject counts for NCI clinical trials in fiscal year 1992. Of 22,483 participants in treatment trials, 12,490 were females and 9,993 were males. Of 9,553 participants in prevention trials, 4,727 were females and 4,826 were males. No information is provided regarding the conditions studied.
Women in Clinical Trials: HIV-Infected Women
D. Murphy (1993) Food and Drug Law Journal 48(2):175-179.
This article summarizes a presentation given by the author at the Food and Drug Law Institute's seminar, ''Women in Clinical Trials of FDA Regulated Products: Who Participates and Who Decides?" on October 5, 1992, in Washington, D.C. She notes that in 1991, 13 percent of reported AIDS cases occurred in females. As of August, 1992, females made up 13.2 percent of the total ACTG trial population. Of the adults in trials, women made up 10.7 percent. She describes the improvement in recruitment of women over time: where women made up 6.8 percent of the population in the first 20 ACTG protocols, but they made up 15.7 percent of the population in the last 20 ACTG trials, with significant enrollment numbers.
Update on Clinical Trials and Pharmaceutical Regimens for Women with HIV Infection
J. A. Korvick and I. Long (1993) In Until the Cure: Caregiving for Women with HIV, A. Kurth, ed.
An analysis from the National Institute of Allergy and Infectious Diseases (NIAID) Division of AIDS. The authors report that the number of females in ACTG clinical trials increased from 2 percent in 1986 to over 18 percent in 1992.
Guideline for the Study and Evaluation of Gender Differences in the Clinical Evaluation of Drugs; Notice
Food and Drug Administration (1993) Federal Register 58 (139):39412 (22 July 1993)
In this notice of proposed policy revision, the FDA noted that the effect of the 1977 guideline that excluded women of childbearing potential from participation in early studies of drugs has been that women generally have not been included in Phase I nontherapeutic studies or in the earliest controlled effectiveness (Phase II) studies except for studies of life-threatening illnesses. The notice includes accounts of FDA surveys of new drug applications (NDAs) undertaken in 1983 and 1989. These are summarized below:
Food and Drug Administration Survey of New Drug Applications (NDAs), 1983
Carried out primarily to assess the inclusion of elderly persons in new drug applications (NDAs), this survey examined the age and gender of patient populations included in 11 pending NDAs. The NDAs were chosen because they were readily available and did not need to be retrieved from storage. The survey showed that nonsteroidal anti-inflammatory drugs were studied predominantly in women (because arthritis is more common in women), that this predominance was slightly less in the case of a pain medication, and that a hypnotic drug and two antibiotics were studied in equal proportions of men and women. The patient populations in the NDAs for two drugs to treat heart disease were about two-thirds male. About two-thirds of the patients in these studies were less than 60 years of age, an age group in which heart disease is more prevalent in men than women. In patients over 70, the gender distribution was about equal. Studies of a drug to treat duodenal ulcer, a predominantly male condition, included about 75 percent males. The two anti-cancer drugs in the survey were studied principally for exclusively male conditions, prostate cancer and testis cancer.
Food and Drug Administration Survey of NDAs, 1988
In an effort to examine selection bias, the FDA surveyed all drugs approved in 1988 (with the exception of 4 orphan drugs, 3 contrast agents for single dose uses, and a topical product for which gender distribution was not available). Study populations for an anti-inflammatory drug and a drug for prevention of vascular spasm after subarach-
noid hemorrhage (both female-predominant conditions) were primarily female. Studies of two cardiovascular drugs included 59 and 67 percent men, reflecting the predominance of angina, "and perhaps hypertension," in men under age 60 (two-thirds of patients were under age 60). An intravenous antibiotic was studied mainly in elderly patients; for unknown reasons, about two-thirds of the patients studied were male. One topical was also studied somewhat more in males for unknown reasons.
Underrepresentation of Women in Clinical Drug Trials
D. L. Schmucker and E. S. Vesell (1993) Clinical Pharmacology & Therapeutics 54:11-15.
The authors performed a quantitative survey of gender distribution in test populations of trials reported in all issues of Clinical Pharmacology and Therapeutics (CP&T) during the periods from 1969 to 1971, 1979 to 1981, and 1989 to 1991, and the British Journal of Clinical Pharmacology (BJCP) during the periods from 1979 to 1981 and 1989 to 1991-an interval encompassing 22 years and including 1,947 articles. Not included in the survey were trials on contraceptives or drugs designed for gender-specific diseases or conditions, case histories, and single-subject trials. The authors found that the percentage of trials reported in CP&T that included men only increased from 27 percent to 38 percent from 1969-1971 to 1989-1991. A similar comparison in the BJCP from 1979-1981 to 1989-1991 yielded a 5 percent increase. In both journals during these same periods, the percentage of women-only trials declined. In neither journal was there a statistically significant increase in the percentage of trials that included both men and women. During these same periods, the number of trials that did not declare subject gender decreased by 57 percent in CP&T and 23 percent in BJCP.
To estimate the frequency of gender-related differences in drug effects, the authors surveyed all clinical trials published in CP&T during 1991 that included both men and women for documented gender differences in drug responses. There were 68 such trials, and none claimed differences in drug response that were attributable to gender. The majority of trials (>60 percent) failed to mention whether or not the data were analyzed for gender differences.
The authors also examined for gender-related differences in efficacy or toxicity all drugs approved by the FDA in 1981 and 1991 and that were listed in the 1992 Physicians Desk Reference (PDR). Data cited under
the headings of Clinical Pharmacology, Contraindications, Precautions, and Adverse Reactions were reviewed for recommendations for restricted use of the drug in nonpregnant women and pregnant women. The survey revealed reservations concerning use during pregnancy, but not in nonpregnant women, for nearly all drugs approved in 1981 and 1991. The authors note that the values are difficult to interpret, however, because absence of any contraindications for drug use in nonpregnant women may reflect (1) no evidence of gender differences, (2) exclusion of women from test populations, or (3) failure to analyze clinical trial data for gender differences. The authors conclude that despite efforts to rectify the underrepresentation of women as participants in clinical trials, this practice has continued during the past decade.
The Right to Participate in Research Studies
M. L. Elks (1993) Journal of Laboratory and Clinical Medicine 122:130-136.
The author reviewed three journals to determine the proportion of women in clinical (nongonadal) studies. In volume 51 of Clinical Pharmacology & Therapeutics (January through June 1992, chosen because of publication of clinical studies of new drugs), 49 studies were reported; 14 (29 percent) included no women and 2 (4 percent) included no men; none of these articles noted this exclusionary status in the titles. The remaining 26 studies had an average of 59 percent male participants. In volume 263 of the American Journal of Physiology: Endocrinology and Metabolism (July through November 1992, chosen because of studies of metabolism/physiology), 32 studies were reported, with 10 (31 percent) including only men (one so stating in the title), 1 (3 percent) including only women, and 4 (12.5 percent) giving no statement of the gender of participants. The 16 remaining studies had an average of 57 percent male participants. In volume 19 of Hypertension (January through June 1992, chosen because hypertension is more common in women than in men), 20 studies were reported; 8 (40 percent) included no women (one title so stated) and 3 were large epidemiologic studies with equal representation. The 9 remaining studies had an average of 64 percent male participants. In the 38 rat studies in this journal, 7 did not state the sex of the rats, 26 (68 percent) used males only, 1 (3 percent) used females only, and 4 (11 percent) included both. None of the titles indicated the gender of the animals studied. The author concludes that frequent systematic exclusion of females has occurred both in human and animal studies. The author also notes that even in most of the both-sex studies, notably more men than women have been included than would be likely by chance.
Inclusion of Women and Minorities in Occupational Cancer Epidemiological Research
Zahm, S.H., Pottern, L.M., Lewis, D.R., Ward, M.H., and White, D.W. (In press). Submitted to Journal of Occupational Medicine.
Of a total of 1,233 studies, 562 (46 percent) included only white men, while the remaining 671 studies (54 percent) included subjects from other race-gender groups. Of these, 35 percent included white women, but only 14 percent presented any analyses of the white women specifically. The proportions with analyses of non-white women (any: 2 percent; detailed: 1 percent) or non-white men (any: 7 percent; detailed: 3 percent) were also small. Studies with detailed analyses of women and minorities tended to use weaker methodologies (i.e., proportionate mortality or cross-sectional design) than the studies of white men and were less able to provide convincing data on the occupational cancer risks of women and minorities.