In Her Own Right: The Institute of Medicine's Guide to Women's Health Issues (1997)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

C HA P T E R 9 Research on Women,Women in Research 9 Research on Women, Women in Research A merican biomedical research today ranks as the finest in the world; the finest, in fact, in the history of the world. But is this massive, prodigiously productive enterprise doing all it can to promote women’s health? For a number of years now, critics have insisted that it does not, that several kinds of bias have kept women and women’s health issues from getting their fair share of the nation’s scientific resources. And because of this, they charge, women experience poorer health care and thus poorer health. The neglect, they allege, takes two main forms. Under the guise of protecting women and children from harm, restrictive rules have excluded reproductive-aged women—the bulk of the adult female popu- lation—from serving as subjects in clinical studies, including those investi- gating diseases, treatments, and medications common among women. In addition, important health concerns like breast cancer, osteoporosis, meno- pause, and even pregnancy and childbirth, have gotten markedly less fund- ing and attention than their significance in women’s lives would justify. And relatively few women belong to the ranks of scientists and policymakers who determine research priorities. But society has moved beyond attitudes that sanction gender- based exclusions and inviolate medical authority. Following the lead of AIDS activists, formerly excluded groups now assert the right to partici- 179 

I N H E R O W N R I G H T pate in trials of the latest experimental treatments. In keeping with femi- nist thinking, women’s advocates now insist on a prominent place for female issues and investigators. Together, these demands have triggered far-reaching reexamination of who participates in research, both as sub- jects and as scientists. “For the nation’s research agenda to be just, it must ensure that medical research promotes the health and well-being of both men and women,” states IOM’s Committee on the Ethical and Legal Issues Relat- ing to the Inclusion of Women in Clinical Studies, which was appointed to examine the policies and practices that govern who gets to participate in and who gets to perform the clinical research that provides so much of our new medical knowledge.1 But its extensive study revealed a number of extrascientific factors that also shape research. “What many perceive to be the relative inattention to the study of health problems experienced primarily by women may arise from subtler (yet powerful) forces such as gender and race biases that permeate both society and scientific research,” the committee believes.2 In addition, the structure and funding of aca- demic medicine have contributed to the neglect of important women’s health issues. HOW UNFAIR HAS SCIENCE BEEN? Clinical studies—research projects that use human beings to learn about medical conditions and the effects of various treatments and interventions—provide much of the new information that helps physi- cians improve care. This work plays a particularly crucial role in deter- mining which treatments work best. The normal course of a disease; the usefulness of particular diagnostic markers; the safety and efficacy of medi- cations, surgeries, and many other kinds of treatments show up most convincingly through systematic observations of large, representative groups of people. The randomized clinical trial—a sophisticated experi- ment that compares the experience of one statistically valid sample of persons (randomly selected to undergo a particular treatment or interven- tion) with a comparable sample of persons randomly selected to be con- trols—is such an effective method that researchers and clinicians alike 180 

C HA P T E R 9 Research on Women,Women in Research consider it the “gold standard” of proof. Treatments proven by this hon- ored technique claim an authority that those without this backing lack. Competing good intentions clash in the question of who should participate in these trials: science’s obligation to protect experimental sub- jects from harm versus people’s desire to gain the benefits of participating in studies. On the one hand, medicine bears the duty, as old as the Hippo- cratic Oath, to “first, do no harm.” And, a series of medical catastrophes during the past century has impressed on researchers just how tricky that duty is to fulfill. For a variety of biological and cultural reasons, women ranked high on the list of the vulnerable victims who seemed to need protection. But modern science, on the other hand, has also taken on the obligation to provide ever better care. And treatments can improve only to the extent that they are tested on individuals like those who will re- ceive them in clinical practice. Even the best therapy may not work as well in persons who differ from the test population in any relevant way, be it biochemically, physiologically, genetically, or behaviorally. Thus, any group—such as women or racial minorities—that is systematically underrepresented in clinical studies runs the risk of receiving treatments tested on people unlike themselves and thus potentially less effective or even harmful. So has biomedical research in fact slighted women? Has their health suffered as a result? Some experts answer with an unequivocal yes. “The historical lack of research focus on women’s health concerns has compromised the quality of health information available to women as well as the health care they receive,” found a U.S. Public Health Service task force in 1985.3 But when the inclusion committee attempted to measure the extent of the neglect, a more complicated picture emerged. Misguided good intentions have in some cases definitely translated into inadequate medical information and resulting inferior care. But in many others, inad- equate data and research design prevent any definitive answer. Biomedical science, ironically enough, has historically paid so little attention to gender issues that even a very extensive review of the literature could not prove systematic underrepresentation of women. So few studies noted the gender of their subjects, analyzed their data by 181 

I N H E R O W N R I G H T gender, or even recruited the numbers of women needed to do those analyses that accurate estimates of the full extent of women’s under- (or in some cases over-) representation proved impossible. Thus, the committee could only conclude that “the available evidence is insufficient to deter- mine whether women have participated in the whole of clinical studies to the same extent as men and whether women have been disadvantaged by policies and practices regarding their participation or a failure to focus on their health interest in the conduct of research.”4 The committee did find two major—and life-threatening— exceptions. Though unable to “establish that gender inequity existed in the whole of past clinical research,” it did find “evidence relevant to this issue in two areas of disease research: AIDS and heart disease.”5 Important research projects of these two major killers of females “either exclude women altogether or include them in numbers too small to yield mean- ingful information about their treatment. . . .”6 Coronary heart disease, “the single most frequent cause of death in women,” and other heart ailments kill 500,000 American women each year and hospitalize 2.5 million. “Yet despite these compelling statistics,” the committee notes, “a perception persists among the public and physi- cians that women do not contract heart disease as often as men and when they do, it is not as serious.”7 Though this misconception both results from and causes gen- der bias, it gained credence from an impeccable research source, the highly influential Framingham project. For 12 years starting in 1948, this pio- neering look at heart disease followed Massachusetts men and women aged between 36 and 68. Three times more males than females died of CHD during those years, leading researchers to the quite reasonable con- clusion that middle-aged men have much greater propensity than women to death by coronary. But many also leaped to the unproved corollary that women therefore enjoy some sort of “protection” against heart disease. Had the researchers trailed their subjects for another decade or so, a more complicated truth would have emerged: whatever “protec- tion” women possess during their reproductive years begins to vanish with menopause. Though they rarely suffer a coronary event before age 65, their heart attack rate rises sharply by 75. “Thus, a failure to include people over 65 in clinical studies of heart disease has serious ramifications 182 

C HA P T E R 9 Research on Women,Women in Research for women,” the committee found, because women “develop the disease later than men and are 20 years older at their first MI [myocardial infarc- tion—a type of heart attack].”8 Probably the most serious of the ramifications is that research- ers, assuming they were studying a disease that mainly threatened men, concentrated their subsequent investigations of both the disease process and potential treatments almost exclusively on one gender. During the next two decades a series of influential clinical studies—the Multiple Risk Factor Intervention Trial (MRFIT), the Coronary Drug Project (CDP), the Lipid Research Clinic, and the Physicians’ Health Study, used tens of thousands of men to analyze such crucial factors as the role of cholesterol and other fats in the bloodstream, the effects of fat-lowering drugs, and the potential of low-dose aspirin to prevent heart attacks. Together, these projects shaped clinical thinking as well as practice. But by excluding women, the studies denied doctors definitive information about how to prevent, diagnose, and treat heart disease in more than half the population at risk. Extrapolating from MRFIT, CDP, and other male-based results ignores important facts about females. Estro- gen, for example, may fight fatty buildup in the arteries. When heart disease does strike a woman, it often manifests itself differently from in a man. The major risk factors—high blood pressure, cigarette smoking, and obesity—hold for both genders, but certain of the familiar male early warning signs do not. In a man, coronary heart disease usually first shows up as a myocardial infarction, but in a woman as the chest pain called angina pectoris. Because women with angina may go years without an MI, many people see the female form of the disease as “less serious,” even though, when an MI does finally strike a woman, she is likelier than a man to die.9 Do later diagnosis and less appropriate treatment contribute to that female higher death rate? No one knows for sure, but studies suggest that the treadmill test and other standard diagnostic tools developed on men may work less well on women. What’s more, women receive fewer of the high-tech invasive procedures like coronary bypass surgery than equally sick men.10 And when women do undergo that operation, they survive it less often, possibly because techniques perfected in male hearts and arteries work less well in women’s smaller organs and vessels.11 183 

I N H E R O W N R I G H T Gender bias stands out even more starkly in AIDS. Researchers initially conceptualized the disease as an affliction of gay males, even though it appeared in women as early as 1981 and since 1986 has been rising faster in them than among men. The focus only gradually widened to include infants and male drug users. Studies that involved women much more often considered them as vectors of their babies’ or their mates’ infections than as victims in their own right.12 Not until 1994 did the National Institutes of Health undertake a major project to map the disease’s course and signs in women. More typical have been studies of how women transmit the virus to sex partners or offspring. One 1990 effort tried to determine whether various methods of administering AZT could cut infection of babies during childbirth. The project’s original design provided no health care for the HIV-infected mothers; a revised protocol continued the moth- ers’ AZT for 6 weeks after birth, but the babies’ for 18 months.13 “A review of the literature yields only a handful of papers focusing on the consequences of the infection in nonpregnant women,” noted a critic in 1992.14 Questions slighted by researchers include such basic ones as how males transmit the infection to females and how women can protect them- selves during sex. Much too little is known about the sexual practices that spread the virus among heterosexuals. This blindness to women as legitimate AIDS sufferers and not simply instruments of other people’s suffering has also denied them treat- ment opportunities available to men. The very definition of the disease developed by the Centers for Disease Control and Prevention and used to apportion disability benefits and spots in treatment programs long ignored conditions common in women. The treatments used in those programs, furthermore, had been tested for effectiveness on men rather than women. Nearly all studies of how people cope with the disease have concentrated on gay men. “Finally,” notes the inclusion committee, “in clinical trials of AIDS drugs, which often may provide significant sources of first-rate medical care and access to experimental treatments for persons with AIDS, the numbers of women participating lags behind expectations for a disease that is increasing the most rapidly among women.”15 Another life-threatening area where research lags is women’s addictions. “The classical studies that have informed our thinking about 184 

C HA P T E R 9 Research on Women,Women in Research alcohol,” Blume observes, such as “George Vallian’s wonderful book, The Natural History of Alcoholism, Chuck Snyder’s very interesting book, Alco- hol and the Jews, should really be entitled Natural History of Alcoholism in Men and Alcohol and Jewish Men, because neither study contained a single subject that was female.”16 The course of the disease in women remains unclear, as does its specifically feminine relationship to depression. Only recently have screening instruments suitable for women become available, although none exists yet that adequately deals with teenage girls. Sometimes missing knowledge means great danger. “There have been many studies of the techniques of detoxification from alcohol, books full. There is none that I could find on the best way to detoxify a pregnant woman,” Blume laments. “Of all the phone calls I get from psychiatrists and other physicians around the country, that is the most common question.”17 Pregnancy is only the most dramatic of the special situations about which researchers know very little. Almost no research exists on the issues that alcoholism raises in women prisoners, students, HIV patients, and seniors, or among lesbians or minority women. “We know that sociocultural factors are important in shaping the development of women’s alcohol problems, but we know nothing about how to apply these factors to case finding, to treatment.”18 HOW HAS SCIENCE BEEN UNFAIR? Treating women unfairly shortchanges their health in two ma- jor ways. First and most obviously, science knows less than it could about both conditions that affect only women and how the two genders experi- ence conditions that affect them both. Does a given disease act identically? Do the different physiologies and biochemistries of males and females— and of females at various stages of the life course—affect diagnosis or response to treatment? Is a drug, surgery, or other treatment proven useful for one gender equally suitable for the other? If studies do not include adequate numbers of both men and women, these questions go unan- swered—and usually even unasked. But asking these questions each and every time is not neces- sary, many experts believe. Both researchers and clinicians generally as- sume that drugs effective on one group of people will work equally well 185 

I N H E R O W N R I G H T in others, barring a compelling biological reason to the contrary.19 “Basi- cally, the question can be rephrased, ‘Will differences in drug disposition or drug response due to gender be clinically significant?’ ” writes Leslie Z. Benet, Ph.D., chairman of the Department of Pharmacy at the University of California, San Francisco. “My answer is, ‘Rarely, if ever.’ ”20 Most drugs available today, Benet notes, have a wide enough margin between a therapeutic dose—the amount or blood concentration needed for a beneficial effect—and the toxic dose—the amount that is poisonous—to accommodate quite a wide range of individual differences. The effects of drugs with narrower margins (known technically as the therapeutic index) between potency and peril tend to vary considerably among individuals, so doctors find the right dose by titrating them indi- vidually for each patient. “It is my hypothesis,” Benet argues, “that the health consequences of the exclusion or underrepresentation of women in clinical studies for narrow therapeutic index drugs will be minimal, since it is immaterial whether women differ significantly from men . . . since in each case the drug will be titrated by the clinician to the appropriate dose or concentration.” And even in wide therapeutic index drugs, “there should be few health consequences for large gender differences . . . since the variability inherent in the male population most likely already encom- passes the difference between the male and female populations.”21 Still, the “known biological, behavioral and social differences between the genders . . . are real and significant,” the inclusion committee insists, and “can affect response to drugs and other treatments, creating the potential for differential health consequences. . . .”22 Benet himself favors including women in trials, because of differences like the recently discov- ered variation in drug metabolism between pre- and postmenopausal women. And male-oriented medicine does cause problems in areas other than heart disease, addiction, and AIDS. Women seem to suffer more adverse drug reactions.23 Even in apparently well-understood conditions, physicians may hesitate to give females drugs or treatments proven safe only in males, depriving them of some of the newest advances.24 More strikingly still, enormously common female conditions like fibroid tumors and preeclampsia of pregnancy remain poorly understood. Women even lack clear medical guidance for dealing with the universal and normal 186 

C HA P T E R 9 Research on Women,Women in Research hormonal changes of menopause. “When it comes to health problems like these involving obstetrics and gynecology, our lack of knowledge is enor- mous,” writes Mary Lake Polan, M.D., chair of obstetrics and gynecology at Stanford University Medical School and member of an IOM commit- tee studying research efforts in academic OB/GYN departments. “Why are babies born prematurely? Why have ectopic pregnancies, those outside the uterus, increased every year since 1970, with a fatality rate of 42 per 1,000 cases? Why hasn’t our understanding of sexually transmitted diseases like herpes, genital warts and AIDS kept up with the sexual revolution? Why? Because our effort to find answers has been half-hearted and inad- equate.”25 WHY EXCLUSION? In keeping women out of the great majority of clinical studies, scientific policymakers intended to do good rather than harm, to decrease rather than increase the sum of human suffering, to protect a vulnerable population—or, more accurately, two vulnerable populations—from what they perceived as great but unknown dangers. Although some clinical research involves nothing more dangerous than comparing the medical records of different groups, much of it entails experiments that may pose unknown hazards. And this century has a sordid history of human experimenta- tion. “Public outcry and tragedy have driven policy development in the area of research on human subjects,” note ethicist Tracy Johnson, M.A., of the Bass & Howes consulting firm and historian Elizabeth Fee, Ph.D., professor of history and health policy at Johns Hopkins School of Hygiene and Public Health.26 In 1938, after a drug called Elixir Sulfanilamide killed 107 people, Congress required, in the Food, Drug and Cosmetics Act of 1938, that manufacturers prove medications safe before bringing them to market. That meant they had to test them on humans. Under the previous Food and Drug Act, passed in 1906, proof of strength and purity had been the only requirements. But human experimentation gained a particularly gruesome reputation in the 1940s when the Nuremberg war crimes trials revealed to a horrified world the extent of atrocities committed by Nazi doctors on 187 

I N H E R O W N R I G H T helpless prisoners in the name of medical science. Growing out of world- wide revulsion at this officially sanctioned savagery, the Nuremberg Code guaranteed to every person acting as a research subject the right of free and fully informed consent to any risks entailed in any study. The code further stipulated that subjects must possess the capacity to give that con- sent, that human experimentation must take place only when potential benefits outweigh risks, and that there must be no other possible method to obtain the desired information. But these provisions did not end exploitation of the vulnerable. In the 1960s and 1970s a shocked American public learned of abuses that bore disturbing parallels to wartime excesses. In 1963 an experiment using live cancer cells on unwitting elderly patients came to light at the Jewish Chronic Diseases Hospital in Brooklyn, New York. Two years later, the Harvard anesthesiologist Henry K. Beecher gained national attention with a speech and subsequent paper in the New England Journal of Medicine describing published studies that did not fulfill the ethical requirement that subjects understand risks. “It must be apparent,” Beecher wrote, “that they would not have been available if they had been truly aware of the uses that would be made of them.”27 Next, with racist overtones reminis- cent of the Nazi period, came revelations about the Tuskeegee Syphilis Study. This infamous government-funded project had since 1932 ob- served the disease’s course in a group of some 400 African American men. Although antibiotics had become the widely available, well-established treatment for syphilis midway through the project, researchers had failed to treat some of these impoverished, uneducated men or even to inform them of the possibility of treatment. Public outrage led to a number of legislative and regulatory reforms tightening procedures and protections for experimental subjects in general. Two other traumatic episodes soon focused attention on the risks to women (and their babies) in particular. Though the tragedies involving thalidomide and DES did not involve research subjects, they drastically heightened fears of drug effects during pregnancy. Together they alerted scientists and terrified the public about the perils of exposing pregnant or even potentially pregnant women to unproved drugs. Both disasters involved freely available medications whose ef- fects during pregnancy, as events tragically revealed, had not been suffi- 188 

C HA P T E R 9 Research on Women,Women in Research ciently tested. In both cases, in fact, early signs of impending trouble had been ignored. In the early 1960s the birth of nearly 8,000 severely de- formed infants, born mainly in Europe, riveted the world’s attention on thalidomide, an over-the-counter sleeping tablet their mothers had taken while pregnant. Though sold in 20 countries, it failed to receive FDA clearance for sale in the United States because of a regulator’s suspicions about early results. Still, although this country missed the disaster’s main brunt, some “investigating” doctors got it from the manufacturer and distributed it here. Despite the small number of cases on this side of the Atlantic, Americans, like people around the world, experienced “a pow- erful emotional impact that created an aversion to involving pregnant women and women of childbearing age in drug research,” the inclusion committee notes.28 Then, as if to underline the point, a second drug-related disas- ter, this time involving large numbers of American families, came to light less than a decade later. The artificial hormone DES, widely prescribed during the 1940s and 1950s to prevent miscarriage, did less visible but still grievous damage. A rare form of vaginal cancer began appearing in “DES daughters”—girls exposed to the drug in utero—as they reached young womanhood. Neither of these calamities resulted, as Johnson and Fee point out, from dangerous research on expectant mothers. The blame lay on “inadequate testing and ignored research results which indicated the poten- tial for reproductive harm in women.” Logically, then, these cases would seem to argue for more, rather than less, research into drug effects in pregnancy, provided it could be conducted with a stringent regard for safety. In the ensuing uproar, though, a “counterintuitive” result won out: strong discouragement for future inclusion of anyone who was, or conceivably could become, pregnant. Ironically, though, because a certain number of pregnant women will always need medical treatment, this protective approach prevents gathering the very information needed to forestall future drug disasters. “Presumably, an all-male study population would not have shielded the manufacturers of thalidomide from liability claims of a drug marketed to women,” Johnson and Fee observe.29 The net result of these calamities was a tightening of regula- 189 

I N H E R O W N R I G H T tions and policies governing participation in research studies, especially by women of “reproductive potential.” The FDA recommended that, except in the case of life-threatening illnesses, no one in that category take part in any early-stage drug trials, which pose the greatest risks of toxicity, re- gardless of her current marital status, condition of pregnancy or nonpregnancy, state of fertility, future childbearing plans, or personal de- sires. That policy also discouraged women’s exclusion from the later-stage testing needed to determine effectiveness and appropriate dosage. And beyond official requirements, drug companies established exclusionary policies of their own. Stunned by the huge liabilities of previous disasters involving pregnant women, they eschewed research subjects who could expose them to any similar catastrophe in the future. TOWARD INCLUSION During the very years that the exclusionary consensus was building, however, almost unnoticed amongst the advancing restrictions, a small movement of dissenters took its first tentative steps. The DES rev- elations coincided with the upsurge of the modern feminist movement. Small groups of activist women began to work for greater control over their own health and health care. Such early advocacy groups as the Bos- ton Women’s Health Collective and the National Women’s Health Net- work started to challenge the assumptions underlying a health care system they saw as paternalistic and unsympathetic to women’s concerns. In 1973 the Boston group’s book Our Bodies, Ourselves brought this view to a nationwide readership. Feminist health activists saw the issues behind scandals over DES and the Dalkon Shield as involving (because of reports that, for example, women were given DES without their knowledge), in Johnson and Fee’s words, “inappropriate inclusion (rather than exclusion).”30 Re- jecting the general panic, they called for more rather than less research into the common but understudied conditions that affect women. Rather than the moratorium on new knowledge that protectionist exclusion im- plied, they sought both more research and greater protections for those who would participate. But the main drive to overturn the nation’s protectionist poli- cies came not from women but from men, specifically, the male homo- 190 

C HA P T E R 9 Research on Women,Women in Research sexuals responding to the dearth of effective treatments for AIDS. Like all drugs under FDA supervision, unproved AIDS treatments were subject to strict safety rules that kept them out of the hands of most patients and many out of this country altogether. From the early years of the epidemic, activists denounced as a cumbersome, deadly waste of time the very drug approval system established by public demand to ensure safety in the wake of past abuses. Like the proverbial general fighting the last war, FDA used a strategy designed to prevent another thalidomide disaster. It only per- mitted drugs to come to market after a long testing process intended to protect subjects and the public from all possible risks. AIDS activists, however, had no interest in being protected. They represented not healthy people seeking to avoid unnecessary injury but desperately ill people seeking relief from an incurable and inexorably fatal disease. They wanted to try new drugs, and they wanted them fast, so they willingly tolerated risks unacceptable to ordinary people. Through the lens of their disease, doomed victims saw experimental drugs not as unknown perils but as potentially lifesaving opportunities, and established procedures not as necessary prudence but as needless impediments. AIDS patients’ only hope, they argued, lay in unproved but promising treat- ments inching their way through the FDA’s obstacle course. Without access to these possibly lifesaving substances, sufferers faced an inevitable doom. In these desperate new circumstances, the AIDS movement in- sisted, old safeguards were less than useless; they were positively harmful. In rejecting protectionism for its own sake, the AIDS activists echoed a watershed document that had appeared several years before the epidemic entered scientific consciousness. “Beneficence . . . requires that . . . we be concerned about the loss of substantial benefits that might be gained from research,” stated the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research in its 1978 Belmont Report.31 Protection was necessary, the report acknowledged, but the true purpose of research is progress that can lessen suffering. To balance risk against benefit, the report enumerated three ethical principles. The first, respect for persons, recognized the right of individuals to be treated as autonomous agents capable of making their own decisions, un- less some specific factor diminishes that autonomy and necessitates protec- tion. The second principle, beneficence, entails the obligation to seek maximum benefit and minimum harm. Finally, the principle of justice 191 

I N H E R O W N R I G H T requires that risks and benefits be fairly distributed among the groups composing the population. That means that women as well as men should participate. In 1987, after more than half a decade of further discussion and dispute, the FDA incorporated these three principles and the AIDS community’s concerns into a new research policy, making experimental treatments for severe or fatal conditions available to properly informed patients earlier in the approval process, before final testing was complete. The specific issues involving women’s participation, however, took sev- eral more years to come to a head. On the scientific front, the change came slowly. As early as 1981, for example, as the first AIDS cases were coming to scientific atten- tion, a little-noticed article in the journal Annals of Internal Medicine had shown that excluding reproductive-aged women from Phase I drug trials tended to keep them out of Phase III trials as well. “That the article stirred so little response reflects the relative consensus at the time among re- searchers and federal policy makers vis-à-vis protectionist policies govern- ing research on women,” Johnson and Fee observe.32 This consensus gradually began to shift, however, and in 1986 the National Institutes of Health introduced a policy of encouraging more use of women as research subjects, but made little progress toward implementation. Demography finally forced the issue. By the late 1980s the huge baby boom generation was moving into midlife. This largest, best- educated, and most professionally successful cohort of women in the nation’s history had entered the age bracket that brings the first signs of aging, including menopause and rising incidence of breast and other re- productive cancers. Now this gigantic, highly self-aware age group, which since babyhood had wrenched the nation’s attention to whatever issues it was currently facing, began coming to grips with how much science had yet to learn about female aging. Their dawning awareness, combined with the general growth of health advocacy in the wake of AIDS, changed women’s place in research from a bureaucratic footnote to a major politi- cal issue. Uproar rather than indifference greeted a 1990 General Ac- counting Office report detailing NIH’s slow pace toward reform. In the light of women’s new role and consciousness, “practices and policies once 192 

C HA P T E R 9 Research on Women,Women in Research presented as protective were now labeled as paternalistic and discrimina- tory,” the inclusion committee observes. “The rationales behind a num- ber of clinical studies that had proceeded without question were now being questioned. Why had the NIH-sponsored Multiple Risk-Factor Intervention Trials of heart disease not included women when women as well as men were dying of heart disease? How could the Baltimore Longi- tudinal Study of Aging include no women when the elderly population in this country is disproportionately female?”33 In September 1990, responding to rising public and congres- sional pressure, NIH established the Office of Research on Women’s Health. From its place within the office of the director of the nation’s premier biomedical research funding agency, ORWH would work, ac- cording to an NIH document “1. to strengthen and enhance research related to diseases, disorders and conditions that affect women and to ensure that research conducted and supported by NIH adequately ad- dresses issues regarding women’s health; 2. to insure that women are appropriately represented in biomedical and behavioral research studies supported by NIH; and 3. to foster the increased enrollment of women in biomedical research—especially in pivotal decisionmaking roles within clinical medicine and the research environment.”34 The NIH Revitaliza- tion Act of 1993 made ORWH a permanent entity and called for in- creased research funding for reproductive cancers, osteoporosis, contra- ception, and fertility. Though vetoed by President Bush in June 1992 over the issue of fetal tissue research, it was reintroduced, passed, and signed by President Clinton, who opposed the ban on fetal research, in 1993. In 1991, meanwhile, new directives from NIH and the Alco- hol, Drug Abuse and Mental Health Administration, the major federal funder of psychological and behavioral research, required representation of women and racial minorities in clinical studies. Now, instead of having to justify including women on the basis of the seriousness of the condition under study, researchers had to justify excluding them. Later that same year, Bernadine Healy, M.D., NIH’s first fe- male director, announced the largest study of women’s health ever under- taken. The $625 million Women’s Health Initiative would, in fact, be the largest study that NIH had ever funded, period. Plans called for WHI to 193 

I N H E R O W N R I G H T enlist numerous research centers to run a major clinical trial, as well as an observational study and community intervention projects. So costly and complex an undertaking naturally attracted much comment and criticism. As controversy grew, NIH, at the behest of the Appropriations Commit- tee of the U.S. House of Representatives, contracted with IOM to assess the proposal’s scientific merit and feasibility. The study committee sug- gested several modifications of the original research design, but still con- cluded that “valuable scientific information could be obtained in the rede- signed study,” although “much of the information could be obtained in better designed, smaller, more focused studies that could have a greater chance of success and probably be less costly.”35 In short, the committee gave its go-ahead for a modified WHI. Important aspects of that decision, however, lay, in the WHI committee’s words, “beyond science,” in the realm of politics and public opinion.36 Try though it might to limit its deliberations to technical issues like cost and methodology, “the committee was constantly faced with broader considerations. Colleagues, advocacy organizations, executive and legislative branch representatives, all approached the committee with rea- sons why the WHI should proceed, and why canceling it would unleash disasters. If canceled, some argued, no one would ever trust the govern- ment again; it would prove that the government does not really care about women’s health; it would so greatly disappoint the community which has gathered around this project that it will be impossible ever to galvanize them again; it would be seen as an unwelcome political blow to those who have pressed for more women’s health research. Others argued that it would be wrong and harmful to women’s health to spend $625 million and find after 14 years that little in the way of useful information had been learned. It would not help women’s health and women’s health research to publicly announce and then misuse an enormous amount of money.”37 Struggling to consider only “the merits of the science,” the committee called the side issues “important” but declined to comment on them further. “With the committee recommendation that the WHI could proceed—though with significant changes of duration and focus—the dis- cussion of these issues should continue,” it suggested38 And even more importantly, “whatever the merits of WHI, the committee has no doubt 194 

C HA P T E R 9 Research on Women,Women in Research about the need for a substantial investment in research on women’s health.”39 MAKING INCLUSION WORK But political pressure and increased funding alone will not suc- ceed in reorienting research to better serve women’s needs. Only a change in the culture surrounding science can fully accomplish that. “As in all scientific endeavors,” notes the inclusion committee, “clinical studies seek to achieve results that are unbiased by the preconceptions of preferences of investigators, staff or study participants. This does not mean that the conduct of science is necessarily objective.”40 Like all human beings, scientists see their experience—includ- ing their professional work—through “lenses” provided by their culture. “Basic, hidden assumptions involving gender—so basic as to be transpar- ent—are ‘lenses’ through which people perceive, conceive and discuss social and scientific reality,” the committee believes. Ancient notions about gender differences and social roles still strongly distort Americans’ percep- tions. “It is important to look at those lenses in order to understand their consequences, instead of looking through them.”41 “Two types of gender-based assumptions” shape the vision of biomedical scientists, the committee notes.42 “Male bias refers to the ‘ob- server bias’ present in scientific inquiry when scientist-observers adopt male perspectives. Male norm refers to the tendency to perceive men’s identity and experience as the characterization or standard of what it is to be a person and to portray female differences, where they occur, as ‘devi- ant.’ ”43 These influences shape both the questions that scientists con- sider important enough to investigate and the conclusions that they reach about them. In the nineteenth century, when female higher education was a controversial innovation, perfectly reputable investigators argued that studying could, by drawing energy from the uterus to the brain, damage fertility. Educated women, they observed, bore relatively few children. As ludicrous as this theory now seems, the same tendency per- sists to view women in terms of their social and reproductive roles rather than as complete individuals in their own right. “Historically,” note 195 

I N H E R O W N R I G H T Johnson and Fee, “medical and public health professionals have conflated women’s health with reproductive and child health”44—a propensity most chillingly illustrated in the AIDS work. Indeed, the very subject matter of certain specialties, domains of knowledge supposedly determined by logical criteria, may reflect mas- culine outlooks. “The fields of sports medicine, psychiatry and occupa- tional medicine would focus on different issues if women’s recreational and nonrecreational activities and stresses dominated,” the committee sug- gests. “Indeed, the specialties themselves might have developed differ- ently, focusing on leisure activities instead of ‘sports,’ or on the hazards of domestic activities instead of the workplace.”45 Had women’s rather than men’s concerns prevailed, questions that engage gynecological researchers might also have evolved differently. Study might have concentrated on such common, debilitating, and inad- equately understood problems as preeclampsia, which complicates preg- nancies and damages babies, or fibroid tumors, “the major cause of hyster- ectomies in this country,” according to Polan.46 Instead, the “substantial body of research directed at gaining control over women’s reproduction” has followed a male-oriented agenda, suggests Susan Sherwin, Ph.D., pro- fessor of philosophy at Dalhousie University in Canada. Only in this area have women received a disproportionately large share of research attention. “For example, almost all contraceptive research has explored means of controlling women’s fertility. Similarly, efforts to relieve infertility have focused on procedures that can be done to women—even when the infertility is associated with such male conditions as low sperm count. As a result, a disproportionate share of the burden, risks, expenses and responsibility for managing fertility now belongs to women, because that is where the knowledge base is. . . . The concentra- tion of attention on women’s reproductive role not only assumes the conventional view that women are, by nature, to be responsible and avail- able for reproductive activities; it also legitimizes, reinforces, and further entrenches such views and the attitudes that accompany them.”47 Even in the “science of women,” therefore, as in all other medical specialties and in the training of medical personnel, the male norm has historically held sway. Many clinical researchers, seeking to keep research designs as 196 

C HA P T E R 9 Research on Women,Women in Research simple as possible, prefer not to take account of monthly hormonal fluc- tuations—in other words, to study men. “One of the reasons that women have traditionally been excluded from clinical studies of conditions that affect both men and women is, as The Washington Post put it, ‘their hor- monal fluctuations [have been] said to “confound” or confuse research results,’ ” reports Debra DeBruin, Ph.D., assistant professor of psychology at the University of Pittsburgh. “Women’s cycles appear not only to be different from men’s physiology, but also to be problematic. Because men are conceived of in gender-neutral terms as paradigmatic persons, re- searchers too often feel they can simply avoid the ‘problems’ caused by women’s cycles by studying only men.”48 The preference for males shows up especially strongly in drug studies because, in the committee’s words, “women have menstrual cycles that produce deviations from the ‘normal’ pattern of drug disposition observable in males.”49 But what appears a mere “confounder” in the laboratory becomes a potential danger in the consulting room. Women older than 60 use half of all cholesterol-lowering medications prescribed in this country, but three-quarters of the clinical trials testing those drugs enrolled only middle-aged men. The “deviations” that complicate re- searchers’ statistical tables thus may produce real complications on the examining table. Doctors therefore face two unsatisfactory alternatives: they can either decline to prescribe for women treatments tested only in men or assume that existing test results apply to everyone. The first de- prives half the population of potentially useful therapies, but the latter exposes them to potential risks. Demographic “deviations” also keep women out of clinical studies. Organizers of the Physician’s Health Study, for example, found that 90% of the appropriately aged medicos were male; the profession included too few middle-aged females to provide the desired statistical power. “The investigators’ reservations regarding the ability to perform such an analysis were sound,” the committee found, but a more imagina- tive approach to including women could have produced useful results. “The enrollment of a specified subgroup does not obligate [investigators] to perform interaction analysis for that subgroup, nor does the analysis need to be definitive if performed. If nothing else, the size and sign of any differences observed for women would have provided some general indi- 197 

I N H E R O W N R I G H T cation of whether the result obtained in men is consistent with that ob- served in, and thus generalizable to, women.”50 Had these researchers wanted to learn about women, in other words, they could have settled for a less-than-perfect but still potentially enlightening project. Indeed, the Boston Nurses Study, a large longitudinal project, is now focusing on women’s health concerns. FINDING A BALANCE Like all human undertakings, the new policy of encouraging use of both genders in all studies has the vices of its virtues. To ensure statistical power, numbers of subjects will need to be larger and studies more expensive. “Scientific considerations suggest that the overarching principle should be inclusion,” the inclusion committee believes.51 But in an era of tight budgets, the need to stretch scarce research funds, to get the most data for each dollar, has become a principle almost as elevated. MRFIT, for example, excluded women precisely for budgetary reasons. Finding the 12,866 men who ultimately participated meant interviewing, taking blood pressure, and doing blood workups on almost 362,000 indi- viduals. “Including women might have produced valuable information,” the committee suggests, “but at a substantial additional cost.”52 Substantially raising studies’ costs, no matter how noble the motive, might significantly reduce their number, which would also mean less knowledge. “The committee is concerned that the policy has gone too far by insisting that each and every clinical trial be designed to ensure sufficient numbers of subjects of both genders to permit subgroup analy- ses. In an era of concern about the nation’s resources, and about expendi- tures on health in particular, we argue that a trial-by-trial application of this requirement is neither good policy nor good science. Clinical trials should include both genders, but requiring scientists to enroll sufficient numbers to ensure the statistical power to detect unsuspected and implau- sible gender differences would produce little additional information at greatly increased cost.”53 Instead of raising the cost and complexity of each and every study, “mechanisms are needed at the national level to ensure that more attention is paid to questions of justice and gender in the setting of re- 198 

C HA P T E R 9 Research on Women,Women in Research search agendas, and at the study level to encourage the appropriate consid- eration of gender-related effects in clinical studies.”54 Few changes could bring about those results more quickly, many believe, than a sharp in- crease in the number of women entering scientific careers and advancing to leadership. THE SCIENCE OF WOMAN An obvious place to look for female research leadership is the nation’s departments of obstetrics and gynecology, where women already account for the majority of residents and junior faculty members. Only pediatrics, public health, and physical medicine claim more female fac- ulty.55 Currently heavily concentrated at the lower academic ranks, this new generation ought to gain increasing power in the years ahead, as they rise to the more senior positions that influence policy in both their own hospitals and medical schools and the specialty at large. Ironically, though, this trend may actually reduce the amount of research conducted in OB/ GYN departments. Though the specialty deals with some of the central issues and experiences that women face throughout their lives, it has traditionally placed a low priority on research. At a time when science is racing toward solutions to some of the most fundamental mysteries of life itself, very basic (and probably less complicated) gynecologic questions remain unan- swered: How do the ovaries, uterus, and other reproductive organs func- tion? How does fertilization happen? What factors affect fetal develop- ment? What determines when labor begins? How can delivery be safer? What will afford safer and more effective contraception and infertility treatments? How can sexually transmitted diseases and reproductive can- cers best be controlled? What is the best way to deal with menopause and premenstrual syndrome? This ignorance translates daily into potentially preventable suf- fering and death. Such “large-scale problems” as the 7% of all babies born at perilously low weights; the 2.6% of all pregnancies dangerously compli- cated by high blood pressure; the rising numbers of ectopic pregnancies, fatal in over 4% of cases; the 10% of married couples who want children but cannot conceive; and the scores of millions of sexually transmitted 199 

I N H E R O W N R I G H T infections all “could be ameliorated” by more research, believes IOM’s Committee on Research Capabilities of Academic Departments of Ob- stetrics and Gynecology.56 So could our nation’s serious dearth of effec- tive contraceptive options. Despite these obvious needs, the heads of OB/GYN depart- ments confided to the committee that financial realities make it “particu- larly difficult to conduct research.”57 Earning the scientific credentials needed to win research funding in today’s extremely competitive envi- ronment can add several more meagerly paid fellowship years to the spe- cialty and subspecialty training that already stretches for as much as eight years beyond medical school. And new OB/GYNs emerge from that training with both very large debts and the potential for very high in- comes practicing one of the most lucrative of all specialties. Medical school departments thus must often offer a great deal—as much as $150,000 a year, according to some reports—to entice promising newcomers to choose academics over private practice. To pay these salaries at the same time that they absorb the high costs of malpractice insurance and the many uninsured women who deliver in teaching hospitals, OB/GYN departments need large incomes from paying patients. Earning them forces faculty to concentrate heavily on clinical care, “often at the expense of investigation.”58 Lack of time and research focus thus creates a vicious circle. If a department cannot afford to support topflight laboratories, it cannot create the environment to nurture young scientists through the long process of training and mentoring that leads to productive research careers. Without young talent entering a department, it never develops the “critical mass” of ability, commitment, stimulation, and opportunity that marks a major research center. Without such a reputation, it is less able to garner re- search funds. This leaves it no choice but to pay its bills through patient care, depriving the faculty of time for the laboratory. “The ethos of a discipline determines its direction,” the com- mittee believes, and much of OB/GYN has traditionally lacked leaders convinced that “research within the discipline is a high priority.” Only a dozen or fewer departments “can be counted as serious research centers”; almost two score receive no federal research funding at all.59 These facts add up to “cause for concern about the current and future state” of OB/ 200 

C HA P T E R 9 Research on Women,Women in Research GYN research. “It is important to expand the number of departments that can succeed in the competitive research arena,” the committee believes.60 Bolstering the supply of physician-researchers is crucial to that goal. One might expect that the recent, overwhelming influx of women into the field should help turn things around. Unlike their male colleagues, female doctors do not generally weigh income heavily in choosing a specialty or career direction.61 Many also feel a strong commit- ment to improving women’s health. But females have traditionally tended much less often than males to become researchers, a fact “likely to in- crease the shortage of research personnel unless special efforts are made to encourage research careers for women and to meet their particular needs.”62 And those needs are substantial, because women face “special obstacles in choosing the laboratory over clinical practice,” Polan notes. The dilemmas of joining the rigid race for tenure with one’s “biological clock” ticking in the background, of doing the grueling work that leads to a secure academic position while raising small children, and of finding one’s way in a male-dominated research culture all discourage would-be investigators. “Providing mentors, flexible work arrangements or extended time to gain tenure all could ease these pressures,” she suggests.63 As departments become increasingly aware of these issues, they have begun taking steps to resolve them. The currently missing cadre of committed and talented young researchers, however, only partially explains one of the central paradoxes of American reproductive medicine. Why has the nation that once led the world in the development of safe and effective contraceptives essentially abandoned the quest for new solutions to today’s ever more pressing needs? Why has the scientific community that brought the world into the age of discretionary parenthood produced no significant advances since the breakthrough developments of the 1960s? Like Sherlock Holmes’s famous dog who did nothing in the nighttime, American contraceptive research has been conspicuous for its silence for over a generation. At least since the 1980s, fewer and fewer young investigators have entered this endeavor. The “undesirable thinness in the field” that this has produced leads in turn to a “sparse literature” on new research, warns IOM’s Committee on Population.64 Clinical studies of new meth- 201 

I N H E R O W N R I G H T ods often fail to get the replicating studies they need to establish sound data on effectiveness, for example. But able new scientists do not ignore or abandon a once-promising field for no reason. A combination of fac- tors, prominent among them that old nemesis of medical flexibility, the liability crisis, has systematically removed the incentives for aggressive contraceptive research and development in this country. In the 1960s heyday of birth control studies, nine major U.S. pharmaceutical companies—such giants as Syntex; Searle; Parke-Davis; Merck, Sharpe & Dohme; Upjohn; Mead Johnson; Wyeth-Ayerst; and Eli Lilly—all maintained active programs in what appeared a fast-moving field. In addition, other companies like A.H. Robins, which produced the infamous Dalkon Shield, also brought products to market. Corporations annually invested millions of dollars in costly, long-term research and development programs aimed at meeting the huge demand for modern contraceptives. Private sources also contributed handsomely; the Com- mittee on Research in Problems of Sex, for example, bankrolled by the Rockefeller family, supported work that led to the Pill. Starting in the 1960s, substantial federal funds also went toward birth control efforts. By the 1990s, only Ortho Pharmaceutical, a Johnson & Johnson subsidiary, continued the corporate effort. Government funding had fallen as well. In place of millions of annual corporate, philanthropic, and federal dollars, total spending had dropped sharply. Young, and even middle- aged, scientists seeking career opportunities were looking elsewhere. Lack of neither promising research leads toward potential male and female con- traceptives nor potentially eager customers drove these firms from the field, however. Rather, it became increasingly difficult to make a profit by getting expensive-to-develop new products into the hands of buyers. The Dalkon Shield disaster hit contraceptive research like an atomic bomb, leveling much of the corporate landscape and contaminat- ing what remained with long-lasting legal and regulatory fallout. Liability insurance, already becoming increasingly difficult to obtain, rose drasti- cally in cost in the wake of the billions of dollars of claims against Robins’ defective product. Companies found it increasingly difficult to justify the expensive, multiyear projects needed to develop and test new contracep- tives, given the unpredictability of scientific success and regulatory ap- proval. Firms began withdrawing from sale effective intrauterine devices 202 

C HA P T E R 9 Research on Women,Women in Research that had none of the faults of the Dalkon Shield, simply because profitably defending lawsuits appeared impossible. Today, the shrunken research field consists primarily of small firms that often lack the resources for major undertakings and of projects supported by nonprofit organizations. In other countries with other regulatory and insurance arrange- ments, progress has continued apace, however. In the United States, manufacturers must present convincing evidence that contraceptives are safe before they can obtain FDA approval. In Britain, on the other hand, products can win approval if the evidence shows that they are not unsafe. This disparity permitted Depo-Provera on the contraceptive market in the United Kingdom but not, based on examination of identical data, in this country. “The difference in outcome seems to have resulted not from scientific differences, but from policy differences (burden of proof) in the two countries,” the population committee concluded.65 The United States cannot begin to meet its people’s contracep- tive needs unless the climate becomes considerably more hospitable, the committee believes. Although manufacturers must satisfy the FDA’s strin- gent safety standards, for example, they cannot rely on that official ap- proval as legal proof of safety or even of a sincere attempt to provide safety. Such rules “can be changed to remove most of their undue nega- tive consequences” without “increasing the health risks of contraceptive users,” the committee believes.66 It suggests that the same should happen to the balance of risks and benefits currently used to judge contraceptives. Because an unwanted pregnancy can impose serious health risks on some women, “methods with fewer side effects are not necessarily safer if they have higher failure rates.” The committee, however, “does not propose reducing the safety requirements,” but rather “adding new criteria” to make safety evaluations “more specific to different groups of users.”67 Despite the various “obstacles” standing in the way of better reproductive health, however, the obstetrics committee nonetheless sees “several encouraging signs,” including an increase in programs to support OB/GYN researchers and a growing engagement among the field’s lead- ership in the need to encourage research. For all their problems, though, these trends, coupled with the general drive to include more women in research, both as subjects and as investigators, hold promise of greater knowledge and better health for women in years to come. Indeed, be- 203 

I N H E R O W N R I G H T cause of “increasing interest in women’s health issues,” the committee believes, “the 1990s offer an unprecedented opportunity” for progress.68 As long-hidden biases come to light and outmoded restrictions fall away, American biomedical research can increasingly become not only the world’s most productive, but also among the world’s most just. NOTES 1. Women and Health Research: Ethical and Legal Issues of Including Women in Clinical Trials, Vol. 1, 28. 2. Ibid., p. 37. 3. Quoted in Women and Health Research, Vol. 1, 43. 4. Women and Health Research, Vol. 1, 49. 5. Ibid. 6. Ibid. 7. Ibid, 64. 8. Ibid. 9. Ibid, 65. 10. Weisman and Cassard, Women and Health Research, Vol. 2, 37. 11. Ibid., 38. 12. Debruin, Women and Health Research, Vol. 2, 130. 13. Ibid. 14. L. Mitchell, quoted in Women and Health Research, Vol. 1, 80. 15. Women and Health Research, Vol. 1, 66. 16. Assessing Future Research Needs: Mental and Addictive Disorders in Women (Transcript), 221. 17. Ibid., 227. 18. Ibid., 228. 19. Women and Health Research, Vol. 1, 101. 20. Benet, Women and Health Research, Vol. 2, 42. 21. Ibid, 42-3. 22. Women and Health Research, Vol. 1, 29. 23. Weisman and Cassard, Women and Health Research, Vol. 2, 37-8. 24. Ibid., 37. 25. Polan, Women and Health Research, Vol. 2, 1. 26. Johnson and Fee, Women and Health Research, Vol. 2, 3. 27. Beecher, quoted in Women and Health Research, Vol. 1, 39. 28. Women and Health Research, Vol. 1, 40. 29. Johnson and Fee, Women and Health Research, Vol. 2, 3. 30. Ibid., 4. 31. Quoted in Johnson and Fee, Women and Health Research, Vol. 2, 5. 32. Johnson and Fee, Women and Health Research, Vol. 2, 1. 33. Women and Health Research, Vol. 1, 43-4. 34. Opportunities for Research on Women’s Health (NIH 1992b), quoted in Women and Health Re- search, Vol. 1, 44. 35. An Assessment of the NIH Women’s Health Initiative, 15. 36. Ibid., 90. 204 

C HA P T E R 9 Research on Women,Women in Research 37. Ibid. 38. Ibid. 39. Ibid., 16. 40. Women and Health Research, Vol. 1, 109. 41. Ibid., 110. 42. Ibid. 43. Ibid., 111. 44. Johnson and Fee, Women and Health Research, Vol. 1, 5. 45. Women and Health Research, Vol. 1, 112. 46. Polan, Women and Health Research, Vol. 2, 1. 47. Sherwin, Women and Health Research, Vol. 2, 14. 48. DeBruin, Women and Health Research, Vol. 2, 136. 49. Women and Health Research, Vol. 1, 113. 50. Ibid., 99-100. 51. Ibid., 104. 52. Ibid., 100. 53. Ibid., 104-5. 54. Ibid. 55. Strengthening Research in Academic OB/GYN Departments, 75. 56. Ibid., 1. 57. Ibid., 7. 58. Ibid. 59. Ibid., 8. 60. Ibid., 83. 61. Ibid., 93. 62. Ibid., 101. 63. Polan, Women and Health Research, Vol. 2, 3. 64. Developing New Contraceptives: Obstacles and Opportunities, 85. 65. Ibid., 99-100. 66. Ibid., 4. 67. Ibid., 2. 68. Strengthening Research, 137. 205