m_topn picture
Atlantic Monthly Sidebar

As originally published in
The Atlantic Monthly

August 1994


The Sex-Bias Myth in Medicine

Though it is commonly believed that American health-care delivery and research benefit men at the expense of women, the truth appears to be exactly the opposite

by Andrew G. Kadar, M.D.

"When it comes to health-care research and delivery, women can no longer be treated as second-class citizens." So said the President of the United States on October 18, 1993.

He and the First Lady had just hosted a reception for the National Breast Cancer Coalition, an advocacy group, after receiving a petition containing 2.6 million signatures which demanded increased funding for breast-cancer prevention and treatment. While the Clintons met with leaders of the group in the East Room of the White House, a thousand demonstrators rallied across the street in support. The President echoed their call, decrying the neglect of medical care for women.

Two years earlier Bernadine Healy, then the director of the National Institutes of Health, charged that "women have all too often been treated less than equally in . . . health care." More recently Representative Pat Schroeder, a co-chair of the Congressional Caucus for Women's Issues, sponsored legislation to "ensure that biomedical research does not once again overlook women and their health." Newspaper articles expressed similar sentiments.

The list of accusations is long and startling. Women's-health-care advocates indict "sex-biased" doctors for stereotyping women as hysterical hypochondriacs, for taking women's complaints less seriously than men's, and for giving them less thorough diagnostic workups. A study conducted at the University of California at San Diego in 1979 concluded that men's complaints of back pain, chest pain, dizziness, fatigue, and headache more often resulted in extensive workups than did similar complaints from women. Hard scientific evidence therefore seemed to confirm women's anecdotal reports.

Men more often than women undergo angiographies and coronary-artery-bypass-graft operations. Even though heart disease is the No. 1 killer of women as well as men, this sophisticated, state-of-the-art technology, critics contend, is selectively denied to women.

The problem is said to be repeated in medical research: women, critics argue, are routinely ignored in favor of men. When the NIH inventoried all the research it had funded in 1987, the money spent on studying diseases unique to women amounted to only 13.5 percent of the total research budget.

Perhaps the most emotionally charged disease for women is breast cancer. If a tumor devastated men on a similar scale, critics say, we would declare a state of national emergency and launch a no-cost-barred Apollo Project-style program to cure it. In the words of Matilda Cuomo, the wife of the governor of New York, "If we can send a woman to the moon, we can surely find a cure for breast cancer." The neglect of breast-cancer research, we have been told, is both sexist and a national disgrace.

Nearly all heart-disease research is said to be conducted on men, with the conclusions blindly generalized to women. In July of 1989 researchers from the Harvard Medical School reported the results of a five-year study on the effects of aspirin in preventing cardiovascular disease in 22,071 male physicians. Thousands of men were studied, but not one woman: women's health, critics charge, was obviously not considered important enough to explore similarly. Here, they say, we have definite, smoking-gun evidence of the neglect of women in medical research--only one example of a widespread, dangerous phenomenon.

Still another difference: pharmaceutical companies make a policy of giving new drugs to men first, while women wait to benefit from the advances. And even then the medicines are often inadequately tested on women.

To remedy all this neglect, we need to devote preferential attention and funds, in the words of the Journal of the American Medical Women's Association, to "the greatest resource this country will ever have, namely, the health of its women." Discrimination on such a large scale cries out for restitution--if the charges are true.

In fact one sex does appear to be favored in the amount of attention devoted to its medical needs. In the United States it is estimated that one sex spends twice as much money on health care as the other does. The NIH also spends twice as much money on research into the diseases specific to one sex as it does on research into those specific to the other, and only one sex has a section of the NIH devoted entirely to the study of diseases afflicting it. That sex is not men, however. It is women.


IN the United States women seek out and consequently receive more medical care than men. This is true even if pregnancy-related care is excluded. Department of Health and Human Services surveys show that women visit doctors more often than men, are hospitalized more often, and undergo more operations. Women are more likely than men to visit a doctor for a general physical exam when they are feeling well, and complain of symptoms more often. Thus two out of every three health-care dollars are spent by women.

Quantity, of course, does not guarantee quality. Do women receive second-rate diagnostic workups?

The 1979 San Diego study, which concluded that men's complaints more often led to extensive workups than did women's, used the charts of 104 men and women (fifty-two married couples) as data. This small-scale regional survey prompted a more extensive national review of 46,868 office visits. The results, reported in 1981, were quite different from those of the San Diego study.

In this larger, more representative sample, the care received by men and women was similar about two thirds of the time. When the care was different, women overall received more diagnostic tests and treatment--more lab tests, blood-pressure checks, drug prescriptions, and return appointments.

Several other, small-scale studies have weighed in on both sides of this issue. The San Diego researchers looked at another 200 men and women in 1984, and this time found "no significant differences in the extent and content" of workups. Some women's-health-care advocates have chosen to ignore data from the second San Diego study and the national survey while touting the first study as evidence that doctors, to quote once again from the Journal of the American Medical Women's Association, do "not take complaints as seriously" when they come from women: "an example of a double standard influencing diagnostic workups."

When prescribing care for heart disease, doctors consider such factors as age, other medical problems, and the likelihood that the patient will benefit from testing and surgery. Coronary-artery disease afflicts men at a much younger age, killing them three times as often as women until age sixty-five. Younger patients have fewer additional medical problems that preclude aggressive, high-risk procedures. And smaller patients have smaller coronary arteries, which become obstructed more often after surgery. Whereas this is true for both sexes, obviously more women fit into the smaller-patient category. When these differences are factored in, sex divergence in cardiac care begins to fade away.

To the extent that divergence remains, women may be getting better treatment. At least that was the conclusion of a University of North Carolina/Duke University study that looked at the records of 5,795 patients treated from 1969 to 1984. The most symptomatic and severely diseased men and women were equally likely to be referred for bypass surgery. Among the patients with less-severe disease--the ones to whom surgery offers little or no survival benefit over medical therapy--women were less likely to be scheduled for bypass surgery. This seems proper in light of the greater risk of surgical complications, owing to women's smaller coronary arteries. In fact, the researchers questioned the wisdom of surgery in the less symptomatic men and suggested that "the effect of gender on treatment selection may have led to more appropriate treatment of women."

As for sophisticated, pioneering technology selectively designed for the benefit of one sex, laparoscopic surgery was largely confined to gynecology for more than twenty years. Using viewing and manipulating instruments that can be inserted into the abdomen through keyhole-sized incisions, doctors are able to diagnose and repair, sparing the patient a larger incision and a longer, more painful recuperation. Laparoscopic tubal sterilization, first performed in 1936, became common practice in the late 1960s. Over time the development of more-versatile instruments and of fiber-optic video capability made possible the performance of more-complex operations. The laparoscopic removal of ectopic pregnancy was reported in 1973. Finally, in 1987, the same technology was applied in gallbladder surgery, and men began to enjoy its benefits too.

Years after ultrasound instruments were designed to look inside the uterus, the same technology was adapted to search for tumors in the prostate. Other pioneering developments conceived to improve the health care of women include mammography, bone-density testing for osteoporosis, surgery to alleviate bladder incontinence, hormone therapy to relieve the symptoms of menopause, and a host of procedures, including in vitro fertilization, developed to facilitate impregnation. Perhaps so many new developments occur in women's health care because one branch of medicine and a group of doctors, gynecologists, are explicitly concerned with the health of women. No corresponding group of doctors is dedicated to the care of men.

So women receive more care than men, sometimes receive better care than men, and benefit more than men do from some developing technologies. This hardly looks like proof that women's health is viewed as secondary in importance to men's health.


THE 1987 NIH inventory did indeed find that only 13.5 percent of the NIH research budget was devoted to studying diseases unique to women. But 80 percent of the budget went into research for the benefit of both sexes, including basic research in fields such as genetics and immunology and also research into diseases such as lymphoma, arthritis, and sickle-cell anemia. Both men and women suffer from these ailments, and both sexes served as study subjects. The remaining 6.5 percent of NIH research funds were devoted to afflictions unique to men. Oddly, the women's 13.5 percent has been cited as evidence of neglect. The much smaller men's share of the budget is rarely mentioned in these references.

As for breast cancer, the second most lethal malignancy in females, investigation in that field has long received more funding from the National Cancer Institute than any other tumor research, though lung cancer heads the list of fatal tumors for both sexes. The second most lethal malignancy in males is also a sex-specific tumor: prostate cancer. Last year approximately 46,000 women succumbed to breast cancer and 35,000 men to prostate cancer; the NCI spent $213.7 million on breast-cancer research and $51.1 million on study of the prostate. Thus although about a third more women died of breast cancer than men of prostate cancer, breast-cancer research received more than four times the funding. More than three times as much money per fatality was spent on the women's disease. Breast cancer accounted for 8.8 percent of cancer fatalities in the United States and for 13 percent of the NCI research budget; the corresponding figures for prostate cancer were 6.7 percent of fatalities and three percent of the funding. The spending for breast-cancer research is projected to increase by 23 percent this year, to $262.9 million; prostate-research spending will increase by 7.6 percent, to $55 million.

The female cancers of the cervix and the uterus accounted for 10,100 deaths and $48.5 million in research last year, and ovarian cancer accounted for 13,300 deaths and $32.5 million in research. Thus the research funding for all female-specific cancers is substantially larger per fatality than the funding for prostate cancer.

Is this level of spending on women's health just a recent development, needed to make up for years of prior neglect? The NCI is divided into sections dealing with issues such as cancer biology and diagnosis, prevention and control, etiology, and treatment. Until funding allocations for sex-specific concerns became a political issue, in the mid-1980s, the NCI did not track organ-specific spending data. The earliest information now available was reconstructed retroactively to 1981. Nevertheless, these early data provide a window on spending patterns in the era before political pressure began to intensify for more research on women. Each year from 1981 to 1985 funding for breast-cancer research exceeded funding for prostate cancer by a ratio of roughly five to one. A rational, nonpolitical explanation for this is that breast cancer attacks a larger number of patients, at a younger age. In any event, the data fail to support claims that women were neglected in that era.

Again, most medical research is conducted on diseases that afflict both sexes. Women's-health advocates charge that we collect data from studies of men and then extrapolate to women. A look at the actual data reveals a different reality.

The best-known and most ambitious study of cardiovascular health over time began in the town of Framingham, Massachusetts, in 1948. Researchers started with 2,336 men and 2,873 women aged thirty to sixty-two, and have followed the survivors of this group with biennial physical exams and lab tests for more than forty-five years. In this and many other observational studies women have been well represented.

With respect to the aspirin study, the researchers at Harvard Medical School did not focus exclusively on men. Both sexes were studied nearly concurrently. The men's study was more rigorous, because it was placebo-controlled (that is, some subjects were randomly assigned to receive placebos instead of aspirin); the women's study was based on responses to questionnaires sent to nurses and a review of medical records. The women's study, however, followed nearly four times as many subjects as the men's study (87,678 versus 22,071), and it followed its subjects for a year longer (six versus five) than the men's study did. The results of the men's study were reported in the New England Journal of Medicine in July of 1989 and prompted charges of sexism in medical research. The women's-study results were printed in the Journal of the American Medical Association in July of 1991, and were generally ignored by the nonmedical press.

Most studies on the prevention of "premature" (occurring in people under age sixty-five) coronary-artery disease have, in fact, been conducted on men. Since middle-aged women have a much lower incidence of this illness than their male counterparts (they provide less than a third as many cases), documenting the preventive effect of a given treatment in these women is much more difficult. More experiments were conducted on men not because women were considered less important but because women suffer less from this disease. Older women do develop coronary disease (albeit at a lower rate than older men), but the experiments were not performed on older men either. At most the data suggest an emphasis on the prevention of disease in younger people.

Incidentally, all clinical breast-cancer research currently funded by the NCI is being conducted on women, even though 300 men a year die of this tumor. Do studies on the prevention of breast cancer which specifically exclude males signify a neglect of men's health? Or should a disease be studied in the group most at risk? Obviously, the coronary-disease research situation and the breast-cancer research situation are not equivalent, but together they do serve to illustrate a point: diseases are most often studied in the highest-risk group, regardless of sex.

What about all the new drug tests that exclude women? Don't they prove the pharmaceutical industry's insensitivity to and disregard for females?

The Food and Drug Administration divides human testing of new medicines into three stages. Phase 1 studies are done on a small number of volunteers over a brief period of time, primarily to test safety. Phase 2 studies typically involve a few hundred patients and are designed to look more closely at safety and effectiveness. Phase 3 tests precede approval for commercial release and generally include several thousand patients.

In 1977 the FDA issued guidelines that specifically excluded women with "childbearing potential" from phase 1 and early phase 2 studies; they were to be included in late phase 2 and phase 3 trials in proportion to their expected use of the medication. FDA surveys conducted in 1983 and 1988 showed that the two sexes had been proportionally represented in clinical trials by the time drugs were approved for release.

The 1977 guidelines codified a policy already informally in effect since the thalidomide tragedy shocked the world in 1962. The births of armless or otherwise deformed babies in that era dramatically highlighted the special risks incurred when fertile women ingest drugs. So the policy of excluding such women from the early phases of drug testing arose out of concern, not out of disregard, for them. The policy was changed last year, as a consequence of political protest and recognition that early studies in both sexes might better direct testing.


THROUGHOUT human history from antiquity until the beginning of this century men, on the average, lived slightly longer than women. By 1920 women's life expectancy in the United States was one year greater than men's (54.6 years versus 53.6). After that the gap increased steadily, to 3.5 years in 1930, 4.4 years in 1940, 5.5 in 1950, 6.5 in 1960, and 7.7 in 1970. For the past quarter of a century the gap has remained relatively steady: around seven years. In 1990 the figure was seven years (78.8 versus 71.8).

Thus in the latter part of the twentieth century women live about 10 percent longer than men. A significant part of the reason for this is medical care.

In past centuries complications during childbirth were a major cause of traumatic death in women. Medical advances have dramatically eliminated most of this risk. Infections such as smallpox, cholera, and tuberculosis killed large numbers of men and women at similar ages. The elimination of infection as the dominant cause of death has boosted the prominence of diseases that selectively afflict men earlier in life.

Age-adjusted mortality rates for men are higher for all twelve leading causes of death, including heart disease, stroke, cancer, lung disease (emphysema and pneumonia), liver disease (cirrhosis), suicide, and homicide. We have come to accept women's longer life span as natural, the consequence of their greater biological fitness. Yet this greater fitness never manifested itself in all the millennia of human history that preceded the present era and its medical-care system--the same system that women's-health advocates accuse of neglecting the female sex.

To remedy the alleged neglect, an Office of Research on Women's Health was established by the NIH in 1990. In 1991 the NIH launched its largest epidemiological project ever, the Women's Health Initiative. Costing more than $600 million, this fifteen-year program will study the effects of estrogen therapy, diet, dietary supplements, and exercise on heart disease, breast cancer, colon cancer, osteoporosis, and other diseases in 160,000 postmenopausal women. The study is ambitious in scope and may well result in many advances in the care of older women.

What it will not do is close the "medical gender gap," the difference in the quality of care given the two sexes. The reason is that the gap does not favor men. As we have seen, women receive more medical care and benefit more from medical research. The net result is the most important gap of all: seven years, 10 percent of life.


Copyright © 1994 by Andrew G. Kadar. All rights reserved.
The Atlantic Monthly; August 1994; The Sex-Bias Myth in Medicine; Volume 274, No. 2; pages 66-70.

m_nv_cv picture m_nv_un picture m_nv_am picture m_nv_pr picture m_nv_as picture m_nv_se picture