Overdosed America (11 page)

What better way to inaugurate the modern era of pharmaceutical marketing than to promise women protection from the ravages of age?
In a 1962 article published in JAMA
, Dr. Robert Wilson, a gynecologist, reported the results of a study in which 304 women had been treated with estrogen. Though 18 cancers were predicted over the time that the women had been taking estrogen, none developed. Wilson concluded that
hormone pills “are prophylactic
to breast and genital cancer.” In his
best-selling book,
Feminine Forever,
published in 1966, Wilson wrote with the authority of a medical scientist, communicating a message of unbridled fear that untreated menopause marked the beginning of a rapid decline in appearance, sexuality, health, and overall quality of life. Dr. Wilson wrote poetically of the tragedy of menopause:

The transformation, within a few years, of a formerly pleasant, energetic woman into a dull-minded but sharp-tongued caricature of her former self is one of the saddest of human spectacles. The suffering is not hers alone—it involves her entire family, her business associates, her neighborhood storekeepers, and all others with whom she comes into contact. Multiplied by millions, she is a focus of bitterness and discontent in the whole fabric of our civilization.

Perhaps Wilson’s own tongue had been sharpened by the support his foundation was receiving from several drug companies, including the manufacturer of Premarin. In 1965 the Wilson Research Foundation received $34,000 (the equivalent of about $175,000 in 2004 dollars) in
contributions from drug companies
, enough to cover Wilson’s expenses while he was writing his book. In total,
Wilson’s foundation
received $1.3 million from drug companies.

The rhetoric was toned down just a bit in an article coauthored by Wilson and his wife, Thelma, a registered nurse, published in the
Journal of the American Geriatrics Society
in 1972 and titled “The Basic Philosophy of Estrogen Maintenance.” The article offered both carrot—“Estrogen produces beauty, the allure which attracts the male. This can no more be resisted than the moth can resist the flame”—and stick: “Breasts and genital organs will not shrivel. Such women will . . . not become dull and unattractive.”
The Wilsons described female aging as a disease
caused by the failure of postmenopausal women’s ovaries to produce estrogen. They hypothesized that this medical condition could be treated with estrogen, the same way that diabetes, caused by the failure of the pancreas to produce adequate insulin, is treated with insulin.

It’s easy to be angry about the harm done to so many women as a result of Dr. Wilson’s work. Yet his harsh rhetoric becomes understandable—even evokes compassion—when seen as his attempt to
transform his own personal trauma
into a positive contribution to medicine. According to his son Ron, Wilson’s own mother had been “the shrew of all shrews.” Ron says that Wilson believed that his mother’s condition “was due to a chemical imbalance, and it was that belief that led him to do his research. He did indeed wish to help women.. . . So I suppose the real thanks for the beginning research must go back to dear ol’ grandma, for if she hadn’t been such a witch, medical ‘progress’ might well have been delayed. I can respect his efforts for women at the time even if his thinking was flawed.”

This personal history set the stage for Wilson to become a well-paid manservant of the drug industry, believing all the while that he was saving the world from the misery that his mother went through and, in turn, had caused. According to his son, Wilson was motivated not by the money he received from the drug companies, but by a desire to help the “cause.” So much so that when Thelma Wilson, who had been taking hormone replacement therapy, developed breast cancer, this fact was hidden even from her son, Ron.

Ron Wilson’s mother’s breast cancer may not have been the only fact that was changed. Ron went on to say, “It is my opinion that some of the original research findings are flawed to downright false. Much of that opinion came out of the mouth of my father, who was not above changing a few figures to help the cause.” Ron says that his father “was used by the various drug firms and when no longer needed he was quickly and quietly cast aside, leaving him with no office practice, no foundation, little family life and few friends.” Wilson turned to alcohol and drugs, and took his own life in 1981. It may be hard to believe that Wilson’s exaggerated claims about Premarin were so effective, but in no small measure owing to his efforts, Premarin became the
most frequently prescribed brand-name drug
in the United States in 1966. It remained
one of the five most frequently prescribed drugs
through 1975. Wilson’s book promoting the use of estrogen for women who had reached menopause, and the hundreds of newspaper and magazine articles that followed, had accomplished their mission.

It was not, however, all smooth sailing in the business of selling eternal youth and femininity. In December 1975, two articles published in NEJM showed that
estrogen therapy increased the risk of cancer
of the lining of the uterus
(endometrial cancer), up to 14-fold
after seven years of treatment. This fear was quashed when, four years later, in 1979, an article published in
The Lancet
showed that
adding another hormone, progestin
, for about 10 days each month to estrogen therapy prevented the changes in the lining of the uterus that predisposed to cancer. Several other
studies soon confirmed that progestin protected women
on estrogen therapy from developing endometrial cancer. Still,
HRT had been linked with cancer
in the public’s mind, and sales plummeted. Only half as many prescriptions for Premarin were filled in 1980 as had been filled in 1975.

An aggressive drug rehabilitation program was needed.
“Marketing a disease
is the best way to market a drug,” notes Dr. Love. And osteoporosis, or thinning of the bones, was a perfect disease to market: There are no symptoms until you develop fractures, so no postmenopausal woman could be sure she was safe. And the criteria were set so that one quarter of all women over 65, and more than half over 75, would be diagnosed with the “disease” if they had bone density tests. There was, however, a lot of work to be done to turn osteoporosis from part of the normal spectrum of skeletal aging into a feared disease.
To educate the doctors
, according to Dr. Love, “the pharmaceutical companies started placing ads in the medical journals showing wheelchairs, x-rays of crooked spines, and pathetic-looking women with dowager’s humps. They funded medical meetings and lectures about osteoporosis.”

The next step was to “educate” the public. In 1985, only 23 percent of women had heard of osteoporosis. But, according to
US News and World Report,
that changed quickly as the result of the efforts of Burson-Marsteller, the
public relations firm
hired by Wyeth-Ayerst.
The campaign was successful
in increasing public concern (some would say unnecessary fear), generating many articles in women’s magazines and culminating in National Osteoporosis Week.
The National Osteoporosis Foundation
was started with drug company support in 1986. Doctors (including me) and patients came to fear that undiagnosed osteoporosis would lead to hipbones’ suddenly snapping with minimal trauma, though in more than 20 years as a busy family doctor I never saw such a thing.

Coinciding with the public’s “education” about osteoporosis was
a 1985 report in NEJM
about the positive effect of estrogen on the risk of heart disease. More than 30,000 postmenopausal women participating in the Nurses’ Health Study, which followed the women for more than three years, showed that nurses who were currently using estrogen had 70 percent less risk of developing coronary heart disease than women who had not used hormones—a dramatic finding.

This was exactly the boost that Wyeth-Ayerst needed to rehabilitate
Premarin sales
. Who would worry about a modest increase in the risk of breast cancer when you could take a pill to beat the odds against osteoporosis, the silent disease; and heart disease, the number one killer? (Much less attention was given to results from the Framingham Heart Study, published in the same issue of the
New England Journal of Medicine,
showing that women who had taken estrogen were 50 percent
more
likely to develop heart disease.)

By 1992, Premarin sales were topping their 1975 peak. One out of five postmenopausal women in the United States was taking hormones. The prestigious
American College of Physicians issued guidelines
to practicing physicians recommending that “all women . . . should consider preventive hormone therapy,” and that 10 to 20 years of therapy were recommended for “maximum benefit.” The American College of Obstetrics and Gynecology also recommended that all postmenopausal women, barring a medical contraindication like breast cancer, should take HRT for life. Bolstered by the recommendations of these professional organizations,
Premarin use increased another 40 percent
over the next three years. In 1995 Premarin once again became the
most frequently prescribed brand-name drug
in the United States. Perhaps the strongest evidence supporting routine HRT was presented
in a 1997 article
published in NEJM showing that “mortality among women who use postmenopausal hormones is lower than among nonusers,” again overriding continuing concerns about the link to breast cancer.

It helps to take a step back and look at the methods used in medical research. The two most common types of medical studies are randomized controlled trials (RCTs) and observational studies. A simple example demonstrates how these types of studies differ and illustrates the inherent strengths and weaknesses of each. Imagine that researchers want to study the impact that running a 10-kilometer road race has on women’s health over a one-year period.

The simplest way to do this would be to set up an observational study. Researchers would wait at the finish line of a local 10K race and ask women if they would be willing to participate in the study. Let’s say 100 women sign up. These women would then complete a questionnaire that would include health habits, personal and family health history, and other personal characteristics that might have an impact on their health, such as marital status, educational level, and family income. This group of women, called the treatment group, would then be compared with a “control group,” also including 100 women, matched as closely as possible on all characteristics—except that they didn’t run in the race. This control group of women would then fill out the same questionnaire.

One year after the race, all the women in the study would be queried about their health. Rates of illness that occurred in the women who ran in the race and in the control group would be calculated. These results would then be statistically “adjusted” for preexisting differences between the groups that might have had an impact on health: If, say, the women who ran in the race were more likely to have a college education or keep up with their preventive health care, the impact of such differences on health could be adjusted for in the statistical analysis. Once all these factors were taken into account, the researchers would then be able to postulate whether running a 10K race had an effect on overall health.

The advantage of this observational study design is that it is inexpensive and can include large numbers of people. Participants don’t have to do anything different from what they are already doing, except fill out the questionnaires. Observational studies take advantage of the differences that occur naturally, without any experimental intervention, such as some women choosing to run in the 10K race and others not.

The disadvantage of this type of study, of course, is that we can never be sure that all the differences between the groups are adequately included in the statistical adjustment. For example, maybe the runners share a strong commitment to investing time and energy in maintaining their health. But perhaps when the researchers designed the questionnaire, they weren’t smart enough to include a question that identified this belief, which could be the real reason why the runners were healthier one year after the race. Without being aware of this difference between the groups, the researchers might incorrectly attribute the runners’ better health to their having participated in the race.

The other way to do this study is a randomized controlled trial, the gold standard of medical research. This study design provides a much more precise way to identify the factors that contribute to a particular outcome. Continuing with the example of the 10K race, researchers would find 200 women who agreed to participate in a study about the health effects of running such a race. The women would then be randomly assigned to the treatment group (to run in the race) or the control group (not to run in the race). The same questionnaire would be filled out at the beginning of the study, and the same inquiry about health events would be made at the end of the year. The differences in health experienced by the women in each group during the year following the race would then be calculated, including statistical adjustments to correct for randomly occurring preexisting differences between the two groups that might have an impact on their subsequent health.

The obvious advantage of this study design is that the differences that make some women more likely than others to run in the race are neutralized by random assignment to the two groups. This greatly reduces the probability that an unidentified preexisting factor is responsible for any differences in health outcome. For this reason, an RCT would be much more likely to identify the true effect of having run in the race. The problem with this kind of study is that it takes more time and is more expensive to do.

It is amazing that before 1998, not one of the claims supporting the benefits of HRT had been substantiated by large RCTs. There were good reasons why the drug companies had not undertaken such studies: these studies are expensive, and when sales are proceeding well, there is little motivation to risk an unfavorable outcome. The Nurses’ Health Study—the source of much of the information about HRT that came out in the 1990s—is an observational study (funded by the National Institutes of Health) designed to reveal the relationship between diet, lifestyle, and hormone therapy (both birth control pills and HRT).