How Doctors Think (28 page)

In the past decade or so, marketing directly to the public prompts people like aging men or postmenopausal women to ask their doctor for a drug even if the drug has not been proven to work for their problem. In the United States, once a drug is approved for sale for a particular purpose, a physician can prescribe it for any clinical condition. The Food and Drug Administration has approved testosterone replacement therapy, using products such as the one Rick Duggan promotes, for patients like Nick Mancini, whose pituitary gland no longer can signal his testes to produce the hormone, or for men with inherited conditions like an extra X chromosome. These are uncommon disorders; the market would number only tens of thousands of patients. But there are nearly forty million men in the United States over the age of fifty. If physicians were to prescribe testosterone to those whose levels are declining, the market could reach billions of dollars. While the FDA forbids drug companies to advertise their products for uses other than those it has approved, they can use other strategies. The ads in
Time
magazine and medical journals were designed to "raise awareness" of the "condition" of testosterone deficiency without naming a specific drug. And, to complement the advertising approach, the drug companies enlisted "opinion leaders" like Bert Foyer who may influence their peers or trainees.

The freedom to prescribe that the FDA grants doctors can have clinical benefits. In my own field of oncology, drugs approved for a specific cancer—cisplatin for testicular carcinoma, or gemcitabine for pancreatic cancer—turned out to have wider application. Many women with ovarian cancer have received a successful platinum-based treatment like cisplatin, and patients with lung cancer or breast cancer have benefited from gemcitabine. A pharmaceutical company may legally carry out clinical trials of a product for conditions beyond those covered by the initial FDA approval. If the data demonstrate benefit, it can then ask the FDA to widen the approved scope of treatment. The problem arises when the marketing is ahead of the medicine—when the data to support treatment are thin, contradictory, or even negative. Then sales depend on opinion leaders who assert benefit despite the absence of proof.

The existence of andropause is unproved. Studies to date on testosterone replacement therapy show no convincing benefits for older men with modestly reduced hormone levels who exhibit the vague symptoms on the questionnaire. Treatment does not significantly increase strength in most muscle groups; compared to a placebo, it neither boosts libido nor increases energy. A panel convened by the National Institutes of Health concluded that the andropause hypothesis had no scientific basis. Nonetheless, the number of prescriptions for testosterone replacement products continues to rise sharply, reaching far beyond patients with well-defined deficiencies like Nick Mancini. Drug companies, whose primary goal is profit, can drive doctors' thinking about what constitutes a malady and how to remedy it.

In 1998, I became enamored with a family of new medications because of my own clinical condition. I had suffered a failed spine surgery and was left with chronic arthritic symptoms that made it impossible for me to pursue my favorite sport, distance running. Every time I began to run, I developed muscle spasms in my lumbar region, with shooting pains into my buttocks. Reluctantly, I gave up running, and although I swam and cycled, I never relinquished a sense of loss. Then a colleague who is a rheumatologist told me about novel anti-inflammatory medications then under development: the
COX
-2 inhibitors, which eventually were marketed as Celebrex and Vioxx. I began to investigate these inhibitors and was gripped by the idea that they might restore me to my favorite sport. My enthusiasm was such that I eventually wrote an article for
The New Yorker
entitled "Superaspirin." The article drew on data that had been recently released from a six-month-long trial of Celebrex in patients with chronic arthritis. And while my
New Yorker
piece included some caveats, overall it heralded a paradigm change in the therapy of arthritis. I ended the article with a fantasy that after taking the
COX
-2 inhibitor, I could lace up my sneakers and run again.

So Dr. Foyer's enthusiasm for testosterone was familiar to me, mirrored in my desire to believe that there would be a way to temper, if not reverse, the degenerative changes in my spine. Of course, the story of the
COX
-2 inhibitors did not end that way. Although some patients clearly benefited from the drugs, their impact, compared with other anti-inflammatory agents, like naproxen and ibuprofen, was not dramatically different; the patients who would most benefit were those who had a history of gastrointestinal bleeding, since the
COX
-2 inhibitors reduced, but did not eliminate, the side effect of stomach irritation seen with other anti-inflammatory drugs. But the notion that they would have no significant toxicity and would usher in a new era in treatment proved wrong. More rigorous studies showed that there was a small but definite incidence of heart attack and stroke, likely due to changes in blood vessels brought about by the inhibition of the
COX
-2 enzyme.

Another of my dreams about Celebrex and Vioxx was that they might help prevent Alzheimer's disease. One hypothesis was that the damage to the brain was caused by inflammation, so that antiinflammatory drugs could be useful. My maternal grandfather, Max Sherman, had played an important role in my life when I was growing up. He had worked in the post office, but told us tall tales about his exploits with Teddy Roosevelt and the Rough Riders. Only years later did I realize that he was much too young to have been a Rough Rider, but at the time it caught my imagination and made him into a living part of history and family glory.

Not long after the death of my grandmother, my grandfather Max's behavior changed. He became sullen and withdrawn, and at first we thought he was depressed. But then his flat affect gave way to periods of aggression, both verbal and physical. My grandfather was one of the sweetest and gentlest men I have ever met, and this kind of behavior was completely foreign to him and to his family. It turned out that he had Alzheimer's disease and ultimately had to be institutionalized. He died unable to recognize any of us. This specter of Alzheimer's, which haunts so many families, certainly haunts mine, so the notion that a safe drug like a
COX
-2 inhibitor could be taken daily for decades and not only restore me to my running but also protect me from the disease that took my grandfather held profound appeal—so profound that it blinded me to critical thinking. As Karen Delgado points out time and again to her patients, it is a common illusion that a drug will arrive that has no toxicity and can, in a near-miraculous way, reverse the consequences of aging. As we now know, certain data on
COX
-2 inhibitors were not initially made public, and while it is understandable to make recommendations based on available information, it is also important to sustain a sense of sobriety and wait for more extensive and long-range assessment before being swept toward a conclusion influenced, in part, by one's personal desires or the seduction of pharmaceutical marketing.

For decades, the bulk of the data concerning estrogen therapy for premenopausal women came from the Nurses' Health Study, begun in 1976 and sponsored by Harvard. This was a so-called observational study, meaning that large numbers of nurses reported what drugs they took, what they ate, and what they did during the day. From those reports, researchers drew inferences about what constituted healthy activity and what did not. Although observational studies can yield useful information, they can be misleading. Hidden biases may prevent subjects from reporting certain factors that affect health or disease. A prospective trial including both treatment and a placebo almost always results in more reliable data than an observational study. The Women's Health Initiative, established in 1991, sponsored by the NIH, and involving more than fifteen thousand women, was a prospective study of hormone replacement therapy and its benefits and risks. The study was to last for fifteen years but was stopped early when an independent board of experts concluded that the hormones estrogen and progestin increased the risk of breast cancer in healthy menopausal women. An increased incidence of coronary heart disease, stroke, and pulmonary embolism also was observed among the women taking hormones, compared with those receiving placebo pills. These results cast serious doubt on what had become conventional wisdom since the Nurses' Health Study.

But even before the Women's Health Initiative results were released in 2002, other data contradicted the idea that aging women should be given estrogen to prevent heart disease, stroke, and Alzheimer's. "It always bothered me that the Framingham Study did not show that estrogen protected women against heart disease," Delgado said. The Framingham Heart Study, a large, long-term study of risk factors for atherosclerosis and heart disease, was an "outlier," since it seemed to contradict the Nurses' Health Study. "It was hanging in my head all those years," Delgado said. But unlike many of her colleagues, she chose not to ignore it. Then the Heart and Estrogen/Progestin Replacement Study (
HERS
) appeared. A drug company sponsored this placebo-controlled trial of estrogen hoping to show that the hormone helped prevent a second heart attack in older women. The results, however, showed the opposite. Yet that negative outcome gave most clinicians no pause in prescribing estrogen. The powerful marketing juggernaut seemed to sweep aside any obstacle in its way.

In early 2006, a
Wall Street Journal
headline declared: "In Study of Women's Health, Design Flaws Raise Questions." A
New York Times
headline read, "Rethinking Hormones, Again." Headlines are meant to catch the eye, but they risk imprinting misinformation in the mind. The
Journal of Women's Health
also published an article using data from the Nurses' Health Study. Although one should not necessarily judge an article by where it appears, there is a pecking order in clinical medicine. The
New England Journal of Medicine
and the
Journal of the American Medical Association
(
JAMA)
are the alpha roosters. In my own specialties, the
Annals of Internal Medicine, Blood,
and the
Journal of Clinical Oncology
are the most prestigious. When researchers have rigorous, groundbreaking data to announce, they try to publish in one of the toptier journals; by the same token, these journals seek out epochal reports to add to their luster.

The Women's Health Initiative and the
HERS
study were published, respectively, in the
New England Journal of Medicine
and
JAMA.
The earlier papers from the observational Nurses' Health Study appeared in these journals as well, but as it became clear that self-reporting and other biases seriously limited the Nurses' Study, its credibility declined. The media are hungry to pursue topics that are not only controversial but draw in readers with desirable demographics; hormone replacement therapy is sure to find an eager audience with disposable income. Journalists accurately reported the information in the news articles: women who started hormone therapy right after menopause reduced their risk of coronary heart disease by about 30 percent. This observation led to the hypothesis that estrogen might best be used in women in their early fifties, at the onset of menopause, and that the average age of the participants in the Women's Health Initiative—sixty-four—could account for the failure of estrogen to protect their hearts.

The articles revealed a subtle insight into the culture of medicine. Where you stand depends on where you sit: your specialty can affect, even determine, your position. In this case, gynecologists and cardiologists came out on different sides. Some gynecologists, who are often the primary care physicians for women, and who had prescribed estrogen as a mainstay of their practice for decades, rushed to embrace the new data from the Nurses' Health Study, even while acknowledging that they were not definitive. Dr. Mary Jane Minkin, a clinical professor of obstetrics and gynecology at Yale, told the
New York Times,
"Personally, in my heart of hearts, I think there is a benefit." Dr. Minkin disclosed that she was a consultant and paid speaker for drug companies that make estrogen, and she took the hormone herself.

Delgado likened Minkin's remarks to Dr. Foyer's. "She is a believer, too." Delgado thought it unlikely that Minkin's "belief" was based on money. But was it based on science? For a patient to hear that a professor at a prestigious medical school like Yale believes "in my heart of hearts" in a treatment so strongly that she takes it herself—that has a powerful impact. "I didn't think these doctors are prostituting themselves for the drug companies," Delgado said. Rather, they are speaking not objectively, but from faith. It is not uncommon to find such believers among physicians.

Many readers do not go beyond an article's headline or its opening paragraph; it is also difficult for laymen to critically assess statements coming from apparent voices of authority. For example, toward the end of the
Times
article, Dr. Frederick Naftolin, the retired chairman of the Department of Obstetrics, Gynecology, and Reproductive Sciences at Yale, disputed the data from the Women's Health Initiative study as counterintuitive. "The relationship between the fall in estrogen and the rise in cardiovascular disease in women is incontrovertible," Dr. Naftolin said. "So why in the world would you not try to find out whether simply maintaining estrogen at the levels of reproductive life could be cardio-protective?"

In response, cardiologists argue that this theory has been discredited. "Atherosclerosis starts well before the age of menopause," said Dr. Deborah Grady, a principal investigator in the
HERS
trial. "On top of that, why would you want a preventive intervention that has a lot of other side effects like blood clots? These people have a theory they don't want to give up on, no matter what." Dr. Richard M. Fuchs, a cardiologist and clinical professor of medicine at Weill Medical College in New York, echoed this view. "There is no good evidence that hormone therapy reduces the risk of heart disease, and there is reasonable evidence to say it increases heart disease and stroke, pulmonary embolism and breast cancer," Dr. Fuchs said. "My advice is all women should try to get off it."