Defining the Future
The story behind the landmark women's health study that is creating exciting breakthroughs
Marilyn Swope, 75, of Zanesville, Ohio, has worn many hats in her life: first-grade teacher, union president, city mayor, quilt shop owner, and great-grandmother. But one of her most cherished roles was guinea pig. A decade ago, Swope saw an article in the newspaper seeking participants for the Women's Health Initiative, a massive federal effort to study the major causes of death, disease, and disability in older women. She felt a responsibility to sign up immediately. Zanesville, about 55miles east of Columbus, had such high incidences of cancer and heart disease, "people always think there is something wrong with the water," says Swope. But she suspected it was more than that. "We are habituated to eating a lot more than what we really need."Swope agreed to enroll in a trial testing the health effects of a low-fat diet and committed the next 10 years of her life to eating in a radically different way. "Maybe it's not going to help you," says Swope. "I was doing it for my family, for my daughters."
Swope probably didn't know it at the time, but that pledge put her smack dab in the middle of a new kind of women's movement. With more than 160,000 participants, the Women's Health Initiative is the largest government trial of its time and is perhaps the most dramatic and ambitious act in a long-running debate on women's health. The study aimed to enroll 1 out of every 200 women ages 50 to 79 and track her for the next seven to 12 years. With its sheer size and scope, as well as an eventual $725 million in funding, the WHI made a defiant statement against scientific trials that excluded women and challenged long-held medical beliefs. "By far and away, it's the largest and most comprehensive longitudinal investigation of women's health issues that's ever been undertaken," says Duncan Thomas, codirector of the Division of Biostatistics at the University of Southern California Keck School of Medicine.
Shock wave. One thing's for sure: The WHI has certainly made a mark. In 2002 and again in 2004, the WHI abruptly halted its two hormone therapy trials, rocking the world of medicine by casting doubt on what most experts thought was a proven tool in preventing heart disease and prompting women everywhere to toss out their estrogen pills. Now new studies are re-examining the role of hormones for younger, premenopausal women (story, Page 66). Earlier in February, the initiative caused another shock wave when researchers announced that in spite of a national obsession with fat-free products, a low-fat diet alone has little effect on cancer. And all the vitamin D and calcium supplements that women were taking don't seem to prevent bone fractures.
Some critics fault the findings on the low-fat diet and vitamin supplements: A lot of money was spent to prove something wrong. Others disagree. "It's very good to put theories to the test--some work, and some don't,"says David Freedman, a statistics professor at the University of California-Berkeley. "It's a big study, but expense is kind of relative given the public health costs of giving wrong advice."
The new research is sure to get another close look during this week's conference examining the future and legacy of the WHI. The findings are more complex than headlines would have you believe, say WHI investigators, and the nuances have yet to be explored. For example, women who took at least 80 percent of their calcium supplements had 29 percent fewer hip fractures than those on placebos. And perhaps down the road, future follow-up may reveal greater benefits to that low-fat diet. Still, some critics blame the WHI for creating confusion about women's most important health concerns. "When you have a study that is this big, the tendency is to think it's going to answer all of our questions--like a Santa Claus bag we should be able to keep reaching into," says Jennifer Hays, a WHI principal investigator at the Baylor College of Medicine.
Uncertainties. Given the paucity of medical research on women, experts say it's only natural that there are still so many uncertainties. In the past, researchers sometimes excluded women of childbearing age to avoid harming fetuses if the women became pregnant. And scientists also claimed men were easier to recruit and retain, that women were unwilling to participate, and that female hormonal cycles would skew results.
But in the early '80s, there was a growing consensus that those perceptions had to change. A few major studies galvanized the cause: the Multiple Risk Factor Intervention Trial, or "MRFIT," which set out to test cholesterol-lowering drugs in 15,000 men, and the Physicians' Health Study, which tracked 22,071 male doctors to examine aspirin's effects on heart disease. In both cases, researchers excluded women because heart disease was considered a men's disease--even though it was and still is the leading killer of women.
Such bias can translate into poor diagnosis and treatment of real women (story, Page 74). Julia Carvalho, for example, lost her mother to a heart attack at the age of 53. "I can remember doctors trying food fads to get her blood pressure down. Once she ate nothing but white rice," recalls Carvalho, now 85, of Carmichael, Calif. The mother of seven was worried about her own heart. But when she tried to learn more about diet or lifestyle changes that prevent heart disease, "every study I read about was a study made on men, not women."
In 1986, the National Institutes of Health began urging research applicants to include women in studies. But when the Congressional Caucus for Women's Issues asked for an evaluation three years later, little had changed. The report served as "a watershed moment that galvanized public opinion and spurred Congress to action," says Republican Sen. Olympia Snowe of Maine, then cochair of the caucus.
Around the same time, physician and former chairman of the research institute of the Cleveland Clinic Foundation Bernadine Healy (now health editor at U.S.News) was appointed the first female director of the NIH. She was in the eye of a perfect storm: Key policymakers and scientists were well aware of the research disparity, and a powerful bipartisan contingent in Congress was eager to support a feel-good women's health cause that had nothing to do with abortion or reproductive rights. Healy decided the time was ripe for a large-scale study of the diseases that beset older women.
To get Congress to approve funding of the project, Healy and her staff had to nail down exactly what the initiative would tackle. Diabetes, which wasn't yet the epidemic it is today, was passed over because important trials on the disease were already in progress. Antioxidant research was also already in the works. Lung cancer, a major killer of women, was rejected because research had already found effective prevention--avoid cigarette smoke and certain pollutants. At the time, it was common for doctors to prescribe hormones to prevent cardiovascular disease and a low-fat diet to stave off breast cancer. But there were no clinical trials to back the advice. Another piece of conventional wisdom that begged examination: Do calcium and vitamin D supplements--which women were popping like crazy--prevent hip and bone fractures?
Tasked with the job of running the WHI was Jacques Rossouw, an NIH gastroenterologist from Cape Town, South Africa, whom fellow researchers call the heart, brain, and soul of the WHI. But he fell short in one respect. "I'm gender impaired for a women's research program," says Rossouw. "The public face should be a woman. It's just good politics and good sense." So Rossouw has been working behind the scenes, while the title of director falls to a series of female scientists.
Convincing. Rossouw and other WHI leaders had to choose 40 principal investigators to run clinical trial centers at hospitals and medical centers in places like Miami, Boston, Honolulu, and Tucson, Ariz. Each investigator's task: Recruit 3,500 women and persuade them to stay the course for the duration of the study. Recruiting a racial and ethnic mix was a prime goal. That was a particularly difficult task, as Joan Johnson, 72, of Islip, N.Y., discovered. "I had been one of these people saying that we of African ancestry are not put in trials and tests, so when they said they were looking for women over 50, I put myself on the line," says Johnson. She had a hard time drafting participants from her church, clubs, and her old sorority. A few women cited the Tuskegee Syphilis Study, in which government researchers left the disease untreated in black men for 40 years. A history of exploitation and ethical lapses in minority health trials had also contributed to a culture of distrust within minority communities.
The WHI leaders went all out to enlist participants: Ads and stories were placed in local media, former Murder She Wrote star Angela Lansbury made public service announcements, and representatives were hired for door-to-door solicitation. Their most fruitful tack turned out to be direct mail; the second or third mailing often resulted in enrollment. "Our goal was to make sure every woman had heard about it at least two or three times before she got that letter," says Baylor's Hays.
The result? By 1998, the WHI had enrolled more than 160,000 women. The women were rigorously screened. All had to be past menopause and between the ages of 50 and 79 and had to commit to staying in the area for at least three years. Women who had had breast cancer were not eligible for any of the clinical trials. Those who had suffered a heart attack in the past six months were banned from the hormone trial, and women who were already on a very low-fat diet were not allowed to participate in the dietary modification study.
"You have to go to several meetings" before you're even assigned to a group, says Lorraine Sandoval-Vigil, 69, a college counselor and retired high school principal in Watsonville, Calif., who joined to help her partner, who was battling breast cancer, switch to a low-fat diet. "They kept asking, are you sure you want to do this, because it's such a long period of time." After blood tests, a medical history, and a physical exam, qualified women chose the hormone therapy, low-fat diet, or both--and a computer randomly assigned them to the experimental or placebo group. At the one-year mark, women were invited to join the calcium and vitamin D study.
Women who weren't able to participate in the clinical trials joined about 93,000 others in the observational study, which monitored women as they went about their daily lives. It didn't involve swallowing pills or a low-fat diet, but it was a commitment nonetheless. After an initial physical checkup, the women were followed for about eight years. They filled out an annual 10-page survey detailing their health, diet, and lifestyle. Tracking down 93,000 women as they traveled, retired, moved, and remarried was a challenge, but by the end of the study, 94 percent of the original participants were still faithfully turning in their surveys. Because it wasn't a clinical trial, researchers were able to cast a wide net, scanning the histories for clues that led to coronary heart disease, stroke, a range of cancers, osteoporosis-related fractures, and diabetes (box, Page 68).
The hormone trial was perhaps the most anticipated project, as it dwarfed all other hormone studies ever undertaken. More than 16,000 women who still had their uteruses were assigned to either an estrogen and progestin pill or to a placebo. An additional 11,000 women who did not have a uterus took estrogen alone or a placebo. The trial was randomized and double blind, meaning neither the participants nor the clinic staff knew which women were on the hormone or the placebo.
Some 49,000 women were enrolled in the low-fat trial, a diet many experts thought could help prevent breast and colorectal cancer. Half of the women were told to cut their fat intake to 20 percent of total calories and to eat at least five daily servings of fruit and vegetables and six of grains. The remainder had no dietary restrictions whatsoever. The dieters met in groups with WHI nutritionists, who taught them how to count fat grams.
Before enrolling in the trial, Maisie Partridge of Roswell, Ga., thinks she was eating about 70 grams of fat a day. That became intolerable after her nutritionist showed her the amount of fat in the typical hamburger and fries. "They used that horrible white fat, Crisco, to illustrate, and by the time they were finished, they had a plate full of fat," says Partridge. The retired marketing specialist cut her diet to 32 grams of fat a day and lost 30 pounds in six months. Since then, she hasn't looked back, preferring simple poached salmon or chicken with fresh vegetables from her garden to her previous fare.
But Partridge, apparently, was something of an anomaly. The diet modification results, published in February, showed that women did not lose weight. Dieters were also not significantly less likely to experience cancer. However, the subset of women who went from a very high-fat diet to a low-fat diet had a 15 percent lower rate of breast cancer.
Since the study was created over a decade ago, some of the nutritional recommendations are distinctly dated. "You take fat-free brownies, and you layer it with instant chocolate pudding made with skim milk, and then use the fat-free Cool Whip on top. Let it set for 24 hours so the moisture really soaks into the brownies,"says Swope of one of her favorite WHI recipes. "Oh, the trifle,"sighs Bette Caan, a WHI primary investigator in Oakland, Calif. "I used to love that. But you eat it once and it tastes good, and eat it again and it doesn't taste that real." If the diet were designed today, says Caan, it would emphasize eating good unsaturated fats and more fresh fruits and vegetables and whole grains and avoiding bad saturated fats and trans fats.
All the data from the trials are funneled to a coordinating center in Seattle, which takes the raw numbers from the 40 centers and shapes them into a larger picture of women's health. Every six months, the data are re-evaluated. The results are passed on to an independent monitoring board of non-NIH scientists and ethicists. The coordinating center is also responsible for safekeeping the blood samples that participants have given over the years. Kept frozen at a facility in Rockville, Md., the samples are a rich source of information. Already, researchers looking at inflammatory biomarkers have noted that hormone therapy resulted in higher levels of C-reactive protein, which is a predictor of heart-disease risk. And an article detailing a connection between women on hormone therapy who experience adverse effects like stroke and heart attack and the presence of certain markers in their blood is in the works. Thus far, all private researchers have collaborated with WHI investigators. But in January, NIH opened the door to this treasure-trove of health data when it began soliciting study proposals. "I think we've seen just the tip of what we can learn from the WHI," says Mark Espeland, professor of public health sciences at Wake Forest University in Winston-Salem, N.C., who has used the study data in his research. "I'm expecting to see many things come out of this database."
The intervention trials ended as planned in March 2005. But about 80 percent of participants opted to continue with the extension study. For five more years, investigators will scrutinize the women's health through surveys. Researchers are intrigued by the notion of being able to compare a woman's blood sample from 1993 with the state of her health in 2010 (column, Page 78).
For many, WHI's legacy goes way beyond blood vials and genetic markers. Three years after Madelyn Glaeden, 67, of Jamesville, Wis., began the low-fat-diet trial, she was diagnosed with breast cancer. Her dieting group companions provided emotional support as she underwent surgery and chemotherapy treatments. "I needed that boost, the energy that the group gave me," says Glaeden. And that's exactly what the WHI was designed to do: give a boost--a big boost--to women's health research.
This story appears in the March 6, 2006 print edition of U.S. News & World Report.