Healthy Skepticism Library item: 18

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

Publication type: news

Parker-pope , T .
The Case for Hormone Therapy: Menopause hormones have been battered by recent studies highlighting their potential dangers. Here's why women might still want to take them.
The Wall Street Journal 2003 Oct 21

Full text:

Last summer, millions of middle-aged and older women woke up to some shocking news: The daily menopause hormones they had come to depend on to even their mood and body temperature, to help them sleep, improve their sex lives, protect their bones and possibly prevent heart disease and Alzheimer’s, seemed to have turned against them.

Doctors conducting a major study of the popular estrogen-progestin combination known as Prempro had suddenly halted the research, citing higher rates of breast cancer and heart problems among Prempro users. In the months that followed, millions of women threw away their hormone pills as an unrelenting barrage of “new” studies warned about the dangers of hormone therapy in general.

But lost amid the headlines and the hysteria was something crucial: the facts.

The fact that nearly all of the revelations came from the same single study of 16,600 women known as the Women’s Health Initiative. The fact that significant questions remain about the type of hormones used, the age of the women who participated and the statistical strength of the data mined from the research.

And the fact that many doctors subsequently concluded that the study told them very little, if anything, about the typical woman who enters their office seeking relief from the symptoms of menopause. As a result, these doctors — and many of the WHI researchers themselves — believe that despite the fears the study raises about breast cancer, heart attacks, strokes and dementia, it’s still reasonable and even wise for women to consider taking the drugs anyway.

“What can we really draw from this study?” asks Alan M. Altman, an assistant professor of obstetrics at Harvard Medical School who, like many doctors on both sides of the debate, has accepted speaking fees from drug firms. “One of the things we can’t say is that it means hormones aren’t good for you. That’s unfortunately the connotation that’s been attached to it.”

While the extent to which women should consider hormone therapy remains controversial, even those who led the WHI study now concede many people have overreacted to the results.

“There may be an element of going overboard,” says Harvard professor JoAnn E. Manson, chief of preventive medicine at Brigham and Women’s Hospital in Boston and lead author on the WHI’s heart report. “I think it’s going too far for the study to be interpreted as saying hormone therapy has absolutely no role in clinical practice any longer.”

Where It Began

To make sense of the WHI study today, you have to understand the thinking at the time work on the study began in 1991. As is the case now, everyone back then pretty much agreed that hormones helped ease the symptoms of menopause and probably increased the risk of breast cancer and blood clots. But at the time, many doctors also believed the drugs protected a woman from all sorts of health problems, particularly heart disease.
FURTHER READING . . .

“ Hormone-Therapy Debate Grows
10/08/03

“ HRT Risks Outweigh Bone-Loss Benefit
09/30/03

“ Hormone Use Holds Risks for Women
08/07/03

“ Study Finds New Risk in Prempro
06/25/03

“ Makers of Other HRT Question Study
06/25/03

“ Study Finds New Hormone Danger
05/28/03

The thinking was that all the supposed added protection against chronic disease far outweighed the risks. As a result, estrogen and progestin combinations were commonly prescribed to women who didn’t have any symptoms, including those who were years past menopause. And many women were advised to stay on hormones for years or even their lifetime.

The main evidence in support of hormone therapy came from an ongoing study of more than 48,000 nurses, which in 1991 showed almost a 50% reduction in heart-attack rates among women who used hormones. The study wasn’t conclusive, but the idea that hormones would protect women’s hearts made sense, because it was known that heart disease sets in a full decade later in women than in men, presumably because a woman’s natural estrogen lends some extra fortification against heart disease.

And so the Women’s Health Initiative set out to see whether hormones really did protect women from developing heart problems and other chronic disease. Armed with $628 million from the National Institutes of Health, researchers at 40 sites around the country in 1993 began recruiting women for the WHI, which was expected to end in 2005.

Two Hurdles

The study, though, contained two huge hurdles that, in retrospect, severely limited its usefulness. The biggest challenge: studying health problems that typically afflict older women. If the study had too many young women, it would take years for them to get old enough to generate enough “events” like heart attacks and fractures to provide a statistically meaningful analysis.

Another problem was finding enough women willing to risk receiving a placebo. Women with severe menopause symptoms — such as hot flashes and night sweats — were specifically discouraged from enrolling in the study because they probably would know almost immediately whether they were getting a sugar pill instead of the real thing.

————————————————————————————————————————

More Questions Than Answers?
Critics say the results of the Women’s Health Initiative, which found higher risks for women who use hormone therapy, may not apply to most women. Here’s a closer look at the study and the critics’ case.
THE WOMEN IN THE WHI CRITICS SAY
Average age: 63 Typical woman seeking help for menopause is in her early 50s
Used Prempro Questions remain about whether results can be extrapolated to all formulations of estrogen and progestin
Used oral drugs Some evidence indicates patch and vaginal therapies don’t pose similar clotting risks
Many appeared unhealthy entering the study Study may have included women with a higher undiagnosed heart disease than expected, making the results less applicable to healthy women
Few had menopause symptoms Women with menopause symptoms might respond differently to hormones, in terms of risks and benefits
Many started hormones 10 to 15 years after menopause Some studies show starting hormone therapy at the onset of menopause offers benefit, but starting late doesn’t
An 81% increase of heart attack in first year Typical women with symptoms have such a low risk to start with, the extra risk usually isn’t a worry
Twice the risk of dementia Minimal risk to young women, and the age and timing of therapy could affect results

Source: Wall Street Journal reporting

————————————————————————————————————————
As a result, the study, at its inception, was designed to be a study primarily of older women well past menopause, and would include only a small portion of younger women, most of whom didn’t suffer from severe menopausal symptoms.

This approach made sense under the circumstances — after all, older women without symptoms were widely being prescribed hormones to stave off disease — and the medical community and the researchers themselves expected the study to produce results strongly in favor of long-term hormone therapy.

But these assumptions also resulted in some significant potential problems with the data. What if older women and younger women react differently to hormones? What if the timing of hormone therapy — whether it’s given just at the time the body stops producing estrogen or whether it’s started years later — makes a difference? With so few women in the 50- to 54-year-old age group, could the study provide any meaningful conclusions about younger women, the group most likely to be taking hormones for menopausal symptoms in the first place?

In the end, the average age of a woman in the WHI was 63. Of the 16,600 women recruited to the study, 67% were over the age of 60, and only 1,700 women were between the ages of 50 and 54.

The Study Is Halted

Fast forward to July 2002, when researchers announced they had stopped most of the study early. Women taking Prempro were, as expected, showing higher rates of breast cancer than the placebo group. Although the women had lower rates of colorectoral cancer and fractures, they had higher rates of stroke, pulmonary embolism and, most surprising, heart disease.

The study had answered exactly the question it set out to: It found that in older women without menopausal symptoms, Prempro didn’t prevent heart disease or offer enough benefits to justify an increased breast-cancer risk.

“It was becoming common clinical practice to put women in their 60s and 70s on hormone therapy for the purpose of preventing many chronic diseases,” says Dr. Manson. “This study answered that question irrefutably. An important practice has been shown to be more harmful than beneficial.”

But the fact that this study answered only a very narrow question about hormone therapy was lost in the media frenzy that ensued. Although the WHI researchers chose their words carefully, they weren’t careful enough. Women and many doctors were left believing that all menopause hormone therapies — not just Prempro — were dangerous and no longer an option for dealing with menopause. Even though another part of the study following women with hysterectomies taking only estrogen wasn’t halted, sales of Wyeth’s Prempro quickly plummeted as women abandoned hormones.

Growing Concerns

But in recent months, as additional data about the WHI study have emerged, critics have become increasingly vocal about what they see as flaws in the study. The strongest concern is that the women in the WHI were simply too old — and started hormone therapy too late in life — for researchers to come to any meaningful conclusion about the value of the treatment for women facing menopause.

“It’s absolutely critical to point out that the Women’s Health Initiative does not speak to younger women,” says Elizabeth Lee Vliet, a Tuscon, Ariz., physician and author of several books on menopause. “You’re looking at a 30-year age difference between the average menopausal woman who comes into my office versus (some of the) women that were in the WHI. It’s like taking a group of old men at risk for prostate cancer and saying it applies to young men.”

Marcia Stefanick, associate professor of medicine at Stanford University and chairwoman of the WHI steering committee, concedes that the WHI “was very much a study about older women.” But she says the study is still useful for younger women. “We have a larger sample of women in their 50s than have ever been studied before,” says Dr. Stefanick.

But the reason the age of the women has become such a flashpoint in the debate is that many experts have long believed that the timing of hormone therapy — whether it begins just as the body loses its own natural hormones or whether it’s delayed — does matter. The theory is that hormones started early can help, but hormones started later may cause harm. And most of the women in the WHI started hormone therapy at least 10 to 15 years past menopause.

Lard-Eating Monkeys

This theory that the timing of hormone therapy makes a difference has been bolstered by studies of menopause in monkeys at the Wake Forest University School of Medicine in Winston-Salem, N.C. After researchers induced menopause in cynomolgus macaques by removing their ovaries, some of the monkeys were left alone, while others were given Prempro. In another set of monkeys, the Prempro was delayed for two years — the equivalent of six years for a woman. All of the monkeys were fed heart-disease-inducing diets consisting of lard, butter and eggs.

After three years, the monkeys that started hormones immediately had up to 70% less plaque buildup in their arteries than the monkeys that didn’t take hormones. And notably, the hormones didn’t help the monkeys who received delayed hormone treatment. The study suggests that Prempro preserves artery health when started early, but it can’t repair the damage once heart disease sets in.

While the monkeys share 94% of their DNA with humans, they’re still monkeys, and it’s impossible to draw any definitive conclusions about women. But the monkey studies do raise a big question about the WHI: Would the results have been different if more women in the study had started hormone therapy earlier?

“What the WHI sheds no light on is the effect of starting therapy as soon as women become estrogen-deficient,” says Thomas B. Clarkson, professor of comparative medicine at Wake Forest, who led the monkey research. “The monkey data would say they’re going to get great beneficial effects if initiated early.”

WHI researchers say the monkey comparison isn’t valid because the animals used in the research were the equivalent of girls in puberty. “For someone to say a monkey study is better than the WHI for demonstrating benefit for heart disease is really kind of insulting as a woman,” says Dr. Stefanick. “I believe a 50-year-old woman is a lot more like a 60-year-old woman than she is like a monkey — especially a pubertal monkey.”

When considering the results of the WHI, age matters in another important way. Younger women have a far lower risk of disease than older women, so any increase in risk simply isn’t as alarming.

This is especially true with a particularly startling conclusion of the WHI: that women have an 81% higher risk of heart attack during the first year of hormone therapy. Though it sounds scary, the average woman in her early 50s has such a low risk of heart attack in the first place that an 81% increase isn’t much to worry about. For those women the slightly higher risk “may be a price worth paying” to relieve menopausal symptoms, notes Dr. Manson.

Age and Dementia

The age issue has also been raised to challenge the WHI’s conclusions about dementia and cognitive function.

The WHI found that Prempro doubled the risk of dementia in women over 65. Why it would trigger a decline in mental function is unclear, but the drug’s clotting effects could interfere with blood flow in the brain, says Jean Wactawski-Wende, co-author of the WHI memory report and an assistant professor at the State University of New York at Buffalo.

But young women are at very low risk for such problems, and some speculate that women who start hormones sooner than most of the women in the study might gain protection from mental decline. In a well-known Utah study, hormone use for 10 years or more was associated with a decline in Alzheimer’s risk. And in a Yale University School of Medicine study published in the September issue of Menopause, middle-aged women on daily estrogen showed improvements in oral reading and verbal memory.

Hormone therapy “has been used in such a cavalier way, and now it’s being dismissed in a cavalier way,” says Yale professor Sally Shaywitz, lead author of the reading study. “But there’s a whole world in between.”

Perhaps the least surprising finding of the WHI was that hormone use increases risk for breast cancer. Though the overall risk is 24% higher among hormone users, that translates to eight more cases each year per 10,000 women, says Rowan T. Chlebowski, a professor of medicine at the Harbor-University of California at Los Angeles Research and Education Institute and lead author on the WHI breast-cancer report.

Overall, the researchers concluded that even after the reduced risks of fractures and colon cancer were factored in, over five years, hormone use would trigger 19 additional life-threatening events each year for every 10,000 women.

But critics of the WHI heart study have become emboldened in recent months by a little-noticed change in the WHI heart data. In August, more than a year after the initial WHI report, the New England Journal of Medicine published the final results of the heart study, after all the tallying and number crunching had been finalized and confirmed by WHI investigators. The confidence interval — a key number that every statistician uses to show that a result is meaningful and unlikely to have happened simply by chance — had slipped. Suddenly, the biggest and most shocking data in the study — that hormone users had a 24% increase in heart risk — no longer met the threshold for statistical significance.

To be sure, the finding fell short of statistical significance by the slimmest of margins. Does it mean that part of the data should be ignored altogether? Probably not. But critics have nonetheless seized on the issue, noting that it raises one more question in a study already plagued by questions.

“The actual statistical significance is open to question,’‘ says Wulf H. Utian, executive director of the North American Menopause Society and a Cleveland Clinic consulting physician. “There is only one year in all the [heart data] where there is a statistical increase, but this has terrified the population.”

Dr. Manson says that even if controversy remains about whether the hormones increase risk to the heart, the main point of the report is that they don’t protect it. She says the assertion that the data miss statistical significance is “splitting hairs,” adding, “Virtually no one can look at these results and say there isn’t a suggestion of increased risk.”

• * *
  The question of what one can and can’t conclude from the WHI continues to polarize the medical community. Questions remain about whether the data on Prempro can be extrapolated to all hormones. Some wonder whether the type of progestin used is the main culprit — fueled by the fact that part of the WHI looking at women on estrogen alone wasn’t stopped. There’s some evidence that patch therapy doesn’t pose the same clotting risks as oral drugs. Others wonder whether the WHI women may have had high rates of undiagnosed heart disease at the start of the study.

But in the months following the first release of the study results, two key areas of consensus about the WHI and hormone therapy have emerged. Most everyone now agrees that older women shouldn’t start taking hormones if they are well past menopause. The research shows clearly that there’s no benefit and that starting hormones late in life may even cause harm.

More surprising, though, is the fact that experts also agree that younger women with distressing menopausal symptoms should still consider taking the drugs, at least for a short time. While the risks noted in the WHI may seem worrisome, any increased risk, particularly for heart disease, would remain exceedingly low for young women. And for women with severe symptoms, the quality-of-life benefits would factor more strongly into the decision.

But while WHI investigators say young women should take hormones for the shortest time possible, not everyone agrees. Critics say the study is too limited to close the book on the potential benefits of long-term hormone therapy.

Some doctors, convinced that the pills do offer protection from chronic disease when started early, still think young women should take hormones long term. And they are advising patients who are already taking hormones to continue, despite the warnings of the WHI study.

Even though WHI investigators disagree with that approach, they also say it was never the intent of the WHI to scare women away from hormones for short-term relief of menopause symptoms.

“Nothing about this study should preclude women who need the hormones to take the hormones unless you’re at very high risk,” says the WHI’s Dr. Stefanick. “Whatever was said 15 months ago, that’s when it was a shock for everybody. We’ve had a year to calm down and get back to the fact that there still is a place for hormones for treating women.”