Long-Term Hormone Therapy May Reduce Alzheimer's Risk for Women

By GINA KOLATA

omen who take hormone therapy after menopause and continue with it for 10 years or more may have a reduced risk of Alzheimer's disease, researchers report.

But scientists, including the study's lead investigator, caution that the study's findings are suggestive, not definitive, in part because of the study's design and because relatively few women who took part developed the disease. For now, medical experts say, there is not enough evidence to tell women, even those at high risk of Alzheimer's, to take estrogen to lower their risk.

The Alzheimer's study, by Dr. John C. Breitner of the Veterans Affairs Puget Sound Health System and his colleagues, involved 1,357 elderly men and 1,889 elderly women living in Cache County, Utah. The researchers questioned the study participants and gave them memory tests for Alzheimer's. Three years later, they tested them again. During that time, 35 men and 88 women developed Alzheimer's. Using statistical modeling, the investigators concluded that estrogen might cut a woman's risk in half, but only if she took it for 10 years or more.

Dr. Breitner emphasized, however, that his study "does not rise to the level of evidence that would warrant physicians prescribing hormone therapy for this indication or for women to start using it for this indication."

Dr. Bill Thies, the vice president for medical and scientific affairs at the Alzheimer's Association, agreed. "Any effect of estrogen on cognitive functioning is not well enough established to make that the determining factor," Dr. Thies said.

The new paper, published today in The Journal of the American Medical Association, comes on the heels of a major reassessment of postmenopausal hormone therapy. In the 1990's, many doctors told women to start taking hormones at menopause and to continue for the rest of their lives to protect their health. But in July, a large federal study reported that hormone therapy involving estrogen and progestin had slight risks — of heart attacks, stroke, blood clots and breast cancer — that were not balanced by its slight benefits, a small reduction in bone fractures and in colon cancer. In response, leading medical groups said healthy women should not use hormone therapy except to relieve disabling symptoms of menopause, like severe hot flashes.

Researchers said that if estrogen reduced the risk of Alzheimer's disease, that might swing the balance, making it advisable for some women to take it.

Alzheimer's studies, however, have had conflicting results. Animal and laboratory studies indicated that estrogen might protect the brain. But clinical studies have failed so far to make the case. Dr. Breitner said his findings might explain the previous studies by positing that there is a critical period, years before Alzheimer's symptoms emerge, when a woman must take estrogen if she is to reduce her risk.

"What's new here is that the benefits, if there are any, are delayed a long time after treatment," Dr. Breitner said in a telephone interview this week.

He explained, however, that his study observed whether women who chose to take estrogen were less likely to develop Alzheimer's disease. One problem is that women who take hormones after menopause are different from those who do not take the drugs. Hormone users tend to be more educated, for example, and healthier. Some of the very features that make them likely to take hormone therapy are also features that, independently, are associated with a lower risk of Alzheimer's disease. Scientists use statistical adjustments to try to correct for the differences between hormone users and nonusers, but they can never be sure they have succeeded.

"The danger is that it is not the use of hormones that is driving this relationship, but it is being the kind of woman who uses hormone therapy," Dr. Breitner said. "There is no way you can get around that except by doing a randomized controlled trial," in which women are randomly assigned to take, or not take, the drugs. Such studies of estrogen and Alzheimer's risk are now under way.

In fact, it was observational studies that helped convince medical experts — incorrectly, as it turned out — that estrogen protected women from heart attacks and strokes.

Such studies can have other drawbacks as well, and some occurred in Dr. Breitner's study. For example, its conclusions were based on just 88 women who developed Alzheimer's in the study period. The researchers subdivided those women into hormone users and nonusers and, among the users, into those taking hormones currently and those who took them previously. They also took into account the duration of the drug therapy. The number of women who fell into any of these groups was small, making the study's conclusions less certain.

In addition, about 20 percent of the women declined to participate. It is possible, said Dr. James Robins, a professor of biostatistics at the Harvard School of Public Health, that they might have included a disproportionate number of women in the early stages of Alzheimer's. Their exclusion may have altered the study's results if they were also more likely — or less likely — than the others to be taking hormones.

Even determining who took estrogen, and when, can be fraught with difficulty. Dr. Norman Relkin, a neurologist at Cornell Medical School, noted, for example, that the researchers ascertained hormone use by asking the study subjects.

"Even a small number of women making a misrecollection could bias the results," Dr. Relkin said.