Women who take daily doses of estrogen after menopause have about a 36 percent greater risk of developing breast cancer than women not taking the hormone, but the excess risk disappears within a year or two if a woman stops the drug, according to a study published today.

"Estrogen replacement therapy" is widely prescribed for women after menopause because studies have established that daily doses of the female hormone can reduce the incidence of heart disease and osteoporosis, a gradual weakening of the bones that can lead to disabling hip and spinal fractures.

But experts have long worried that giving older women estrogen might increase the frequency of breast cancer, which kills about 44,000 American women annually. However, a number of studies have failed to find such an increase.

Now, the Nurses' Health Study, an ongoing assessment of the health and lifestyles of more than 120,000 female registered nurses, has found that taking estrogen after menopause causes a moderate but significant increase in breast cancer rates. Previous studies apparently failed to see the increase because they looked mainly at past users of the drug.

However, for most women the benefits of estrogen treatment probably outweigh the modest increase in breast cancer risk, said Meir Stampfer, an associate physician at Boston's Brigham and Women's Hospital and a coauthor of the study, published today in the Journal of the American Medical Association.

"I wouldn't recommend that a woman change her current practice based on these data alone," Stampfer said. He added that a woman and her doctor should weigh her medical history and personal risk factors for heart disease, breast cancer and osteoporosis in deciding whether to take the hormone.

Heart attacks kill about 247,000 American women each year, according to the American Heart Association. Stampfer said more than two dozen studies have shown that taking estrogen reduces a woman's risk of heart disease by about 50 percent.

"It's very, very reasonable for different women in different circumstances to come to different conclusions," he said.

Researchers enrolled 121,700 nurses in the study in 1976 and questioned them regularly about their habits and health over the following decade. Stampfer said about 23,000 of the women had completed menopause when the study began, and by 1986, that number had grown to about 50,000. The proportion of post-menopausal women who were taking estrogen varied during the 10-year study from less than 20 percent to as high as 33 percent.

Statistical analysis of the cancer cases that developed revealed that if a woman was currently taking estrogen, her chance of developing breast cancer was 36 percent greater than that of a woman who had never taken the drug.

Stampfer said this elevated risk disappeared within two years after a woman stopped the drug, regardless of how long she had taken it.

When the researchers examined the relation of estrogen with other possible risk factors for breast cancer, they got a surprise: Estrogen appeared to raise breast cancer risk only in women who also drank alcohol. Stampfer said he could not explain the finding, and that it could be a statistical aberration. If it is real, it could reflect some metabolic interaction between alcohol and the hormone.