Taking the drugs around the time of menopause carries significantly more risk for the disease than starting them later, after a gap of at least five years, the study found. And while the study confirms that breast cancer risk rises with increasing time on the drugs, even women who took them for less than five years had more breast cancers than nonusers.

The analysis, published Jan. 28 in the Journal of the National Cancer Institute, looked at data from the Million Women Study, which recruited a quarter of all British women ages 50 to 64 in the late 1990s through 2001. Though not a clinical trial in which participants are randomly assigned to either a treatment or control group — considered the most telling and bias-free approach — the size of this study makes it a key addition to the Women's Health Initiative. In that large U.S. trial, the study of estrogen-progestin HRT was halted in 2002 because of evidence that the treatment, while reducing colon cancer and hip fractures, promoted heart disease, stroke and blood clots — as well as breast cancer.

Millions of women still take hormones

Since then, millions of women have stopped taking hormones, and many physicians have stopped recommending long-term hormone therapy for prevention of cardiovascular disease and osteoporosis, a common practice in the 1990s. But millions of American women continue taking hormone replacement to relieve hot flashes, vaginal dryness and other menopausal symptoms. Clinical guidelines generally advise taking hormones at the lowest dose and for the shortest time possible — typically around the time menopause begins — when symptoms strike.

Assessing the risk

Most women's absolute chance of getting breast cancer is small. In the British study, the incidence of breast cancer among women in their 50s who never used hormone therapy was 0.30 percent, doubling to 0.61 percent in women who took an estrogen-progestin combination before or within five years of menopause.

Also reassuring is the finding, affirming earlier studies, that when hormone users stopped taking the drugs, their risk fell to that of nonusers within a few years.