Breast cancer's link to alcohol assailed.

Breast cancer's link to alcohol assailed

Ties between alcohol consumption and breast cancer tightened last year when two studies found that even two to three drinks per week significantly increase the risk of the disease among women (SN: 5/9/87, p.292). But those apparent statistical bonds may be unraveling, according to recent scientific reports. Researchers now say the link between breast cancer and drinking alcohol is weak, if present at all.

In a study by the American Health Foundation in New York City, Randall E. Harris and Ernst L. Wynder used personal intterviews to assess alcohol consumption and other risk factors among 1,467 women with breast cancer and 10,178 age-matched female hospital patients used as controls. Study participants, who came from 20 hospitals throughout the United States, were part of a larger, ongoing study designed to assess tobacco-related diseases.

After adjusting for confounding variables--including those known to be associated with breast cancer risk, such as a woman's age at first pregnancy -- the scientists say they found no solid evidence that any amount of alcohol increases breast cancer risk.

Harris and Wynder concede in the May 20 JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION that their results "do not entirely rule out a weak association...in certain subgroups [such as leaner women]," but add that their results fail to provide "compelling evidence that alcohol has a role in the genesis of [breast cancer]." They explain that it is difficult to separate socioeconomic factors' effects on drinking habits from their effects on reproductive histories and other potential cancer risks.

Those conclusions are echoed by data from the Centers for Disease Control (CDC) in Atlanta, where Susan Y. Chu and her co-workers interviewed roughly 3,000 women with breast cancer, plus an equal number of controls. Earlier this year at an American Cancer Society seminar in Daytona Beach, Fla., Chu reported that "overall, the CDC study found no relationship between drinking alcohol in the last five years [prior to the study] and breast cancer risk."

Like Harris and Wynder, Chu acknowledges such studies have inherent problems and rely on subjects' recall of drinking habits from years past. Rather than the usual approach of estimating the amount of alcohol per day or week, it may be better to consider total years of drinking, Chu says. She says studies that find an alcohol/cancer link should not be ignored, but emphasizes that "positive findings are differnt from conclusive results."