CHICAGO (AP) — Whether sipping beer, wine or whiskey, women who drink just three alcoholic beverages a week face slightly higher chances for developing breast cancer compared with teetotalers, a study of more than 100,000 U.S. nurses found.

The link between alcohol and breast cancer isn't new, but most previous studies found no increased risk for breast cancer among light drinkers. The new research provides compelling evidence because it followed so many women for up to almost 30 years, experts said.

Still, the study only shows an association between alcohol and breast cancer; it doesn't prove that drinking causes the disease. There could be some other reason light drinkers appeared to be at higher risk — maybe they were less active than nondrinkers or had unhealthy diets, said Dr. Susan Love, a breast cancer expert and author who runs a Santa Monica, Calif.-based research foundation.

Women in the study who averaged three to six drinks a week throughout the study had a 15 percent higher chance of developing breast cancer than nondrinkers. That risk means, for example, that among women in their 50s, who on average face a 2.38 percent risk for breast cancer, light drinking would result in 4 additional cases of breast cancer per 1,000 women

Risks increased by 10 percent for every 10 grams of alcohol consumed daily. That's equal to a little less than one 12-ounce bottle of beer, a 4-ounce glass of wine or a shot of whiskey. The increasingly elevated risks were a little higher than seen in other research. It made no difference whether the women drank liquor, beer and wine.

Given research suggesting that drinking moderate amounts of alcohol including red wine may protect against heart disease, deciding whether to avoid alcohol is a personal choice that should be based on a woman's other risks for breast cancer and heart disease, the researchers said.

The study appears in Wednesday's Journal of the American Medical Association. It began in 1980, asking healthy, mostly white nurses aged 30 to 55 to fill out periodic questionnaires about lifestyle and risk factors for cancer and heart disease. Follow-up ended in 2008 or when women died or were diagnosed with cancer.

The researchers took into account other cancer risk factors, including age of menstruation and menopause, family history, weight and smoking, and still found a link with alcohol.

The strongest risks were seen with cumulative consistent alcohol use throughout the study. Increased risks also were seen in binge drinkers — women who consumed at least three drinks daily in a typical month. The results do not apply to women who may have partied hard during week-long vacations but otherwise rarely drank, said lead author Dr. Wendy Chen, a researcher at Brigham and Women's Hospital and Dana-Farber Cancer Institute and an assistant professor at Harvard Medical School.

"No one should feel guilty about one particular week or two," Chen said.

The results don't mean women can avoid breast cancer by not drinking, and they don't answer whether women can lower their risk if they stop drinking, said breast cancer specialist Dr. David Winchester, chief of surgical oncology with NorthShore University HealthSystem in Evanston, Ill.

Drinking alcohol "is definitely not one of the leading explanations" for why breast cancer develops, he said. "It's one of many contributing factors."

Cancer researcher Jo Freudenheim noted that the risks linked with alcohol, shown in this study and others, are much lower than those associated with smoking and lung cancer.

The study "doesn't change the picture; it just brings it into a little sharper focus," said Freudenheim, head of social and preventive medicine at the University at Buffalo.