Long-term coffee consumption is associated with lower stroke risk in women who don't smoke, according to a 24-year follow-up study reported in Circulation: Journal of the American Heart Association.

Researchers also say regular coffee drinking may be associated with a modest reduction in stroke risk in non-smoking women. The research, which used Nurses' Health Study data, found that compared with women who drank less than one cup of coffee a month, the risk of all types of stroke was -- 20 per cent less in women drinking four or more cups/day; 19 per cent less in women drinking two to three cups/day and 12 per cent less in women drinking coffee five to seven times a week.

Previous analyses - including a 2006 report from the Nurses' Health Study - raise the possibility that coffee may help protect against diabetes and does not appear to raise the risk of heart attack, researchers said. However, the few studies on stroke had contradictory findings, said Esther Lopez-Garcia, lead author of the study and assistant professor of preventive medicine at the Universidad Autonoma de Madrid, Spain.

Researchers from Spain and Harvard Medical School in Boston analyzed the impact of coffee consumption on stroke risk over 24 years. The subjects were 83,076 women who began the study in 1980 with no history of stroke, heart disease, diabetes or cancer.

Every two to four years, the women completed food frequency questionnaires about their diet, including their consumption of coffee, tea, decaffeinated coffee and caffeinated soft drinks. Researchers used a woman's average coffee consumption from all available reports (prior to a stroke or death) for the analysis.

Between 1980 and 2004, 2,280 strokes were documented: 1,224 ischemic (caused by blockage of a blood vessel feeding brain tissue); 426 hemorrhagic (caused when a blood vessel feeding brain tissue bursts); and 630 of undetermined type.

To assess the role of coffee consumption, the researchers adjusted for several factors known to influence stroke risk, including age, smoking status, body mass index, physical activity, alcohol intake, menopausal status, use of hormone replacement therapy, use of aspirin and diet. This type of analysis can only account for known factors but cannot consider risk predictors as yet unidentified.

Among other findings, coffee was not associated with either raised or lowered stroke risk in the subgroups of women with high blood pressure, diabetes or high cholesterol.

Researchers said women who drink a lot of coffee also tend to smoke. The difference between smokers and non smokers was noted -- among women who had never smoked or quit the habit, drinking four cups or more of coffee a day was associated with a 43 percent reduction in stroke risk; and among smokers, drinking four cups or more was associated with only a 3 per cent reduction in risk.

"The potential benefits of coffee cannot counterbalance the detrimental effects smoking has on health," Lopez-Garcia said.

Other caffeinated beverages, such as tea and caffeinated soft drinks, as well decaffeinated coffee, were not associated with any change in stroke risk.

"This finding supports the hypothesis that components in coffee other than caffeine may be responsible for the potential beneficial effect of coffee on stroke risk," she said. "Antioxidants in coffee lower inflammation and improve blood vessel function."

While possibly good news for current coffee drinkers, the authors said their findings don't provide enough evidence to recommend that women start drinking coffee for its health benefits.

"I would also add that the beneficial effects of coffee can only be applied to healthy people," Lopez-Garcia said. "Anyone with health problems that can be worsened by coffee (insomnia, anxiety, hypertension or heart problems) should talk to their doctor about their specific risk."

Stroke is the third leading cause of death in the United States after diseases of the heart and cancer.