Ask the Doctors

Dear Doctor: News coverage of a study recently suggested that frequent, moderate drinking could ward off Type 2 diabetes. Could this be true?

Dear Reader: You're right to distrust alcohol. Excessive use can increase the risk of multiple cancers as well as cirrhosis of the liver, nerve damage both peripherally and within the brain, weakened heart muscle and abnormal heart rhythms. It also can lead to traffic deaths and an addiction that destroys lives. However, small to moderate alcohol consumption has been shown to decrease the rate of heart attacks -- and also may help control blood sugar.

A recent Danish study analyzed 2007 and 2008 survey data on alcohol consumption from 28,704 men and 41,847 women. Survey participants were separated into different groups: no history of alcohol consumption; prior consumption but none in the last year; less than one day of alcohol consumption per week; one to two days of alcohol consumption per week; three to four days of alcohol consumption per week; and five to seven days of alcohol consumption per week.

After almost five years, the data showed that lifetime abstainers had a higher rate of Type 2 diabetes than other groups. The lowest rate was seen among those who drank alcohol three to four days per week. In fact, compared to people who consumed alcohol less than once per week, those men were 27 percent less likely to develop diabetes, and the women were 32 percent less likely to develop diabetes.

But before you take another sip of wine, let us parse the data a little. In the study, female abstainers were twice as likely to be obese when compared to women who drank alcohol three to four days per week or five to seven days per week. In men, this rate was about 50 percent greater. Similarly, male and female abstainers were twice as likely to be physically inactive when compared to the frequent alcohol users. Moreover, abstainers had a greater family history of diabetes and were more likely to be less educated. (The authors said they adjusted for these factors and, to some degree, they did.)

A 2012 European study found similar results. After controlling for the fact that people who consumed no alcohol were more likely to be inactive, obese and less educated, the authors showed that women who drank 1 unit of alcohol per day were slightly less likely to develop diabetes compared to women who drank 1 unit every two to eight days. For men, the connection between alcohol consumption and lowered diabetes risk was considered less significant.

Lastly, an analysis of 20 studies in 2009 showed that, in men, the protective benefit of alcohol was greatest at 22 grams of alcohol (about 1 1/2 glasses of wine) per day. In women, 24 grams of alcohol per day was most protective. But the authors didn't adjust their findings based on weight, physical activity or educational level.

Note that the European studies analyzed populations in which consumers of alcohol had generally higher rates of physical activity, which reduces the risk of diabetes; that may not hold true in the United States. But in summary, alcohol consumption -- though it has a myriad of problems -- does seem to be linked to a slight decrease in the risk of diabetes.

(Send your questions to, or write: Ask the Doctors, c/o Media Relations, UCLA Health, 924 Westwood Blvd., Suite 350, Los Angeles, CA, 90095. Owing to the volume of mail, personal replies cannot be provided.)

More like Ask the Doctors