One of the most contentious issues in the vast literature about alcohol consumption has been the consistent finding that those who don't drink tend to die sooner than those who do. The standard Alcoholics Anonymous explanation for this finding is that many of those who show up as abstainers in such research are actually former hard-core drunks who had already incurred health problems associated with drinking.

But a new paper in the journal Alcoholism: Clinical and Experimental Research suggests that — for reasons that aren't entirely clear — abstaining from alcohol does tend to increase one's risk of dying, even when you exclude former problem drinkers. The most shocking part? Abstainers' mortality rates are higher than those of heavy drinkers. (See pictures of booze under a microscope.)

Moderate drinking, which is defined as one to three drinks per day, is associated with the lowest mortality rates in alcohol studies. Moderate alcohol use (especially when the beverage of choice is red wine) is thought to improve heart health, circulation and sociability, which can be important because people who are isolated don't have as many family members and friends who can notice and help treat health problems.

Read more: http://www.time.com/time/health/article/0,8599,2014332,00.html#ixzz0yF4l8EZl