Real evidence is needed before we extol alcohol’s benefits

With his Sept. 10 op-ed column, Jeff Jacoby asks, “Why can’t alcohol labels tout benefits?” Here’s why.

The evidence for benefits is not “irrefutable,” as one observer suggests. First, studies suggest that any benefit is for older men with heart attack risks, not women or younger people. Second, study conclusions are about those who already drink; results might be different for the few who would start drinking for health.

Third, people who report “moderate” drinking also report other unrelated healthy characteristics, such as more education, fewer tooth extractions, more physical activity, getting colon cancer screening, and not smoking. Clearly drinking does not cause these. Fourth, people often stop drinking when they get sick; thus abstinence gives the false appearance of being unhealthy. Finally, in one study drinking was associated with health benefits in other family members, a spurious effect.


Real evidence is needed from randomized studies, as we demand for all drugs. We thought estrogen reduced heart disease; randomized trials found otherwise.

Alcohol is a carcinogen, and, even in low amounts, it is one of the only reversible risk factors for breast cancer. We don’t recommend that people take carcinogens for their health in any amount.

Alcohol is a leading cause of preventable death. Any recommendation must consider that.

On Thursday I will participate in a debate on this issue at Boston University. But it shouldn’t be much of a debate.

Dr. Richard Saitz

The writer chairs the department of community health sciences at Boston University School of Public Health.