Does a Drink a Day Really Keep the Doctor Away?

Abstract & Commentary

Synopsis: In the right person, a drink a day may actually help keep the doctor away but additional studies, especially in women, are required before definitive recommendations can be made.

Source: Muntwyler J. Lancet 1998;352:1882-1885.

Patients with a history of myocardial infarction (MI) are at high risk of mortality from reinfarction and sudden death;1 however, numerous advances in drug therapy as well as risk factor modification with lifestyle changes have been extremely effective in reducing these mortality rates. In addition, in high-risk coronary artery disease populations such as middle-aged or older men with multiple cardiovascular risk factors, light to moderate alcoholic intake has previously been demonstrated to confer significant benefit.2,3

Muntwyler and associates from the Division of Preventive Medicine and Cardiovascular Disease at the Harvard Medical School published an excellent paper in Lancet. Of the 90,150 men in the Physicians’ Health Study who provided information on alcohol intake and had no prior history of cancer, stroke, or liver disease, 5358 had suffered a previous MI. After adjustment for serious potential confounders, moderate alcohol intake was associated with a significant decrease in total mortality in the 920 deaths among these men. Those physicians who drank 1-4 drinks per month had a relative risk of total mortality of 0.85, whereas those who drank 2-4 drinks per week had a mortality risk of 0.72 and those whose intake was as high as one drink per day had a risk of 0.79. These values clearly demonstrate that those patients who had previously suffered an MI and who consume small to moderate amounts of alcohol have a distinctly lower total mortality rate compared with men who rarely or never drank alcohol.

Comment by Harold L. Karpman, MD, FACP

In several primary prevention studies, light to moderate alcohol intake has been demonstrated to be associated with a decrease in the risk of mortality mainly in older subgroups—suggesting that this positive effect may have been due to the effect of alcohol in that group of patients who are at the highest risk of death from cardiovascular disease. However, since all patients in secondary prevention studies are at a relatively higher risk for cardiovascular death than exists in patients who have never suffered a cardiovascular event, it was not surprising to find that the improved mortality rates occurred regardless of age in those patients who consumed light to moderate amounts of alcohol.

Muntwyler et al have demonstrated an unexpected risk reduction for the 23% of deaths from noncardiovascular causes in a high-risk, secondary-prevention cohort of 5358 men who had previously suffered an MI and who consumed light to moderate amounts of alcohol.4 Although there was more than a three-fold excess of cardiovascular over noncardiovascular deaths, a light to moderate intake of alcohol resulted in a slight but clinically important decrease in total mortality compared to those men who never or who rarely drank alcohol. The shape of the association curve for alcohol intake and cardiovascular mortality suggested that the maximum potential benefit of alcohol intake was reached by drinking 2-6 drinks per week and that both the death rate and the shape of the association curve was similar in men with and without previous MIs.

It has been previously suggested that the beneficial effects of light to moderate alcoholic intake upon reducing coronary risk may be because of the HDL level increases in this population of men and/or because of a lower risk of thrombosis. If additional studies continue to demonstrate beneficial effects of alcohol in secondary prevention, the absolute benefit would almost certainly be greater in secondary than in primary prevention populations and, as has been demonstrated in trials of antiplatelet agents and cholesterol-lowering drugs, these differences could far outweigh any hazard of drug or alcohol intervention. Because of the intoxicating and habituating properties of alcohol, its intake cannot be clearly prescribed for patients as a general measure for the secondary prevention of coronary artery disease unless one has a thorough knowledge of the patient’s past history and a realistic assessment of each patient’s potential for the development of alcohol-related problems. Specific recommendations should be tailored to each individual patient, taking into account the potential risk of alcohol intake in that patient relative to the likely cardiovascular benefits that one would expect to achieve as a result of a small to moderate intake of alcohol on a regular basis.

The bottom line: In the right person, a drink a day may actually help keep the doctor away but additional studies, especially in women, are required before definitive recommendations can be made.


    1. McGovern PG, et al. N Engl J Med 1996;334:884-890.

    2. Thun JT, et al. N Engl J Med 1997;337:1705-1714.

    3. Shaper AG. Lancet 1988;ii:1267-1273.