Further Reading


Priorities, Vol 8 No 1, pp. 24-28, 1996. (Published by American Council on Science and Health, New York)

Should physicians recommend alcohol to their patients?

Stanton Peele1


Whenever I have visited a physician over the last decade, the following scenario has been replayed: we discuss my cholesterol levels (total, LDL, and HDL). We review diet guidelines and other medical recommendations. Then I say, "Don't forget to remind me to drink a glass or two of wine daily." Invariably, the physician demurs: "That hasn't been proven to protect you against atherosclerosis."

My doctors, all of whom I have respected and liked, are wrong. Evidence has established beyond question that alcohol reduces coronary artery disease, America's major killer. This result has been found in the Harvard Physician and Nurse studies and in studies by Kaiser Permanente and the American Cancer Society (ACS). Indeed, the evidence that alcohol reduces coronary artery disease and mortality is better than the evidence for the statin drugs, the most potent cholesterol-reducing medications.

Drinking to excess does increase mortality from several sources, such as cancer, cirrhosis, and accidents. But a series of studies in the 1990s -- including those conducted in conjunction with Kaiser, ACS, and Harvard -- in the U.S., Britain, and Denmark, have found that moderate drinking reduces overall mortality.

Nonetheless, many object to the idea that doctors should inform their patients that moderate drinking may prolong life. They fear that such advice will justify the excessive drinking some patients already engage in, or they worry that encouragement from doctors will push people who cannot handle alcohol to drink.

The view that people are so stupid or malleable that they will become alcohol abusers because doctors tell them that moderate drinking is good for them is demeaning and self-defeating. If people can't regulate their own diets, drinking, and exercise, then doctors should avoid giving patients any information about their health behavior, no matter how potentially helpful.

Not only can people handle such information on lifestyle, it offers the primary and best way to attack heart disease. Of course, doctors may also prescribe medications. These medications rarely solve underlying problems, however; and they often cause adverse side effects that counterbalance their positive effects. Because they are not a cure, courses of medication, once begun, are rarely discontinued.

People are the best regulators of their own behaviors. Even those who drink excessively often benefit when doctors provide straightforward, accurate information. Clinical trials conducted by the World Health Organization around the world showed that so-called brief interventions, in which medical personnel advised heavy drinkers to reduce their drinking, are the most successful therapy for problem drinking.2,3,4

But far more Americans drink less, not more, than would be most healthful for them. To fail to inform these patients about the benefits of moderate drinking is both counterproductive and dishonest. Physicians may ask, "How much alcohol do you drink," "Is there any reason that you don't drink (or drink so little)," and (to those without religious objections, previous drinking problems, etc.), "Do you know that one or two glasses of wine or beer a day can be good for your health, if you can safely consume them?"

Here are the data about alcohol and mortality:

  1. In 1995 Charles Fuchs and his colleagues at Harvard found that women who drank up to two drinks a day lived longer than abstainers. Subjects were 85,700 nurses.5
  2. In 1995, Morten Gronbæk and colleagues found that wine drinkers survived longer than abstainers, with those drinking three to five glasses daily having the lowest death rate. Subjects were 20,000 Danes.6
  3. In 1994, Richard Doll and his colleagues found that men who drank up to two drinks daily lived significantly longer than abstainers. Subjects were 12,300 British doctors.7
  4. In 1992, Il Suh and colleagues found a 40 percent reduction in coronary mortality among men drinking three and more drinks daily. The 11,700 male subjects were in the upper 10 to 15 percent of risk for coronary heart disease. Alcohol's enhancement of high density lipoproteins was identified as the protective factor.8
  5. In 1990, Paolo Boffetta and Lawrence Garfinkel found that men who drank occasionally -- up to two drinks daily -- outlived abstainers. Subjects were over a quarter of a million volunteers enrolled by the American Cancer Society.9
  6. In 1990, Arthur Klatsky and his colleagues found that those who drank one or two drinks daily had the lowest overall mortality rate. Subjects were 85,000 Kaiser Permanente patients of both genders and all races.10

These data -- from large prospective studies of people of both sexes, different occupations, several nations, and varying risk profiles -- all point to alcohol's life-sustaining effects. This phenomenon is now so well accepted that the U.S. Dietary Guidelines released in January 1996 recognize that moderate drinking can be beneficial.

The levels of drinking at which alcohol lowers death rates are still open to dispute. The new U.S. guidelines indicate that men should not drink more than two drinks per day and women should not exceed one per day. But the British government has set its limits for " sensible drinking" at three to four drinks for men and two to three drinks for women [these drinks are somewhat smaller, however]. That abstemiousness increases the risk of death, however, can no longer be doubted. Moreover, alcohol operates at least as effectively as pharmaceuticals to reduce the risk of death for those at high risk for coronary disease.8, 10,11

At one point, researchers questioned whether people who had quit drinking due to previous health problems inflated the mortality rate among abstainers. But this position can no longer be maintained. The studies described above separate drinkers who have quit drinking and who had pre-existing health problems from other nondrinkers. The benefits of drinking persisted with these individuals omitted.

At some point, ranging from three to six drinks daily, the negative effects of drinking for cancer, cirrhosis, and accidents catch up to and surpass alcohol's beneficial cardiac impact. Moreover, women under 50 -- who have relatively low rates of heart disease and relatively high rates of breast cancer mortality -- may not benefit from drinking.5,6, 10 That is, unless they have one or more cardiac risk factors.

Even younger women with such risk factors benefit from light to moderate drinking. And, we must remember, most American women and men have such risk factors. (Fuchs et al. found about three-quarters of the nurses in the Harvard study had at least one.) Remember, over all ages, American women are ten times as likely to die of heart disease (40%) as of breast cancer (4%).

Why, then, do Americans -- physicians, public health workers, educators, and political leaders -- refuse to recognize alcohol's benefits? We might also ask why the United States banned the manufacture, sale, and transportation of alcoholic beverages from 1920 to 1933. It is probably too obvious to mention that alcohol has never been banned -- or prohibition even seriously discussed -- in France, Italy, Spain and a number of other European nations.

What is it about America and some other nations that prevents them from considering that alcohol may be good for people? These so called "temperance" nations, which see alcohol in a highly negative light.12 This is true even though nations with higher alcohol consumption have lower death rates from coronary heart disease (see Table 1). Oddly, temperance nations-- despite concentrating on alcohol problem prevention and treatment -- actually have more drinking problems than those in which alcohol is socially accepted and integrated.

This occurs even though temperance nations drink less alcohol. But they drink a higher percentage of their alcohol in the form of spirits. This drinking is more likely to take place in concentrated bursts among men at sporting events or in drinking establishments. This style of drinking contrasts with that in wine-drinking nations, which encourage socialized drinking among family members of both genders and all ages at meals and other social gatherings. These cultures do not teach people that alcohol is an addictive drug. Rather, moderate drinking is modeled for children and taught to them in the home. Furthermore, these cultures accept that drinking may be good for you. We should too.


Table 1. Temperance, alcohol consumption, and cardiac mortality
Alcohol Consumption (1990) Temperance Nations a Non-Temperance Nations b
total consumptoin c 6.6 10.8
percent wine 17.7 43.7
percent beer 53.1 40.4
percent spirits 29.2 15.9
AA groups/million population 170 25
coronary mortality d (males 50-64) 421 272
a Norway, Sweden, U.S., U.K., Ireland, Australia, New Zealand, Canada, Finland, Iceland
b Italy, France, Spain, Portugal, Switzerland, Germany, Denmark, Austria, Belgium, Luxembourg, Netherlands
c Liters consumed per capita per year
d Deaths per 100,000 population

Source: Peele S. Utilizing culture and behavior in epidemiologic models of alcohol consumption and consequences for western nations. Alcohol & Alcoholism, in press.


  1. Stanton Peele received a fee of $300 for this article from the American Council on Science and Health. He received no other financial support.
  2. Babor TF, Grant M, eds. (1992)Project on identification and management of alcohol-related problems. Geneva: World Health Organization.
  3. Miller WR, Brown JM, Simpson L, et al. (1995) What works: a methodological analysis of the alcohol treatment literature. In Hester RK, Miller WR, eds. Handbook of alcoholism treatment approaches. Boston: Allyn and Bacon, 12-44.
  4. Heather N. (1995) Brief intervention strategies. In Hester RK, Miller WR, eds. Handbook of alcoholism treatment approaches. Boston: Allyn and Bacon, 105-122.
  5. Fuchs CS, Stampfer MJ, Colditz GA, et al. (1995) Alcohol consumption and mortality among women. N Engl J Med. 332: 1245-50.
  6. Gronbæk M, Deis A, Sorensen TIA, et al. (1995) Mortality associated with moderate intake of wine, beer, or spirits. BMJ. 310:1165-9.
  7. Doll R, Peto R, Hall E, et al. (1994) Mortality in relation to consumption of alcohol: 13 years' observations on male British doctors. BMJ. 309:911-8.
  8. Suh I, Shaten BJ, Cutler JA, Kuller LH. (1992) Alcohol use and mortality from coronary heart disease: the role of high density lipoprotein. Ann Intern Med. 116:881-87.
  9. Boffetta P, Garfinkel L. (1990) Alcohol drinking and mortality among men enrolled in an American Cancer Society prospective study. Epidemiology. 1:342-8.
  10. Klatsky AL, Armstrong MA, Friedman GD. (1990) Risk of cardiovascular mortality in alcohol drinkers, ex-drinkers and nondrinkers. Am J Cardiol. 66:1237-42.
  11. Levine GN, Keaney JF, Vita JA. (1995) Cholesterol reduction in cardiovascular disease. N Engl J Med. 332:512-21.
  12. Peele S. (1993) The conflict between public health goals and the Temperance mentality. Am J Public Health. 83:805-10.