Beneficial Effects of Consumption of Low Levels of Alcohol

Comments by Michael Gough, Ph.D., On behalf of The Competitive Enterprise Institute, filed with the Bureau of Alcohol, Tobacco and Firearms



Beverage alcohol has both beneficial and adverse effects, and both effects are so well accepted that a recent analysis of the effects of alcohol consumption in a half million U.S. residents stated:

We expected death rates from certain conditions to increase with alcohol consumption: cirrhosis of the liver; alcohol-related cancers (those of the oral cavity, pharynx, esophagus, liver, and larynx, but excluding the salivary glands and nasopharynx; accidents and other external causes; and possibly breast cancer in women, cancer of the colon or rectum, pneumonia, and hemorrhagic stroke. Decreased mortality was expected from coronary heart disease, total stroke, and possibly "other circulatory conditions" combined. No association was expected with all "other cancers" or "other causes of death."1 (References and International Classification of Disease, ICD, numbers omitted).

The beneficial effects of alcohol consumption are seen in people who consume about 1 to 2 alcoholic drinks daily, and the alcohol-related diseases are seen in people who drink more.2 Clearly, there is a balance between the two effects, but the beneficial effects of moderate alcohol consumption are so well recognized that the 1992 article "The Primary Prevention of Myocardial Infarction" included it as one of nine possible ways to reduce the death toll from coronary heart disease (CHD), which accounts for about one of every three deaths. "The estimated reduction in the risk of myocardial infarction associated with mild-to-moderate consumption of alcohol is 25 to 45 percent."3

The combination of epidemiological studies with current understanding of the biological basis of the beneficial effects of moderate alcohol consumption presents a convincing picture. The prestige of the researchers — some of whom are among the most respected epidemiologists and statisticians in the world — who have demonstrated the beneficial effects of low-level consumption of alcohol and the publication of their results in highly regarded journals underlines the credibility of the reported beneficial effects of moderate alcohol consumption.

The scientific evidence for the beneficial effects of moderate alcohol consumption is stronger than indicated in the current federal dietary guidelines for the consumption of alcohol, and stronger statements are justified.



A one-page resume for Michael Gough is attached. For the purposes of this paper, his experiences since 1978 at the congressional Office of Technology Assessment, at Resources for the Future, the Cato Institute, and other private sector organizations where he has analyzed health risk research are important. In addition, he chaired the U.S. Department of Health and Human Services Advisory Committee to the United States Air Force Study of the health of the men who sprayed 90 percent of the Agent Orange used in Vietnam. That study (still on-going) is one of the largest, and probably the most, expensive epidemiological studies ever done, and Dr. Gough is thoroughly familiar with the design, execution, and analysis of large scale epidemiology studies. Such studies are the foundation for understanding the effects of alcohol on humans.



Epidemiologic studies, taken altogether, show that moderate consumption of alcohol reduces death rates below those seen in nondrinkers. Higher levels of consumption cause an increase in "alcohol augmented diseases." In some populations the effects of those diseases are sufficient to drive the death rates among heavier drinkers above the rates seen in nondrinkers; in some populations they have smaller effects.

The beneficial effects are seen in both men and women and in populations of people 35 and older. No benefits are seen in people in their 20s and 30s because mortality from the diseases that alcohol protects against is so low at those ages that any beneficial effect cannot be seen. Based on understanding of the biological basis for the protective effects of alcohol, it is likely that moderate alcohol consumption in the 20s and 30s is important to the beneficial effects seen in later years.

Dose-Response to Alcohol

Figure 1 from Gaziano,4 summarizes the results of alcohol consumption on human health. In it, total mortality (better described as "mortality from all causes") is described by a "J-shaped curve" (panel a). Age-adjusted mortality decreases at low and moderate drinking levels, and then rises above the age-adjusted mortality rate of nondrinkers at higher drinking levels. As will be seen below, mortality from all causes in some populations follows a J-shaped curve, and, in others, it follows a "U-shaped" or "L-shaped curve."

Source: Gaziano. 1995 at p. 2.

Decreases in cardiovascular disease (CVD) mortality account for the decrease in mortality rates seen at low to moderate drinking levels. The "U-shaped curve" (panel c) shows that alcohol consumption over a relatively wide range is associated with lower CVD mortality, but the CVD mortality rate increases at higher exposures and eventually reaches the rate in nondrinkers.

Coronary heart disease (CHD) is a subset of CVD, and the decrease in CHD, which is described by an "L-shaped curve" (panel d), largely explains the reduction in CVD. CHD mortality decreases at low levels of alcohol consumption, and it does not increase at higher levels.

The increase in total mortality seen at higher alcohol consumption levels is related to increases in alcohol-related cancers (panel b), accidents and other external causes, suicides, cirrhosis and some other diseases. For example, Doll et al. showed that, in men between 50 and 90, deaths from all causes were lowest in men who consumed about 2 drinks a day. The reduced death rates were accounted for by decreases in CHD deaths (and from deaths from other causes as well, see below). Total mortality, in part from what Doll calls "alcohol augmented diseases" increased at higher consumption levels, but it did not reach the level seen in nondrinkers until alcohol consumption was about 6 drinks a day.5

General Acceptance of the Beneficial Effects of Moderate Consumption of Alcohol

The studies of the effects of alcohol consumption and health vary one from another in details, but they are sufficiently consistent for recommendation-making organizations in the United States6 and in the United Kingdom7 to conclude that moderate consumption of alcohol protects against heart disease and reduces total mortality. "Moderate" is not always carefully or consistently defined, but the largest U.S. study8 found that "those who consumed up to one or two drinks of alcohol daily had lower overall mortality rates than nondrinkers," and essentially all the studies find beneficial effects in that consumption range.

Because details of the observations about the effects of alcohol differ between men and women and between groups at high and low risk of heart disease, some details from Thun et al. (1997) are presented below. That study9 contained controls for the effects of smoking, did not include people who had quite drinking10 because of illness in the nondrinking (comparison) group, and collected information about the stability of drinking habits through questionnaires distributed to study participants. Like all studies, it had some limitations. Alcohol consumption was based on self-reports (but this is true for essentially all alcohol studies), there were few very heavy drinkers in the study, and there was no information about binge drinking (this too is true for most alcohol studies). All in all, however, it is the largest and one of the best, if not the best, conducted and analyzed studies.

Men. The left side of figure 2 shows the lower probability of death for men who consume one to two alcoholic drinks daily. Nondrinking nonsmokers have a 26 percent probability of dying at between 35 and 69 years of age; in contrast, drinking nonsmokers have a lower, 22 percent, probability of dying. As expected, the probability of death is far higher in smokers — nearly double that of nonsmokers — but the beneficial effects of moderate alcohol consumption are still evident. The probability of death among nondrinking smokers is 46 perent and among drinking smokers, it is 43 percent.

Source: Thun et al. 1997 at p. 1712.

Moderate alcohol consumption reduced deaths from all CVD by 30 to 40 percent in men, and that reduction was consistent at all levels of alcohol consumption. The total mortality reduction was lessened by increased mortality from cirrhosis and alcoholism and cancers of the mouth, throat areas, and liver ("alcohol augmented causes"11), all of which increased three- to seven-fold in men who reported consuming more than four drinks daily.

In terms of all cause mortality, the reduction in CHD mortality is more important than the increase in alcohol augmented deaths because of the different mortality rates from the two sets of diseases. CHD is responsible for about 45 percent of all male deaths (in Thun et al. 1997), and the alcohol-augmented deaths are responsible for about 6 percent of all deaths. The consumption of 1 to 2 alcoholic drinks per day reduced CHD mortality about 30 percent. There was no increase in alcohol augmented causes of death at levels of consumption less than 3 drinks per day, and at no consumption level was the increase in alcohol augmented causes sufficient to drive the overall mortality higher than that seen in nondrinkers.12

Women. Figure 2 shows that moderate drinking reduces mortality among women smokers and nonsmokers, but to a lesser extent than seen in men. Expected mortality among drinking nonsmoking women between 35 and 69 years of age is 14 percent as compared to the 17 percent expected in nondrinking nonsmoking women. Among smoking women, the expected mortality of drinkers was 28 percent as compared to 30 in the nondrinkers.

The alcohol augmented diseases that were seen to increase in men who drank increased to about the same level in women who consumed more than 4 drinks daily. In addition, breast cancer also increased in women who reported more than one drink daily. In terms of overall mortality, the increase in all alcohol augmented diseases, including breast cancer, was more than offset by the 40 percent reduction in CVD mortality. Another difference between men and women is that the alcohol-related increase in mortality from external causes — unintentional injuries and suicides, primarily — that was seen in men who drank more than 4 drinks daily was not seen in women.

Men and Women. As shown in table 2, the beneficial effects of moderate alcohol consumption are seen in both sexes. The overall reduction in mortality for both men and women was about 20 percent at 1 to 2 drinks per day.13 Thun et al. (1997) characterized their study as showing that "moderate alcohol consumption slightly reduced overall mortality."4 Their choice of the word "slightly" is not explained, but whatever word is used, the mortality reduction was 20 percent, significantly lower than the mortality seen in nondrinkers.

Doll et al. (1994) and Thun et al. (1997) as well as Klatsky et al. (1992)15 noted that moderate alcohol consumption was associated with a reduction in mortality from CVD and from all causes not known to be associated with alcohol as well. The finding of beneficial effects on non-CVD mortality is unexplained. Thun et al. (1997) say,

An unexpected finding was the lower mortality among drinkers than among nondrinkers from the aggregate of the causes we had originally postulated would be unrelated to drinking…. This reduction, which was more evident among men than among women, needs further investigation.16

High and Low Risk Groups

Thun et al. (1997) divided their study population into four groups on the basis of their ages and risk of CVD. The two age groups were younger people between the ages of 30 and 59 and older people between 60 and 79. Both age-groups were divided into two groups based on whether they were at low or high risk of CVD, based on information collected on questionnaires.

Total mortality among the younger age group at low risk of CVD displayed a J-shaped curve (see figure 1). In this group, consumption of 1 to 2 drinks a day reduced total mortality but consumption of 3 or more drinks drove total mortality rates above those seen in nondrinkers. The explanation for this response is that the risk of CVD in this group is sufficiently small that reducing it does not offset the increase in alcohol augmented mortality that occurs at greater than 3 drinks per day.

In the younger age group at high risk of CVD and in the older age group at low risk of CVD, there was a U-shaped dose response curve for total mortality (see figure 1). In both groups, total mortality dropped with one drink, and although mortality fluctuated at higher consumption levels and moved higher at 3 drinks per day, it never reached the level seen in nondrinkers. In both these groups the risk of CVD mortality is so high that reducing it more than offsets the risks from mortality from all other causes.

The beneficial effect of alcohol is even higher in the older, at high risk of CVD group, where the mortality dose response curve is L-shaped (see figure 1). Mortality falls at one drink per day, and it remains at the lower level or drifts downward at higher consumption levels. In this group, CVD is such a common cause of death that reducing its toll more than offsets any alcohol augmented deaths.

The Effects of Alcohol in Young Adults

Although Thun et al. (1997) showed that moderate consumption of alcohol had beneficial effects in people over 35, they had no data to address its effects on younger people. They refer to other studies that show that injuries and other external causes are the primary causes of mortality in young men less than 29 years of age and to a study that showed a direct relationship between deaths from all causes — primarily accidents, violence, and suicide — in Swedish soldiers. There were, however, no reports of adverse health effects — as opposed to adverse effects from external causes — among young people who consumed moderate amounts of alcohol.

Neither are there reports of beneficial effects of alcohol in people below 35. Moreover, there are not likely to be such reports because mortality from CVD is so low in people of those ages that a truly huge study population would be necessary to observe reductions of even 30 or 40 percent, if it were possible at all.

The absence of a demonstrated effect at young ages does not necessarily mean that alcohol consumption at those ages is not beneficial. By analogy, the National Cancer Institute urges everyone, including children, to eat five fresh fruits and vegetables daily. There is no clear impact of eating fruits and vegetables on cancer rates in children. There is however, a clear impact of life-long consumption of fruits and vegetables on cancer risks in middle and old age. Similarly, several of the Institutes at the National Institutes of Health urge that everyone eat healthy diets with reduced fats and sugar and increased grains and fruits and vegetables. Here again, no payoff is expected in young people, very few of whom die from diet-related diseases. The payoff comes in middle and old age.

It may be that low to moderate consumption of alcohol during the 20s and 30s, which continues on into older ages, is critical to the beneficial effects seen in CVD at older ages. Thun et al. (1997) touch on this issue, when they say,

it is not known how long moderate alcohol consumption must continue for this benefit to occur. Alcohol consumption beginning in middle age might suffice, while averting much of the risk of accidents and cancer associated with drinking.17

In the absence of more information, it can as easily be said that alcohol consumption at the younger ages, continuing into middle and old age, is necessary for the beneficial effects. The information about the biological basis for the beneficial effects suggests that this might be the case.


The Biological Basis for the Beneficial Effects

Two different mechanisms account for the beneficial effects of moderate alcohol consumption. Drinkers have higher concentrations of a blood clot dissolving enzyme (tissue plasminogen activator), which reduces heart attack risk. There is an immediate effect of drinking alcohol on lowering the heart attack rates, which is probably related to the effect on alcohol on reducing blood clot formation.18

Far better studied, and of more importance to the question of whether moderate consumption of alcohol at young ages is beneficial, is the relationship between moderate alcohol consumption and high density lipoprotein (HDL) cholesterol. HDL cholesterol is the "good" cholesterol, which binds to cholesterol and brings it back to the liver for elimination or reuse and reduces cholesterol concentrations in other tissues and organs including blood vessels.

Moderate alcohol consumption increases the levels of both HDL2 and HDL3, both of which reduce the risk of heart attacks.19 There is every reason to believe that high levels of HDL2 and HDL3 reduce concentrations of cholesterol at all ages, and moderate alcohol consumption in the 20s and 30s, which increases HDL cholesterol levels, may be important in reducing the effects of high cholesterol and staving off CVD later in life.


The Studies, Authors, and Places of Publication

Many of the authors of the studies of alcohol and health are the leaders in their fields of epidemiology, medicine, and other disciplines.

Richard Doll, the first author of the "British doctors study"20 and one of the authors of the study of alcohol and health in a half-million U.S. residents,21 is probably the foremost epidemiologist in the world.

His colleague, Richard Peto, who contributed to the same papers, is a world-renowned statistician and heads the Clinical Trials Unit at Oxford University.

Charles Hennekens, an author on the "prevention paper,"22 the major study about women and alcohol,23 and a study about the biochemical mechanisms by which alcohol protects against heart attacks,24 and numerous other papers is the John Snow Professor of Medicine at Harvard Medical School.

Eric B. Rimm, an author of papers about the protective effects of alcohol regardless of whether it is consumed in wine, beer, or spirits,25 a study of alcohol and coronary diseases in male health professionals,26 and a review article about alcohol and health,27 and his colleague Meir J. Stampfer, who is a contributor to many of the papers referenced here, are on the faculty of the Harvard School of Public Health.

Other authors are from well-known medical schools and universities.

The studies have been published in journals with rigorous publication standards. Those journals in the United States include The New England Journal of Medicine, JAMA, and in England, BMJ, and The Lancet.

The reputations of the authors and the journals provide assurance that the results reported in the studies about the beneficial effects of alcohol have been subjected to careful analysis and reviews. In addition, the consistency of results observed in studies in different countries, in different populations, and in different age groups provides additional confidence about the observed beneficial effects.


The Federal Dietary Guideline for Alcohol

The current federal guideline for alcohol consumption states, "Current evidence suggests that moderate drinking is associated with a lower risk for coronary heart disease in some individuals."28 The guidelines then provide warnings about certain people who should not drink. From the studies that have been done, there seems to be little reason to qualify the statement by use of the word "some."29

Every group of people studied by Thun et al. (1997) from ages 35 on, whether or not at high risk from CVD, benefited from moderate alcohol consumption. There is no evidence for immediate beneficial effects on younger adults who consume moderate amounts of alcohol because the risks from CVD are so low in those people that no beneficial effect could be observed. Based on knowledge of the biological basis for the beneficial effects of moderate alcohol consumption, it is, however, more likely than not that such consumption in younger people contributes to the beneficial effects seen at older ages.

The guidelines understate the beneficial effects. As shown in a number of studies, moderate alcohol consumption is associated with reduction in all cause mortality in people older than 35. The beneficial effects are not limited to "a lower risk for coronary heart disease."30

The Competitive Enterprise Institute, in an earlier petition to the Bureau of Alcohol, Tobacco, and Firearms proposed the following statement as a label to inform consumers of the benefits of moderate alcohol consumption: "There is significant evidence that moderate consumption of alcoholic beverages may reduce the risk of heart disease." The CEI label more accurately reflects the information about moderate consumption of alcohol.


ATF's Current Position

ATF31 denied CEI's petition32 to grant a rule that would allow labels that inform consumers of the beneficial effects of moderate consumption of alcohol to be attached to wine bottles. In its denial and in its most recent court filing,33 ATF raised several objections to the claim that moderate alcohol consumption is medically beneficial.

ATF's denial itemized several diseases that are known to be associated with high consumption of alcohol. There is no argument that heavy consumption of alcoholic beverages can cause serious health consequences. CEI accepts that those diseases are more likely to occur in heavy drinkers, but CEI does not interpret the available evidence to indicate that moderate consumption of alcohol increases the risks of those diseases.

In its denial, ATF states,

…at this time, there is not significant scientific evidence to support an unqualified conclusion that moderate alcohol consumption has net health benefits for all or even most individuals.34

This statement apparently ignores the decreases in mortality that are consistently seen in study populations that are described in the studies referenced in this paper. Certainly, there are some people who should avoid alcohol: Children and adolescents, pregnant women and women trying to conceive, recovering alcoholics, drivers and operators of machinery, and people taking medications. With the exception of those groups, who comprise a minority of the population, there does not appear to be a group of adults that does not benefit from moderate alcohol consumption. Moreover, women who are neither pregnant or trying to become pregnant and drivers and operators of machinery and people taking medicine would be expected to benefit from moderate alcohol consumption when their situations are different.

ATF's denial of CEI's petition refers to a paper by Mary C. Dufour, Deputy Director, National Institute on Alcohol Abuse and Alcoholism.35 She identifies several groups of people who should avoid alcohol. Her advice parallels that given by authors of most of the papers referenced here. Indeed, the warnings to pregnant women and drivers and operators already appear on alcoholic beverage containers, and warnings about adverse effects from combinations of alcohol and medications appear on many medicines. All of the groups except recovering alcoholics that should avoid alcohol are informed about their risks. It seems reasonable that a person who considers himself a recovering alcoholic knows that he cannot drink.

Dufour refers to Fuchs et al. (1995) to argue that consumption of less than a drink a day by young women leads to an increase in all cause mortality. However, none of the increases in all cause mortality associated with alcohol consumption in young women 34 to 39 and 40 to 49 years of age was statistically significant.36 Moreover, the larger study by Thun et al. (1997) found a decrease in overall mortality in both men and women over 35 years of age.

Middle-aged and older people who enjoy drinking in moderation, Dufour says, "are likely to experience net health benefits."37 This comment, although narrower in scope, parallels the arguments made by CEI.

ATF's brief quotes from a paper by Michael H. Criqui in which he says, "In fact, in a study of 128,934 participants in a pre-paid health plan, the overall reduction in mortality was limited to those 60 years of age or older."38

The study of 128,934 people referred to by Dr. Criqui was written by Arthur L. Klatsky et al.39 No statement about overall mortality in people of different age groups was found in that paper, and it is unclear from where Dr. Criqui derived his statement. In the abstract of their study, Klatsky et al. state:

Lighter drinkers were at lower risk from cardiovascular diseases, especially coronary heart disease (relative risk at 1 to 2 drinks per day = 0.7; CI, 0.6 to 0.9), independent of baseline risk, with the greatest reduction of risk in older persons. Lighter drinkers over 60 years of age also had a slightly lower risk for noncardiovascular death, but this finding was not independent of baseline coronary heart disease risk.40

Klatsky et al.'s (1992) paper shows that moderate consumption of alcohol reduces CVD and CHD mortality. The relative risk from all causes for people of all ages in the study who consumed up to 2 drinks per day was 0.9, or a 10 percent reduction from the risk seen in nondrinkers. The risk of all cause mortality at 1 to 2 drinks per day was lower than in nondrinkers at all ages between 40 and 70, but not at younger ages.



The available evidence about the beneficial effects of moderate alcohol consumption justifies a stronger statement than now appears in the federal dietary guidelines. Moderate alcohol consumption appears to benefit all people, except those groups who should avoid alcohol, at all ages above 35. No beneficial effects would be expected at ages 21 to 35, but it appears likely that moderate drinking at those ages may contribute to lower mortality later in life.

The available evidence contradicts AFT's statement that "there is no significant scientific evidence to support an unqualified conclusion that moderate alcohol consumption has net health benefits for all or even most individuals."41 Indeed, with the exception of those well-defined groups of people who should avoid alcohol, there is clearly convincing evidence for the health benefits of moderate alcohol consumption.


1Michael J. Thun, Richard Peto, Alan D. Lopez, Jane H. Monaco, S. Jane Henley, Clark W. Heath, and Richard Doll, "Alcohol Consumption and Mortality Among Middle-Aged and Elderly U.S. Adults," The New England Journal of Medicine 337: 1705-1714 (1997). 2Thun et al. 1997; Richard Peto, Richard Doll, Emma Hall, Keith Wheatley, and Richard Gray, "Mortality in Relation to Consumption of Alcohol: 13 Years' Observations on Male British Doctors," BMJ (British Medical Journal) 309: 911-918, 1994. Royal College of Physicians, Royal College of Psychiatrists, and Royal College of General Practitioners, Alcohol and the Heart in Perspective: Sensible Limits Reaffirmed (London, UK: Chameleon Press, June 1995.) 3JoAnn E. Manson, Heather Tosteson, Paul M. Ridker, Suzanne Satterfield, Patricia Hebert, Gerald T. O'Connor, Julie E. Buring, and Charles H. Hennekens, "The Primary Prevention of Myocardial Infarction," The New England Journal of Medicine 326: 1406-1416, at p. 1412. 4J. Michael Gaziano, "Alcohol and Coronary Heart Disease," BELLE (Biological Effects of Low Level Exposures) Newsletter 4: 1-5 (1995). 5Doll et al. 1994, figure at p. 915. 6U.S. Department of Agriculture, U.S. Department of Health and Human Services, Nutrition and Your Health: Dietary Guidelines for Americans (Washington, DC: U.S. Department of Agriculture, 1995). 7Royal College of Physicians et al. 1995. 8Thun et al. 1997. 9Thun et al. 1997 at p. 1713. 10Inclusion of ex-drinkers who had quit drinking because of medical reasons in the nondrinking group could bias the results because of the possibly higher mortality of those individuals. 11Doll et al. 1994. 12Thun et al. 1997 at p. 1710. 13Thun et al. 1997, figure 1 at p. 1710. 14Thun et al. 1997, at p. 1705. 15Arthur L. Klatsky, Mary Anne Armstrong, and Gary D. Friedman, "Alcohol and Mortality, " Annals of Internal Medicine 117: 646-654 (1992). 16Thun et al. 1997 at p. 1713. 17Thun et al. 1997 at p. 1713. 18Paul M. Ridker, Douglas E. Vaughan, Meir J. Stampfer, Robert J. Glynn, and Charles H. Hennekens, "Associaton of Moderate Alcohol Consumption and Plasma Concentration of Endogenous Tissue-Type Plasminogen Activator," JAMA (Journal of the American Medical Association) 272: 929-933 (1994). 19See Gaziano 1995. 20Doll et al. 1994. 21Thun et al. 1997. 22Manson et al. 1992. 23Charles S. Fuchs, Meir Stampfer, Graham A. Colditz, Edward L. Giovannucci, JoAnn E. Manson, Ichiro Kawachi, David J. Hunter, Susan E. Hankison, Charles H. Hennekens, Bernard Rosner, Frank E. Speizer, and Walter C. Willett, "Alcohol Consumption and Mortality Among Women," The New England Journal of Medicine 332: 1245-1250 (1995). 24Ridker et al. 1994. 25Eric B. Rimm, Arthur Klatsky, Diederick Grobbee, and Meir J. Stampfer, "Review of Moderate Alcohol Consumption and Reduced Risk of Coronary Heart Disease: Is the Effect Due to Beer, Wine, or Spirits?" BMJ (British Medical Journal) 312: 731-736 (1996). 26Eric B. Rimm, Edward L. Giovannucci, Walter C. Willett, Graham A. Colditz, Alberto Ascherio, Bernard Rosner, and Meir J. Stampfer, "Prospective Study of Alcohol Consumption and Risk of Coronary Disease in Men," The Lancet 338: 464-468 (1991). 27Meir J. Stampfer, Eric B. Rimm, and Diana Chapman Walsh, "Commentary: Alcohol, the Heart, and Public Policy," American Journal of Public Health 83: 801-804 (1993). 28U.S. Department of Agriculture, U.S. Department of Health and Human Services. 1995 at p. 40. 29The National Institute on Alcohol Abuse and Alcoholism (NIAAA) (see letter from Stephen W. Long, Director, Office of Policy Analysis, National Institute of Alcohol Abuse and Alcoholism, to William T. Earle, Deputy Associate Director, Bureau of Alcohol, Tobacco, and Firearms, January 10, 1997) apparently agrees that the benefits of moderate alcohol consumption are not restricted to some members of the population.

The Bureau of Alcohol, Tobacco, and Firearms (ATF) had included the following two sentences in a draft response prepared in response to a petition from the Competitive Enterprise Institute (CEI): "The NIAAA has reviewed this letter, and has concurred in ATF's evaluation of the relevant scientific and medical data. Most significantly, the NIAAA agrees that the evidence supports a conclusion that while moderate consumption may reduce the risk of coronary heart disease in certain individuals, the benefits and risks associated with moderate alcohol consumption will vary with each individual consumer."

NIAAA asked that the two sentences, which included the statement that "benefits and risks associated with moderate alcohol consumption will vary with each individual consumer" be deleted. The deletion is consistent with the idea that all consumers of moderate amounts of alcohol will benefit. 30CEI, "Petition To ATF For A Rule Allowing Alcoholic Beverage Labels To Carry Statements On The Health Benefits Of Moderate Alcohol Consumption" (May 9, 1995). 31Letter from Gerard W. LaRusso, Chief, Alcohol and Tobacco Programs Division, ATF to Sam Kazman, General Counsel, CEI, January 13, 1997, at p. 8. 32CEI petition to ATF, May 9, 1995 33ATF, "Memorandum In Support Of Defendants' Motion For Summary Judgment On Counts One And Two Of The Ameded Complaint (Oct. 6, 1998), CEI v. Rubin" (D.D.C. Civ. Action No. 96-2476) 34Letter from Gerard W. LaRusso, January 13, 1997. 35Mary C. Dufour, "Risks and Benefits of Alcohol Use Over the Life Span," Alcohol Health and Research World 20: 145-150 (1996). 36Fuchs et al. (1995) at p. 1248. 37Dufour at p. 150. 38Michael H. Criqui, "Moderate Drinking: Benefits and Risks," Alcohol and the Cardiovascular System (Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism, 1996) pp. 117-123 at p. 118. 39Klatsky et al. 1992. 40Klatsky et al. 1992 at p. 646. 41LaRusso letter, January 13,1997, at p. 8.


Michael Gough, Ph.D. December 7, 1998