1997). Moderate alcohol-drinking reduces the risk of death from those cardiovascular causes, on the average, by approximately 20–40% (Doll, 1997; Thun et al., 1997). A reduction in cardiovascular disease mortality will translate into a reduction in total mortality in many populations because cardiovascular disease is by far the leading cause of death in middle and old age. The inverse association between alcohol consumption and cardiovascular disease risk is causal: ethanol has been shown in short-term experimental studies to increase the serum concentration of high-density lipoprotein cholesterol (Rimm et al., 1999), and it also appears to affect platelet function and other components of clotting and fibrinolysis (Hendriks et al., 1994; Meade et al., 1987; Renaud et al., 1992).

Quantifying Net Public Health Benefit

The positive and negative effects of alcohol on mortality raise the question, “Is alcohol consumption good for health?” The answer is conditional. The net benefit of alcohol consumption in a population depends on age distribution of the population, because the ratio of mortality from conditions that are prevented by alcohol to mortality from conditions that are made more common by it varies greatly with age. The net benefit also will vary with the population prevalence of factors that predispose to (or protect from) cardiovascular disease, and they might differ in men and women.

Optimal public health guidelines on alcohol consumption are not the same across or even within populations, because the importance of cardiovascular disease and injuries or trauma varies significantly with age and sex as well as from one society to another. For instance, in Sub-Saharan and Latin American countries, the ratio of deaths from coronary heart disease to deaths from violence is close to 1.0, and sometimes even less than 1.0 among men (Murray and Lopez, 1996). Groups inherently at high risk from the detrimental effects of alcohol (such as adolescents and young adults, binge drinkers, and those with lower socioeconomic status) in which deaths from injuries (including motor vehicle injuries), violence, and other external causes are high, have not been included in epidemiologic studies that analyze the alcohol/mortality relationship. For example, among U.S. men aged 15–29, deaths from injuries and other external causes account for 75% of all deaths, compared with 4% from cardiovascular diseases (Schoenborn and Marano, 1988). In another study involving Swedish military recruits in the same age range, a linear increase in



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement