Original articles
The Level of Alcohol Consumption at Which All-Cause Mortality Is Least

https://doi.org/10.1016/S0895-4356(99)00076-1Get rights and content

Abstract

Moderate consumers of alcohol have lower mortality than either nondrinkers or heavy drinkers. This systematic review aimed to quantify the level of alcohol consumption (termed the nadir) at which the lowest mortality occurs. Twenty cohort studies reported analyses of all-cause mortality for at least three categories of alcohol consumption, giving a total of 60,224 deaths among men and 74,824 deaths among women. The nadir in each study was estimated for men and women separately in units per week, where 1 unit is 9 g of alcohol. The estimated nadirs varied substantially between countries. Combined nadirs were estimated for U.S. men (overall nadir 7.7 units per week, 95% confidence interval [CI] 6.4–9.1), U.K. men (12.9 units per week, 95% CI 10.8–15.1), and U.S. women (2.9 units per week, 95% CI 2.0–4.0). The nadirs were not found to be increased in studies of older persons and apply for ages 50 to 80 years.

Introduction

Moderate consumers of alcohol have a reduced risk of all-cause mortality compared with nondrinkers and heavy drinkers 1, 2, 3, 4 because of a protective effect of alcohol consumption against coronary heart disease (CHD) 1, 5, 6, 7 combined with a harmful effect on other diseases. Knowledge of the level of alcohol consumption associated with the lowest mortality is important in formulating public health messages [8], but no quantitative overview of the evidence has been performed.

Some authors have said that such an overview is not possible in view of the different ways in which alcohol consumption is measured in different studies and differences by gender, age, and country in the true relationship between mortality and alcohol consumption 3, 9. However, if quantitative conclusions are being drawn, then it is important to use the best available methods. This article aims to estimate the nadir—the level of alcohol consumption at which mortality is minimized—for each study separately and then to estimate an overall nadir allowing for differences by gender, age, and country.

Section snippets

Methods

MEDLINE was searched for papers published up until the end of 1995 whose titles contained both (1) “alcohol”, “drinking”, “beer”, “wine”, or “spirits” and (2) “mortality”, “death”, or “coronary heart disease”. This yielded reports from 29 cohort studies in developed countries that related all-cause mortality to amount of alcohol consumed. Studies were excluded if mortality was not reported for men and women separately 10, 11 or if fewer than three groups of alcohol consumption were reported

Results

Table 1 lists the 20 studies and the main features of the analyses used. A total of 60,224 deaths were observed among men, and 74,824 deaths among women. The adjusted risk of death is plotted against alcohol consumption for each study in Figure 1. The horizontal scales are the same for all except the Italian study, enabling comparisons of the nadirs to be made.

For men, 12 of the 19 estimated nadirs lie below 10 units per week, and 3 lie above 20 units per week (Figure 2). The estimated nadirs

Discussion

This meta-analysis has estimated the level of alcohol consumption at which mortality is lowest (the nadir). If the relationship is causal and not biased or confounded, and if any subgroups with different nadirs can be correctly identified, then the value of the nadir is relevant to the formulation of sensible drinking messages.

Various techniques have been used to avoid bias and confounding. First, all results were adjusted for age, and all but five studies were adjusted for smoking and other

Acknowledgements

I am grateful to Simon Thompson and Paul McKeigue for their helpful comments on earlier drafts of this paper, and especially to Stuart Pocock for many helpful discussions and ideas.

References (52)

  • R. DerSimonian et al.

    Meta-analysis in clinical trials

    Control Clin Trials

    (1986)
  • A.G. Shaper et al.

    Alcohol and mortality in British menExplaining the U-shaped curve

    Lancet

    (1988)
  • P. Anderson et al.

    The risk of alcohol

    Addiction

    (1993)
  • R. Beaglehole et al.

    Alcohol, cardiovascular diseases and all causes of deathA review of the epidemiological evidence

    Drug Alcohol Rev

    (1992)
  • C.D.J. Holman et al.

    Meta-analysis of alcohol and all-cause mortalityA validation of NHMRC recommendations

    Med J Aust

    (1996)
  • M. Marmot et al.

    Alcohol and cardiovascular diseaseThe status of the U shaped curve

    BMJ

    (1991)
  • M. Maclure

    Demonstration of deductive meta-analysisEthanol intake and risk of myocardial infarction

    Epidemiol Rev

    (1993)
  • A.L. Klatsky

    Epidemiology of coronary heart disease—influence of alcohol

    Alcohol Clin Exp Res

    (1994)
  • Inter-Departmental Working Group

    Sensible Drinking

    (1995)
  • J.C. Duffy

    Alcohol consumption and all-cause mortality

    Int J Epidemiol

    (1995)
  • P.A. Scherr et al.

    Light to moderate alcohol consumption and mortality in the elderly

    J Am Geriatr Soc

    (1992)
  • T. Gordon et al.

    Drinking and mortalityThe Albany study

    Am J Epidemiol

    (1987)
  • L.O. de Labry et al.

    Alcohol consumption and mortality in an American male populationrecovering the U-shaped curve—findings from the Normative Aging study

    J Stud Alcohol

    (1992)
  • O. Suhonen et al.

    Alcohol consumption and sudden coronary death in middle-aged Finnish men

    Acta Med Scand

    (1987)
  • J. Rehm et al.

    Effects on mortality of alcohol consumption, smoking, physical activity, and close personal relationships

    Addiction

    (1993)
  • P. Boffetta et al.

    Alcohol drinking and mortality among men enrolled in an American Cancer Society prospective study

    Epidemiol

    (1990)
  • Cited by (0)

    View full text