Page last updated: Monday, May 12, 2008
Alcohol consumption and risk of heart failure in the Physicians' health study
The authors of a recent study state that heart failure (HF) is the leading cause of hospitalization among the elderly, and 1 in 5 adults aged 40 years will develop HF in their lifetime. Data on the effects of moderate alcohol consumption on the risk of HF have been sparse and inconsistent. This study sought to evaluate the association between moderate alcohol consumption and incident HF. A total of 21,601 participants of the Physicians’ Health Study I who were free of HF and provided data on alcohol intake at baseline were prospectively followed up from 1982 to 2005. Incident HF cases were ascertained through annual follow-up questionnaires and validated with the use of Framingham criteria.

During an average follow up of 18.4 years, 904 incident cases of HF occurred. The crude incidence rates of HF were 25.0, 20.0, 24.3, and 20.6 cases per 10,000 person-years for alcohol categories of <1, 1 to 4, 5 to 7, and > 7 drinks per week, respectively. There was no evidence for a strong association between moderate alcohol consumption and HF without antecedent coronary artery disease.

The authors conclude that although heavy drinking should be discouraged, their data indicate that moderate drinking may lower the risk of HF. The lack of an association between moderate alcohol intake and HF without antecedent coronary artery disease suggests that possible benefits of moderate drinking on HF may be mediated through beneficial effects of alcohol on coronary artery disease.

Comments by R. Curtis Ellison The results of this analysis support what many, but not all, recent prospective epidemiological studies have shown: a reduction in the risk of HF among moderate drinkers in comparison with non-drinkers or, as in this study, occasional drinkers. The reduction in risk is usually found to be greater for HF patients who have evidence of antecedent coronary heart disease (CHD), suggesting that much of the protection is from a reduction in the risk of a myocardial infarction or other evidence of CHD.

In an earlier paper from the Framingham Study, the risk for HF was lower for both CHD-associated and non-CHD associated disease; in the present study, cases without preceding CHD was divided into 2 groups, those “without antecedent MI” and those “without antecedent CAD.” For subjects in the category of >7 drinks/week, the hazard ratios for HF for these two groups were 0.66 and 0.84, respectively. While the 95% CI for both of these hazard ratios included 1.0, the direction was the same as the overall results (a reduction in risk). A weakness of the study is that only baseline estimates of alcohol consumption were available, so effects of changes in alcohol intake over time could not be assessed.

We note that the estimate of HF for subjects consuming > 7 drinks/week without antecedent MI (HR = 0.66) was essentially the same as the overall risk for subjects consuming > 7 drinks/week (HR = 0.62). Since the majority of cases (20 of 23) of HF among subjects consuming >7 drinks/week did not have an antecedent MI, there may well be other protective effects of moderate alcohol consumption against HF beyond its effects on CHD. Thus, we are not convinced that the data support the conclusion of the authors that “possible benefits of moderate drinking on HF may be mediated through beneficial effects of alcohol on coronary artery disease.” There may also be benefits other than through the effects of alcohol on the risk of CHD.

Article: Luc Djoussé L, Gaziano JM. Alcohol consumption and risk of heart failure in the Physicians’ Health Study I. Circulation 2007;115:34-39.

no website link
All text and images © 2003 Alcohol In Moderation.