Letter by Stranges et al Regarding Article, “To Drink or Not to Drink? That Is the Question”
To the Editor:
In a recently published article in Circulation,1 Drs Kloner and Rezkalla reported an extensive review of the literature in the attempt to answer the pending question on whether or not patients and the general public should be advised to drink moderate amounts of alcohol on the basis of the existing evidence of beneficial effects on total mortality and cardiovascular disease. This is a question of considerable clinical and public health significance that is still waiting for a definitive answer.
Although we appreciate the efforts made by the authors1 in their scrutiny of the literature, we believe that this review failed to adequately acknowledge some of the methodological limitations in the studies cited that have given rise to controversy regarding their findings, particularly the reference group selection and the lack of an evaluation of the drinking measures or attention to the role of drinking pattern. We agree with the authors’ conclusion that moderate drinking is likely to confer cardiovascular benefits, but it should be acknowledged that the magnitude of these benefits may be biased by the choice of reference categories, because it has been argued persuasively that former drinkers (quitters) may well be less healthy for reasons unrelated to the fact that they are not currently drinking.2
A number of the studies cited were not originally designed to examine the effects of alcohol consumption on health and did not use standard quantity-frequency questions to assess intake. Moreover, most of these studies did not adequately measure drinking pattern, which is a problem not only because it renders uncertain the actual volume of alcohol consumed but also because the pattern of drinking is likely to influence the effect of alcohol consumption on health. Clearly, the effects on health of having 7 drinks on a Saturday night are likely to differ from those of having a glass of wine with dinner every evening of the week. However, the volume measures used in many of the studies cited define both as 1 drink a day or moderate drinking. Substantial evidence from studies designed to look specifically at drinking pattern now verifies that the way alcohol is consumed (ie, frequency and intensity of consumption, binge drinking, drinking with/without meals) may modify the underlying association between alcohol consumption and both total mortality and cardiovascular disease.3–4 Thus, a recommendation that moderate drinking protects against cardiovascular disease and total mortality without considering the risks associated with “unhealthy” patterns of drinking could potentially lead to unbalanced advice given to individuals and the population at large with regard to drinking habits. Finally, cardiovascular disease and mortality represent end points influenced by a lifetime of risk exposures. Alcohol consumption varies considerably over the course of a lifetime, and the relation of this variability to health is poorly understood.5
Drs Kloner and Rezkalla acknowledge that large, randomized trials of moderate drinking pose ethical, behavioral, and logistic challenges. We submit that further research is needed on the association between lifetime drinking patterns and health before such trials are undertaken.
Kloner RA, Rezkalla SH. To drink or not to drink? That is the question. Circulation. 2007; 116: 1306–1317.
Stranges S, Wu T, Dorn JM, Freudenheim JL, Muti P, Farinaro E, Russell M, Nochajski TH, Trevisan M. Relationship of alcohol drinking pattern to risk of hypertension: a population-based study. Hypertension. 2004; 44: 813–819.