Once again moderation is stressed and no consensus on cardio-protective benefits of moderate drinking.  

Moderate drinking is usually defined as: ≤ 1 drink per day for women (≤ 2 per day for men) in the absence of binge drinking. One drink is defined as 12 oz of regular beer, 5 oz of wine (12% alcohol), and 1.5 oz of 80-proof spirits, all equivalent to about 15 g of alcohol.


Alcohol and CVD: The Tippling Point

If you drink alcohol, do so in moderation. If you don't drink, don't start.

So says the current advice of the American Heart Association in relation to alcohol and prevention of cardiovascular disease.

When it comes to the purported cardioprotective benefits of moderate alcohol consumption, the scientific world is divided into believers and skeptics and there are abundant data to support both points of view.

As reviewed on Medscape, the 2014 World Cancer Report, issued by the World Health Organization's (WHO's) International Agency for Research on Cancer, concluded that no amount of alcohol is safe, at least when it comes to cancer risk.

This prompted a critique of the WHO report by the members of the International Scientific Forum on Alcohol Research in which they disputed the "paternalistic blanket condemnations against alcohol," noting that "WHO seems to deliberately ignore the overwhelming scientific evidence showing that light-to-moderate consumption of alcohol not only reduces overall mortality but is usually not associated with an increased risk of cancer."

Is moderate drinking truly cardioprotective or is it merely a marker of a healthy lifestyle? Will we ever know for certain in the absence of randomized clinical trial data that will never be forthcoming, given the ethical and practical challenges entailed? For the nonbelievers, the evidence will always be on shaky ground, and adjustments for confounding will never fully cement the cracks.

Among the plausible explanations for alcohol's cardiovascular effects include reductions in platelet aggregation and thrombotic markers such as fibrinogen, increases in HDL cholesterol (by about 8%),[1] and anti-inflammatory effects such as lowering C-reactive protein levels.[2] But epidemiologists have been led down the observational garden path before, notably with hormone therapy[3] and vitamin E,[4] 2 therapies widely purported to have cardiovascular benefits that did not hold up in randomized trials.[5,6] Documented effects on surrogate markers like HDL-C are also no guarantee of a reduction in hard events as seen in trials with niacin.[7,8]

The "goldilocks" amount of alcohol is said to be 1 to 2 drinks daily. However, amounts in or near this level have been associated with an increased risk for hypertension,[9] an effect the pro-alcohol lobby say is explained by heavier drinkers who underreport their intake.[10] This uncertainty about the true drinking status of study populations is at the crux of the debate on alcohol and cardiovascular health.

Dr. Rob Califf (Duke University Medical Center) believes that there is a good chance that alcohol reduces cardiovascular events, but in an email he cautioned that "it's important to separate pure speculation supported by partial data from proven fact with clear evidence."

How Is "Moderate" Defined?

Moderate drinking is usually defined as: ≤ 1 drink per day for women (≤ 2 per day for men) in the absence of binge drinking. One drink is defined as 12 oz of regular beer, 5 oz of wine (12% alcohol), and 1.5 oz of 80-proof spirits, all equivalent to about 15 g of alcohol.

The J-Shaped Curve and Sick Quitters

The J-shaped curve for alcohol was first proposed in the 1970s, reflecting the observation that both teetotalers and heavy drinkers tended to die earlier than moderate drinkers.

In what the authors claim to be the most comprehensive meta-analysis to date, Ronksley and colleagues[11] reviewed 84 papers (culled from more than 4000 published from 1950 through September 2009) and concluded that alcohol consumption in the 2.5 to 14.9 g/day range was consistently associated with a 14% to 25% risk reduction in multiple cardiovascular outcomes (including CVD mortality) compared with abstaining.

Unlike the usual J-shaped curve, this analysis showed a 25% to 35% reduced risk of coronary heart disease (CHD), not only for light to moderate consumption but also for heavier intakes. Overall there was a neutral association with stroke mortality, which the researchers attributed to the fact that alcohol was associated with a slightly lower risk for ischemic stroke but an increased risk for hemorrhagic stroke.[11]

A re-analysis of the 84 papers included in this meta-analysis by a group of alcohol skeptics determined that all but two had serious methodological problems.[12] Among the perceived weakness of many studies on alcohol is the inclusion of former drinkers, so-called "sick quitters," in the abstainers cohort.[13] When Ronksley and colleagues excluded former drinkers or evaluated them separately, there was no change in the association of moderate drinking with a reduced incidence and mortality from CHD. These former drinkers did not appear to have an increased risk for cardiovascular events, but did have a higher risk of death from cardiovascular disease (CVD) or CHD.[11]

In contrast, another analysis using stricter criteria for ensuring the purity of the abstemious cohort, determined that studies that excluded former or occasional drinkers from the nondrinkers did not show a higher risk of all-cause mortality (7 studies) or CHD mortality (2 studies) among abstainers compared with "light" or "moderate" drinkers.[14]
Dishonest Teetotalers

Of course, excluding former drinkers requires one to accurately identify them. Caldwell[15] reviewed data from the 1958 British Birth Cohort Study that asked almost 9500 individuals about their drinking status at 5 timepoints from age 16 years until age 45, and found that 60% of those who claimed to be never-drinkers at 45 years of age had reported drinking in earlier follow-up studies (25% were past daily or weekly imbibers).

The British were able to spot this discrepancy because they assessed alcohol intake at multiple timepoints, but many studies use only a single baseline measure. Australian epidemiologists tackled this issue in a 2014 meta-analysis[16] on alcohol and total mortality by omitting studies that measured alcohol intake at only 1 timepoint -- only 9 cohort studies published during 1991-2010 (comprising 62,950 participants and 10,490 deaths) met their inclusion criteria.

Their findings showed weak evidence for an inverse association with mortality at lower intake levels (1-29 g/day) and an increased mortality risk for intakes over 40 g/day, compared with abstention. The authors cautioned that few studies reported findings for women and that "nonstandardized methods [were] used to capture and analyze exposure."

Excluding former drinkers is also problematic given the argument that a true "intention-to-treat" analysis would count former drinkers among the imbibers; weeding out these less healthy drinkers biases the sample. In a pooled analysis of 14 waves (1997-2010) of the US National Health Interview Survey (NHIS), it was shown that excluding former drinkers exaggerates the difference in health status between abstainers and drinkers, especially for men.[17]

Red Wine and the 60 Minutes Bump

It was back in 1991 that 60 Minutes first espoused the wonders of red wine in a segment on the so-called French Paradox. At the time, reporter Morley Safer noted that the benefits were "all but confirmed."

Postprandial alcohol has been shown to increase insulin sensitivity and glucose metabolism,[18] and wine-drinking tends to occur with meals. Even among the moderate-alcohol believers, opinions are split on whether red wine offers more cardioprotection than other types of alcohol.

When asked for comment, Dr. Curtis Ellison (Boston University School of Medicine), who is the Scientific Co-director of the International Scientific Forum on Alcohol Research (and was featured in the original 60 Minutes story), believes that most of the protection is from the alcohol but notes that "wine drinkers have a greater reduction in risk than consumers of beer, and usually much lower than consumers of spirits." While he believes that some of this added protection is due to the polyphenols in wine, he cautions that "we are comparing people, not beverages; but in animal experiments, almost always the effects from wine and/or its polyphenols are greater than the effect just of alcohol."

Reversal on Resveratrol?

Of all the nonalcohol components of wine, the polyphenol resveratrol is probably the most studied and is even sold in supplement form with the promise of a longer life. This antioxidant, found in the skin of grapes, was back in the headlines recently when the investigators for the Aging in the Chianti Region study reported no association between increasing quartiles of resveratrol (measured from 24-hour urine samples) and cardiovascular risk (or cancer risk) at 9 years among 783 men and women aged 65 years or older at enrollment.[19]

Resveratrol metabolites can be detected in the urine of people who consume 1 glass of wine per week if the last drink was consumed 3 days previously, or in those who consume 3 glasses of wine per week if the last drink was consumed 5 days previously.[20] In an interview with heartwire, the lead author of InCHIANTI, Dr. Richard Semba, claimed that this should "close the book" on the health benefits of dietary-acquired resveratrol.

Resveratrol devotees counter that the doses in wine are far lower than the concentrations shown to have anti-inflammatory effects in human studies. The resveratrol content in wine varies depending on the grape, with pinot noirs having the highest content. And yes, indeed, the concentrations in wine are much lower (1.2 to 13 mg/kg) than those tested in lab studies (110-150 mg).[20]

Marker of a Healthy Lifestyle in Selected Populations

Moderate alcohol consumption is a powerful general indicator of optimal social status, according to the nonbelievers.[21,22] Abstainers are typically older, eat a less healthy diet, and exercise less than moderate drinkers.[23,24] This association of moderate alcohol consumption with social status is especially seen in Western cultures.
The INTERHEART study, conducted in 50 countries, showed that regular alcohol consumption was protective overall but not among native South Asians.[25] Similarly, a prospective study of almost 4500 Indian men found no inverse relationship between moderate alcohol intake and CVD; in fact, a slight increased risk was seen.[26] Even within the United States, the Atherosclerosis Risk in Communities Study did not show a cardioprotective effect in black men in contrast to what was seen in white men.[21]
To address the potential confounding from unhealthy abstainers, some researchers have excluded participants with current or prior ill health in prospective studies, but critics say that this approach not only reduces the sample size considerably but might even artificially create a J-shaped curve, masking a true linear relationship.[27]

Counseling on Alcohol

Given the lack of consensus opinion and abundant confounding, what is the appropriate advice to give people about alcohol and CVD? Would anyone counsel a teetotaler to indulge, or suggest that a beer drinker favor the grape over the grain?
As the author of a paper on the role of confounding in alcohol research noted, "Even assuming [that] the benefits of low-dose consumption are real...alcohol is not a good candidate as a population-based prevention strategy, to say the least."[28]

Dr. Ellison, who has spent much of his career exploring the topic, concedes that many factors affect the choice to drink alcohol, including religious and social influences. His advice: "If you decide to consume alcohol, we know the pattern of drinking that is associated with the lowest risk of adverse events and the greatest chance of having health benefits. And that is small amounts on a regular basis (most days of the week), with no binge drinking. Further, if it is wine with meals, the benefits are even greater."

Dr. Califf agrees that modest drinking may be good for your health but advises having social support to avoid excessive drinking.

Dr. Stephen Devries, Director of the Gaples Institute for Integrative Cardiology, believes that alcohol may be one of the most cardioprotective elements of the Mediterranean-style diet, "but overall, I don't believe that alcohol consumption is an essential part of a protection program. For those who don't wish to include alcohol in their lifestyle or for whom there might be a special concern about addiction, there are plenty of other potent prevention strategies that can be employed."

Dr. Melissa Walton-Shirley (Cardiology Associates, Glasgow, Kentucky) concurs and goes further: "Alcohol is more trouble to a society than its proposed worth. When one takes into account the cost of alcoholism on family life; drunk driving; and treatment for liver disease, atrial fibrillation, stroke, cancer, dementia, and cardiomyopathy, its romanticized reputation as a medicinal is definitely diminished."

The scientific world may be divided on the health benefits of moderate alcohol consumption, but there is remarkable consensus on the harms of excessive drinking. That said, the debate on alcohol and health will not die out anytime soon. A recent segment on 60 Minuteson the 90+ study,[29] one of the largest studies on the oldest old (more than 1400 nonagenarians so far), touted moderate wine consumption (and dessert) as one of the keys to a long life.

Dr. Ellison reports that Forum members receive no compensation for their contributions. He receives some unrestricted donations from companies in the alcoholic beverage industry to operate the Website at Boston University. Donors have no role in selecting emerging papers that the Forum reviews, the discussions among Forum members of new papers, or the final critiques produced by the Forum. Donors learn about the critiques only when they are published on the Forum Website and made available to the public. He personally receives no grant support, speaker's fees, or other support from the industry and has no equity in such companies.

Drs. Califf, Devries, and Walton-Shirley report no relevant conflicts.



Antiaging Atlanta
12600 Deerfield Parkway

Alpharetta, Georgia 30004

Phone: 404-247-2167