Latest scientific news 25 February 2021

Benefits and hazards of wine/alcoholic beverages, the J-curve: from scientific evidence to individual and public health

Two well-known scientists and a family physician1 published three extensive reviews confirming the latest scientific evidence of the J-curve and discussing the methodological challenges, including the Mendelian randomisation (MR) but also possible practical implications of these results at individual and public health level2-4 . The following is a short summary of the very detailed discussion in these three publications.

The cardiovascular benefits of moderate drinking were observed already in the 1960’s, but the results were initially disputed by the scientific community. Only in the 1980s, many studies from all over the world examined the relationship between moderate drinking and cardiovascular health identifying a potentially protective level of alcoholic beverages consumption between the extremes of abstinence and drinking in excess. The graphic representation of this relationship is J-shaped.  A growing number of studies have followed since then and have shown similar results in different individuals (older and younger adults and those with chronic diseases) and populations for cardiovascular disease, type 2 diabetes and total mortality.

Most importantly, the results from the observational studies were shown to make sense biologically: in moderate drinkers, favorable changes in blood parameters predictive of coronary heart disease risk (such HDL cholesterol, fibrinogen levels involved in blood coagulation and atherosclerosis), a higher insulin sensitivity and a lower-level inflammation in the body (atherosclerosis is a chronic, low grade inflammatory disease) were observed.

Despite the vast amounts of studies, these observed benefits of moderate wine/alcohol consumption have been questioned in the past three years and media headlines have been claiming that “There is no safe amount of alcohol”, thus denying the above-mentioned positive health effects.

Methodological considerations

Criticism has focused on the methodological challenges encountered in the observational studies, such as: imprecise alcohol intake (self-reported), the composition of the reference group (“sick quitters” 5), potential biases and confounding factors.

Despite these possible methodological problems, the authors conclude that this does not seem to undermine the findings on the J-shaped relationship between coronary heart disease and alcoholic beverages observed in epidemiological studies.  Such observational evidence indicates indeed a causal association and is strongly supported by short-term studies where the consumption of wine/alcoholic beverages was associated with cardiovascular risk factors as well as many experimental data in animals. However, the existence of a J-curve is being challenged by some degree of uncertainty.

Only a long-term randomized controlled trial, where healthy adults are dinking wine/alcoholic beverages and then waiting for the development of CV, would be the best evidence but it might not be feasible.

Why do large meta-analyses come to different conclusions and dispute the existence of a J-curve?

Many studies examining the effects of alcoholic beverages lack information on drinking patterns such as the frequency of drinking (regular moderate versus binge drinking), the type of beverage (wine, beer or spirits) and whether it is consumed with or without meals. For example, in the Mediterranean drinking patternThe Mediterranean diet, which includes moderate consumption of wine, is considered ..., wine is consumed in modest amounts as an accompaniment to food which lowers the peak blood alcohol concentration.

Furthermore, there may be important differences associating the consumption of alcoholic beverages with CVD than when relating alcohol to cancer. For example, binge drinking in women is associated with an increased risk in breast cancer, while measuring only the “alcohol intake per week” often does not account for binge drinking. The need to evaluate the pattern of drinking rather than just the average intake has been emphasized since weekend drinking has a stronger effect on breast cancer risk than spreading the consumption of similar amounts throughout the week. Indeed, higher blood alcohol levels may be the key in raising the level of acetaldehyde, the main alcohol break-down product and a known cancer-causing agent.

The distinction of the drinking pattern is, thus, particularly important for cancer related studies:  peak blood levels of alcohol (binge drinking of large amounts in a short time) represent a more influential factor than the average weekly consumption.

Mendelian randomisation

Newer attempts using Mendelian randomisation (MR) as an alternative way of obtaining unbiased “truth” about the association between alcoholic beverages and CVD and using genetic factors as indices of alcohol intake have been suggested. As more genetic factors are discovered and used to learn the true relation between alcohol consumption and health, the knowledge will progress markedly in the future. Nonetheless, relying just on an estimate of average intake, however accurate it may be, is insufficient to study this relation – many environmental and lifestyle factors are important in modifying the health effects of drinking and must be considered when making conclusions or recommendations on alcohol and health.

And even though many genetic factors which can be included in MR analyses have been already identified, it is obvious that results from a variety of studies must be considered when attempting to judge the overall health effects of alcohol consumption. This is especially the case because type of beverage, drinking patterns (regular moderate vs binge drinking, rate of consumption, with or without food), smoking and other lifestyle habits, diet and many other environmental factors can modify the effects of alcohol consumption. Thus, the combination of data from observational studies, clinical trials, animal experiments as well as MR analyses will be needed to improve our knowledge between the relation of alcohol intake, health and disease. It remains a continuing challenge.

How can this scientific evidence around the J curve be translated into practical life?


J curve and public health

From the perspective of a health professional (family physician), health communication should emphasize the nadir of the J curve as a health range for the general population. Then, the focus might be on reducing excessive intake. “When the evidence is clear, public health leaders should embrace the left side of the J curve and point out the pursuit of the nadir as the goal.”

He also suggests that advice to the public regarding alcohol consumption should always include other important lifestyle factors that affect health: smoking, obesity, diet and exercise. However, regardless of other healthy lifestyle factors, the presence of light to moderate regular consumption of alcoholic beverages in subjects, when added to the analyses, provides significantly greater protection against CVD than seen for the adherence to all the other lifestyle factors alone.

J curve and individual health

Skovenborg et al. further explain that a sensible interpretation of the message of the J curve would be: a consumption of more than a few drinks per day is undesirable from the standpoint of health for most people. “Whether drinking low to moderate amounts may be desirable or undesirable depends on individual characteristics and any advice regarding consumption of alcohol should be adjusted to factor in the risks and potential benefits for each individual patient considering age, sex, family history, personal drinking history and specific medical history. Instead of waiting for the perfect consensus on the evidence, what is required is a synthesis of common sense and the best available scientific facts by a knowledgeable health professional to make a balance judgement.”

 Notes and references:

  1. Prof. Morten Groenbaek, Director at the National Institute of Public Health in Copenhagen, Denmark;Prof. Curtis Ellison, Department of Medicine, Boston University School of Medicine, Boston, USA; Dr. Erik Skovenborg, Family physician, Aarhus, Denmark.
  2. Ellison, R.C., M. Grønbæk, and E. Skovenborg, Using Mendelian randomization to evaluate the effects of alcohol consumption on the risk of coronary heart disease. Drugs and Alcohol Today, 2021. 21(1): p. 84-95.
  3. Skovenborg, E., M. Grønbæk, and R.C. Ellison, Benefits and hazards of alcohol-the J-shaped curve and public health. Drugs and Alcohol Today, 2020. 21(1): p. 54-69.
  4. Grønbæk, M., R.C. Ellison, and E. Skovenborg, The J-shaped curve-conceptual and methodological challenges. Drugs and Alcohol Today, 2020. 21(1): p. 70-83.
  5. Often, people who give up drinking do so because of health conditions, which may or may not be caused by alcohol. This means that in some studies, people that drink light-moderate amounts of alcohol are often healthier than non-drinkers. This si called sick quitters effect