This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.
Alcohol and cardiovascular disease: the J-shaped curve
Decades of many observational studies, animal experiments and even some clinical trials have led to the remarkably consistent finding that regular, moderate alcohol consumption has a favourable, protective effect on cardiovascular disease, and on overall mortality.
This is the basis of the J-shaped curve – compared to non-drinkers, mortality decreases with moderate alcohol consumption (a few units/day) before increasing with heavier drinking. So it was somewhat surprising that two publications in January denied its existence.
The first was a policy brief issued by the World Heart Federation¹. It concluded, ‘Contrary to popular opinion, alcohol is not good for the heart’, and ‘Research in the latest decade has led to major reversals in the perception of alcohol in relation to health in general and CVD in particular’.
So how can this be? What is this research that has led to such a major reversal? The policy brief draws heavily on a paper published in The Lancet in 2018². This was a combined analysis of data from over 500,000 alcohol drinkers. They were divided into eight groups, depending on how much they drank, ranging from less than 25g/week to over 350g/week. Among these drinkers, the lowest risk of all-cause mortality was among those consuming 100g/week, equivalent to about 6 175ml glasses of wine each week. This was widely publicised at the time as ‘the level of acceptable risk’, with mortality risk rising with higher levels of consumption.
But where is the data on non-drinkers for comparison? The evidence was hidden away within a large supplementary Appendix. This revealed the truth: that teetotallers experienced a whopping 30% more cardiovascular events than those drinking moderately, even higher than the heavier drinkers. For all-cause mortality, the teetotallers were 20% worse off than the moderate drinkers. So in short, those consuming 100g/week had the minimum risk for cardiovascular events, but what the paper neglected to point out was that it was that level of consumption that reduced the risk in the first place.
Incidentally, the WHF Policy Brief also claimed that, ‘No randomised controlled trials (RCTs) have confirmed health benefits of alcohol’. This is untrue. A meta-analysis (a summary of research data) of 44 randomised control trials published in the BMJ³ found favourable changes in biological mechanisms known to be related to cardiovascular disease when people drank alcohol at moderate levels.
The second (4 ) was a study with the unequivocal title, ‘Alcohol – The myth of cardiovascular protection’. This drew on data from over 350,000 participants in the UK Biobank study, and they were classified into beer, wine and spirit drinkers. As expected, the drinkers of all three types of drink enjoyed a lower risk for all cardiovascular events compared to never-drinkers. So how is this consistent with the headline of the study? As in the first study, the researchers performed their analyses on drinkers only, removing the teetotallers. The researchers effectively used lighter drinkers and not teetotallers as the reference group. Even then, wine consumption was associated with reduced risk.
It is unquestionable that heavy alcohol consumption and binge drinking carry significant health risks. But equally unquestionable is that for those who choose to drink alcohol, moderate consumption has health benefits, as both these studies actually show.
Despite what these studies claim, the J-shaped curve is alive and well.
Richard Harding is a former UK Government scientist who worked in food safety and nutrition. He was a member of the Department of Health’s Working Group on Sensible Drinking in 1995, and gave evidence on alcohol and health to the House of Commons Select Committee on Science and Technology in 2011.
- Wood AM et al, Risk thresholds for alcohol consumption combined analysis of individual participant data for 599,912 current drinkers in 83 prospective studies, Lancet 2018, 391, 1513-23. https://doi.org/10.1016/S0140-6736(18)30134-X
- Brien SE et al, BMJ 2011;342:d636, https://doi.org/10.1136/bmj.d636
- Schutte R et al, Alcohol – The myth of cardiovascular protection, Clinical Nutrition 41 (2022) 348e355, https://doi.org/10.1016/j.clnu.2021.12.009
Related news