Longevity Briefs: Is A Little Alcohol Really Healthier Than None At All?

Longevity briefs provides a short summary of novel research in biology, medicine, or biotechnology that caught the attention of our researchers in Oxford, due to its potential to improve our health, wellbeing, and longevity.

A note on generative AI use: Current AI tools do not meet our standards for reliably providing accurate information concerning health and longevity. We do however use AI tools in a limited capacity to facilitate our work and communication of information. Read our FAQ for more details.

The problem: We’ve all heard it before: a little bit of alcohol is good for you. Many studies have suggested that people who drink a small amount of alcohol tend to be healthier than those who drink none at all. However, people who abstain from alcohol often do so as a result of other health problems. A significant percentage of non-drinkers are former alcoholics who have already done considerable damage to their health prior to quitting. People who develop chronic health conditions also tend to stop drinking alcohol. This means that as people grow older and develop health problems, the least healthy among them drop out of the drinking population, meaning that those who still drink represent a healthier than average group. These factors must be carefully controlled for if we are to have an accurate picture of the impact of alcohol on health and longevity.

The discovery:

In this study, researchers conducted a meta-analysis of 107 existing studies involving over 4.8 million participants in total. They found that across studies deemed to be of good quality (those that excluded former drinkers from the abstainer group, whose participants were exclusively under the age of 55 and were followed up past the age of 55), people drinking 1.3-25 grams of ethanol per day on average (between 1 glass/week and two glasses/day) were 2% less likely to die during the study followup than abstainers. This was not a statistically significant difference. In other words, when they compared drinkers to people who had never been drinkers, and in an age group that are unlikely to have quit alcohol due health problems (which as mentioned, leaves behind a healthier-than-average sample of drinkers), there is no significant difference in risk.

If either of these criteria were not fulfilled (former drinkers were not excluded or over-55s were included) then the relationship was a classic ‘J-shaped’ curve: people who drank a little had lower mortality than those who did not drink at all, while further drinking increased mortality. However, when only good quality studies were taken into account, the start of the J-curve was flattened.

The implications:

This study argues that the apparent benefit of low level alcohol consumption is a mirage caused by poor quality studies that fail to properly control for confounding factors. This is not a new idea – the merits of low level alcohol consumption have been disputed for a long time, but media and scientific journals alike are biased towards reporting positive findings, and we ourselves are biased towards believing what we want to believe. On the subject of bias, it should be noted that multiple authors of this study did receive travel support from IOGT-NTO, a Swedish non-profit temperance society, so some potential conflict of interest does exist.