Studies that suggest alcohol can be good for you in moderation – and even extend your life – may be based on flawed research methods, according to a team of scientists bent on setting this ubiquitous misconception straight.
This paper is the most recent installment in what is becoming something of a saga between the University of Victoria-led team and the International Scientific Forum on Alcohol Research (ISFAR).
Psychologist Tim Stockwell and epidemiologist Jinhui Zhao, both from the University of Victoria in Canada, led the most recent study to verify whether the claim – and the research supporting it – holds up under scrutiny.
Their meta-analysis cross-checks 107 scientific studies that investigated the association between alcohol consumption and the risk of death from any cause. That array includes data on a total of 4,838,825 participants and 425,564 recorded deaths.
They split the studies by quality, for comparison, to illuminate where bias may have led data interpretation astray.
‘High quality’ studies were defined as those in which the average age of the cohort was 55 years or younger, where follow-up extended beyond age 55, and where former or occasional drinkers were excluded from the reference group.
This stratification weeded out studies affected by ‘abstainer bias’, where a person’s inclusion in the control group because they abstain from drinking alcohol doesn’t take into account their lifetime drinking habits.
The risk is that people affected by long-term drinking, who have quit for health reasons, can be included in the control group, even though they are not representative beacons of long-term abstinence.
And comparing the longevity of a one-wine-a-day drinker with that of a now-sober but long-term alcoholic isn’t the same as comparing that moderate drinker with someone who never touches the stuff.
Studies where participants are older are more likely to be affected by this issue, since there’s more lifetime within which current abstainers may have once been drinkers, accumulating associated health issues that throw off the average all-cause mortality in the abstainer group.
“That makes people who continue to drink look much healthier by comparison,” says Stockwell.
The meta-analysis confirmed that the average relative risk of death from any cause was much higher among low-volume (i.e.,’moderation’) drinkers, in studies that Stockwell and team deem ‘high-quality’ across several criteria.
“There is simply no completely ‘safe’ level of drinking,” Stockwell says.
The high-quality studies included those that were not affected by abstainer bias in the reference group, but also those that used younger cohorts; that excluded sick participants, assessed less than 30 days of alcohol use, did not control for smoking status, and did not control for socioeconomic status.
The lower-quality studies, those with older participants, and which did not distinguish between former drinkers and lifelong abstainers, the researchers found, were the ones linking alcohol consumption to longevity.
Studies on alcohol and mortality often control for smoking, as it’s linked to both drinking and higher mortality. But smoking might also influence how alcohol affects health. For example, alcohol can lead to smoking and increase cravings in heavy drinkers, and reduced alcohol use has been linked to quitting smoking.
Some studies found no lower mortality risk for low-volume drinkers when accounting for smoking. Stockwell and Zhao suggest future research should clarify whether adjusting for smoking falsely suggests moderate drinking is beneficial.
The authors note that the relationship between socioeconomic status (SES) and alcohol-related mortality also needs further study. One 2023 study found that high alcohol consumption increased mortality risk in low-SES individuals, less so in medium-SES individuals, and not at all in high-SES individuals.
“If you look at the weakest studies,” Stockwell says, “that’s where you see health benefits.”
Stockwell and Zhao have posed this question before – several times – and been met with criticism. While it’s easy to shrug off censure from those with clear financial motives, it’s more difficult to dismiss critique from fellow scientists.
“The present study is a response to the criticism that we should have excluded studies from our meta-analysis that did not meet pre-registered quality criteria,” the authors write, referring to criticism from ISFAR scientists published earlier this year.
ISFAR’s directors are former clinical pharmacologist Creina Stockley and nutrition researcher Henk Hendriks, who between them have received grants, funding, and paid consultancies from a variety of alcohol industry bodies, though it is unclear where funding for ISFAR comes from.
Stockwell’s team also disclosed receiving previous research funding from “government alcohol monopolies”, as well as travel support from IOGT-NTO, a not-for-profit nongovernmental organization that advocates against alcohol use.
This study was published in the Journal of Studies on Alcohol and Drugs.
