For decades, researchers have tried to answer two seemingly simple questions: Is moderate alcohol consumption harmful? And does it provide any health benefits?
A new study claims the answer to both questions is “No.” The researchers found that men who drank up to 45 grams of alcohol a day (around three glasses of wine) and women who drank up to 25 grams (around two glasses) had approximately the same lifespan as people who never drank. However, these moderate drinkers did not appear to enjoy any net health benefits when considering all causes of death. Consistent with previous evidence, heavy drinking (more than around three drinks a day for men, and more than two for women) was associated with significantly increased risk of early death, and women faced greater health risks at lower levels of consumption than men.
The findings prompted a wave of headlines across the media, some of them oversimplifying the risks and benefits of drinking to the point of being confusing or misleading. While the conclusions may encourage other researchers to improve how future alcohol studies are conducted, it’s unclear how much the study contributes to our evolving understanding of the specific links between moderate wine consumption and health.
What You Need to Know Beyond the Headlines
- • The study looked at all types of alcohol; it did not break out wine, beer and spirits.
- • The study examined “all-cause mortality”—death due to any cause—not the risks and benefits related to individual diseases and other causes of death, including accidents.
- • The study categorized “moderate drinking” as up to around three glasses per day for men and up to two per day for women—higher than the U.S. recommendation.
- • To correct for biases in past studies, researchers broke abstainers into two groups: those who never drank and those who used to drink, but quit.
- • They also corrected for other lifestyle factors—such as diet, exercise and socioeconomic status—that may have skewed results of past studies.
- • For this analysis, researchers divided people into two broad age groups, people ages 19–55 and 56–78. Many alcohol studies focus on older drinkers, potentially making drinking seem healthier than it is, while younger drinkers are more likely to die of alcohol-related injuries.
The study, “Association Between Daily Alcohol Intake and Risk of All-Cause Mortality,” was published on March 31 in JAMA Network Open and led by researchers at the University of Victoria’s Canadian Institute for Substance Use Research. It received partial funding from the Canadian Centre on Substance Use and Addiction, an NGO, “as a subcontract for a Health Canada grant to develop guidance for Canadians on alcohol and health,” though that department of the government played no role in the study. (Earlier this year, Canada lowered its recommended alcohol consumption guidelines to two drinks or less per week, dramatically less than the current U.S. guidelines of no more than two drinks per day for men and one per day for women.)
The study’s central finding is that moderate drinkers live about as long as people who never drink. There are two valid ways to interpret that statement. On one hand, moderate amounts of alcohol are not harmful and do not shorten drinkers’ lifespans. On the other, moderate amounts of alcohol don’t seem to help prolong life. Since moderate drinkers live about as long as those who never drink, but no longer, the study implies that there are no health benefits of moderate alcohol consumption, at least when looking at all causes of death lumped together. That finding contradicts decades of research—including many of the studies included in the analysis—that have linked low and moderate alcohol consumption to improved health. (Research on specific diseases has linked moderate drinking most strongly with improved cardiovascular health.)
While studying all-cause mortality is good for measuring the effects of methodological biases, “it’s not a good way of getting accurate estimates of the real risk from drinking.”
Dr. Tim Stockwell—a lead author of the study, former director of the Canadian Institute for Substance Use Research and emeritus professor of psychology at the University of Victoria—told Wine Spectator that the analysis “does undermine [past research linking alcohol to heart health, but] it doesn’t eliminate it. This isn’t the final story … we’ve just identified a weakness.”
He also notes the complexity of studying alcohol’s effects on the body: “There are lots of moving parts here, and these aren’t [randomized controlled trials]—these are observational studies. This is life, and there are so many factors influencing our health. It’s really, really hard to tease out what’s due to alcohol and what isn’t.”
The study is a meta-analysis, meaning it presents new statistics based on existing data from multiple studies; it looked at 107 studies conducted between 1980 and 2021 involving a total of over 4.8 million participants between the ages of 19 and 78. The scientists performed a series of adjustments to those analyses to better account for what they say are biases in the methodology of past studies on all-cause mortality and alcohol. According to Stockwell, “These all-cause mortality studies are really imprecise.” Statistically adjusting for biases, the researchers claim, shows that any apparent benefits of moderate alcohol consumption were actually due to methodological flaws.
One of those potential flaws is known as abstainer bias or former drinker bias, often called the “sick quitter” phenomenon. This is the idea that some people in the abstainer control group are actually former drinkers. (Some purported abstainers may also be very occasional drinkers who report never drinking.) Many of these people may have given up alcohol because of health problems, whether due to heavy drinking or something else. Therefore, comparing current drinkers to a non-drinker control group that includes less-healthy former drinkers makes moderate drinkers seem healthier than they are.
While researchers have long been aware of abstainer bias, few have examined its impact on alcohol studies. Of the 107 studies used for this meta-analysis, only 21 did not exhibit any form of abstainer bias. Stockwell and his colleagues hope that other scientists will go to greater lengths to eliminate this bias in future studies.
Another issue with alcohol studies—like all epidemiological, or observational, research—is the difficulty of controlling for other lifestyle factors that influence health. These confounding variables include socioeconomic status, diet, exercise, race, smoking status and more. Many skeptics of wine’s health benefits claim that wine drinkers tend to be healthier to begin with, with better diets, more active lifestyles and better access to healthcare. Though many researchers attempt to control for those factors to isolate the health effects of alcohol consumption, it’s impossible to fully avoid their influence. Of the 107 studies examined, 91 controlled for smoking and 69 adjusted for socioeconomic status, but only 44 controlled for exercise, 28 accounted for race and just 17 did so for diet. Stockwell and his colleagues also noted that future studies should enroll younger drinkers and follow them into old age to reduce bias.
What does the study mean for people who enjoy a glass or two of wine with dinner? The research team examined links between any alcohol consumption and any cause of death. Studies have repeatedly shown that wine drinkers enjoy better health outcomes compared to beer and spirits drinkers. Since this study lumped all alcohol together, it may not reflect potential health benefits due to the polyphenols in wine.
Acknowledging that there have been “confusing messages flying around,” Stockwell noted, “We do not think these kinds of studies (on alcohol and death from all causes) are the best way to develop guidance on low-risk drinking—even though in the past these kinds of studies have been used for that purpose.” He says recent research has indicated increased, if small, risk at much lower levels of consumption than previously thought.
By assessing risk of dying from any cause, the study grouped together various diseases—such as cardiovascular disease, cancer and diabetes—as well as accidental causes of death. Such accidents are largely tied to binge drinking, which all authorities agree has devastating health consequences and is more common among younger drinkers. It’s not possible to use the meta-analysis’ figures to parse how drinking is related to any given health condition. Stockwell calls for further studies on specific diseases, saying that while studying all-cause mortality is good for measuring the effects of methodological biases, “it’s not a good way of getting accurate estimates of the real risk from drinking.”
These aspects of the study may limit its applicability to the lives of people who mostly or exclusively drink wine, never binge drink and have concerns about their risk of developing particular illnesses—especially cardiovascular disease, which remains the number one cause of death among Americans, according to the Centers for Disease Control and Prevention. Stockwell embraces the need for clearer and more sophisticated ways of “communicating risk in a way that people can get their heads around,” he says. “What’s the scale and scope? How do I decide? What does it mean for me?”
When asked what he hopes current moderate drinkers take away from the study, Stockwell reflected, “I think it’s the duty of the scientific community to be as transparent and rigorous as possible in communicating what we know about potential negative or positive effects of drinking alcohol … so that people can make up their own minds.”
Want to learn more about how wine can be part of a healthy lifestyle? Sign up for Wine Spectator‘s free Wine & Healthy Living email newsletter and get the latest health news, feel-good recipes, wellness tips and more delivered straight to your inbox every other week!