You've probably heard that a little booze a day is good for you. I've even said it at parties. "Look at the French," I've said gleefully over my own cup. "Wine all the time and they still live to be not a day younger than 82."

I'm sorry to say we're probably wrong. The evidence that alcohol has any benefit on longevity or heart health is thin, says Dr. Timothy Naimi, a physician and epidemiologist at Boston Medical Center.

He and his colleagues published an analysis 87 of the best research studies on alcohol's effect on death from any cause inthe Journal of Studies on Alcohol and Drugs on Tuesday. "[Our] findings here cast a great deal of skepticism on this long, cherished belief that moderate drinking has a survival advantage," he says.

In these studies, the participants get sorted into categories based on how much alcohol they think they drink. Researchers typically size up occasional, moderate and heavy drinkers against non-drinkers. When you do this, the moderates, one to three drinks a day, usually come out on top. They're less likely to die early from health problems like heart disease or cancer and injury.

But then it gets very tricky, "because moderate drinkers tend to be very socially advantaged," Naimi says. Moderate drinkers tend to be healthier on average because they're well-educated and more affluent, not because they're drinking a bottle of wine a week on average. "[Their] alcohol consumption ends up looking good from a health perspective because they're already healthy to begin with."

To make things worse, Naimi says that some of the non-drinkers in these studies weren't always dry. "People in poor health tend to quit drinking," he says. In the studies, those who abstained from alcohol altogether were lumped together with those who quit later in life, bringing down the overall health of the entire group.

Naimi and his colleagues sorted the lifetime nondrinkers from the quitters, controlled for socioeconomic class and reanalyzed the data from all 87 studies. "When we accounted for these biases, moderate drinkers had no survival advantage," Naimi says. "It became a wash either way." Those who imbibe one or three glasses a day appeared to do no better than those who never touch alcohol.

What's more, Naimi says the group that did the best were the people who had on average one drink every 10 days. "It's almost homeopathic amounts of alcohol. So it's extremely unlikely that it's the alcohol that's making them look good," he says. The heavy drinkers had the shortest lives on average, which should come as no surprise.

The meta-analysis severs one arm of reasoning why alcohol might be beneficial, says Dr. Jennie Connor, a physician and researcher at the University of Otago in New Zealand who was not involved in the work. "They've very successfully investigated one of the many series of problems in this reasoning," she says. "People used studies to say, 'Hey, look there's a benefit [to alcohol], and when we talk about harms we have to balance that against the benefit.' That's always been a fallacious argument, and it's really been reduced to nothing now."

The Centers for Disease Control and Prevention has also given the purported health benefit of alcohol a wary eye. The agency recommends people have no more than one drink a day for women and two a day for men. This refers to the amount imbibed on one day and isn't meant to be taken as an average over several days. The Dietary Guidelines for Americanssuggests people who don't drink alcohol should not start for health reasons.

Naimi says the biological hypothesis for why alcohol could be good for health has also been eroding. "[Alcohol] raises your good cholesterol. That's the main biological argument," he says. "But the whole idea that [good cholesterol] causes heart health is going away now." A recent study published in Science showed that people with naturally higher levels of good cholesterol had more cardiovascular diseases than those without.

But still, the notion that alcohol extends our lifespans is a persistent one. According to Connor and Traci Toomey, an alcohol policy researcher at the University of Minnesota, studies supporting the link between moderate drinking and long life have been used to argue against certain alcohol harm reduction policies. "Usually there are proposals of having [health] labels," Toomey says. "[The industry] controls a lot of the messaging around alcohol."

In Connor's eyes, that messaging has kept the notion alive in spite of recent work showing moderate drinking does not provide a cardiovascular benefit. "Even if it were true," she says, "the idea that people are going to drink only one or two drinks a day as medication for the rest of their lives is just ridiculous. We've been sold an idea that's incoherent."

Copyright 2021 NPR. To see more, visit https://www.npr.org.