It’s fitting at this time of year, when many abstain from beer and wine for “Dry January,” to consider the shifting public conversation about drinking alcohol. The World Health Organization, for example, warns that even small amounts of alcohol can be harmful to health. News headlines often stress the risks.
Two researchers from Harvard’s School of Public Health and Medical School argue that this dire messaging—that no amount of alcohol is safe—oversimplifies the existing knowledge about alcohol consumption.
“Nobody thinks that lots of alcohol is good, but there are legitimate questions about small amounts,” says Kenneth Mukamal, a professor of medicine who studies how drinking and other lifestyle factors influence cardiovascular disease. With two decades of practice as a primary care physician, Mukamal understands the need to offer patients sound advice. “I want to be as accurate as possible,” he says. “The biggest frustration for me…is the implication that the science is settled so clearly and securely that we don’t actually need to get the right answer to this.”
Mukamal and professor of medicine Eric Rimm, a nutritional epidemiologist who has studied the impact of alcohol on cardiovascular health for more than 35 years, acknowledge that many people should not drink at all. These include pregnant women, people recovering from alcoholism, and those who can’t control how much they drink. Mukamal and Rimm also emphasize that any benefits from alcohol appear to come from limited doses; they agree with current U.S. dietary guidelines, which recommend no more than two drinks a day for men, and no more than one drink a day for women. (A drink is considered 12 ounces of regular beer, 5 ounces of wine, or 1.5 ounces of distilled spirits.)
When Rimm first came to Harvard in the 1980s as a graduate student in public health, he was intrigued by previous findings that people who drank modest amounts of alcohol experienced fewer heart attacks than those who drank heavily or didn’t drink at all. Some scientists wondered if people who drank modestly might have healthier diets or other beneficial habits. Others theorized that teetotalers might have more heart attacks than drinkers because they were less healthy; perhaps they were former heavy drinkers advised by their doctors to stop drinking. Rimm explored this “sick quitter” hypothesis in his doctoral thesis, using data from the thousands of people enrolled in Harvard’s Health Professionals Follow-Up Study. His goal was to tease out what specific factors accounted for moderate drinkers’ lack of heart attacks. That study, published in the Lancet in 1991, suggested that alcohol consumption alone—not diet or other factors—was responsible for about a 30 percent lower risk of heart attack. In the decades since, further studies have confirmed that light drinking can increase levels of high-density lipoprotein, HDL (the so-called “good” cholesterol), and reduce blood sugar and insulin levels, lowering diabetes risk.
But for many cancers, the story is different, Rimm says; drinking even modest amounts raises the risks of breast, colon, and esophageal cancer. Doctors may advise people at particular risk for breast cancer, for example, to limit their drinking, he adds.
Despite these links to cancer, Rimm and Mukamal still wonder if small amounts of alcohol could extend lives. “The number-one killer of women, by far, is heart disease,” Rimm says. “Unfortunately, heart disease kills 10 times more women than breast cancer.” The complex effects of alcohol consumption, Mukamal says, point to the need for more precise answers. This requires both observational studies, like the health professionals study Rimm relied on, and large randomized trials in which one group of people consumes a limited amount of alcohol over time, and another group doesn’t drink for a similar period. (Such randomized trials, considered the gold standard of research, are regularly used to study new drugs.)
Rimm would like more studies on what constitutes healthy drinking patterns. Based on research he and Mukamal have already done, he suspects it’s important for those who drink to space their consumption out across the week—every other day, for example—instead of consuming large amounts on weekends. Previous studies also support consuming alcohol with meals.
But large, randomized studies on drinking have been difficult to pursue, the researchers say. They blame a lack of funding and polarized attitudes about alcohol consumption. Nearly a decade ago, Mukamal was the lead investigator for a National Institutes of Health (NIH) study on the impact of alcohol on heart disease, involving participants at two sites in the United States and two more in Nigeria and the Netherlands. The New York Times published a story at that time suggesting that the study presented ethical problems because it was co-funded by donations made to the NIH foundation by alcoholic beverage companies. “The article said that the whole study was bought by industry,” Rimm says. Mukamal notes that ethics committees from the four study sites and Beth Israel Deaconess Medical Center, which was coordinating the project, all granted their approval. Still, the NIH ended the study.
Mukamal says he has never personally received funding from any industry groups, and rarely drinks alcohol himself: “The spirits at my house are leftovers from our wedding 30 years ago.” He and Rimm believe that public attitudes about alcohol are often shaped less by science and more by culture and personal experiences, such as having a family member addicted to alcohol.
The colleagues are closely watching a randomized trial on alcohol consumption now under way in Spain, hoping it may help answer some of their unresolved questions. In the meantime, they advise following current U.S. guidance. “You don’t have to drink,” Mukamal points out, “but if you would like to, don’t ever have more than one in a day if you’re a woman, or two in a day if you’re a man. Full stop.”
