You might have stumbled across some headlines recently declaring that drinking alcohol will lower your risk of type 2 diabetes. “Looking to Ward Off Diabetes? Drink Alcohol, Study Suggests,” proclaimed one article. “Regular Drinking Can Reduce Risk of Developing Diabetes,” another stated. If that seems a little off to you, you’re right. While research has found that moderate drinking is generally OK for you, it’s a stretch to say that it’s actually good for you—and that includes lowering your diabetes risk.
Here’s what you need to know about that recent diabetes study.
That study, published in the journal Diabetologia, was based on data from the Danish Health Examination Survey, which followed 28,704 men and 41,847 women for about five years. During that time, participants completed self-reported questionnaires to get information on how often they drank alcohol, how often they engaged in binge drinking, and how much wine, beer, and hard alcohol they drank weekly. The researchers also got information on how many of these people developed diabetes.
Here’s what they found: During a follow-up, 859 men and 887 women developed diabetes. The researchers found that the people with the lowest risk of developing diabetes were men who had 14 drinks per week and women who had 9 drinks per week. According to their data, these groups were less likely to develop diabetes than people who had one drink or less per week. They also examined the types of alcohol people were drinking and found that those who drank wine had the lowest diabetes risk; and in men, beer was also linked with a lower risk of developing diabetes. Hard alcohol didn’t seem to sway the risk one way or another for men, but women who had seven or more servings of spirits a week had an increased risk of developing diabetes.
But here’s a really important point: This study didn’t find that drinking actually lowers your risk of diabetes.
Instead, the study found that the people they examined with the lowest risk of developing diabetes also happened to be moderate drinkers. “It is an observation, not an experiment,” Robert M Cohen, M.D., an endocrinologist at UC Health and professor of medicine in the University of Cincinnati College of Medicine, tells SELF. “It is not valid to say that the reported amount of alcohol intake caused the difference in diabetes risk. There is a suggestion, however, that they may in some way be related.”
It’s not completely out of nowhere: Previous research, including one meta-analysis published in the journal Diabetes Care, also found an association between light to moderate alcohol use and a lower diabetes risk. But again, that’s just an association.
“Red wine and wine in general contains polyphenols, which are shown to help improve insulin resistance,” Mary Vouyiouklis Kellis, M.D., an endocrinologist at the Cleveland Clinic, tells SELF. Alcohol also seems to influence a hormone called adiponectin, which could help influence a person’s insulin sensitivity, lowering their risk of diabetes, she says. One study published in the journal Diabetes Care found that adiponectin levels increased with moderate drinking in middle-aged men.
There are a few other limitations worth noting here.
The fact that the data came from self-reported questionnaires is a “significant red flag,” Marc Leavey, M.D., an internist at Baltimore’s Mercy Medical Center, tells SELF. People are notoriously bad at reporting how often they use substances, including reporting what they have, how much they have of it, and how often they have it, he explains.
“No one wants to think of themselves as outside the norm so it is extremely likely that the respondents under-reported the amount of alcohol that they consumed or, worse yet, under-reported their diagnosis of diabetes,” diabetes expert Albert Tzeel, M.D., M.H.S.A., Regional Medical Director for Senior Products at Humana — North Florida, tells SELF. It’s also not great that the study only examined a Danish population, which doesn’t have as wide an ethnic makeup as a country like the U.S., Dr. Tzeel says. When researchers only study one ethnic group or a group of people predominantly from the same ethnicity, it’s hard to make generalizations for others. African Americans, Mexican Americans, American Indians, Native Hawaiians, Pacific Islanders, and Asian Americans are at a higher risk for diabetes, according to the American Diabetes Association—and the most recent study didn’t cover those at-risk populations.
It’s also important to note that people who drink moderately may also have other lifestyle factors that could cause a lowered risk of diabetes, like eating a healthy diet or exercising regularly. Or it could be that people who don’t drink are more likely to abstain because they have other health conditions that put them at a greater risk of diabetes.
