Harvard scientists conducted a large-scale study that showed that moderate alcohol consumption reduces levels of “bad” cholesterol (LDL) and increases “good” cholesterol (HDL) .
These findings are supported by data from medical records of nearly 58,000 adults in Japan , who ranged from total abstinence to moderate drinking .
Limiting alcohol consumption is important because a variety of health problems, such as accidental injuries, liver disease, and high blood pressure, have been linked to excessive drinking. But new research from Harvard University suggests that it's not all bad. In cases of moderate consumption, there is a benefit in that "bad" cholesterol is lowered.
Researchers conducted an experiment involving nearly 58,000 Japanese adults, using a database of medical records obtained during routine checkups. They found that when people went from non-drinkers to drinkers during the study, their levels of the “bad” cholesterol – low-density lipoprotein cholesterol, or LDL – decreased.
“Good” cholesterol, or high-density lipoprotein cholesterol, or HDL, increased when participants drank alcohol, according to the study
Interestingly, HDL levels increased so much that they actually surpassed the improvements typically seen with medication, the researchers note.
People who went from no drinks to 1.5 drinks per day or less had a 0.85 mg/dL (bad) LDL cholesterol decrease and a 0.58 mg/dL (good) HDL cholesterol increase, compared to non-drinkers who never started drinking. Those who went from zero to 1.5-3 drinks per day had a 4.4 mg/dL (bad) LDL cholesterol decrease and a 2.49 mg/dL (good) HDL cholesterol increase.
For reference:
- The optimal LDL level for adults is less than 100 mg/dL;
- The optimal HDL level is 60 mg/dL or higher.
Importantly:
Higher LDL levels can increase your risk of heart disease, stroke, peripheral artery disease, and other health problems.
Higher HDL levels have a protective effect against cardiovascular disease. Although some of the changes reported in the study were small, the researchers note that in some cases they may be significant.
For example, a 5 mg/dL increase in LDL is enough to increase the risk of a cardiovascular event by 2–3 percent.
The researchers ran three different models to adjust for a variety of factors, including basic ones such as age, sex, body mass index, as well as diseases such as hypertension and diabetes, and lifestyle factors such as exercise, eating habits, and smoking. All models showed the same associations.
They also divided the data by the type of alcoholic beverage people consumed—wine, beer, sake, and other alcoholic beverages. The results were consistent across all categories.

