TUESDAY, Aug. 15, 2017 -- Light to moderate drinking can lower your overall risk of premature death and, specifically, your odds of dying from heart disease, a new study reports.
Moderate drinkers -- men who have one or two drinks a day, and women who have one drink a day -- have a 29 percent decreased risk of heart-related death and a 22 percent reduced risk of death from any cause, compared with teetotalers, the study findings showed.
This study is the latest to examine whether alcohol is good or bad for you. The researchers found that light drinkers (fewer than three drinks a week) also receive some protection -- a 26 percent reduced risk of death from heart disease and a 21 percent overall lower risk of premature death, according to the report.
But the relationship between alcohol and death risk is a "J-shaped curve," in which too much drinking can be detrimental to health, said study co-author Dr. Sreenivas Veeranki.
Heavy drinkers are 27 percent more likely to die of cancer and 11 percent more likely to die early overall, the researchers found. Regular binge drinking one or more days a week also increased risk of early death, about 22 percent for cancer-related causes and 13 percent overall. (Binge drinkers consume excessive amounts of alcohol in a short time period.)
"If you're an alcohol consumer, drink with caution. Drink lightly," said Veeranki, an assistant professor in preventive medicine and community health at the University of Texas Medical Branch. "If you're an alcoholic, consume lower amounts on a less-frequent amount of days. If you're not an alcoholic, don't start, obviously."
For this study, Veeranki and his colleagues reviewed data from more than 333,200 people who participated in the federally funded U.S. National Health Interview Surveys from 1997 to 2009. The survey includes questions on alcohol use.
The investigators linked the survey data to the National Death Index, which showed that about 34,700 survey participants have since died. Of those people, just over 8,900 died from heart-related causes and 8,400 died from cancer.
Prior lab studies have shown that alcohol can lower "bad" LDL cholesterol and increase "good" HDL cholesterol in the blood, Veeranki said. Alcohol also appears to reduce the formation of plaques in blood vessels that can block arteries and cause strokes.
Veeranki noted that studies in the past have provided conflicting advice regarding drinking and health, in part because of unintended biases in the data. In this study, the researchers tried to account for all these potential sources of bias.
For example, the researchers accounted for the possibility that former drinkers might be misclassified as people who have abstained all their lives, Veeranki said. The study team also tried to account for the possibility that people who develop heart disease or cancer are told to quit drinking, which could skew results.
Dr. Eugene Yang, a clinical associate professor of medicine at the University of Washington, said that even with these efforts, any such study won't be able to control all of the variables.
For example, the surveys used for the new study rely on people self-reporting how much they regularly drink. "That already creates a bias in the study that you can't necessarily compensate for, no matter how you do the analysis," said Yang, who is also a member of the American College of Cardiology's Prevention Council.
Because of that, Yang said, "we can't really be that certain" about the potential health benefits of occasional drinking.
"If somebody is asking me if they should start drinking, my answer would be that some studies have shown some benefit, other studies have not shown a benefit. And because the level of evidence of a study like this is not the strongest, I don't advocate people start drinking just for the possibility of a cardiovascular benefit," Yang said.
The new study was released online Monday in the Journal of the American College of Cardiology.
For more on alcohol and heart health, visit the American Heart Association.
Copyright © 2017 HealthDay. All rights reserved.