Alcohol advice often comes in simple sound bites. Any amount is harmful. A glass of wine may be good for your heart. Moderate drinking is probably fine.
A new global analysis helps explain why the reality is more complicated. Researchers found that alcohol consumption was associated with higher risks for all 10 cancers examined, with risks rising as people drank more. The study also found links with liver disease and pancreatitis. But for several heart, metabolic and neurological conditions, the relationship varied depending on how much alcohol people consumed.
The study, published in Nature Health, was a systematic review and meta-analysis of 843 cohort and case-control studies published through 2023. That means the researchers analyzed a large body of existing observational evidence rather than assigning people to drink different amounts of alcohol in a clinical trial.
The findings show associations, not proof that a particular amount of alcohol directly caused or prevented a specific disease. That distinction is especially important when interpreting the study’s more complicated findings about low-to-moderate drinking.
Researchers from the Institute for Health Metrics and Evaluation at the University of Washington examined alcohol consumption in relation to 20 health outcomes. These included cancers, cirrhosis and other chronic liver diseases, pancreatitis, type 2 diabetes, heart disease, stroke, atrial fibrillation and dementia.
To evaluate the strength of the evidence, the researchers used the institute’s Burden of Proof framework. The method accounts for differences among studies and emphasizes the most conservative estimate supported by the data. Each alcohol-related health outcome received a rating of zero to five stars based on the strength and consistency of the evidence.
The cancer findings were among the clearest. Alcohol consumption was associated with increased risk for cancers of the breast, colorectum, esophagus, larynx, lip and oral cavity, pharynx, liver, stomach, pancreas and prostate.
For several cancers, elevated risk appeared even below 10 grams of pure alcohol per day. For comparison, a standard drink in the United States contains about 14 grams of pure alcohol. That is roughly the amount in a 12-ounce beer with 5% alcohol, a 5-ounce glass of wine with 12% alcohol or 1.5 ounces of 80-proof liquor.
“For cancer, the evidence is consistent and unambiguous: risk rises with any level of alcohol intake,” said senior author Dr. Emmanuela Gakidou, a professor in the Department of Health Metrics Sciences at the Institute for Health Metrics and Evaluation.
The analysis also found that alcohol consumption was associated with higher risks of cirrhosis and other chronic liver diseases, pancreatitis and atrial fibrillation, an irregular heart rhythm. At higher levels of consumption, alcohol was associated with increased risk across all 20 health outcomes examined.
The findings were more complicated for several other conditions. For type 2 diabetes, Alzheimer’s disease and other dementias, ischemic heart disease and two types of stroke, the researchers found J- or U-shaped patterns. In other words, observational studies sometimes linked low-to-moderate drinking with modestly lower risk compared with not drinking, while higher consumption was associated with increased risk.
That does not mean people should begin drinking to protect their heart or brain.
Observational studies can be difficult to interpret because people who drink moderately may differ from people who do not drink in ways that affect health. Some nondrinkers may have stopped consuming alcohol because of illness. Other factors, including income, diet, access to health care and physical activity, can also influence results.
Earlier research has also found conflicting evidence about whether alcohol protects against ischemic heart disease. Studies based on self-reported alcohol consumption have sometimes identified an apparent benefit at low-to-moderate intake, while studies using genetic information to reduce certain types of bias have not found the same association.
“The science on alcohol and health is genuinely complex,” Gakidou said. “For some cardiometabolic and dementia outcomes, studies suggest small reduced risks at low-to-moderate consumption, but those associations became weaker and reversed at higher levels of drinking. Rather than interpreting these results as an endorsement of drinking, they lay out a complex map of where the evidence is strong, weak or mixed.”
The analysis does not identify one universal threshold below which alcohol is risk-free. It also does not mean that every drink carries the same level of risk for every person. Individual health history, medications, drinking patterns and the amount of alcohol in a serving all matter.
A heavy pour or high-alcohol beer may also contain more than one standard drink, even if it comes in a single glass or can.
The research was supported by the Bill & Melinda Gates Foundation. The study also received financial support from AARP and the Alzheimer’s Disease Data Initiative.