Like many of my friends, I enjoy a glass of wine with dinner or a beer with friends. But a recent U.S. surgeon general report warning that any alcohol consumption is unsafe made me pause. Should I embrace “Dry January,” or even quit altogether?
Recently, at least eight respected medical societies and government agencies have issued reports on alcohol and health. All agree on one point: moderate to heavy drinking causes serious health and social harms. Where they diverge — sometimes sharply — is in guidance for people deciding between abstinence and occasional drinking.
As a public health scientist and occasional light drinker, I decided to review the evidence for myself.
In 2023, Health Canada advised limiting alcohol to no more than two drinks per week, citing “overwhelming evidence” of harm. The same evidence was used by the World Heart Federation to recommend abstinence. Yet the Canadian report relied largely on committee judgments rather than an independent systematic review, and amid the volley of support, criticisms, and reactions, the Canadian government still has not formally adopted the guidelines.
Cancer risks
A 2025 progress report from the American Association for Cancer Research has been cited as evidence for reducing or eliminating drinking based on links to cancers of the breast, liver, and colon. However, the 270-page report devoted only three paragraphs to alcohol, largely focused on heavy consumption.
A widely cited meta-analysis of 222 studies involving about 92,000 light drinkers and 60,000 nondrinkers found that most individual studies showed no increased cancer risk from light drinking; a small increase emerged only after pooling results. Many studies relied on case-control designs vulnerable to recall bias. By contrast, a meta-analysis of 60 prospective studies found no association between low-level drinking and overall cancer risk, aside from a weak link to breast cancer.
A 2025 report by the National Academies of Sciences, Engineering, and Medicine reached similar conclusions. Reviewing studies published since 2010, the panel found small associations between moderate drinking and breast cancer, weaker evidence for colorectal cancer, and no consistent links to cancers of the mouth or throat. Critics noted that two scientists on the initial committee list had prior industry ties. But the National Academies’ vetting process flagged and removed them before deliberations began. The two in question did not participate in committee discussions, and they did not vote.
Ironically, this episode increases the report’s credibility by demonstrating the academy’s intolerance for conflicts of interest.
By contrast, the surgeon general’s report relies largely on secondary sources and assumes that cancer risk rises with every sip, without identifying a threshold at which harm meaningfully begins.
Heart disease: reversal and reconsideration
For decades, observational studies suggested that moderate alcohol consumption — particularly wine — might reduce heart disease risk. That view has recently been challenged, though the evidence remains nuanced.
Much of the shift stems from analyses of the U.K. Biobank study. Heavier drinkers had higher mortality than lighter drinkers, but the study excluded lifelong abstainers — many of whom stop drinking because of illness or medication use. Notably, the same study showed that a glass of wine with dinner was associated with lower mortality.
In July 2025, an American Heart Association expert committee issued a comprehensive statement suggesting that low-to-moderate drinking — up to one drink daily for women and two for men — is not associated with major cardiovascular harm and may offer modest protection. The National Academies review similarly concluded that moderate drinkers had lower rates of fatal and nonfatal heart attacks and strokes than abstainers. The European Heart Association, in contrast, reviewed the same evidence and concluded that alcohol provides no cardiovascular benefit.
Absolute risk matters
Public health discussions often emphasize relative risks, which can exaggerate harm. If 2 out of 1,000 nondrinkers in comparison to 3 out of 1,000 light drinkers develop a disease, the relative risk is 1.5 — a 50% increase. Yet this statistic can be misleading. The absolute difference is just 1 additional case per 1,000 people. In other words, the risk to an individual increases by only about 0.1%.
In the U.K. Biobank study, compared with occasional drinkers, moderate drinkers increased their absolute risk of death by about 1.8% over a dozen years, while heavy drinkers faced increases closer to 6%. In another study, women drinking fewer than seven drinks per week had a 0.56% chance of developing breast cancer over 10 years, compared with 0.64% among women drinking seven or more drinks weekly. The increase in absolute risk was 0.08%.
So what should I do?
Framing alcohol use as simply “safe” or “dangerous” oversimplifies a complex evidence base. Moderate drinking may slightly increase breast cancer risk but not most other cancers, while light drinking may modestly reduce cardiovascular risk.
Reasonable people can interpret these findings differently. Several considerations shaped my own thinking.
First, many studies focus on a single cause of death — heart disease or cancer — rather than overall survival. My personal goal is to live longer; I am indifferent to which diagnosis ultimately appears on my death certificate. On this point, the National Academies concluded that moderate drinkers are slightly less likely to die from any cause than abstainers, a finding that carries weight for me.
Second, when deciding between abstinence and occasional light drinking, the most relevant comparison for me is between people who drink lightly and those who never drink. Most studies report higher mortality among abstainers, but this finding comes with an important caveat: Some people avoid alcohol because of underlying illness or medication use.
To address this bias, some recent studies exclude lifelong abstainers altogether. The unintended consequence is that these analyses cannot answer the question most relevant to me — whether healthy occasional light drinkers reduce their risk of death by becoming abstainers.
When examined in absolute rather than relative terms, my increased risk of premature death associated with responsible light drinking appears to be very small—probably well under one percent.
Finally, nearly all of the reports focus on the risks of cancer, heart disease, and liver disease. What is largely missing is the simple reality that many occasional drinkers — including me — enjoy the taste of a good wine and feel that it enhances both food and conversation. The same surgeon general who warned about alcohol has also highlighted alienation and social isolation as major public health threats. For me, responsible, light drinking may slightly increase some health risks, but it can also promote social connection and shared experience.
Ultimately, the choice to drink or abstain reflects personal values. If minimizing breast cancer risk is paramount, reducing alcohol makes sense. If social enjoyment and the possibility of a slightly longer life matter more, an occasional glass of wine with dinner may fit comfortably into a balanced lifestyle. And for someone who is concerned about their relationship with alcohol, Dry January might offer a valuable opportunity to reset and reflect.
As a public health scientist, I have learned that firm prescriptions rarely capture life’s complexity. Recognizing that I may be accepting a small risk, I am sticking with an occasional glass of wine with dinner during Dry January.
Robert M. Kaplan is a senior scholar at the Stanford School of Medicine’s Clinical Excellence Research Center. He is a former associate director of the National Institutes of Health and former chief science officer of the U.S. Agency for Healthcare Research and Quality.
