Does Drinking Alcohol Shorten Your Life?

Drinking alcohol does shorten your life, on average, and the evidence has grown considerably stronger in recent years. A large 2023 meta-analysis of 107 studies found that once researchers corrected for common biases in how earlier studies were designed, even low-volume drinking showed no survival benefit compared to never drinking, while consuming roughly two or more standard drinks a day significantly raised the risk of death from all causes.¹ The picture is more complicated than “any sip kills you,” but the old notion that a glass of wine a day adds years to your life has not held up well under modern scrutiny.

Why the “Moderate Drinking Is Good for You” Idea Collapsed

For decades, the conventional wisdom in public health was that the relationship between alcohol and death followed a J-shaped curve: abstainers died a bit sooner than light drinkers, light drinkers lived the longest, and heavy drinkers died earliest. Hundreds of studies appeared to confirm this pattern. The problem was that most of those studies had a flaw baked into their design. When researchers grouped people as “non-drinkers,” that category included former drinkers who had quit because they were already sick. Naturally, this group had worse health outcomes, which made the moderate drinkers look healthier by comparison.

An umbrella review of systematic reviews published through early 2022 found that over 70% of the meta-analyses on alcohol and mortality did not exclude former drinkers from the reference group, leaving them vulnerable to this so-called “sick-quitter effect.”² When the 2023 JAMA Network Open meta-analysis explicitly adjusted for this bias and other study-quality issues, the apparent survival advantage for moderate drinkers shrank to a statistically nonsignificant finding. In the fully corrected model, people drinking about one to two standard drinks a day showed no meaningful reduction in death risk compared to people who had never drunk at all.¹

That does not mean every older study was wrong or dishonest. Some research that tracked drinking patterns over time, rather than just average volume, still found a modest mortality advantage for people who drank lightly and non-heavily about twice a week.³ But the weight of evidence has shifted: the protective story was almost certainly inflated, and the remaining signal, if it exists at all, is small and confined to very specific patterns of use.

What Genetics Tells Us About Cause and Effect

Observational studies can show associations but struggle to prove that alcohol itself is the cause of any outcome. People who drink moderately also tend to exercise more, earn higher incomes, and have better access to healthcare. Untangling alcohol’s effect from those lifestyle factors is difficult in traditional research. That is where a technique called Mendelian randomization has become valuable. It uses people’s inherited genetic variants that influence how much they tend to drink as a kind of natural experiment, sidestepping the lifestyle confounding that plagues observational work.

A Mendelian randomization study published in 2024 estimated that genetically predicted alcohol consumption shortened lifespan by roughly one year per unit increase in logged weekly drinks, with the effect persisting even after adjusting for smoking and education level. In men specifically, the estimated loss was closer to a year and a half.⁴ The study found no evidence that alcohol provides any longevity advantage for either men or women.

A separate Mendelian randomization analysis examined cause-specific deaths and found a straightforward linear relationship: for each additional roughly eight grams of daily alcohol (a bit less than one standard drink), the odds of death from all causes rose by about 27%, cardiovascular death by about 30%, cancer death by about 20%, and digestive disease death by roughly double.⁵ When these researchers tested for a J-shaped curve, they found none. The relationship was linear: more alcohol, more risk, with no protective dip at low levels. This directly contradicted the conventional observational data, which had shown the familiar J shape in the same population.⁵

The Heart Health Question

The strongest argument for a health benefit of alcohol has always been the heart. Moderate drinking raises HDL cholesterol (the “good” kind) and influences blood-clotting factors in ways that look cardioprotective on paper. A meta-analysis of experimental studies estimated that about 30 grams of alcohol a day (roughly two standard drinks) raised HDL cholesterol and shifted clotting markers in a direction associated with about a 25% reduction in coronary heart disease risk, at least based on those biomarker changes alone.⁶ Earlier reviews also noted that the protective association seemed specific to coronary artery disease, which gave it biological plausibility.⁷

This is real physiology, and researchers aren’t making it up. But the Mendelian randomization data cited earlier tells a different story once you look at the net effect on actual death rather than just favorable biomarkers. Alcohol’s boost to HDL exists, but it is apparently offset, or more than offset, by damage elsewhere in the cardiovascular system and the rest of the body. The genetic analyses found that cardiovascular mortality went up, not down, with higher genetically predicted drinking. One likely reason is that alcohol also raises blood pressure and increases the risk of abnormal heart rhythms, which erode the coronary benefit.

A dose-response meta-analysis of over ten million participants found a linear increase in the risk of atrial fibrillation with increasing alcohol consumption in men. Women showed a somewhat different pattern, with risk rising sharply once intake exceeded about 1.4 drinks per day.⁸ Data from the long-running Framingham Study found that consuming more than about three drinks a day was associated with a roughly 34% higher risk of atrial fibrillation, though moderate drinking below that threshold didn’t show a significant association.⁹ Atrial fibrillation itself substantially raises the risk of stroke and heart failure, so the math on net cardiac benefit gets complicated fast.

Alcohol and Cancer

Where the heart picture is nuanced, the cancer picture is blunt. When your body processes alcohol, the first breakdown product is acetaldehyde, a compound that is directly carcinogenic. It damages DNA and promotes tumor growth, particularly in tissues that are directly exposed to it: the mouth, throat, esophagus, and stomach.^{10 11}

Breast cancer has one of the most consistent dose-response relationships with alcohol. Moderate drinking has been linked to roughly a 30 to 50% increase in breast cancer risk in case-control and cohort studies.¹² The threshold is low: while fewer than three drinks per week does not appear to meaningfully increase risk, three to six drinks per week may already produce a small elevation.¹³ A meta-analysis found that at roughly two drinks a day, the relative risk of breast cancer compared with non-drinkers was in the range of 1.4 to 1.7, depending on study type.¹⁴

Cancer risk is one area where the genetics-based studies and the traditional observational data agree: more drinking means more cancer. The Mendelian randomization analysis described earlier found a 20% increase in cancer death risk per roughly eight additional grams of daily alcohol.⁵ There is no safe threshold in the data for cancer the way there might be for heart disease.

What Alcohol Does to Your Liver Over Time

The liver bears the heaviest metabolic burden from drinking, and alcohol-associated liver disease follows a predictable progression. It typically starts with fatty liver, advances to inflammation and scarring, and can end in cirrhosis or liver cancer. Fatty liver on its own is largely reversible if a person stops drinking, but cirrhosis is not, and it dramatically shortens life expectancy.^{15 16} Not everyone who drinks heavily develops cirrhosis; genetics, nutrition, obesity, and co-infection with hepatitis viruses all influence individual vulnerability. But the overall direction is clear and is one of the major pathways through which alcohol shortens lives.

Brain Health and Dementia

Prolonged heavy drinking causes structural and functional brain damage, and this is not limited to stereotypical “alcoholic” levels of intake. Heavy use has been consistently linked to cognitive decline, brain shrinkage, and an increased risk of all types of dementia.¹⁷ The mechanisms include direct neurotoxicity from alcohol and acetaldehyde, nutritional deficiencies (particularly thiamine), and vascular damage. Alcohol-related dementia can mimic Alzheimer’s disease clinically but sometimes partially reverses with sustained abstinence, which distinguishes it from most neurodegenerative dementias.¹⁸

Drinking Pattern Matters, Not Just Volume

Two people can drink the same total amount of alcohol per week and face very different risks depending on how that drinking is distributed. Concentrating a week’s worth of alcohol into one or two sessions is more dangerous than spreading it out. A 20-year follow-up study of older moderate drinkers found that those who also had periodic episodes of heavy drinking had more than double the odds of dying over the study period compared to those who drank the same total volume but never binged.¹⁹

Beyond chronic disease, alcohol is a major driver of acute, injury-related deaths. Falls, drownings, car crashes, violence, and suicide are all substantially more likely when someone is intoxicated.²⁰ This matters especially for younger people, for whom injury is a leading cause of death and where binge drinking is most common. Much of alcohol’s life-shortening effect in people under 40 comes not from cirrhosis or cancer but from car accidents and violence.

Why It Hits Some People Harder

Biological sex changes the equation substantially. Women reach higher blood alcohol concentrations than men even when doses are adjusted for body weight, partly because women have less total body water to dilute the alcohol and metabolize it differently in the stomach.²¹ Women also have less of a particular stomach enzyme responsible for breaking down alcohol before it reaches the bloodstream, especially at higher concentrations.²² This means each drink delivers a bigger physiological hit to women, which is one reason public health guidelines in many countries set lower limits for women.

Age makes a difference too. A study using data from the Health Professionals Follow-up Study and the Nurses’ Health Study found that the apparent coronary benefit of moderate drinking was smaller on an absolute basis in younger adults than in middle-aged and older adults, because younger people have so much less coronary disease to prevent in the first place.²³ At the same time, older adults are more vulnerable to falls, drug interactions, and the cognitive effects of alcohol.²⁴ So for young adults, the risk-benefit math tilts overwhelmingly toward risk (injuries, no meaningful cardiac upside), while for older adults, the risks and any residual benefits coexist in a more complex way but increasingly lean negative.

The Alcohol-Harm Paradox and Socioeconomic Inequality

One of the more striking findings in alcohol epidemiology is that people from lower socioeconomic backgrounds suffer more alcohol-related illness and death than wealthier people, even when they drink the same amount or less. This pattern is called the alcohol-harm paradox.²⁵ A systematic review found that people with low socioeconomic status had 1.6 to 3.6 times the risk of all-cause mortality and 3.8 to 5.2 times the risk of alcohol-attributable mortality compared with higher-status individuals at comparable drinking levels.²⁶

The reasons are not fully understood but likely involve clustering of risk factors: less nutritious diets, higher smoking rates, less access to healthcare, more hazardous occupational exposures, and different drinking patterns (more binge drinking, less consistent food intake while drinking). A large U.S. cohort study found that the apparent protective association between light drinking and heart disease death was significantly stronger for high-income individuals than for low-income ones, suggesting that whatever benefit exists is partly a proxy for the lifestyle advantages that come with wealth.²⁷

Alcohol Plus Smoking Is Worse Than Either Alone

If you both smoke and drink, the risks do not merely add up; they multiply. Acetaldehyde from alcohol and carcinogens from tobacco hit many of the same tissues, particularly the mouth, throat, and voice box. A systematic review and meta-analysis found that heavy drinkers who also smoked heavily had roughly 35 times the risk of head and neck cancer compared to people who did neither.²⁸ The combined effect on oral cancer risk specifically was about 36-fold, and on laryngeal cancer close to 39-fold. A separate meta-analysis estimated that the combined effect of alcohol and smoking on head and neck cancer death was nearly four times greater than what you would expect from simply adding the two individual risks together.²⁹ For people who drink and smoke, this synergy is arguably more dangerous than either habit in isolation.

Alcohol and Biological Aging

Beyond specific diseases, there is evidence that alcohol accelerates biological aging at the cellular level. Telomeres, the protective caps on the ends of chromosomes, shorten naturally as you age, and shorter telomeres are associated with earlier death and age-related disease. A Mendelian randomization study found that genetically predicted higher alcohol consumption was associated with shorter telomere length, suggesting a causal relationship rather than just a correlation. The proposed mechanisms include oxidative stress from alcohol metabolism, which generates reactive oxygen species that damage telomeres, and a reduction in protective antioxidants like glutathione.³⁰ This telomere-shortening effect represents a pathway through which alcohol could shorten lifespan that is independent of any single disease diagnosis.

How Much Countries Think Is Safe Varies Wildly

If you look to government guidelines for clarity, you might be surprised by how inconsistent they are worldwide. A review of low-risk drinking guidelines found only 58 out of 194 countries had issued them at all, and those were mostly high-income nations. What counts as a “standard drink” ranged from 8 to 20 grams of pure alcohol depending on the country. Weekly low-risk limits for women ranged from 15 to 170 grams, and for men from 15 to 294 grams.³¹ That is an enormous spread. At one end, some countries effectively advise no more than one or two drinks a week; at the other, some consider over two drinks a day reasonable for men. These guidelines often lag behind the latest research, and many were established when the J-curve was still considered reliable.

Why We Can Metabolize Alcohol at All

Given how damaging alcohol is, it seems odd that our bodies are equipped to handle it. The answer goes back millions of years. Genomic evidence suggests that our hominid ancestors have been exposed to dietary ethanol from fermenting fruit for at least 10 to 12 million years.³² A key mutation in the gene for the main alcohol-metabolizing enzyme, shared with chimpanzees and gorillas, appears to date to a period when tropical forests were contracting and ancestral apes began foraging fallen fruit on the forest floor, fruit that was often fermenting and ethanol-rich. Being able to efficiently metabolize this ethanol gave those apes access to a calorie source that competitors could not safely exploit.³³

The concentrations of ethanol in rotting fruit on a forest floor, though, are nothing like those in a bottle of wine or a cocktail. Our enzyme systems evolved to handle small, incidental exposures, not the concentrated and frequent doses that modern drinking habits deliver. In a sense, our ability to metabolize alcohol is not an endorsement that it is good for us; it is a relic of an ecological strategy for eating spoiled fruit. The mismatch between that evolutionary legacy and the modern availability of distilled and fermented beverages goes a long way toward explaining why alcohol causes so much harm despite our apparent biological tolerance for it.