Is Moderate Drinking Actually Healthy? What the Latest Research Says

The short answer: no, moderate drinking isn't good for you. The idea that a daily glass of wine protects your heart came from research with serious design flaws, and newer studies that correct for those flaws find no meaningful health benefit from low-level drinking. In January 2023, the World Health Organization went further and stated plainly that "no level of alcohol consumption is safe for our health."

That doesn't mean one drink will hurt you in any dramatic way. It means the older "moderate drinking is healthy" story has quietly collapsed, and public health guidance has shifted with it. If you drink because you enjoy it, that's a personal choice. But drinking for your health is no longer supported by the evidence.

Where the "moderate drinking is healthy" idea came from

For decades, observational studies kept finding the same pattern: people who had one or two drinks a day seemed to live longer and have fewer heart attacks than people who didn't drink at all. This produced the famous J-shaped curve, where light drinkers appeared healthier than both abstainers and heavy drinkers.

That finding drove headlines about red wine, resveratrol, and the "French paradox." It also shaped official guidance for years.

The problem is how the "abstainer" group was defined. Many of those studies lumped lifelong non-drinkers in with people who had quit drinking, often because they were already sick. Former heavy drinkers with damaged livers, recovering alcoholics, and people told by doctors to stop for medical reasons all got counted as "non-drinkers." Compared to that group, of course light drinkers looked healthier. They weren't sick enough to have been told to stop.

Researchers call this the "sick quitter" bias. Once you correct for it, the supposed benefit shrinks or disappears.

What newer studies changed

A 2023 systematic review and meta-analysis published in JAMA Network Open by Zhao and colleagues re-examined 107 studies covering nearly 5 million people. When they adjusted for the sick-quitter problem and other biases, they found no significant reduction in all-cause mortality for low-volume drinkers compared with lifetime non-drinkers. Women drinking 25 grams or more of alcohol per day (roughly two standard US drinks) had a significantly higher risk of dying from any cause.

Zhao J, Stockwell T, Naimi T, Churchill S, Clay J, Sherk A (2023). Association between daily alcohol intake and risk of all-cause mortality. JAMA Network Open.

DOI 10.1001/jamanetworkopen.2023.6185 ↗

The Global Burden of Disease Study, published in The Lancet in 2018 by Griswold and colleagues, analyzed data from 195 countries and concluded that the level of alcohol consumption that minimizes health loss is zero. That study looked at 23 alcohol-related health outcomes together, not just cardiovascular risk in isolation.

The WHO followed in January 2023 with a statement in The Lancet Public Health saying no safe amount has been established. Canada's updated guidance, released the same month by the Canadian Centre on Substance Use and Addiction, moved from the previous recommendation of up to 15 drinks per week to a warning that risk begins climbing after just two drinks per week.

The US hasn't gone that far yet. Current Dietary Guidelines for Americans still define moderate drinking as up to one drink per day for women and up to two for men, but the guidelines explicitly state that people who don't drink shouldn't start for any reason, including health.

How much is "moderate," and why the definition matters

A standard drink in the US contains 14 grams of pure alcohol. That's:

• 12 ounces of regular beer (about 5% ABV)

• 5 ounces of wine (about 12% ABV)

• 1.5 ounces of distilled spirits (about 40% ABV)

The catch: most pours aren't standard. A generous glass of wine at dinner is often closer to 8 or 9 ounces, which is nearly two standard drinks. A craft IPA at 7.5% ABV in a 16-ounce pint is roughly two drinks, not one. Home pours of spirits tend to run heavier than a jigger.

This matters because "I only have a glass of wine with dinner" can easily be two or three standard drinks in practice, which pushes someone from the low-risk category into a range where the health signal gets louder.

Risks that show up even at low levels

Some harms don't need heavy drinking to appear.

Cancer. The International Agency for Research on Cancer classifies alcohol as a Group 1 carcinogen, the same category as tobacco and asbestos. Alcohol is causally linked to cancers of the mouth, throat, larynx, esophagus, liver, colon, rectum, and female breast. For breast cancer specifically, a 2015 analysis in the BMJ by Cao and colleagues found that even light drinking (up to one drink per day) was associated with a modestly increased risk in women. The US National Cancer Institute estimates that about 5.5% of new cancers globally are attributable to alcohol.

Blood pressure. A 2023 meta-analysis published in Hypertension by Di Federico and colleagues, using data from seven studies and nearly 20,000 adults, found a linear dose-response relationship between alcohol intake and systolic blood pressure, with no threshold below which the effect disappeared. Even one drink a day was associated with measurable increases over time.

Sleep. Alcohol is sedating, so people often assume it helps sleep. It doesn't. It suppresses REM sleep in the first half of the night and causes fragmented, lighter sleep in the second half as it metabolizes. A 2018 study in JMIR Mental Health by Pietilä and colleagues, using wearable data from over 4,000 people, found that even low amounts of alcohol reduced sleep quality scores, with the effect worsening as intake rose.

Atrial fibrillation. A 2020 study in the New England Journal of Medicine by Voskoboinik and colleagues showed that abstinence in regular drinkers with atrial fibrillation significantly reduced arrhythmia recurrence. Related work has shown even moderate drinking can trigger AFib episodes in susceptible people.

Who faces higher risk at any dose

The average-risk picture hides a lot. Certain groups face steeper risk from any amount of alcohol:

• Women. Body composition and enzyme differences mean women reach higher blood alcohol concentrations from the same dose, and breast cancer risk rises with even light drinking.

• People over 65. Slower metabolism, more medications in the mix, and higher fall risk change the calculation.

• Anyone with a family history of alcohol use disorder. Genetic risk is real, and lighter drinking can escalate.

• People on medications that interact with alcohol. This includes many antidepressants, benzodiazepines, opioids, some blood thinners, and acetaminophen at higher doses.

• People with liver disease, pancreatitis, certain cancers, or a history of arrhythmia.

• People who are pregnant or trying to conceive. No amount has been established as safe during pregnancy.

How to think about your own drinking level

The question isn't really "is one drink dangerous?" It's "what am I getting from this, and what is it costing me?"

A few honest checks:

1. Count standard drinks, not glasses. Measure a pour once. Most people underestimate.

2. Track a typical week. Not your best week, not your worst. A normal one.

3. Notice what alcohol is doing for you. Sleep aid? Social lubricant? Way to shut off after work? Those are the jobs it's doing, and there are usually better tools for each.

4. Watch for creep. Weekly totals tend to rise slowly over years. What was two drinks with dinner at 30 can quietly become four by 45.

If you drink daily or heavily and want to cut back, don't just stop cold. Alcohol withdrawal in dependent drinkers can cause seizures and, in severe cases, delirium tremens, which can be fatal. Talk to a doctor first.

Medication-assisted options exist for people who want structured help reducing or quitting. Naltrexone, for instance, is an FDA-approved medication that reduces alcohol cravings by blocking the reward response. It requires a prescription and evaluation by a licensed clinician; it isn't something you can decide to take on your own.

FAQ

Frequently asked questions

Is a glass of red wine a day still considered good for your heart?+

No. The cardiovascular benefit shown in older studies has largely disappeared once researchers corrected for methodological problems like the sick-quitter effect. The American Heart Association does not recommend drinking wine or any alcohol to prevent heart disease.

What about the Mediterranean diet, which includes wine?+

The health benefits of the Mediterranean diet come primarily from the food pattern (vegetables, fish, olive oil, legumes, whole grains), not the wine. Studies that separate the wine from the rest of the diet don't find that wine is doing the heavy lifting.

Is beer or wine safer than liquor?+

The type of alcohol matters less than the total grams of ethanol consumed. A standard drink of beer, wine, or spirits delivers roughly the same amount of alcohol and carries similar health risks at the same dose.

If I only drink on weekends, am I in the clear?+

Not necessarily. Concentrating a week's worth of drinking into two nights (binge drinking, defined by the NIAAA as 4+ drinks for women or 5+ for men in about two hours) carries its own risks, including higher accident risk, cardiac events, and long-term liver strain.

How long after quitting do health markers improve?+

Blood pressure and sleep quality can improve within weeks. Liver enzymes often normalize within a few months in people without advanced liver disease. Cancer risk declines more slowly over years. A 2018 study in BMJ Open by Mehta and colleagues found measurable improvements in insulin resistance and liver fat after just one month of abstinence in moderate drinkers.