Is red wine good for your heart?

The evidence is weaker than it used to look. NASEM found lower cardiovascular mortality among moderate drinkers with only moderate certainty, and its review did not compare beverage types. Mendelian randomization studies, which use genetics to reduce confounding, found no protective effect from light or moderate drinking. NIAAA advises clinicians not to tell non-drinkers to start drinking for health reasons.

How many drinks per week is considered heavy drinking?

NIAAA defines heavy drinking as 15 or more drinks per week for men and 8 or more per week for women. It also counts 5 or more drinks on any single day for men, or 4 or more for women. Those single-day thresholds match its definition of binge drinking, which brings blood alcohol concentration to 0.08% or higher.

Does the type of alcohol change the cancer risk?

IARC classifies alcoholic beverages, ethanol, and acetaldehyde as Group 1 carcinogens without distinguishing between beer, wine, and spirits. The NASEM report also did not compare health outcomes across beverage types. What matters in the current evidence is the grams of ethanol consumed, not the drink it arrives in.

If I stop drinking, does my risk go back down?

Partly, and it takes time. IARC found sufficient evidence that stopping or reducing alcohol lowers the incidence of oral cavity and esophageal cancers, with more limited evidence for laryngeal, colorectal, and breast cancers. The National Cancer Institute notes that risk may take years to approach that of a never drinker, and that it is never too late to stop.

Why did the Dietary Guidelines remove the drink limits?

HHS and USDA released the 2025-2030 edition in January 2026 with a general instruction to consume less alcohol for better overall health, and without the previous numeric caps. Two federal scientific reviews published in 2025 had reached different emphases about light drinking. Several medical organizations have publicly objected to the removal of the numbers.

How Much Alcohol Is Actually Safe?

The short answer: No amount of alcohol is proven risk free. US agencies once capped "moderate" drinking at 2 drinks a day for men and 1 for women, but the 2025-2030 Dietary Guidelines dropped those numbers and now say only to drink less. Federal reviews published in 2025 reached different conclusions about light drinking, so the honest answer is that the safe threshold is contested. Everyone agrees risk climbs as intake climbs, and that for some people the safe amount is zero.

What counts as one standard drink?

In the United States, a standard drink contains 14 grams of pure ethanol, or about 0.6 fluid ounces. The National Institute on Alcohol Abuse and Alcoholism (NIAAA) lists the equivalents:

12 fluid ounces of regular beer at about 5% alcohol by volume
5 fluid ounces of table wine at about 12% alcohol
1.5 fluid ounces of 80-proof distilled spirits at about 40% alcohol
8 to 10 fluid ounces of malt liquor or hard seltzer at about 7% alcohol
3 to 4 fluid ounces of fortified wine such as port or sherry

Real pours are usually bigger than the chart. NIAAA notes that a 750 ml bottle of 12% table wine holds five standard drinks, and a 750 ml bottle of 80-proof spirits holds about 17. Craft beers average above 5% alcohol, and some reach 8% to 9%. NIAAA hosts a drink size calculator and a cocktail content calculator on its Rethinking Drinking site for this reason. Any weekly count is only as accurate as the pour behind it.

Do US guidelines still set a daily limit?

Not in numbers. The Dietary Guidelines for Americans, 2025-2030 is the current edition, published in January 2026. Its guidance on alcoholic beverages tells adults to consume less alcohol for better overall health. It no longer states the older daily caps of 2 drinks for men and 1 for women that appeared in the 2020-2025 edition.

Those older numbers still circulate widely, and NIAAA still uses its own pattern definitions:

Binge drinking: a pattern that brings blood alcohol concentration to 0.08% or higher. For a typical adult that is 5 or more drinks for men, or 4 or more for women, in about 2 hours.
Heavy drinking: 5 or more drinks on any day or 15 or more per week for men. For women it is 4 or more on any day or 8 or more per week.
High-intensity drinking: at least twice the binge threshold, so 10 or more drinks for men or 8 or more for women.

NIAAA writes that there is no guaranteed safe amount of alcohol for anyone, and that current research points to added risk even at low levels. The World Health Organization goes further: no form of alcohol consumption is risk free. WHO attributed 2.6 million deaths worldwide to alcohol in 2019, including 401,000 cancer deaths.

Why do federal reviews disagree?

Two US reviews fed the guidelines update in 2025, and they emphasized different things.

The National Academies of Sciences, Engineering, and Medicine (NASEM) ran systematic reviews on eight health outcomes. It concluded with moderate certainty that moderate drinking is associated with lower all-cause mortality and lower cardiovascular mortality than never drinking. It also concluded with moderate certainty that moderate drinking raises female breast cancer risk. No finding reached high certainty, and the committee offered no dietary advice.

The Alcohol Intake and Health Study, run through the federal ICCPUD committee, modeled lifetime risk cause by cause instead. It reported that the risk of dying from alcohol use begins at low levels of average use. In its model, men and women hit a 1 in 1000 lifetime risk of an alcohol-attributable death above about 7 drinks per week. The 1 in 100 threshold arrives above roughly 8.5 drinks per week for both sexes.

The gap is methodological. NASEM compared drinkers with never-drinkers in observational data. ICCPUD built a risk model and treated no level as protective. The same lens is useful for reading other habit research, as covered in the evidence behind longevity habits.

What does the cancer evidence show?

This is the least contested part of the topic.

The International Agency for Research on Cancer has classified alcoholic beverages, ethanol, and acetaldehyde as Group 1 carcinogens, meaning carcinogenic to humans, since 1987. IARC links alcohol to seven cancers: oral cavity, pharynx, larynx, esophagus, colorectum, liver, and female breast.

The January 2025 Surgeon General's Advisory on alcohol and cancer risk put figures on that. It estimated 96,730 alcohol-related cancer cases in the US in 2019, and about 20,000 alcohol-related cancer deaths a year. Alcohol ranks third among preventable causes of cancer, after tobacco and obesity. Only 45% of Americans recognize it as a cancer risk factor, against 89% for tobacco.

The advisory also published absolute lifetime risk. For women, the lifetime risk of an alcohol-related cancer was about 16.5% at under 1 drink per week, 19.0% at 1 drink a day, and 21.8% at 2 drinks a day. For men the same three figures were 10.0%, 11.4%, and 13.1%. NIAAA reports that even one drink per day is associated with a 5% to 15% higher breast cancer risk in women compared with not drinking.

Cutting back appears to help. IARC found sufficient evidence that stopping or reducing drinking lowers oral cavity and esophageal cancer incidence. The National Cancer Institute notes it may take years for risk to approach that of never drinkers. Routine screening by age is worth discussing with a clinician regardless.

What happened to the heart benefit?

The old J-curve showed light drinkers with less heart disease than abstainers. Two problems weakened that picture.

The first is abstainer bias, often called the sick-quitter effect. Many older studies put lifelong non-drinkers in the same group as people who quit because they were already ill. That makes abstainers look unhealthy and flatters drinkers by contrast. NASEM excluded studies with that flaw.

The second is genetics. Mendelian randomization studies use inherited variants that affect alcohol metabolism, which sidesteps lifestyle confounding. A 2024 systematic review of those studies in Frontiers in Epidemiology found genetically predicted alcohol intake causally associated with hypertension, atrial fibrillation, and myocardial infarction. It reported no evidence supporting protective effects of light to moderate drinking. Blood pressure is one of the clearer links, and hypertension.md covers that ground in more detail.

NASEM still found lower cardiovascular mortality with moderate certainty, so the methods have not been reconciled. NIAAA is unambiguous on one practical point: clinicians should not advise non-drinking patients to start drinking for their health.

When is the safe amount zero?

NIAAA lists situations where avoiding alcohol entirely is safest:

Before driving, operating machinery, or doing anything needing alertness and coordination
While taking prescription or over-the-counter medicines that interact with alcohol
With certain medical conditions, including liver disease
While recovering from alcohol use disorder, or when unable to control the amount
Under age 21
During pregnancy, or when pregnancy is possible

The CDC is direct about pregnancy: there is no known safe amount of alcohol use during pregnancy, and no safe time during pregnancy to drink. All types can be harmful, including red or white wine, beer, and liquor. CDC also notes it is never too late to stop.

People with a family history of alcohol use disorder should be mindful of their intake. NIAAA's Alcohol Treatment Navigator is a free tool for finding evidence-based care. If you track liver enzymes or other metabolic markers, alcohol intake belongs in that conversation with your clinician.

The bottom line

A US standard drink is 14 grams of pure alcohol, and most real pours hold more. The federal caps of 1 and 2 drinks a day are gone from the current Dietary Guidelines, replaced by an instruction to drink less. NASEM found lower mortality among moderate drinkers with moderate certainty, while the ICCPUD model found risk rising from low levels with no protective floor. The cancer evidence points one way: alcohol is a Group 1 carcinogen, and lifetime cancer risk rises measurably from under 1 drink per week to 2 drinks per day. For pregnancy, driving, and certain conditions and medicines, zero is the only figure the agencies give.

Last updated: September 2026. This article is for informational purposes only and does not constitute medical advice. Talk with a licensed clinician about your own drinking, your medications, and any liver, cardiac, or cancer risk factors you have.