Alcohol and Health: What the Evidence Says
What the Dietary Guidelines, WHO, NCI and large studies say about alcohol, cancer, heart health, pregnancy and medicines, and where the evidence disagrees.
Content last updated . This page has not been medically reviewed.
Alcohol is a drink, a social habit, and a substance that affects the body in many ways. Headlines have swung between "a glass of red wine is good for you" and "no amount is safe." This guide sets out what official bodies and large studies say, where they agree, where they disagree, and why. It is information, not advice to start, stop or change drinking. If you have questions about your own drinking, or you cannot cut down when you want to, talk with a clinician; support exists. For an alcohol-free alternative see alcohol-free beer and wine.
Read this first
- Alcohol is a proven cause of cancer. The International Agency for Research on Cancer (IARC) has classed alcohol as a Group 1 carcinogen since 1987, and the National Cancer Institute says risk rises with the amount drunk. [5] The WHO calls alcohol an established carcinogen and says there is no form of alcohol consumption that is risk-free. [6]
- The US Dietary Guidelines do not recommend starting to drink for any reason. For adults who do drink, the limit is 2 drinks or less in a day for men and 1 drink or less in a day for women, and "drinking less is better for health than drinking more." [1]
- Pregnancy: do not drink. The Dietary Guidelines say it is not safe to drink any type or amount of alcohol during pregnancy, and the NHS advises avoiding it. [1] [8]
- The "moderate drinking is protective" idea is contested. Some analyses find lower death rates among moderate drinkers; others find that this vanishes when study biases are handled. The sources below disagree, and none of them is a randomized trial. [2] [3] [4] [9]
- Medicines: alcohol can interact with many medicines, from sleep aids to blood thinners and pain relievers. Ask a pharmacist. [10]
What counts as "a drink"
The Dietary Guidelines define one US drink equivalent as 14 grams (0.6 fluid ounces) of pure alcohol. That is 12 fluid ounces of regular beer (5 percent alcohol), 5 fluid ounces of wine (12 percent), or 1.5 fluid ounces of 80-proof spirits (40 percent). The limit is a daily cap, "not intended as an average over several days." [1] Binge drinking is defined there as 5 or more drinks for the typical adult male or 4 or more for the typical adult female in about 2 hours, and the guidelines say it should be avoided. [1]
Other countries use other measures. The NHS (UK) advises men and women not to drink more than 14 units a week on a regular basis, spread over 3 or more days. A 175 ml glass of 12 percent wine is 2.1 units and a pint of 5.2 percent beer is 3 units. [7] A UK unit and a US drink are not the same size, so numbers from different countries cannot be compared directly.
Alcohol also carries calories. In US data cited by the Dietary Guidelines, alcoholic beverages supply about 5 percent of adults' calories, or about 9 percent among those who drink. [1]
What official bodies say
| Body (date) | Statement |
|---|---|
| Dietary Guidelines for Americans 2020-2025 | Do not start drinking for any reason; if drinking, 2 drinks or less in a day for men, 1 or less for women; some people should not drink at all: pregnant or possibly pregnant, under the legal age, those with certain medical conditions or taking medicines that interact, those recovering from alcohol use disorder or unable to control intake. Alcohol raises cancer risk, and for some cancers the risk rises even below 1 drink a day; "Caution, therefore, is recommended." [1] |
| WHO (fact sheet, June 2024) | About 2.6 million deaths in 2019 were attributed to alcohol. Alcohol was linked to 4.4 percent of cancers diagnosed worldwide in 2019. "There is no form of alcohol consumption that is risk-free." [6] |
| National Cancer Institute (reviewed May 2025) | Strongest links: cancers of the mouth, throat, voice box, esophagus, liver, breast and colon and rectum. Risk rises with amount; even light drinking may raise risk of some cancers. NCI also cites a recent US Surgeon General advisory that calls for reconsidering the alcohol limits. [5] |
| NHS (UK, reviewed August 2024) | Not more than 14 units a week regularly, spread over 3 or more days, with several drink-free days if cutting down. [7] |
We tried to read the Surgeon General advisory itself but the page could not be opened, so we rely on NCI's description of it. [5]
What research says
| Question | What the sources found | Strength of evidence | What was studied |
|---|---|---|---|
| Cancer risk | Strong, dose-related links for several cancers. For women, NCI cites about 17 alcohol-related cancers per 100 women drinking less than one drink a week, about 19 at one drink a day and about 22 at two; for men about 10, 11 and 13. Relative risks cited include 1.04 (light), 1.23 (moderate) and 1.6 (heavy) for breast cancer. [5] | Established (IARC Group 1) | Many cohort and case-control studies |
| Light drinking and breast cancer | A meta-analysis found that, after correcting for drinker-misclassification bias, low-level drinkers (under 21 g a day) had an odds ratio of 1.085 (95% CI 1.015 to 1.160). [11] | Promising but limited | Pooled observational studies |
| Death from any cause | A 2023 meta-analysis of 107 cohort studies (about 4.8 million people, 425,564 deaths) found no significant lower mortality for occasional or low-volume drinkers compared with lifetime non-drinkers; risk was significantly higher at 45 g a day and above, and female drinkers overall had significantly higher mortality than female lifetime non-drinkers (RR 1.22). [2] The National Academies' 2024 review found, with moderate certainty, an association between moderate drinking and lower all-cause mortality compared with never drinking, but noted limits such as abstainer bias (comparison groups that mix lifelong non-drinkers with former drinkers) and said no finding reached high certainty. [4] | Mixed (the two reviews disagree) | Observational cohorts only |
| A "safe" weekly amount | In a pooled analysis of 599,912 current drinkers from 83 studies (public funders), the lowest risk of death from all causes was at or below about 100 g of alcohol a week (about 7 US drinks, our conversion). Stroke, heart failure and some other cardiovascular outcomes rose with intake; heart attack was the exception, with lower risk at higher intake. The authors concluded that recommended limits should be lower than in most guidelines. [3] | Mixed | Individual-participant data, 19 high-income countries |
| Heart disease, genes as a natural experiment | A study of 261,991 people used a gene variant that makes people drink about 17 percent less. Carriers had lower systolic blood pressure (0.88 mm Hg), and 10 percent lower odds of coronary heart disease (OR 0.90, 95% CI 0.84 to 0.96). [9] | Promising but limited | Mendelian randomization, people of European descent |
| Liver disease, injuries, alcohol use disorder | The Dietary Guidelines say excess alcohol raises the risk of liver disease, cardiovascular disease, injuries and alcohol use disorders. [1] | Established at higher amounts | Guideline review |
What the National Academies review did not do is also important: it gave no dietary advice, did not compare types of alcohol, and for several outcomes, including colorectal, oral, pharyngeal, esophageal and laryngeal cancers, reached no conclusion. It was requested by Congress for the next Dietary Guidelines and sponsored by the USDA Food and Nutrition Service. [4]
Why the studies disagree. The Dietary Guidelines state that higher average drinking goes with a greater risk of death from all causes, and that "emerging evidence suggests that even drinking within the recommended limits may increase the overall risk of death." [1] Observational studies can be tripped up when "non-drinkers" include people who quit because they were already ill; the National Academies list this as abstainer bias, and the 2023 meta-analysis attempts to adjust for it. [2] [4] Gene-based studies [9] are one way to reduce that problem.
Red wine, other types. We did not read a source that compared red wine or other drink types and found a benefit; the National Academies say their review did not compare types. [4]
If you drink: practical points
- Compare what you pour with the standard sizes above (12 ounces of beer, 5 ounces of wine, 1.5 ounces of spirits). [1]
- Keep to daily caps, not weekly averages, if you follow US guidance. [1]
- Count alcohol calories. [1]
- Do not mix with medicines without asking a pharmacist. [10]
- If you do not drink, the Dietary Guidelines say not to start for any reason. [1]
Related pages
Frequently asked questions
Is any amount of alcohol safe?
It depends on whose wording you use. The WHO says there is no form of alcohol consumption that is risk-free. [6] The Dietary Guidelines set limits for adults who choose to drink, but say drinking less is better than drinking more and that some people should not drink at all. [1]
Is a glass of red wine good for the heart?
Some large observational analyses find lower heart-attack risk in moderate drinkers, but the same pooled analysis found higher risks for stroke and heart failure, [3] and a gene-based study suggested that people genetically prone to drink less had lower coronary heart disease risk. [9] We read no randomized trial showing that drinking prevents heart disease.
Does alcohol cause cancer?
Yes, for several types. IARC classed alcohol as a Group 1 carcinogen in 1987, and the National Cancer Institute lists cancers of the mouth, throat, voice box, esophagus, liver, breast and colon and rectum as most strongly linked, with risk rising with the amount. [5]
Can I drink while pregnant or breastfeeding?
The Dietary Guidelines say it is not safe to drink any type or amount of alcohol in pregnancy, and the NHS advises avoiding it. [1] [8] For breastfeeding, not drinking is the safest option; the Dietary Guidelines say up to 1 drink in a day is generally not known to be harmful, especially with at least a 2-hour wait before nursing. [1]
Why do studies give different answers about moderate drinking?
Because they are observational and use different comparison groups. The National Academies point to abstainer bias, underreported intake and other limits; the 2023 meta-analysis that adjusted for such biases found no significant benefit for low-volume drinkers. [2] [4]
More details: evidence, limits and cautions
The details behind this page: cautions, limits of the evidence and where sources disagree. Open any note to read it.
Drinking and pregnancy, breastfeeding, children
- Pregnancy: the Dietary Guidelines say it is not safe to drink any type or amount of alcohol during pregnancy; women who become pregnant should stop immediately; those trying to become pregnant should not drink, because alcohol can harm the baby at any time in pregnancy, even before a woman knows she is pregnant. [1] The NHS says drinking in pregnancy can cause long-term harm to the baby. [8] The WHO links drinking in pregnancy to fetal alcohol spectrum disorders and a higher risk of miscarriage, stillbirth and premature delivery. [6]
- Breastfeeding: the Dietary Guidelines say not drinking is the safest option; moderate intake (up to 1 standard drink in a day) is generally not known to be harmful, especially if she waits at least 2 hours after a single drink before nursing or expressing milk. [1]
- Under the legal age: the Dietary Guidelines list people under the legal drinking age among those who should not drink. [1]
Medicines, health conditions and other cautions
- Medicines. NIAAA's guide says alcohol can cause nausea, drowsiness, fainting and loss of coordination with certain medicines, and can raise the risk of internal bleeding, heart problems and breathing difficulty. Examples it lists: sedatives and sleep aids (drowsiness, slowed breathing, overdose risk), opioid painkillers, warfarin (bleeding), pain relievers such as ibuprofen (stomach bleeding) and acetaminophen (liver damage), certain antidepressants (MAOIs), and some diabetes medicines (low blood sugar). Older adults are at particular risk. The list is not complete. [10] See Supplements, sedatives, sleep medicines and alcohol.
- Cancer and smoking. NCI says people who both smoke and drink have much greater risks of cancers of the mouth, throat, voice box and esophagus than people who do only one, and that people with cancer should ask their health care team about alcohol. [5]
- Older adults. The Dietary Guidelines say older adults may feel alcohol's effects faster, with higher risk of falls and injuries. [1]
- Driving and machinery: the Dietary Guidelines say people who plan to drive or do skilled activities should avoid alcohol. [1]
- Alcohol use disorder: if you cannot control how much you drink, see a clinician. Read more in Alcohol use disorder research.
- Allergy: we did not read a source on allergens in alcoholic drinks, so check labels and ask a clinician. For trouble breathing or swelling of the face or throat after any food or drink, call 911 or your local emergency number.
References
- U.S. Department of Agriculture and U.S. Department of Health and Human Services. Dietary Guidelines for Americans, 2020-2025. Chapters 1, 3, 5. Read 2026-10-24.
- Zhao J et al. Association between daily alcohol intake and risk of all-cause mortality: a systematic review and meta-analyses. JAMA Netw Open 2023 (abstract read via Europe PMC). Read 2026-10-24.
- Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599,912 current drinkers in 83 prospective studies. Lancet 2018 (abstract read via Europe PMC; public funders listed). Read 2026-10-24.
- National Academies of Sciences, Engineering, and Medicine. New report reviews evidence on moderate alcohol consumption and health impacts. News release updated 17 December 2024 (we read the release, not the full report). Read 2026-10-24.
- National Cancer Institute. Alcohol and cancer risk. Reviewed 2 May 2025. Read 2026-10-24.
- WHO. Alcohol fact sheet. 28 June 2024. Read 2026-10-24.
- NHS. Calculating alcohol units. Last reviewed 27 August 2024. Read 2026-10-24.
- NHS. Foods to avoid in pregnancy. Last reviewed 15 June 2026. Read 2026-10-24.
- Association between alcohol and cardiovascular disease: Mendelian randomisation analysis based on individual participant data. BMJ 2014 (abstract read via Europe PMC). Read 2026-10-24.
- National Institute on Alcohol Abuse and Alcoholism. Harmful interactions: mixing alcohol with medicines. Revised 2014. Read 2026-10-24.
- Methodological biases in estimating the relationship between alcohol consumption and breast cancer: the role of drinker misclassification errors in meta-analytic results. 2014 (abstract read via Europe PMC). Read 2026-10-24.