Alcohol and health: is there a safe amount?

"A glass of red wine is good for your heart."

"Everything in moderation."

"One beer never hurt anyone."

For years alcohol occupied a very particular place in the conversation about health. We knew that drinking too much was harmful, yet small amounts, red wine especially, were sometimes presented almost as a form of cardiovascular prevention.

Today the picture looks considerably less favourable.

Current data cannot identify an amount of alcohol at which we can say the health risk is zero. The link between alcohol and cancer is particularly well documented.

That does not mean the risk is the same after one glass of wine a month and after several drinks every day.

The key relationship is this:

the more alcohol and the more often it is consumed, the higher the risk.

What does a "safe amount" actually mean?

The phrase can be misleading.

To call a given amount of alcohol safe, we would need evidence that below a specific level, drinking it does not increase the risk of any health harm.

We cannot currently point to such a threshold.

The WHO stresses that the available data do not allow us to establish a dose of alcohol entirely free of health risk, and that for cancer no threshold has been demonstrated below which the carcinogenic effect stops.

The latest, fifth edition of the European Code Against Cancer goes further still. In its 2025 cancer-prevention recommendations the message is very simple:

avoid alcoholic beverages.

It is worth distinguishing:

"safe", meaning no increase in risk

from

"less risky", meaning a lower probability of harm than a larger amount of alcohol.

Low intake may mean a low absolute risk for a given person. That is still not the same as zero risk.

Why does alcohol increase cancer risk?

The substance responsible for alcohol's effects is ethanol.

The International Agency for Research on Cancer classifies alcoholic beverages as a group 1 carcinogen, meaning there is sufficient evidence of a carcinogenic effect in humans.

This does not mean alcohol is "equally dangerous" as every other substance in that group. The classification describes the certainty of the evidence for a carcinogenic effect, not the size of the risk at any given exposure.

Alcohol has a confirmed causal link with at least seven cancer sites:

• oral cavity,
• pharynx,
• larynx,
• oesophagus,
• liver,
• colorectum,
• breast in women.

One of the mechanisms is the formation of acetaldehyde during alcohol metabolism. This compound can damage DNA and interfere with the normal repair of genetic material.

Oxidative stress and alcohol's effect on hormonal balance also play a role.

Does a single glass increase risk too?

The risk associated with small amounts is far lower than with heavy drinking.

But it is not zero.

A 2023 meta-analysis covering more than 100 cohort studies confirmed a dose-response relationship between alcohol consumption and cancer risk. At low intake there was no significant increase in the risk of all cancers analysed together, but an increased risk was shown for breast, colorectal and oesophageal cancer, among others.

This is precisely why the message:

"one glass of wine is safe"

is too categorical.

It is better to say:

one glass means far lower exposure than several drinks a day, but we cannot call it entirely free of risk.

According to data cited in the latest edition of the European Code Against Cancer, even light and moderate alcohol consumption accounted for roughly 23,000 new cancer cases in the European Union in 2017. Around 11,000 of these were breast cancer.

What about the famous glass of wine "for the heart"?

This is one of the most persistent myths about alcohol.

For many years observational studies showed a characteristic J-shaped curve.

People who drank heavily had a higher risk of disease and death.

But people consuming small amounts of alcohol sometimes fared better than abstainers.

At first glance the conclusion seemed obvious:

a small amount of alcohol protects the heart.

The problem appeared once researchers looked more closely at who the "abstainers" actually were.

In some older studies, a single group contained both people who had never drunk and people who had drunk in the past but stopped, for example because of health problems.

This is the so-called sick-quitter effect.

If we compare a sicker group with relatively healthy people who drink small amounts of alcohol, alcohol can look more protective than it really is.

A large 2024 review found that most earlier reviews of alcohol and mortality had not properly eliminated this problem. In the analysis with the lowest risk of bias, no protective effect of low intake was observed, and risk rose with the amount of alcohol.

A 2023 meta-analysis of almost 4.8 million people produced similar results. After accounting for key sources of bias, low alcohol intake was not associated with significantly lower mortality compared with people who had never drunk.

Perhaps it was not alcohol protecting moderate drinkers

There is one more problem.

People who drink small amounts of alcohol may differ from non-drinkers in many ways:

they have a different economic situation,

they eat differently,

they exercise more often,

they have different social relationships,

they smoke less often,

they use medical care more often.

In a study published in JAMA Network Open in 2022, people drinking light to moderate amounts of alcohol also had more favourable lifestyle behaviours.

Once these were taken into account, part of the observed "protective" cardiovascular effect of alcohol disappeared.

Analyses using genetic data, meanwhile, pointed to an increased risk of hypertension and coronary artery disease as alcohol consumption rose.

This is why current data do not justify starting to drink alcohol in order to protect the heart. It is a good illustration of what preventive medicine means in practice: decisions based on current data rather than on a repeated belief.

And red wine and resveratrol?

Red wine contains polyphenols, including resveratrol.

That does not make alcohol a health food.

Polyphenols are also found in grapes, berries, cocoa, nuts, olive oil, coffee, tea and many other plant products.

We do not need ethanol to obtain them. In this respect alcohol is no different from other components credited with protective effects – what counts is the overall pattern of eating, which we cover in the article on the anti-inflammatory diet.

If someone enjoys a glass of wine, they can drink it for the taste and the pleasure, accepting the risk that comes with it.

It is much harder to justify drinking it "for health".

Alcohol and blood pressure

Alcohol's effects are not limited to cancer.

A 2023 meta-analysis of cohort studies showed a nearly linear relationship between alcohol consumption and a rise in blood pressure.

Compared with non-drinkers, mean systolic pressure was higher over time already at an intake of around 12 g of alcohol per day, and the difference grew with the dose.

A more recent 2024 meta-analysis also showed a rising risk of developing hypertension with increasing alcohol intake, particularly above roughly 12–24 g of ethanol per day.

This matters, because alcohol tends to be skipped over in conversations about blood pressure. If you are working on lowering yours, cutting back on alcohol deserves the same weight as the other elements of a diet for hypertension.

Alcohol and heart rhythm disorders

Regularly drinking larger amounts of alcohol also increases the risk of atrial fibrillation.

A meta-analysis covering more than 10 million participants showed that an increase of one drink per day was associated on average with roughly a 6% higher risk of atrial fibrillation. The relationship differed between women and men.

So not every glass leads to cardiac problems.

But the argument:

"alcohol is good for the heart"

is considerably harder to defend today than it was a decade or so ago.

Alcohol and the liver

Here the relationship is even more intuitive.

The liver is responsible for metabolising most of the alcohol entering the body.

Regular exposure increases the risk of steatosis, inflammation, fibrosis and cirrhosis, among others.

A 2024 meta-analysis covering more than 5 million participants showed a clear dose-response relationship. At an intake of around 25 g of alcohol per day, the risk of cirrhosis was roughly 1.8 times higher than in people who had never drunk, and it rose very steeply as the dose increased.

The risk also depends on sex, genetics, body weight, the presence of metabolic disease, viral infections and drinking patterns.

There is therefore no number at which we can guarantee any individual that their liver will stay healthy. Liver enzymes are one of the parameters worth including in the preventive tests worth doing once a year.

Alcohol helps you fall asleep. It does not necessarily help you sleep

An evening glass of wine can bring a sense of relaxation and make falling asleep easier.

That is only part of the story.

A 2025 meta-analysis of 27 studies showed that alcohol changes sleep architecture. Even relatively small doses affected REM sleep, and the disruption intensified with the amount of alcohol.

So:

"alcohol helps me fall asleep faster"

is not the same as:

"I sleep better after alcohol".

This matters especially for people who start using alcohol as a way to wind down before bed. If you wake up tired despite eight hours of sleep, evening alcohol is one of the first things worth checking.

Is there one amount of alcohol that suits everyone?

No.

Risk does not depend on the number of drinks alone.

What matters includes:

• age,
• sex,
• health status,
• medications taken,
• genetic predisposition,
• drinking patterns,
• frequency,
• amount per occasion,
• smoking,
• liver disease,
• cancer risk,
• pregnancy.

The Global Burden of Disease analysis published in The Lancet in 2022 produced interesting results.

The authors showed that the balance of alcohol-related risk can differ depending on age and on the disease profile of a given population. Among people aged 15–39 the level of consumption associated with the lowest risk was very close to zero. In some older populations, a model accounting for specific cardiovascular diseases indicated the possibility of a very small amount associated with lower total risk.

This does not change the fact that for cancer prevention the most favourable amount of alcohol remains zero.

How much is actually in "one drink"?

This matters, because saying:

"I only have two glasses"

tells us little if we do not know how big they are.

In Poland and many European countries, one standard drink means roughly 10 g of pure ethanol.

That amount is found approximately in:

• 250 ml of beer at 5%,
• 100 ml of wine at 12%,
• 30 ml of vodka at 40%.

Which means that:

500 ml of beer is about 2 standard drinks,

200 ml of wine is about 2 standard drinks,

and a large cocktail in a bar can contain more than one serving of alcohol.

So "one glass" does not always mean "one standard drink".

Is it better to have several drinks in one day than one a week?

No.

What counts is not only the total amount of alcohol but also the pattern of drinking.

Drinking a large amount on one occasion carries additional dangers:

injuries,

accidents,

alcohol poisoning,

heart rhythm disturbances,

loss of control over behaviour.

So the rule does not work:

"I have a weekly allowance, so I can drink it all on Saturday."

Two drinking patterns can deliver a similar amount of ethanol over a month while carrying a different risk profile.

Who should avoid alcohol completely?

There are situations in which we are no longer talking merely about reducing risk.

Alcohol should be avoided by, among others, people who are:

• pregnant,
• trying to conceive,
• about to drive or operate machinery,
• taking medicines with significant alcohol interactions,
• living with conditions in which alcohol can worsen their health,
• having difficulty controlling how much they drink.

In pregnancy, no safe amount and no safe timing of exposure have been established. Current recommendations remain unambiguous: it is best to avoid alcohol entirely.

Does this mean everyone should stop drinking?

From the point of view of minimising health risk, no alcohol is the safest choice.

That does not mean the conversation about alcohol should come down to frightening people over a single glass of wine.

Risk is not binary.

Someone who has a glass of wine a few times a year is in a completely different situation from someone drinking half a bottle of wine every day.

The WHO captures this relationship well:

the less alcohol, the safer.

So even if someone does not want to give up alcohol entirely, reducing the amount still matters.

How to reduce the risk if you do drink

There is no way to turn alcohol into a health-promoting product.

You can, however, reduce exposure.

What helps includes:

• drinking less often,
• reducing the amount per occasion,
• avoiding episodes of heavy drinking,
• choosing non-alcoholic drinks at some social occasions,
• paying attention to glass and drink sizes,
• not using alcohol as a way of coping with stress or sleep problems.

If someone currently drinks four times a week and cuts back to one evening, they do not need to reach complete abstinence to reduce their exposure.

Reduction is a beneficial change too.

Summary

Is there a safe amount of alcohol?

If by "safe" we mean an amount that does not increase any health risk, current data cannot identify one.

The evidence is particularly strong for cancer. Alcohol is a confirmed carcinogen, the risk appears at small amounts too, and it rises with consumption.

At the same time, this does not mean every amount of alcohol carries the same threat.

One glass now and then and several drinks every evening are completely different levels of exposure.

So the most important message does not need to be complicated:

if you do not drink, you do not need to start "for your health".

If you do drink, less means lower risk.

And if the goal is to minimise health risk as far as possible, the most favourable amount of alcohol remains zero.

Frequently asked questions

Is one glass of wine a day healthy?

There is no basis for recommending daily wine for health. Some earlier observational studies suggested a lower risk of heart disease among moderate drinkers, but newer analyses show that part of this association may stem from methodological bias and lifestyle differences.

Is a single glass of wine safe?

A single small amount of alcohol means a low absolute risk for most healthy adults, but we cannot describe it as entirely safe. For cancer risk, no level of exposure free of risk has been demonstrated.

Is red wine healthier than beer or vodka?

For carcinogenic effects, what matters most is the amount of ethanol delivered. The WHO and IARC stress that beer, wine and spirits all increase cancer risk. The polyphenols in red wine do not offset the effect of the ethanol itself.

Is 0.0% alcohol a better choice?

If it genuinely contains 0.0% alcohol, it avoids ethanol exposure. For products labelled "alcohol-free" it is worth checking the label, since depending on local rules and the product they may contain small amounts of alcohol.

Can you offset alcohol with exercise and a healthy diet?

No. Physical activity and a healthy diet are beneficial, but they do not neutralise the carcinogenic effect of alcohol. A healthy lifestyle does not create a "credit" that can later be spent on alcohol.

Does alcohol help with stress?

It can briefly reduce tension, but it does not address the cause. Regularly using alcohol as a way of coping with stress can also encourage an increase in how often and how much a person drinks.

Does alcohol help you fall asleep?

It can shorten the time it takes to fall asleep, especially at higher doses, but it also disrupts sleep architecture. A 2025 meta-analysis of 27 studies showed worsening of REM sleep that intensified with the amount of alcohol.

Can you have the occasional glass of wine in pregnancy?

No safe amount of alcohol in pregnancy and no safe timing of exposure have been established. The current recommendation is complete abstinence, including for people planning a pregnancy.

If I do not drink, is it worth starting for my heart?

No. Current data give no grounds for starting to drink alcohol to prevent cardiovascular disease. Analyses using genetic data point instead to a rising risk of hypertension and coronary artery disease as consumption increases.

This material is educational and does not replace an individual medical consultation. If you have difficulty controlling how much you drink, it is worth speaking to a doctor or an addiction specialist.

References

1. Anderson B.O. et al. Health and cancer risks associated with low levels of alcohol consumption. The Lancet Public Health. 2023;8:e6–e7.
2. World Health Organization Regional Office for Europe. No level of alcohol consumption is safe for our health. 2023.
3. International Agency for Research on Cancer. European Code Against Cancer, 5th Edition. 2025–2026.
4. WHO/Europe, IARC. Joint statement on raising awareness of the link between alcohol and cancer. 2023.
5. Bae J.M. Cancer risk based on alcohol consumption levels: a comprehensive systematic review and meta-analysis. Epidemiology and Health. 2023.
6. Sarich P. et al. The association between alcohol consumption and all-cause mortality: An umbrella review of systematic reviews using lifetime abstainers or low-volume drinkers as a reference group. Addiction. 2024;119:998–1012.
7. Zhao J. et al. Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. 2023;6:e236185.
8. Biddinger K.J. et al. Association of Habitual Alcohol Intake With Risk of Cardiovascular Disease. JAMA Network Open. 2022;5:e223849.
9. Cecchini M. et al. Alcohol Intake and Risk of Hypertension: A Systematic Review and Dose-Response Meta-Analysis of Nonexperimental Cohort Studies. Hypertension. 2024;81:1701–1715.
10. Cecchini M. et al. Alcohol Intake and Blood Pressure Levels: A Dose-Response Meta-Analysis of Nonexperimental Cohort Studies. Hypertension. 2023.
11. Jiang H. et al. Alcohol consumption and atrial fibrillation risk: An updated dose-response meta-analysis of over 10 million participants. Frontiers in Cardiovascular Medicine. 2022.
12. Llamosas-Falcón L. et al. How does alcohol use impact morbidity and mortality of liver cirrhosis? A systematic review and dose-response meta-analysis. Hepatology International. 2024;18:216–224.
13. Gardiner C. et al. The effect of alcohol on subsequent sleep in healthy adults: A systematic review and meta-analysis. Sleep Medicine Reviews. 2025;80:102030.
14. GBD 2020 Alcohol Collaborators. Population-level risks of alcohol consumption by amount, geography, age, sex, and year: a systematic analysis for the Global Burden of Disease Study 2020. The Lancet. 2022;400:185–235.
15. National Centre for Addiction Prevention (Poland). Psychoactive substances and health. The standard drink.