Alcohol and your health: the honest, updated evidence
The old reassurance that a daily glass is good for your heart has largely fallen apart. Once researchers correct for a basic flaw, that many "non-drinkers" are really people who quit because they were already ill, the apparent benefit of moderate drinking mostly vanishes. A 2023 review of 107 studies and nearly 4.8 million people found no real drop in death rates for light drinkers, and the WHO now says there is no completely safe level. That does not make a glass of wine a crisis. It means the honest picture is a dose-response one: less is lower risk, and the UK guideline of 14 units a week, spread out, is a sensible ceiling rather than a target.
Key facts
- A 2023 meta-analysis of 107 cohort studies (about 4.8 million people) found no significant reduction in all-cause mortality for people drinking under roughly 25g of alcohol a day, once the analysis accounted for former drinkers who had quit through ill health.1
- The UK Chief Medical Officers advise no more than 14 units a week for everyone, spread over three or more days, with several drink-free days. One unit is 8g of pure alcohol; 14 units is about six pints of average beer or six 175ml glasses of wine.56
- Breast cancer risk rises by about 7 percent for every extra 10g of alcohol a day, with no safe threshold below which the effect disappears.3
- In 2023 the UK recorded 10,473 alcohol-specific deaths, the highest on record, and 76 percent were from alcohol-related liver disease, up 46 percent since 2012.8
Why "moderate drinking is good for you" collapsed
For decades, surveys seemed to show a J-shaped curve: light drinkers lived longer than both heavy drinkers and teetotallers. The problem sits in that last group. Many lifelong-abstainer categories were quietly padded with people who had stopped drinking precisely because they were unwell, frail or on medication. Next to this sicker reference group, moderate drinkers looked healthy, but the comparison was skewed from the start. This is the sick-quitter effect, and a systematic review found that more than 70 percent of the major studies over the past 30 years failed to exclude former drinkers, biasing the results in alcohol's favour.2
When that flaw is corrected, the protective signal shrinks or disappears. The 2023 JAMA Network Open meta-analysis by Jinhui Zhao and colleagues pooled 107 cohort studies and found that, after adjusting for study design and the age and sex of cohorts, light to moderate drinkers had no meaningfully lower death rate than lifetime abstainers. Raised mortality appeared at higher intakes, from a lower threshold in women.1 A separate Global Burden of Disease analysis concluded that for adults aged 15 to 39 there is no health benefit at any level, while any cardiovascular upside for older adults is small and easily outweighed by other risks.4
Evidence strength: the mortality data are observational, so they cannot prove cause and effect. But the consistency across large meta-analyses, the clear biological mechanisms, and the disappearance of the "benefit" once the sick-quitter bias is removed all point the same way. In 2023 the WHO stated plainly that no level of alcohol consumption is safe for health, and the International Agency for Research on Cancer classes alcohol as a Group 1 carcinogen, the same category as tobacco and asbestos.7
The dose-response risks, plainly
Alcohol does not flip from harmless to harmful at a single line. Most of its risks climb gradually with the amount you drink, which is why the message is about dialling down rather than a pass or fail.
Cancer
This is the clearest harm and the least known. Alcohol is causally linked to at least seven cancers: mouth, throat, voice box, oesophagus, breast, liver and bowel. Around 17,000 UK cancer cases a year, roughly 4 percent of the total, are attributable to drinking.9 For breast cancer the dose-response is well measured: a pooled reanalysis of 53 studies found risk rose about 7.1 percent per 10g of alcohol a day, and roughly one in 13 UK breast cancers is linked to alcohol.3 The mechanism is direct: the body turns alcohol into acetaldehyde, which damages DNA, and alcohol also raises oestrogen.
Blood pressure, sleep, mood and liver
- Blood pressure. A 2023 dose-response meta-analysis in Hypertension found the link between alcohol and rising blood pressure was essentially linear, with no safe threshold: systolic pressure was about 1.25mmHg higher at 12g a day and 4.9mmHg higher at 48g a day, versus none.10 Our piece on blood pressure explains why even small shifts matter.
- Sleep. A nightcap helps you fall asleep but wrecks the second half of the night. A meta-analysis of 27 studies found alcohol suppresses REM sleep early on, then triggers fragmented, restless sleep and a REM rebound as it clears, even at low doses.11
- Mood. Alcohol is a depressant that can blunt anxiety in the moment, which is part of why the relationship runs both ways. Genetic (Mendelian randomisation) studies suggest that more frequent drinking raises the risk of major depression, and disrupted sleep and low mood feed each other.12 If you drink to cope with low mood or anxiety, that is worth a kind conversation with your GP rather than a private worry.
- Liver. Early fatty liver and even some fibrosis can improve or reverse if drinking is reduced in time, but liver disease is usually silent until it is advanced. Alcohol-related liver deaths in the UK have risen 46 percent since 2012.8 The encouraging side is that the liver is forgiving when you act early.
The UK low-risk guideline
The point of the guideline is not abstinence; it is keeping risk low for people who choose to drink. It treats men and women the same, which surprises people who remember the older split limits.
| Guidance point | What it means |
|---|---|
| Weekly limit | No more than 14 units a week, for both men and women |
| Spread | Spread over 3 or more days; avoid saving units for one heavy session |
| Drink-free days | Have several alcohol-free days each week |
| 14 units looks like | About 6 pints of 4% beer, or six 175ml glasses of 13% wine, or 14 single spirits |
| Pregnancy | Safest not to drink at all |
Alcohol is calorie-dense too: 7 calories per gram, close to pure fat, so 14 units carries roughly 1,000 hidden calories a week.6 Tracking your real intake is often the single most useful first step, and our stack builder can help you fit changes into a wider plan.
Practical, non-judgmental ways to cut down
Cutting down is not all-or-nothing, and small changes compound. People who took part in Dry January in a University of Sussex study were still drinking less six months later, with more drink-free days and fewer units per session, and the effect was strongest among those who used a tracking app for support.13 Things that genuinely help:
- Set drink-free days first. Picking two or three alcohol-free days a week is easier to stick to than a vague "less". The free NHS Drink Free Days and MyDrinkaware apps make this concrete.
- Make the easy choice the default. Keep less alcohol in the house, switch to smaller glasses or lower-strength drinks, and alternate each alcoholic drink with water.
- Name your reason. Better sleep, money saved, mood, a family history of breast or bowel cancer. A specific reason beats willpower.
- Watch the pour. A large glass of strong wine can be three units on its own. Measuring at home recalibrates what "one drink" really is.
If you are physically dependent, suddenly stopping can be dangerous. If you shake, sweat or feel anxious or sick until your first drink of the day, do not stop abruptly. Speak to your GP or call Drinkline free on 0300 123 1110 first, as you may need a supported, gradual reduction. Cutting down should never put you at risk.
- Given my drinking and my family history, what is my realistic cancer and liver risk?
- Could a simple blood test or a FibroScan check on my liver be worthwhile?
- I sometimes drink to manage stress, low mood or sleep. Can we look at that together?
- If I am drinking heavily, do I need a supported plan to cut down safely?
- Could cutting back help my blood pressure or other medications work better?
References
- Zhao J, Stockwell T, Naimi T, et al. Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Netw Open. 2023;6(3):e236185. pmc.ncbi.nlm.nih.gov.
- Institute of Alcohol Studies. The sick-quitter effect: alcohol-related death has been underestimated over the last 30 years. 2024. ias.org.uk.
- Hamajima N, Hirose K, Tajima K, et al. (Collaborative Group on Hormonal Factors in Breast Cancer). Alcohol, tobacco and breast cancer: collaborative reanalysis of individual data from 53 epidemiological studies. Br J Cancer. 2002;87(11):1234-1245. nature.com.
- GBD 2020 Alcohol Collaborators. Population-level risks of alcohol consumption by amount, geography, age, sex, and year. The Lancet. 2022;400(10347):185-235. thelancet.com.
- UK Chief Medical Officers. UK Chief Medical Officers' Low Risk Drinking Guidelines. 2016. gov.uk.
- NHS. Alcohol units. Live Well. nhs.uk.
- World Health Organization. No level of alcohol consumption is safe for our health. 2023. who.int.
- Office for National Statistics. Alcohol-specific deaths in the UK: registered in 2023. 2025. ons.gov.uk.
- Cancer Research UK. Alcohol and cancer: how does alcohol cause cancer? cancerresearchuk.org.
- Di Federico S, Filippini T, Whelton PK, et al. Alcohol Intake and Blood Pressure Levels: A Dose-Response Meta-Analysis of Nonexperimental Cohort Studies. Hypertension. 2023;80(10):1961-1969. ahajournals.org.
- Ebrahim IO, Shapiro CM, Williams AJ, Fenwick PB. Alcohol and sleep I: effects on normal sleep. Alcohol Clin Exp Res. 2013;37(4):539-549. pubmed.ncbi.nlm.nih.gov.
- Fang Y, et al. Risk of major depression increases with increasing frequency of alcohol drinking: a bidirectional two-sample Mendelian randomization analysis. Front Public Health. 2024;12:1372758. pmc.ncbi.nlm.nih.gov.
- de Visser RO, Piper R. Short- and longer-term benefits of temporary alcohol abstinence during Dry January. University of Sussex / Health Psychology evaluation. 2019. sussex.ac.uk.
This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.