A glass of red wine being poured against a neutral background

Is Moderate Drinking Good for Your Heart? What the Evidence Shows

9 min read

For years the message was reassuring. A glass of red wine with dinner was supposed to be good for your heart, and moderate drinkers seemed to live longer than people who did not drink at all. It became one of the most repeated ideas in health. The trouble is that the studies behind it had a flaw baked in, and once researchers corrected for it, the protective glass of wine mostly disappeared.

This is not a scare piece. It is a look at what the largest and most careful reviews now show, and what that means for someone who enjoys a drink and wants to make an informed choice. The short version is that the heart protection we were promised does not hold up in the strongest studies, the risk climbs steadily as intake rises, and the safest approach does not rely on alcohol for any heart benefit.

4.8 million
Adults pooled across 107 studies. Low-volume drinking showed no significant protection against death from any cause. JAMA Network Open, 2023.
14 units
The NHS weekly low-risk limit for men and women, spread over 3 or more days
NHS guidance
0.93
Death risk at low-volume intake, a fall that was not statistically significant
Zhao, JAMA Network Open 2023
1.61x
Death risk for women at 65 g a day or more, relative to non-drinkers
Zhao, JAMA Network Open 2023
7 cancers
Cancer types alcohol causes. It is rated a Group 1 carcinogen
WHO, 2023

Where the good news story came from

The famous J-shaped curve did exist in the data. Plot health against drinking and moderate drinkers often sat lower than teetotallers, which looked like protection. The problem was who ended up in the non-drinker group. Many people quit alcohol because they are already unwell, or because past heavy drinking damaged their health. Lumping these former drinkers and sick people in with lifelong abstainers made the non-drinker group look worse than it should, and made moderate drinkers look healthier by comparison.

This is known as the sick-quitter effect. Moderate drinkers also tend to be wealthier, more active and better fed than average, and those advantages, not the alcohol, explain much of their apparent edge. When researchers separate lifelong non-drinkers from former drinkers and account for these differences, the protective dip shrinks or vanishes. The heart-friendly glass of wine was largely a statistical shadow.

What the biggest review found

The most thorough test came in 2023, when Zhao and colleagues published a meta-analysis in JAMA Network Open. They pooled 107 cohort studies covering 4.8 million adults and, crucially, adjusted for the study flaws that had inflated earlier results. At low-volume intake, up to about 24 g of alcohol a day, the risk of death from any cause was 0.93 times that of non-drinkers. That looks like a small benefit, but the confidence interval crossed 1.0 and the result was not statistically significant. In plain terms, low drinking was not linked to living longer.

Above that level, mortality climbed. The authors reported clear excess deaths once intake became heavy, and the rise started at a lower dose in women than in men. The chart below traces that dose-response for the group where the increase appears soonest. The pattern is consistent with other large recent analyses. More alcohol tracks with higher mortality, and no part of the curve dips meaningfully below the non-drinker line.

Death risk rises with heavier drinking Women, all-cause death vs non-drinkers Non-drinker reference, 1.0 0.93 1.21 1.34 1.61 Up to 24 g 25-44 g/day 45-64 g/day 65+ g/day Relative risk of all-cause death vs non-drinkers. Low band is pooled for both sexes.

Source: Zhao and colleagues, JAMA Network Open 2023, 107 cohort studies of 4.8 million adults. Hazard ratios for women vs non-drinkers. For men, the clear rise began nearer 45 g a day.

Risk arrives sooner for women

One detail from the same review matters for how people apply it. Women reached a statistically significant rise in death risk at a lower dose than men. For women, the review found clearly higher mortality from about 25 g a day, roughly three UK units, with the death risk 1.21 times that of non-drinkers in that band and rising to 1.61 times at 65 g a day or more. For men the significant rise began nearer 45 g a day, at 1.15 times the risk, reaching 1.34 times at the same top band. Body size, water content and how quickly alcohol is broken down all play a part, so the same drink lands harder on a smaller frame.

What the numbers mean in UK units

Grams per day are hard to picture, so it helps to convert. One UK unit is 8 g of pure alcohol. A standard 175 ml glass of wine is about 2 units, and a pint of beer is 2 or 3 units depending on its strength. The NHS advises no more than 14 units a week for men and women, spread over 3 or more days. That is about 6 pints of average-strength beer or 10 small glasses of lower-strength wine across a week.

The doses where the review found clearly higher death risk sit well above that guideline. The chart below puts the thresholds and the guideline on one scale, in units per week. The 25 g a day mark for women works out at around 22 units a week, and the 45 g a day mark for men at around 39 units a week. The NHS limit of 14 is deliberately set below these points, because the risk begins to climb below these marks, even where a single study cannot prove the rise beyond doubt.

The UK guideline sits below the risk marks Alcohol per week in UK units. 1 unit = 8 g. 14 22 39 NHS guideline Women risk mark Men risk mark Risk marks are the daily intakes above, converted to units per week.

Source: NHS unit guidance and Zhao and colleagues, JAMA Network Open 2023. Daily gram thresholds converted at 8 g per UK unit.

The part the wine story left out

Even if light drinking did nudge heart risk down a little, that is not the whole ledger. In 2023 the World Health Organization stated that no amount of alcohol is safe for health, and that the risk to a drinker starts from the first drop. Alcohol is rated a Group 1 carcinogen by the International Agency for Research on Cancer, the same category as tobacco and asbestos, and it causes at least seven types of cancer, including bowel cancer and breast cancer. The risk to the breast in particular rises at intakes many people would call moderate.

The WHO also noted that any possible heart benefit of light drinking has not been shown to outweigh the cancer risk carried by the same intake. So the honest picture is a trade-off that leans the wrong way. A small and uncertain heart effect sits on one side, a clear cancer signal on the other.

If you drink, how to lower the risk

None of this means a glass on a special occasion will undo your health. Risk is about pattern and amount over time, not a single evening. If you choose to drink, a few habits keep the risk toward the low end of the range the studies describe.

1
Stay under 14 units a week
The NHS guideline is the same for men and women. Counting units, rather than drinks, keeps the total honest, because a large glass of wine can be about 3 units on its own.
2
Spread it out
If you regularly drink close to 14 units, spread them over 3 or more days. Saving a week of units for one night raises the harm from that session, even when the weekly total is the same.
3
Keep several drink-free days
Two or three alcohol-free days a week give your liver a break and make it easier to keep the weekly total down. Drink-free days are one of the simplest ways to cut your average.
4
Do not start for your health
The evidence gives no reason to take up drinking, or to drink more, for a health benefit. If you do not drink, the science offers nothing to persuade you to begin.
5
Protect the rest of your habits
Sleep, training and diet do far more for your heart than any drink. Alcohol late at night also disrupts sleep, which chips away at recovery and the quality of the next day's session.

Common mistakes

A few beliefs keep the old story alive. The first is treating red wine as medicine, when the polyphenols it contains are present in far larger amounts in grapes, berries and tea without the alcohol attached. The second is trusting the pour at home, where a generous glass often holds more than 2 units, so two glasses quietly become a large share of the weekly guideline. The third is banking units, going dry all week to justify a heavy weekend, which the research links to worse outcomes than the same amount spread out. The last is reading no safe level as no point in cutting down. Lower is genuinely better across the range, so every drink you skip counts in your favour.

Frequently asked questions

Is a glass of red wine good for my heart?
Probably not in the way the old headlines claimed. The apparent benefit came mainly from flawed comparison groups. When the 2023 JAMA Network Open review corrected for that, low-volume drinking showed no significant drop in death risk. Any small heart effect is outweighed by other harms.
How much alcohol is safe to drink?
The WHO position is that no level is completely free of risk. The practical UK advice is to keep within 14 units a week for men and women, spread over 3 or more days. Less than that is better still, and not drinking carries the lowest risk of all.
Why do the limits hit women sooner?
Women tend to reach higher blood alcohol from the same drink, due to smaller body size and lower water content. In the 2023 review, a clearly higher death risk appeared from about 25 g a day for women and about 45 g a day for men.
Does alcohol really cause cancer?
Yes. The International Agency for Research on Cancer rates alcohol a Group 1 carcinogen, and the WHO reports it causes at least seven cancer types, including bowel and breast cancer. The risk rises with intake and has no proven safe threshold.
I only drink at weekends. Is that fine?
It depends on the amount. Packing a week of drinks into one or two nights counts as heavy drinking for those sessions, which the evidence links to worse outcomes than the same total spread out. Keep the weekly figure low and share it across more days.

The bottom line

The heart-protecting drink was a comforting idea built on shaky ground. The best current evidence, drawn from millions of adults, finds no meaningful benefit at low intake and a steady rise in risk as drinking climbs, sooner for women than for men. Add the clear cancer link, and the case for drinking on health grounds falls apart. You do not have to give up alcohol to be healthy. The realistic goal is to keep it modest, spread it out, and stop expecting it to do your heart any favours.

Build a routine your heart actually benefits from
Training, sleep and food move your heart health far more than any drink ever could. DT Fitness London builds evidence-based training and nutrition around the life you already lead. Book a consultation at www.dtfitnesslondon.com.

Sources

Back to blog