You have seen the headlines before. They usually resurface every few months while scrolling the news over breakfast, or halfway down your feed between a friend’s holiday photos and someone’s sourdough.

“No amount of alcohol is safe” they say. And for a second or two, your glass of Grenache looks faintly accusing.

So let’s talk about where those headlines actually comes from, because the story underneath it is even more interesting than the headlines themselves.

So let’s talk about where those headlines actually come from, because the story underneath it is even more interesting than the headlines themselves.

It goes back to a 2018 study published in The Lancet, part of an enormous global project measuring disease and risk, which concluded that the amount of drinking that does least harm is none at all.

What almost nobody reported was that the study never published the numbers telling you how many actual people we’re talking about, and The Lancet’s own press office had to go back to the authors and ask for them before it went out (which tells you something in itself!).

Here is what those numbers looked like. If you line up 100,000 people who drink one glass a day, over a year 918 of them will develop one of 23 conditions that alcohol is known to play some part in, and if you line up another 100,000 who drink nothing at all, 914 of them will develop one as well.

The difference they found was four people in a hundred thousand, across an entire year.

These are illnesses with many causes, where alcohol is only ever one of them – the list runs from cancers and liver disease all the way to road injuries and tuberculosis, which tells you how broadly the net was cast. (Also, my favourite detail is that the study’s “non-drinkers” were actually allowed up to 0.8 drinks a day – yes, really!).

Those four extra people are real and we’re not dismissing them, but the finding of this research was never that a glass of wine is dangerous. It was that risk climbs the more you drink.

Since then the experts have been arguing with each other rather loudly.

In 2025 the US National Academies of Sciences reviewed the evidence for Congress to help set America’s dietary guidelines, and found that moderate drinkers were less likely to die, of anything at all and of heart disease in particular, than people who never drink. Before you assume those with an agenda had a hand in it (hold your judgement!), this is one of the most rigorous scientific bodies in the world.

But that isn’t the last word either, because another report prepared for the same government came to almost the opposite conclusion. Critics of the first review also spotted that when you compare moderate drinkers with light drinkers, rather than with people who don’t drink at all, the benefit quietly disappears. There is a whole field of research that follows people’s genes rather than their drinking habits, which dodges most of the usual problems, and it keeps finding no benefit at all. Two respected reviews, the same evidence, entirely different verdicts, and anyone who tells you this one is settled is most likely trying to sell you something.

So where does that leave you and your glass of Grenache? With the most interesting study of the lot, as it happens.

In March 2026, researchers presented the largest analysis yet at the American College of Cardiology: more than 340,000 British adults from the UK Biobank, followed for an average of over thirteen years. Heavy drinking fared badly – no surprises there – with a 36 percent higher risk of dying from cancer.

But when the researchers looked at what people drank, a divide opened up. Moderate wine drinkers had a 21 percent lower risk of dying from cardiovascular disease than those who never or rarely drank. Beer, cider and spirits drinkers got no such reprieve. Even at low intake, they carried a 9 percent higher risk.

Why should wine give different outcomes from other alcoholic drinks? The researchers put forward two ideas.

The first is a chemical explanation. Things like the polyphenols and kindred compounds that are present in wine are not in other alcoholic drinks like vodka, gin or beer.

The second is a matter of when wine tends to be drunk. We usually enjoy a glass of wine when eating food, at a table, in company, over an unhurried evening. Wine drinkers, the study noted, also tended to eat and live better in general. All of which is worth stating without any hint of smugness: this does not make wine a health tonic. But it does suggest that the way you drink and what you drink matters every bit as much as whether you drink at all.

Of course, wine is not risk free. And no one is saying that it is. (Including us.)

Heavy drinking came off poorly in every study named here, and no one can seriously claim that wine carries no risk at all. (Crossing the road has risk. Everything has risk!)

The point, the one the anti-alcohol headlines keep failing to adequately explain, is that heavy drinking and a glass of wine with dinner are not remotely the same thing, and a headline that lumps the two together doesn’t paint the true picture.

Good science is rarely simple, and the research on alcohol is still very much arguing with itself. So if someone tells you the science is settled, they’re probably oversimplifying. Wine deserves honesty. So does the science.

We hold our wine to high standards. The science deserves the same!

Sources: the “no safe level” conclusion, and the 918 versus 914 figures, come from the GBD 2016 Alcohol Collaborators, published in The Lancet (2018), with the absolute risks released through the journal’s press office; the 23 outcomes are conditions in which alcohol is a risk factor, not conditions caused by alcohol alone. The congressionally commissioned review is the National Academies of Sciences, Engineering and Medicine, Review of Evidence on Alcohol and Health (2025); its conclusions have been contested, notably in the Journal of Studies on Alcohol and Drugs (2025). The 340,000-participant analysis was presented at the American College of Cardiology Annual Scientific Session (March 2026): a 36 percent higher risk of cancer death among heavy drinkers, a 21 percent lower risk of cardiovascular death among moderate wine drinkers, and a 9 percent higher risk among low-intake beer, cider and spirits drinkers. At the time of writing it had not been published in a peer-reviewed journal.