For centuries, alcohol has enjoyed a strange position in both health science and society at large. It’s been openly acknowledged as harmful in excess, potentially addictive, hard on the liver, and socially destructive- yet somehow totally taken for granted as a fact of life, uniquely exempt from scrutiny when consumed ‘responsibly.’

In this hallowed safe zone of moderation, nutrition science was there to reassure us that a glass of red wine per day wasn’t just tolerable, but even virtuous, heart-healthy and basically medicinal.

Then, almost overnight, the whole mood has radically shifted.

Headlines and government agency announcements have replaced former pleas for balance with a totally different message: that no level of alcohol is safe. As if listening closely, younger generations have moved away from drinking en masse. The familiar notion that drinking is an entirely normal part of socializing and might even extend our lives in the process is decisively on the wane.

But alcohol didn’t suddenly become poison, or fall out of favour for no reason. It’s possible that the only thing to fall apart was the story we told ourselves about it before.

Read on to see why this cultural shift has happened, and what the science really says about the actual health effects of alcohol.

In this article, we’ll discuss

• How drinking in moderation became nutritional common sense
• Why newer research unravelled that story
• What the science actually says about alcohol and health
• Why anecdotes and ‘Blue Zones’ don’t change anything
• What a balanced and non-alarmist takeaway looks like

The rise (and collapse) of “healthy drinking”

The idea that a little bit of alcohol might actually be good for you didn’t just come out of nowhere. Toward the end of the 20th century, researchers observed an interesting pattern in epidemiological studies: individuals who consumed alcohol moderately often appeared healthier and tended to live longer than those who abstained completely. This now infamous ‘J-shaped curve’ showed that, while light-to-moderate drinkers sat in the lowest risk category, heavy drinkers were at the red end of the spectrum, and non-drinkers appeared in second place.

Add to all of this the perception of the so-called French Paradox; low rates of heart disease in wine-drinking populations, and the narrative practically wrote itself. Alcohol, especially wine, wasn’t just a vice; it was a wholly legitimate and vaguely European lifestyle choice. The problem is that this story relies heavily on habit; we have been conditioned over the course of generations to view drinking as something that happens. Our parents drank, and so did theirs, so why shouldn’t we?

The rest of the story, the scientific bit, came from epidemiology. In the world of nutrition science, observational data (which shows correlation but not causation) is particularly sensitive to bias and is generally considered to be the lowest quality of research. As contemporary researchers have continued to examine and reexamine the growing pool of evidence, a very different story has emerged.

The methodology problem that changed everything

The real correction to the idea that drinking moderately was healthy didn’t begin with a single damning meta-analysis. Instead, it started with a shift in how researchers framed their questions about the health impacts of alcohol.

This shift is most clearly evident in the Global Burden of Disease (GBD) analyses, published in the prestigious Lancet journal, beginning in 2018. The hugely ambitious and comprehensive study gathered hundreds of large datasets across dozens of countries to examine how different levels of alcohol consumption affected total health loss- measured in disability and death- rather than focusing on isolated outcomes like cancer or heart disease.

The researchers asked a simple but profound question: across all diseases and causes of death, what level of alcohol intake most effectively minimizes adverse health outcomes? Their answer was clear and striking: zero. The J-shaped curve had vanished.

This stark conclusion didn’t come from new data, but from recognition of flaws in earlier observational research- most notably something called abstainer bias; many of the older studies compared drinkers to non-drinkers, a group that often included former drinkers who had given up alcohol due to illness.

When the relatively healthy light drinkers were compared to a mixed group including people who stopped drinking due to poor health, the light drinkers were (artificially) made to look robust indeed. But once lifelong abstainers were separated adequately from former drinkers, the advantage evaporated. Light drinking didn’t appear protective at all, but was, at best, neutral. From there, the higher the intake, the higher the risk. Other large meta-analyses have since confirmed the same findings.

So alcohol didn’t lose its benefits; they were just never really there to begin with.

Alcohol and cancer risk

Since 2018, more systematic reviews have shown that breast, colorectal, esophageal, and other types of cancer risk extend into light drinking specifically. This is where ‘no safe level’ sentiments come from, and this type of language is now routinely echoed by the WHO, CDC, and the National Cancer Institute.

The framing of alcohol as a carcinogen, like cigarettes, makes many people understandably uncomfortable. After all, the increase in absolute risk for someone who has a drink on special occasions is small enough to remain insignificant.

But public health doesn’t operate at the level of anecdotes or small numbers, it operates at scale. And when millions of people drink (even lightly), small relative personal risks translate into a clearly visible share of total alcohol-attributable cancers. In Europe, for example, approximately half of all alcohol-related cancers occur in light or moderate drinkers.

But what about the cardiovascular benefits?

For years, the strongest defence of light drinking rested on heart health. And it’s true, some research exists to show that alcohol can raise the ‘good’ HDL cholesterol, improve insulin sensitivity, and improve clotting factors; all things that sound reassuring on paper. But when scientists examine actual clinical outcomes, the picture changes considerably.

One recent meta-analysis found a near-linear relationship between alcohol intake and hypertension risk, with risk increasing above 12 grams of alcohol per day- about one standard drink. Separate meta-analyses looking at atrial fibrillation (AF) found that each additional drink per day raised AF risk by about six per cent, like clockwork. This matters because AF isn’t just any lab value; it refers to real arrhythmias with real downstream consequences, such as stroke.

Sadly, any modest cardiovascular upside has to be carefully weighed against blood pressure, arrhythmia, and stroke risks. It’s a tradeoff that looks less and less favourable with every study published.

But what about the Blue Zones, and my grandparents?

Sooner or later, alcohol apologetics will appeal to Blue Zones, or to their unstoppable grandfather who drank brandy every night and lived to 95. But while emotionally compelling, such anecdotal examples remain scientifically limited.

Blue Zones, those mysterious parts of the world where people die especially old and especially happy, are not alcohol trials. They’re unique and complex cultural ecosystems characterized by strong communal bonds, high physical activity, ultra-low UPF intake, and relatively modest caloric intake. Wine, when present, is typically enjoyed in small amounts, with meals, and in social settings. And of course, many long-lived people don’t drink at all.

Regardless, while such anecdotes loom large in the collective imagination, they are increasingly being seen as the exception rather than the rule. Somewhat analogous to cigarette smoking, that family elder who never missed a nightcap is increasingly seen to live long despite their vice, not because of it.

What does alcohol actually do when it enters the body?

Though referred to as a ‘spirit,’ there is nothing mystical about ethanol. When it enters the body, it isn’t metabolized like a nutrient or even something benign. Instead, it is immediately recognized as a toxin that needs to be neutralized and cleared.

Ethanol is quickly converted into the intermediary compound acetaldehyde, which is highly reactive and generates oxidative stress, disrupting regular cell cycles and potentially damaging DNA. It’s essential to recognize that this process occurs identically, regardless of whether the alcohol originated from inexpensive malt liquor or the finest Italian vintage. From a metabolic perspective, there is no “healthy” version of alcohol; it’s really all the same toxin.

So, is drinking less just better?

Alcohol isn’t uniquely evil, and it’s also not the health food we once managed to pretend it was. The most defensible position is that light-to-moderate drinking is neutral at best, with real, dose-dependent risks that quietly accumulate the more we drink.

Stripping alcohol of its imagined benefits may lead to positive changes in behaviour. People drink less when they stop thinking of wine as medicine. And fewer drinks, at a population level, means fewer cancers, fewer arrhythmias, and fewer alcohol-related harms in general.

As a calculated risk, there will probably always be a time and place for a drink. It’s hard to even imagine humanity without it. However, we seem to have permanently left the era where alcohol is considered beneficial- or even harmless- in small doses. And from the perspective of public health, it’s hard not to see that as a win.

FAQ: Alcohol and Health

Is one drink a day safe? It depends on what you consider safe. Cancer risk appears to increase even at low intake levels, while other health risks rise more clearly with higher intake. The safest intake is zero.

Is red wine better than other drinks? In terms of cancer, not really. Once alcohol content is accounted for, studies comparing wine types find no meaningful difference in cancer risk.

Should non-drinkers start drinking moderately, for health reasons? No major guidelines recommend this- and no evidence supports it at all. If you don’t drink, the healthiest move is to keep it that way.

Can quitting alcohol reduce cancer risk? Yes! Risk declines over time after quitting, although it may take years to return to baseline.

Is it ok to drink occasionally? For most people, occasional drinking is unlikely to cause disaster. The main thing is to abandon the idea that it’s doing you a favour.