How Much Alcohol Should You Drink? Part 1

It seems like a simple question. And, for several decades, there was a simple answer: A little alcohol is good for you. Studies showed that light drinkers have better cardiovascular health and live longer than both heavier drinkers as well as folks who abstain.

The figure below tells the tale. The more you drink, the lower the risk of health problems – up to a point. Beyond that point, drinking more increases risk.

Over the past two decades, a gloomier perspective has emerged: Light drinking confers no health benefits. It may even undermine your health. The worst-case scenario – e.g., the World Health Organization's statement that "no level of alcohol consumption is safe" – is shown in the figure below.

This newsletter was prompted by recent news reports reflecting the gloomier perspective. None of the reports recommend abstinence. But they all debunk the supposed benefits of light drinking, and they all insist that drinking in moderation may be dangerous. As the New York Times admonished two weeks ago, "Drink less; live longer."

In this two-part series, I'll be addressing two questions:

This week, I discuss whether we should abandon the idea that light drinking can be healthful.

Next week, I'll discuss whether we should assume that light drinking is unhealthy (and, if so, how much alcohol, if any, can be safely consumed).

Why is this important?

For thousands of years people have been drinking alcohol and noticing that overindulgence is dangerous. Only during the past century have we developed statistical methods (and raw computing power) that allow us to study large groups of people and distinguish between the effects of, say, zero vs. one or two drinks per day.

—If you're among the roughly two-thirds of Americans who drink on occasion, you need to know how much alcohol can be consumed safely, and whether light drinking is actually good for you.

—If you enjoy drinking enough that you're willing to take risks, you need to know what the risks are if you overindulge, so you can figure out the risk-benefit ratio that works for you.

—If you don't drink, you need to know whether you're missing any benefits. (Meanwhile, you're being affected by those who drink to excess. Alcohol misuse is estimated to cost the U.S. about $249 billion dollars per year, and some of that money is your tax dollars supporting prevention, treatment, and the management of everything from traffic accidents to domestic violence.)

I'll be drawing clear, practical conclusions here, but you have to expect some nuance and uncertainty too. Two thousand years ago, Pliny the Elder noted that "There is no topic more difficult to handle, or more full of detail [than] whether wine does good to people rather than harming them”, and, as you'll see, that description still seems reasonable, in spite of our sophisticated statistics and staggeringly powerful computers.

Healthy drinking and the French Paradox

By the turn of the 21st century, many scientists had come to believe that light drinking promotes cardiovascular health and longer life.

This idea captured the public imagination following a 60 Minutes episode on the so-called "French Paradox" (i.e., low rates of cardiovascular disease among French people in spite of diets high in cholesterol and saturated fats).

Near the end of this 1991 episode, Morley Safer claimed that it has been "all but confirmed" that alcohol – particularly red wine – protects French drinkers from all those pâtés and cheeses and dishes laden with butter and cream. The following year, sales of red wine in the U.S. increased by 40%.

The demise of "healthy" drinking

Although there had always been a few dissenters, during the past two decades evidence against the protective benefits of light drinking has accumulated.

A key problem with the older studies is that they relied on comparisons between abstainers and light drinkers, but these two groups don't just differ in how much alcohol they consume. Some abstainers avoid alcohol because they have other health problems. Some abstainers are actually former alcoholics who've already damaged their health. In short, light drinkers may be healthier than abstainers, but it's not because of the alcohol.

This point was succinctly demonstrated in a 2023 meta-analysis of 107 studies conducted between 1980 and 2021. The meta-analysis, published in JAMA Network Open, showed that when abstainers are simply compared to light drinkers (less than about two drinks per day), the light drinkers tend to live longer. However, when abstainers who formerly drank are set aside, the difference in longevity vanishes. Light drinkers don't live longer than those who never drank.

As for the French Paradox, there's no clear evidence that alcohol (or red wine in particular, or, even more specifically, a particular polyphenol called resveratrol that's found in red wine) can, by itself, in ordinary circumstances, offset the cardiovascular effects of diets high in cholesterol and saturated fats. A glass of red wine will not save you from McDonald's.

Finally, some studies suggest that the "Mediterranean Alcohol Drinking Pattern" (light drinking, restricted to meals, with no binging) promotes better health than abstinence, but it's unclear whether the benefits, if any, are specific to wine, and whether they depend on factors such as following a "Mediterranean diet", socializing while drinking, drinking at a slower rate, and/or being an older person. There are a lot of question marks. I wouldn't assume yet that the MADP is especially healthful.

Some good news

If you consider yourself a light or moderate drinker, I can almost hear you saying "Awwww. I thought this was good for me."

Perhaps it is.

The studies I've been alluding to all rely on aggregate data. Abstainers, or people who average less than a drink or two per day, are compared to a group of people who drink a little bit more, and no differences are found in rates of heart disease, in life expectancy, or in some other health outcome.

Aggregate data is a mixed blessing. It tells you something about groups of people but not much, if anything, about individuals.

For instance, Luxembourg and Japan are almost identical in GDP per capita. If you divide GDP (the total value of goods and services in each country) by the number of citizens, you get a strikingly similar value for each country (about $25,000 USD). But this doesn't mean that you, personally, would be equally affluent in each country. In fact, from day one, you'd almost surely be richer in one of those countries, depending on your profession, your city of residence, and so on. Some people would be better off in Luxembourg, others in Japan, still others would fare about the same in each country.

By analogy, saying that abstainers and light drinkers have the same health outcomes tells you nothing about the prospects for individuals.

For instance, reducing stress is a well-established way of improving mental and physical health. If you're a light drinker, and you find it relaxing, alcohol consumption may be good for you.

The devil is in the details though. Even if you have no physical conditions or take medicines that preclude drinking, a low level of consumption isn't automatically a good way to reduce stress:

—Light drinking that simply takes your mind off your troubles for a few hours can be helpful on occasion, but the benefits may not be sustainable if you're just avoiding your stressors rather than actively coping with them.

—Abstaining from alcohol most of the time, then binge drinking on occasion to de-stress, has no health benefits. (I'll discuss recent data on the risks next week.)

—Light drinking can be relaxing and make it easier for a stressed person to fall asleep, but studies show that the quality of sleep may be diminished.

—If you're abstinent after a history of Alcohol Use Disorder or otherwise struggling with overconsumption, light drinking would likely create problems including, among other things, greater rather than less stress.

In the end, aggregate data can't capture your particular physiology and lifestyle with enough granularity to yield specific predictions. What benefits one person might not help someone else.

For instance, I've always been an early riser – one of those annoying people who naturally falls asleep around 11 p.m. and wakes up, alert and refreshed, around 5 or 6. Whenever something keeps me up late for several nights in a row, I'll find myself starting to get drowsy later at night and waking up later in the morning, and I don't feel quite as good. How to reset? For me, drinking a glass of wine around 10 p.m. works perfectly. An hour later, I'm drowsy, and (as some studies would predict) I wake up the next morning at the usual early hour feeling fine.

I'm not saying this hack would work for everyone. Some people, maybe. Whether or not light drinking benefits you likely depends on who you are and the specifics of how and when you drink.

Conclusions

I see no data ruling out the possibility that if you're healthy and well-rested (and not taking medications that interact with alcohol), occasional or even daily consumption of small quantities (e.g., one beer or glass of wine) can be good for your health.

In recent reports, journalists have thrown out the baby with the bathtub gin, claiming not only that light drinking is bad for you but that there aren't benefits either. I can't blame them really. More and more studies are showing that abstainers and light drinkers have similar health outcomes. But what the journalists often miss is that the studies rely on aggregate data. Light drinking may actually be good for you (meaning you personally).

Of course, light drinking could be good and bad for you simultaneously – reducing stress, for instance, while undermining arterial health. We haven't reached the end of the story yet. Next week I'll discuss whether even the lowest levels of alcohol consumption are risky, as often claimed nowadays, or whether there's a sweet spot where light drinking could be beneficial without elevating risk.

Thanks for reading!