A new paper published in The Lancet led to headlines like “There’s No Safe Amount of Alcohol.” But the truth is more measured.

Share story

Last week a paper was published in The Lancet that claimed to be the definitive study on the benefits and dangers of drinking. The news was apparently not good for those who enjoy alcoholic beverages. It was covered in the news media with headlines like “There’s No Safe Amount of Alcohol.”

The truth is much less newsy and much more measured.

It’s important to note that this study, like most major studies of alcohol, wasn’t a new trial. It was a meta-analysis, or a merging of data, from many observational studies. It was probably the largest meta-analysis ever done to estimate the risks from drinking for 23 different alcohol-related health problems.

The researchers also combined almost 700 sources to estimate the most accurate levels of alcohol consumption worldwide, even trying to find drinking that might otherwise be missed (from tourism, for instance). They then combined all this data into mathematical models to predict the harm from alcohol worldwide.

They found that, overall, harms increased with each additional drink per day, and that the overall harms were lowest at zero. That’s how you get the headlines.

But, and this is a big but, there are limitations here that warrant consideration. Observational data can be very confounded, meaning that unmeasured factors might be the actual cause of the harm. Perhaps people who drink also smoke tobacco. Perhaps people who drink are also poorer. Perhaps there are genetic differences, health differences or other factors that might be the real cause. There are techniques to analyze observational data in a more causal fashion, but none of them could be used here, because this analysis aggregated past studies — and those studies didn’t use them.

We don’t know if confounders are coming into play because this meta-analysis could only really control, overall, for age, sex and location. That’s not the researchers’ fault. That’s probably all they could do with the data they had, and they could still model population-level effects without them.

But when we compile observational study on top of observational study, we become more likely to achieve statistical significance without improving clinical significance. In other words, very small differences are real, but that doesn’t mean those differences are critical.

The news warns that even one drink per day carries a risk. But how great is that risk?

For each set of 100,000 people who have one drink a day per year, 918 can expect to experience one of the 23 alcohol-related problems in any year. Of those who drink nothing, 914 can expect to experience a problem. This means that 99,082 are unaffected, and 914 will have an issue no matter what. Only 4 in 100,000 people who consume a drink a day may have a problem caused by the drinking, according to this study.

At two drinks per day, the number experiencing a problem increased to 977. Even at five drinks per day, which most agree is too much, the vast majority of people are unaffected.

I’m not advocating that people should ignore these risks. They are real, but they are much smaller than many other risks in our lives, and much less than the headlines would have you believe, especially at the levels that most agree are permissible.

This is a population-level study, arguably a worldwide study, but the results are being interpreted at an individual level. They are merging, for instance, the 23 alcohol-related health issues together. But not everyone experiences them at the same rate.

For diabetes and heart disease, for instance, the risks actually go down with light or moderate drinking. The authors argue that this result is overrun, however, by risks for things like cancer and tuberculosis, which go up. But for many individuals, the risks for diabetes and heart disease are much higher than those for cancer and tuberculosis.

Motor-vehicle accidents caused by drinking are clearly worse in some places than others. The study had data only for the United States and extrapolated that worldwide. It’s not clear that everyone is at the same risk in reality.

This message shouldn’t get lost in any argument: There is no debate, and this study confirms once again, that heavy drinking is really bad for you.

The population-level average of daily drinks is 1.9 for women and 3.2 for men, according to the study. This is above U.S. government health recommendations, but it is probably skewed by those who drink heavily,a relatively small percentage of all people who drink.

Of course, some drink too much without knowing it. For this study, a drink was defined as 10 grams of pure alcohol, as much as you might get in one ounce of spirits (a small shot glass) that is 40 percent alcohol; 3.4 ounces of wine that’s 13 percent alcohol; or 12 ounces of beer that’s 3.5 percent alcohol. Many people consume more than that and consider it “a drink.”

Moderate alcohol consumption is associated with certain improved risk factors for health (which this study confirms), but this is not the same as recommending that people drink. I don’t, and I don’t know of any health-related professional organizations that do.

But just because something is unhealthy in large amounts doesn’t mean that we must completely abstain. A chart in the study showed rising risks from alcohol from 0 to 15 drinks.

Consider that 15 desserts a day would be bad for you. I am sure that I could create a chart showing increasing risk for many diseases from 0 to 15 desserts. This could lead to assertions that “there’s no safe amount of dessert.” But it doesn’t mean you should never, ever eat dessert.

Food is not medicine. Neither is alcohol. Alcoholism is terrible. There’s a balance, and we could spend lifetimes arguing over where the line is for many people. The truth is we just don’t know. If these studies are intended to drive population-level policy, we should use them as such, to argue that we might want to push people to be wary of overconsumption.

Too many people interpret them individually, however, with panic-inducing results.

We may have wrung about as much good as we can out of these observational studies. Another meta-analysis is unlikely to tell us anything new. If we want fresh knowledge, a big randomized controlled trial of light to moderate drinking may be needed.

One was in the works recently, but it was shut down after The New York Times reported on ethical concerns about the way it was being pitched to and funded by industry. That doesn’t mean that a different trial wouldn’t be of public interest and worth public funding. If we want to get better answers to the harms and benefits of light to moderate drinking, that may be our only course.