Unexplainable - Is a little alcohol bad for you?

Episode Date: September 8, 2025

We spoke to two researchers who disagree about the answer to this question. But they do agree about why it's so hard to answer to begin with. Guests: Dylan Scott, senior correspondent at Vox; Kenneth Mukamal, physician and academic researcher at the Beth Israel Deaconess Medical Center; Timothy Naimi, director of the University of Victoria’s Canadian Institute for Substance Use Research For show transcripts, go to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠vox.com/unxtranscripts⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠For more, go to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠vox.com/unexplainable⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠And please email us! ⁠⁠⁠[e-mail]⁠⁠⁠We read every email.Support Unexplainable (and get ad-free episodes) by becoming a Vox Member today: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠vox.com/members⁠⁠⁠⁠⁠⁠⁠⁠Thank you! Learn more about your ad choices. Visit podcastchoices.com/adchoices

Frozen lasagna, medium power, 15 minutes. Sounds like Ojo time. Let's play. Feel the fun with Play Ojo. The online casino with all the latest slot and live casino games. What you win is yours to keep with no wagering requirements. Instant payouts and no minimum withdraws. Hey, I just won.

Woohoo. Feel the fun. Play Ojo. Honey, forget about the lasagna. Let's celebrate. 19 plus Ontario only. Please play responsibly. Concerned about your gambling or that of someone close to you. Call 1866-531-2600 or visit Connexontera.ca.

Visit BetMGM casino and check out the newest exclusive. The Price is Right Fortune Pick. BetMDM and GameSense remind you to play responsibly. 19 plus to wager. Ontario only. Please play responsibly. If you have questions or concerns about your gambling or someone close to you, please contact connects Ontario at 1-866-531-2,600 to speak to an advisor, free of charge.

BetMGM operates pursuant to an operating agreement with Eye Gaming Ontario. Dylan Scott, a senior correspondent and senior good pal of mine. Bird Pinkerton, pal of mine, and co-host of Unexplainable. You have a mystery for me, I believe, about alcohol. I do. So if you're anything like me, you know, I think we both grew up in the 90s, you have heard some very mixed messages about what drinking moderate amounts of alcohol can do for your health. Just in time for the weekend, a new warning about the dangers of alcohol and cancer.

It feels like every week you read the news and there's a new study about alcohol and it's saying the exact opposite of the study you were reading about last week. Several studies have shown moderate drinking may be good for your overall health, but no amount of alcohol is good for your health. Less drinking means less risk. So I decided like, I want to get to the bottom of this for my. and for our audience, like, I want to actually understand what to think about the health effects of drinking moderate amounts of alcohol. Okay. And so what did you find?

Like, what are the health effects of drinking moderate amounts of alcohol? Well, this is unexplainable. So I found it was unexplainable. But in a way that I think is really interesting and revealing. Because this story of alcohol's health effects that are imperfect attempts to measure it is I think the story of how it's really, really hard to isolate and pin down the health effects

of all kinds of things that we really would want answers about these things that we interact with and consume all of the time. Okay, let's go. You know, I think to start, we should establish that the science is pretty settled when it comes to heavy drinking, alcohol is bad for you. It has neurological effects.

It causes liver disease. And we've known that for hundreds and hundreds of years. How much is too much, though, I guess? Or, like, what is a lot of alcohol? This was one of the things that, you know, surprised me when I looked into this. And I think it would surprise a lot of people, which, is that what normal people think is a lot of drinking and what scientists and doctors think is a lot of drinking, there's a pretty big gap there. But looking at the scientific definition,

for some people, the limit is two drinks per day. Anything above that is heavy drinking. And for other people, it can be just one drink per day and anything above that is heavy drinking. And like, you know, a drink can be a pretty small amount. Even one can of beer, if it has really high alcohol content, can be more than one drink. And so if you're drinking more than that limit each day, the researchers that I talked to were in agreement that that's going to be a problem. There are heart risks. There are cancer risks. It's damaging your liver. And you would do better to drink less. But what if you are drinking, like, just a glass of wine every two days or whatever? I mean, this is where things get contentious.

This is where all of the argument is. So let's start with sort of the pro-alcohol side. And we can start with a researcher who has made the case that maybe some small amount of alcohol might actually be good-ish for you. My name is Ken Mukumal. I'm a primary care physician at Beth Israel Dekees Medical Center. I do think it's worth pointing out that Ken has been in the news in the past, and he has faced some criticism from his peers about his research allegedly being influenced by the alcohol industry.

Like back in 2018, he was leading a study on alcohol's health effects for the NIH, and that study got shut down because the NIH found some suggestion that there was bias in the study's design. But, you know, even with that history, I thought it was really helpful to talk to Ken and to get his perspective on the research. And I also thought it was interesting, like he staked out this position, but he doesn't really seem to have a personal investment in drinking alcohol. Although a size-able part of my research career has been related to alcohol, we hardly drank it all in my house. My dad was a chemical engineer, and if he happened to open a beer. It was like, you know, a major event in the house. And in fact, I continued harder to drink at all.

But even so, part of the reason he wanted to study alcohol is that, as he said, he's actually a primary care doctor. So he's not like just a cardiologist by training who's only worried about your heart. He's not only a liver doctor or a kidney doctor. He's worried about the whole person. And for him, alcohol was a really good fit because alcohol touches on so many. aspects of people's lives. Decades ago, as he was starting to think about this, he developed this question about whether some amount of alcohol might be helpful. This was right around the time that in clinical medicine more generally, when I was going

through training at that time, we were moving into the idea that when people had heart attacks, that one of the ways that we could either treat them or prevent them was to use things like giving people baby aspirin, right? And that eventually turned into, like, you know, lure all over the world. aspirin is a blood thinner and heart attacks happen basically because the flow of blood to the heart is blocked in some way, and so thinning out the blood might help. And indeed, the New England Journal had published a sort of small paper in the early 1980s showing that aspirin prolongs how long it takes somebody to clot their blood, and if you give them aspirin plus alcohol, it was markedly longer than even just with aspirin alone. So I'll be honest, I'm thinking to myself, well, if that's true, and we're spending all this time, you know, trying to get people aspirin and stuff, like, why aren't we washing down that aspirin with a little bit of alcohol? If that data from the early 19 days was right, and we're seeing that alcohol has a little bit of bloodthening effect, and we're looking for additional blood thinners, like, it seems like the natural next thing to do.

And his boss actually didn't totally shoot this idea down. And, you know, Ken told me, like, there are better blood thinning options than alcohol. So the idea wasn't, like, let's, you know, give everybody who's having a heart attack a shot of whiskey. But, you know, he was still interested, like, could naturally thinning your blood this way, you know, prevent heart attacks just over the long term? And could there be other positive impacts that alcohol might have on your health somehow? So if you're, like, having a glass a day, is your blood just like, a little bit thinner, and maybe that prevents heart attacks. This more like kind of long-term, small-dose exposure, prophylactic approach. Yes, that's what he was curious about. And he, you know,

went down this long rabbit hole for many years after that looking for some answers. He was mostly looking at, you know, these big data sets, these things that are called cohort studies or large population-based studies. More or less, the idea is quite simple, right? We're going to take lots of people, we're going to ask them some universal standard questions, including things like, how often do you drink alcohol? When you drink, how much do you have? What's the most that you have? And then using that information, we're going to then sort people into bins, and then wait and see what happens. Now, these are not experiments. I'm going to emphasize that, because that's going to, I think, come back to haunt us a little bit. These are not situations where

I ask you to drink a certain amount or ask you not to drink a certain amount and manipulate it the way that, say, I would do with mice in a laboratory. These are just, you chose to drink or you didn't choose to drink, and I'm going to follow along and see what happens. And in our case, they involved anywhere from, you know, a few thousand to tens of thousands of people followed for anywhere from three or five years to 25 years. And so, you know, Ken's looking at these studies where people have reported their alcohol consumption, and then he's looking at data that shows their health outcomes, and he's going through the numbers, and he found a slight link in small doses to some potential health benefits from drinking alcohol. To the extent that there's sort of a simple message that I could offer, it's that where we see, well, I'll call, quote, benefits of alcohol.

These are not proven. It's definitely causal, but to the extent that we see any, it's predominantly for things like diabetes and heart disease. So he's saying that there may be a slight benefit, a slight positive health effect for your heart from low-level drinking. If there's any benefit, there's benefits from drinking small amounts frequently. and in fact, binge drinking, for the most part, obliterates any of the benefits that you see from frequent drinking. He doesn't sound super sure about this, though, we're, like, confident.

Like, when he says, if there's any benefit, he's not making, like, a particularly definitive statement. No, he's not. And, I mean, this is what I found the most interesting after digging in. into this. Ken actually spent a lot of our time in our conversation, talking less about like the specific data he found and what it showed, and more about like the problems with the data itself. This is not high quality evidence in the same way that you would want, say, to have a new drug approved. It's just not. What's wrong with the evidence? That is after the break. It's all about you. and when you fly with Virgin Atlantic in their upper class cabin,

they take the VIP treatment to the next level. With a private wing to check in in your own security channel at London Heathrow, you can glide from your car to their clubhouse, a destination in its own right in 10 minutes or less. On board, you can treat yourself to your own private suite to stretch out in with lots of storage space, a lie flat bed, and delicious dining from beginning to end. Just be sure to leave room for dessert.

their mile high tea with all the little cakes and sandwiches is a showstopper. Go to virginatlantic.com to learn more. Square knows that in hospitality, efficiency is everything. That's why their system lets you take payments. Track sales, handle inventory, manage staff, send invoices, and keep up with finances all in one place. Fly through orders with zero mistakes. Get the data you need and keep everything working together.

So you're ready for whatever's next. Learn more about their customizable with plans that, Up.com. The moment you've been waiting for is here. GMC's truck month is on. For a limited time, get 0% financing for 72 months on the 2026 GMC Sierra 1500 crew cab pro graphite. Feel the strength of GMC Sierra's 5.3-liter V8 engine. Elevate your confidence with a factory 2 inch lift and off-road suspension. Ready for whatever lies ahead. Power, capability, confidence, all at 0% during GMC's truck month. Don't wait. Visit your local GMC dealer today and make it yours. Dylan Scott. Before the break, you basically told me that our data on alcohol is bad. Why is that?

Well, it might be most helpful to start with like what would high quality data even look like. And in an ideal world, what you would do is set up a randomized control trial. Like you'd have two groups of people. and one group would drink and the other group wouldn't drink, and you would follow them both over time, and you'd track what happened with their health. And at the end of this trial, you would be able to measure what the health effects of alcohol actually were. And that is the gold standard that we use

for approving pharmaceutical drugs, for example. I can see why it would be pretty impossible to get like a thousand people, or whatever into a randomized control trial and basically be like, you have to drink two bottles of IPA every day for the rest of time. And, you know, you can't drink any alcohol, but you have to drink two bottles of orange juice or something. Like, that's not realistic in the world we live in. No, no, exactly.

So instead, we have the setup we were describing before the break. We have these observational studies that just ask, people about their real lives, you know, how much alcohol they drink, and, you know, we try to follow them over the course of their lives just as they unfold naturally, try to deduce some health effects from that observational data. But that kind of study, it has some pretty big problems. And I'll just try to do a thumbnail sketch about what most of the evidence is about and what its problems are, and you can, like, edit it out of your thing if you want to. This is Tim Namy.

He is the director of the University of Victoria's Canadian Institute for Substance Use Research. He has been doing research since the 1990s, and at first glance, it would seem that he fundamentally disagrees with our first researcher Ken. Like, Ken is out here saying, there's a possibility that a small amount of alcohol might be good for your health. And Tim thinks the opposite. The science is very consistent for 40 years that like when it comes to alcohol, that less is better. He is basically saying there is no amount of alcohol that is good for you. It's not that people shouldn't enjoy alcohol or shouldn't drink alcohol. It's just that there are many things in life that we do that come with some risks.

And it's okay, right? But it's just like, I guess as a scientist and a public health person, I like to call BS when I think there's been a bit of BS. And I think we've seen over the past 20 years that that kind of health halo around, particularly wine and particularly so-called moderate drinking, has kind of fallen away. At no level of consumption, was there a significant protective effect? When we talked to Tim and he was explaining to me why he doesn't think that glass of red wine

has any kind of health halo, I realized that he was actually saying a lot of the same things that I was hearing from Ken. Because, you know, as it turns out, they're both looking at the same kind of data, and they both agree on this key central point. I think the evidence base is pretty crappy. In fact, you know, a lot of Tim's argument for why he's calling BS on the idea that a small amount of alcohol could have some health benefits is because of his frustration with the data sets that we're drawing on. He says that, for one thing, a lot of studies out there aren't asking questions that are specific enough. Most of what informs our information about the relationship between alcohol and, say, heart disease is based on what we call big cohort studies.

These are studies that are designed to ultimately study like heart disease, cancer, but they weren't really designed with alcohol in mind. Why is that important? It's important because typically they're only asking people at one point, like at the beginning of the study, for example, like, okay, how many days a week do you drink and how much do you drink on the days you drink? And then from that, they calculate an average consumption. We know there's a big health difference between one drink every night versus having seven drinks in a week and they were all consumed within two hours on Saturday night. But, you know, depending on the way the researchers phrase their questions or just the information they choose to collect, in theory, both of those could be recorded as I had seven

drinks this week. And that imprecision makes it hard in general to interpret some of the data we have on alcohol and its health effects. That does seem like fixable, though, right? In the design of either future studies or current studies? I mean, sure. You know, we could design a perfect survey question to ask about people's alcohol use. But even if we did, there are other things that are not so easily fixed,

like the lack of randomness in studies like this. Because, you know, you're not, when you're not randomizing things truly, like when you're just having people report how much they're drinking and then letting them go about their lives, there are a lot of ways that the cultural context in which they live could affect the data that you're getting. You know, we could go back, I think, to like the idea that moderate drinkers are seeing some positive health effects to their drinking.

As Tim pointed out, you know, moderate drinkers actually tend to fit a certain profile, and they share other things in common. The problem is the so-called moderate drinkers, especially in developed countries, tend to be very socially advantaged. These people also have higher education, they have higher incomes, their children smile at them, they go to the gym, they drive BMWs. I mean, this is a big problem, right? So basically, Tim's argument here is that maybe it's not that little bit of alcohol consumption that's good for you. It's that the people who tend to have lots of other healthy things going for them in their lives tend to be the ones who are drinking only a little bit. And that might kind of bump up the numbers, you know, showing positive health effects from their drinking. So it makes it appear like, oh, moderate drinking, you know, gives you straight teeth, and it gives you a good job, and it gives you a BMW.

But in fact, it's the BMW and the nice job that lets you buy a nice bottle of wine. And on the flip side, you know, he said we also need to be really careful about how we think about and study non-drinkers. Those people are actually in many cases, not never drinkers, but are in fact former drinkers. Former drinkers, we know, are very unhealthy. They tend to be unhealthy. And it may be because they drank a lot or had bad effects of alcohol and they stopped drinking. But what we know about former drinkers is, for a variety of reasons, they tend to not do well. Some folks might not be drinkers at the time of the study, but they might have been heavy drinkers before the study even started. But at the time of the study, when the researchers are collecting all this data, they would report that they actually don't drink. So this could really influence the results that this study ultimately reports.

And I've seen, you know, other studies have pointed out that like some of the people who are really, really heavy drinkers, they die at such a young age that they don't end up getting included in these studies. And so, you know, all of these things are the problems with observational studies and research. relying on observational data to draw firm conclusions about the health effects of different things. Unless you're controlling for all of this stuff, it becomes really hard to isolate for the specific health effects of alcohol. But like, my understanding is that you can find ways to correct data for stuff like this, right? Like, if you ask people who join your study, did you or have you had a problem with alcohol, can't you kind of control for all those factors,

I guess, or like wait the study accordingly? I mean, you can certainly try. And, you know, Tim Namy is a co-author on a recent really big meta-analysis of all the prior research of cardiovascular effects from alcohol use. And that's exactly what they were trying to do. You know, they were trying to slice and dice this data to account for all. kinds of different things, you know, gender and age and race and even this abstainer bias that they were cognizant of. But, you know, there's an art to doing that kind of work. You know, different

people could choose to weigh different variables to different degrees. And that's what leads to some of these differences of opinion that we've seen between different scientists. Is there any hope of getting better data? Like, I don't know, I personally would like, like to have better data on this thing that's pretty ubiquitous in modern society. Yes, I agree. And there are researchers who are trying to use things like genetics to improve the data we have. You know, they're studying people who have a genetic tendency to drink or not to drink and looking at their health outcomes. And then we've got other researchers who are trying to set up studies to measure the specific biological mechanisms that might be.

I'd explain some of the benefits or, you know, negative consequences that we're seeing. And so, you know, I think that would make people feel more confident at least that, like, okay, it's actually the alcohol that's doing something here. And it's not just that, you know, these people have to happen to have certain drinking patterns or their health outcomes are actually being determined by, you know, having a BMW or their kid is smiling at them, whatever it may be. But, you know, overall, I mean, I thought, look, I thought Ken made a compelling point that even though this is hard, even though the answers may not be perfect, we could and should try to do better. At least for me, the fact that something is hard is not a reason not to do it,

especially when something as important as this, right? A lot of people drink alcohol. Over half of Americans are drinking alcohol. How is it that we don't know exactly what the hell this stuff does if half of Americans are doing it? And the fact that it is hard is a cop-out answer, I think, for not having good information. Were you surprised by this at the end that we don't have better answers? You know, it honestly, it simultaneously did and did not surprise me.

You know, I've reported on health for a long time, reported a lot on how people's behavior affects their health. And this is a recurring theme that, you know, understanding how people's behavior affects their well-being is really difficult. With caffeine or chocolate or sugar, you know, these substances that we're consuming all the time and we want to understand what they're doing to our bodies, we've run into this problem again and again, where it's really hard to set up a proper randomized trial that would really give us the answers that we want. And so instead we're left with this imperfect observational data that leaves us a little bit uncertain about exactly what conclusions. We should be drawing with it. What should people do in the face of this imperfect data?

Should they start drinking? Should they stop drinking entirely? I think the most important thing is that people just be aware of the facts that we've been covering on this episode. Because as we've said, the researchers who we talk to, we're all in agreement that once you're drinking heavily, according to the scientific definitions, you're inviting some pretty serious health risks. And I don't think anybody would say, you know, you should start drinking because it's going to be good for your health. But I don't think anybody we talked to would say, like, absolutely everyone, you know, we need to go back to prohibition. Everybody should stop drinking. Like, I myself enjoy drinking.

But I am approaching 40 now. I write about health care. I've been following this research. And I've, you know, I do try to moderate my drinking even more these days because I am aware of all this. this stuff. But I still get to enjoy that glass of wine with dinner or have a beer when we're grilling out on the back patio. I try don't worry about it too much. It's really, I think, about just informed consumption. People should be informed. They should know what they're getting into. And beyond that, we just got to make our own choices. Dylan has a lot more reporting on

alcohol and health. He has especially been digging into the influence that industry has on what we think about moderate drinking and its health effects. So we will link to his recent piece on that in the transcript. And you can hear him on Today Explained tomorrow talking about the alcohol research that RFK Jr.'s Health Department has buried. In the meantime, this episode was produced by me, Bird Pinkerton. It was edited by Julia Longoria. Meredith Hodnott runs our show. Noam Hassanfeld makes our music. Christian Ayala did the mixing and the sound design for this episode, Melissa Hirsch checked the facts. Jorge Just is a wonderful fact. And I am always, always grateful to Brian Resnick for co-creating the show. If you have thoughts about this

episode, we are at Unexplanable at Vox.com. And I love hearing from you all. If you want to support the show, perhaps, and help us keep making it, please join our membership program. It is at Vox.com. slash members, and you will get ad-free podcasts. You'll get unlimited access to Vox journalism, and you'll be supporting this show. In fact, if you sign up because you love Unexplainable, please tell our bosses that. It would make a very big difference. You can also support us by leaving a nice rating or a review, or just by telling people in your life to listen to the show. Unexplainable is part of the Vox Media Podcast Network, and we will be back. on Wednesday.