Hello. I’m Kelly Thomas. I am the director of scientific content at TEDMED, and I am joined today by Baku Jenna. Babu, please, share a bit about your background and the exciting things you’ve been working on since we last saw you for your TEDMED talk a few years ago.

Yes. It’s been a long time. My name is Babu Jena. I’m an economist and a physician.

I’m a professor at Harvard Medical School, and I see patients at Massachusetts General Hospital.

And I host a podcast called Freakonomics MD and wrote a book with Christopher Worsham called Random Acts of Medicine.

Yeah. Bapu, I loved reading your book and just the simple way you explained, natural experiments, and I think it’s informative and exciting to read for people in medicine and and not in medicine. Curious to know what steered you towards, like, applying scientific method to natural experiments. Maybe you can explain a little bit about the difference between a natural experiment and a randomized clinical trial to reveal these patterns in health care that helps us understand what may work and what may not work.

So I’m an economist and a physician. And in medicine, the the typical way that we decide whether or not something works or doesn’t work and we decide whether or not to use that in in people to treat disease is a randomized controlled trial. You take, a bunch of people and you randomize some of them to get one treatment and the other group to get a different treatment or it could be even a placebo. And that allows you to figure out whether something works to figure out the causal effect of that treatment.

And in economics, in in in sort of the area that I work, which is health economics, it’s sometimes hard to do these sorts of randomized controlled trials for a lot of different reasons. And so we often try to do is find situations where people are exposed almost by chance in the real world, by nature, to one path of care versus another. And we call that a natural experiment. So it’s not a randomized experiment at the hands of an investigator, someone who’s running a trial, but it’s essentially nature saying, alright. This group of people were kind of exposed randomly to one path of life or one path of care and the other group to a different path.

Mhmm. When we saw you for your TEDMED talk and also in your book, one of the natural experiments you looked at was what happens to emergency care on days of marathons for runners and nonrunners, especially people who are going into cardiac arrest or having a heart attack. And, you know, for me, that’s a pretty unusual set of events to put together. I’m curious to know what sparked that idea and and how did that unfold.

You know, when when, a few years ago, we had this study. It was published in the New England Journal of Medicine, and it was titled, I think, Marathons and Mortality. And if I were to say those words together, you might think, alright. People who are running the race keeling over in the street because it’s hot outside or they’re exhausted, whatever it may be.

And in fact, if you look at people who run marathons and you measure their cardiac biomarkers after running a marathon, those levels are elevated, almost mimicking the pattern of what you’d see if someone actually had heart injury. And it’s not surprising because marathons are really difficult to do. It’s like a difficult thing for the body to do. In a few years ago, my wife was running a race.

It wasn’t a marathon, but it was a race that started in one part of Boston, went, nearby the hospital where I work, and then to a different part of Boston, and then went all the way back to where it started. And she wanted me to watch her on that race route, and that’s what I talked about in that TEDMED talk a few years ago.

And, I couldn’t get to the hospital to park, which is what what I was planning to do. And I told her that a few hours later and she says to me, well, what happened to everybody that needed to get to the hospital that day? And that was sort of an offhand comment that she made, but it made us curious as to what happened. And and what we found was that if you look at people who live near marathon routes, on the day that a marathon is being performed in a city, those people, the elderly people, so these are people above the age of sixty five, they are more likely to die from cardiac arrest or a heart attack. And the reason why is because they just can’t get to the hospital on time because the roads are blocked.

What are the implications of this study? Should cities that are hosting marathon just be aware of it or is there something, something more?

Yeah. That’s a good question. So I think they should definitely be aware of it.

You know, in our city, Boston, where I live now, there was a Taylor Swift concert a couple of months ago. And, it was in Gillette Stadium where, the Patriots play, and it was know, it’s hard to get out of the stadium. I didn’t go to the concert, but I’m told it was really hard to get out of that area. And you could you could certainly imagine people who are older who live in that area might have had difficulty getting to the hospital if they needed to go.

And so the solution here though is not to cancel marathons, is not to cancel Taylor Swift concerts or July four celebrations, whatever it may be. The solution is is two things. One is to to make sure that people and and cities are aware of this problem. I think to some extent, emergency medical services are aware.

But when we did our research, we talked to a bunch of EMS, emergency medical service agencies, and many of them were not aware of this particular issue. They’re certainly aware of the problem that people who are attendees at these events might have medical problems, and so you need to be able to respond quickly for those folks. But they weren’t really thinking about what might be the spillover effects on people who live in the nearby communities who might not be able to get to the hospital as quickly as possible. So I think that’s the first implication is just to be aware.

So if you live at home or in a nursing home and it’s a marathon day and you’re having chest pain, you probably wanna get seen earlier or make a call earlier than you otherwise might have just recognizing that there might be delays. Mhmm. I think there’s a second implication which is is quite a bit broader, which is in medicine, one of the biggest questions that we struggle with and really in in life, if you’re a parent or have a loved one who has any medical problem, is how quickly do you need to act? So, you know, we’ve got two kids.

And if it’s two AM in the middle of the night and one of our kids has a fever and a headache, I’ve got to think to myself, do I need to bring them into the emergency department right then? Do I need to call the pediatrician right then? Can I wait until morning, give them some Tylenol, see how they respond?

Or if you’re a young intern working in the hospital and a nurse calls you and says the patient in in room thirty four is short of breath. You’re gonna go there immediately, but you need to escalate that care within minutes, within thirty minutes, maybe an hour. Maybe you could see them for the rest of the day, see how they turn out. That question is hard to solve because you would never conduct a randomized trial where you said, look, we’ve got a thousand people with chest pain.

Five hundred of them we’re gonna have seen immediately, and five hundred we’re gonna say, alright, you know, listen to a podcast and and show up to the ED in, you know, thirty minutes or two hours or whatever. You would never do that trial. It wouldn’t be ethical. And yet, it is important for us to know how much it is the case that delays in care matter and for what kind of conditions they matter.

And the marathon study showed us that there is a situation where people were, by chance, exposed to delays in care because in their city on a given day, there happened to be a marathon. And while we looked at particular cardiac conditions, you might look at all sorts of other medical conditions to see how responsive health outcomes are to small delays in care and how that varies by the type of medical problem that you have.

Mhmm. Makes so much sense. And one of the other natural experiments looked at, COVID spread during, birthday celebrations, especially when there are worse small children in the household. And as you know, as well as I do, that birthdays are a big deal. And kids want to celebrate, and so do the parents because it is really a marker of parenthood as well.

So can you just give some color around that study?

Yep. And by by the way, the TEDMED talk that I that I gave a few years ago, it was it was right at the start of the pandemic. And I remember we were being told, this is maybe February or something like that, not to shake people’s hands, but instead give elbow bumps and, you know, foot bumps. That didn’t that didn’t slow the spread of the virus, unfortunately, but it’s it was a good try. Right. But, you know, a few months well, almost a half year later, we were in the throes of the pandemic.

And our daughter was turning, I think, five or six at that time. It was in in the in the wintertime.

And we were thinking about whether or not to do a birthday party for her in person with people that we really knew very well and trusted or to do it via Zoom. And we ultimately ended up doing it by Zoom, and we had a magician perform a magic show virtually, which was actually a real, a real hit.

So and it’s more than I thought would be possible, but, certainly came through. But it got me thinking about this idea that people had been talking about at the beginning of pandemic quite a bit, which is what is it that led to disease spread? You know, was it these large super spreader events, you know, large conferences, conventions, get togethers, meeting people in bars and restaurants, people that you did not really know very well? Is that the driver? Or was it really more small social gatherings even with people that you might know and trust and who you think, alright, these folks are being pretty responsible. They’re not going around much. They’re they’re wearing masks, etcetera.

Maybe we can get together, in a safe way. And that’s a hard question to study. Why? Because you need lots of different types of data to do that.

You need to have information on a lot of people, and you need to know whether or not they’re gathering or not gathering, when that’s happening, if at all. You need to know whether or not they develop COVID nineteen at some point in time after they did or did not gather. And then you’ve gotta solve this thorny empirical problem, which is that gathering isn’t random. The people who chose to gather at that time were different than the people who didn’t.

So is it the gathering that led someone to get COVID nineteen or was it a decision not to mask or to fly or whatever else it might be? And the way we got around that was really based on the birthday.

The idea was that birthdays are something that people want to celebrate. That’s the first insight.

And then the second is that data that we typically use for my kinds of research in health care have information about someone’s birthday and also diagnosis of COVID nineteen. And what we showed was that if you look at we looked at hundreds of thousands of households across the US. And if we look at households in the same city in the same week, those households in which a household member had a birthday were more likely to have a COVID nineteen diagnosis in the subsequent several weeks compared to otherwise identical households in that same city in that same week where no member of the household had a birthday.

And birthdays are random, so this is a natural experiment. Right? It’s not like COVID is coming at a particular time when someone happens to have a birthday. If that happens, it’s because people were gathering and increasing the likelihood of getting COVID nineteen.

And so the first thing that we showed was that these small social gatherings with presumably people that you know and trust were, perhaps leading to COVID nineteen infections.

Second finding, which is quite interesting, and it goes to your point, Kelly, was that the effect was larger when a child in the household had a birthday compared to when an adult in the household had a birthday. And it makes sense because it’s easier to say, you know, I’m an adult. I don’t need to have a birthday party. It’s harder to say that for, for a child. And that’s what we found.

And then the last thing which I thought was particularly interesting was that this sort of birthday effect, if you will. The birthday effect was equally large in highly blue and highly red parts of the country. So even at that point in the pandemic and even now where there was a lot of sort of political polarization around what people in different political parties were, quote, unquote, willing or not willing to do to help slow the spread, of the pandemic. Even there was a lot of rhetoric around those issues.

If you look at actual decisions, not what people say, but what they did, what we found was that the birthday effect was equally large in highly Republican versus Democrat areas of the country. And to me, that said two things. One is that you have to look at what people do, not what they say they’re going to do. And this is a a principle that economists feel quite strongly about.

It’s why we we try to look at actual behavioral data as opposed to to survey data. And the second thing was sort of a broader point, which is that, you know, when we think about the pandemic, you know, we kinda think about all the differences that people had and the way they thought about how to address the pandemic. But the unifying feature across everyone is that there’s not a single person who didn’t pay some sort of price. Right?

And we focus a lot on the differences as opposed to the similarity of that shared and and difficult experience. So I think that was sort of an interesting takeaway for me from that research and then also, you know, now having lived and worked through the pandemic.

I’m so glad that you’re exploring these types of experiments and and, teaching everyone else to think about them as well. And I really appreciate you sitting down with TEDMED today, and we hope to speak to you again.

Thank you.