When David Epstein was a sophomore-year varsity runner at Columbia University, his former training partner, a few years his junior, collapsed and died suddenly during a high school track meet.
"Normally, a fallen runner draws only slight curiosity from a track-savvy crowd," Epstein writes in his new book The Sports Gene. "But Kevin was a state champion, and the dusty, green rubber floor of the track was no place for a state champion to be lying on his back, shuddering." Kevin Richards, it was later discovered, had a genetic disease called hypertrophic cardiomyopathy—the most common cause of natural sudden death among athletes.
As Epstein tried to make sense of Kevin's death, he also became "really, really curious": Were DNA patterns truly so powerful that they could seemingly preordain some athletes for superstardom while they destined his friend for tragedy?
Now a senior writer for Sports Illustrated, Epstein has been investigating the effects of genes on athletic ability ever since. His book examines whether it's genetics or training that most affects who becomes a sports legend and who becomes a recreational player, and closely inspects common, often misconstrued theories about gender, race, and just how many hours it takes to master a craft.
Epstein talked to me recently about why Usain Bolt is so fast, a little-discussed reason so many boxers and football players end up with brain injuries, and why it's dangerous to believe that intellect and athleticism—brains and brawn—are mutually exclusive.
Responses have been edited for clarity and length.
You've written about sports medicine and sports science before. At what point did you say to yourself, "Well, this is becoming a book"?
I started thinking about these things, not quite in genetics terms but in nature-vs.-nurture terms, back when I was in high school. We had a big Jamaican population in my high school, and that made track and field a really popular sport. That phenomenon made me curious. And later, I was running in college at Columbia against a lot of Kenyan guys—who I discovered were largely from the same part of Kenya. And since I would do the same training as the guys I was running with, I would have expected that we would all cross the line in the race at more or less the same time. But it wasn't like that at all: Instead, it would get more and more different. I started to marvel at that, and I got interested in training theology and those things.
Then, my sophomore year of college, my former training partner dropped dead during a track meet. It turned out he had this genetic disease related to sudden cardiac death, and that's what got me into writing about sports science in the first place—wanting to write about sudden cardiac death in athletes.
Your book lays out a lot of heavy-duty science, but what's so effective about it, I think, is that it also puts a memorable, sometimes already-recognizable human face on it. You talked to so many athletes and read up on so much research—which one of these cases you studied taught you the most?
Ooh. The contrast between Donald Thomas and Stefan Holm, the high jumpers, probably—that really stuck with me.
Stefan Holm is this highly dedicated Swedish high jumper who had started inching up year after year after year until he was a world champion and an Olympic gold medalist. I think most people think of jumping as a thing you either have or you don't, but the degree to which he transformed himself, I didn't know was possible. He was this epitome of the "nurture" approach. He had some talent innately, just not as much you'd have thought.
And then to see him get beat by Donald Thomas in 2007, a guy who doesn't even really like the high jump, who had only been doing it for about a year, but who's just got the right genes and the right build to respond incredibly well to training. It just made me realize that there are so many different paths to the same basic physical outcome.
One of the most intriguing things you talk about in The Sports Gene is why it seems like Jamaican runners just seem to be really fast.
So there's a nature and nurture aspect to Jamaica. Early on, I just looked back through pictures of the Olympics and it dawned on me that, in the last couple of years, Jamaicans are ludicrously overrepresented in world-class sprinting—and especially because of Usain Bolt, more people have realized. It's a small country, so to see them dominate world sprinting is just kind of unbelievable.
As I started tracing some of the backgrounds of these athletes, where they were coming from, it was interesting: These athletes of Jamaican descent were coming from many different countries—the Netherlands, Portugal, Nigeria. But their ancestries were always traceable to what's actually a small region of Western Africa, the Bight of Biafra. It has some of the world's highest malaria rates. African Americans and West Africans who are from—or whose ancestors are from the same small area in West Africa where Jamaicans were captured and then taken from in the 17th century, these incredibly high malaria zones—are actually incredibly under-represented in endurance events. These athletes are really over-represented in explosive running, but their national marathon records are awful!
Turns out that West Africans and people of West African descent have low hemoglobin—low levels of this protein that carries oxygen that's really important for endurance running.
For a long time, it was thought to be a nutritional problem: They weren't getting enough iron, some people thought. Some of that was because doctors often looked at basic medical norms that came from white people and compared everyone to those, so that would have implied that if you didn't match that, something was wrong with you. But the fact is, people from malarial regions just have lower hemoglobin. It's a genetic adaptation: People with lower hemoglobin are less likely to get malaria.
There was some evidence, too, to suggest that because they have less capability to produce energy with oxygen, there was a shift in their metabolism, in their forms of energy production. And another part of it is that people with ancestries that predispose them to having long limbs, that's better for running. All other things being equal, having longer legs relatively helps a lot.
But today, I think the particular difference with Jamaicans is this: They have this population that has a lot of talented individuals, but they also have this amazing town sprinting program. It's very much like what we do for college football in the U.S.—we have people who are out of their minds over their local college football team, over their school. Only there it's high-school track.
When I went to the National High School Championships in Jamaica, I started talking to the coaches who were there at the warmup track. I kept asking them about their recruiting. I kept asking, "Well, what do you do? Are you allowed to go meet their parents? Do you talk to the kids?"
And they kept saying, "We're not allowed to give their parents refrigerators." I was like, I didn't even ask about that!
Later I learned that coaches were bribing parents with refrigerators to get the fastest kids—starting when they were, like, 10 years old—to come to their high schools. So it's very, very difficult for a fast child to slip through unnoticed in Jamaica. They make the most of their talented kids.
You mentioned that within the field of genetics, sometimes race is blatantly left out in the interest of political correctness. You wrote in your book that some researchers will "do backbends" to not talk about race.
But you don't do that. You go there. At what point did you decide to just go for it?
I lost so much sleep over this. I literally almost backed out of writing this book, because the issues of race and gender got me so nervous. Eventually my agent and one of my colleagues convinced me to just do it, in the best way I could.
But I remember being at the 2012 American College of Sports Medicine Conference, talking to the head of the physiology department at a major research university. The head of the department was telling me that he had data on ethnic differences in response to a certain dietary supplement during an exercise program, and that he would never publish it. He didn't want to get into that issue.
I heard this a number of times: He was worried it would be extrapolated into saying somehow that there were also innate intellectual differences between black and white people. When I heard that, I said, "That is a huge problem. That means science could be disappearing into the filing cabinet, into the garbage cans, because people aren't willing to take this on."
And that's when I thought, I have to do this. I'm not going to do that same thing and leave it on the cutting-room floor.
When people ignore the fact that low hemoglobin is a genetic adaptation to malaria and that West Africans are low in hemoglobin, they find that when they start giving them iron supplements, they start dying from malaria. So to ignore racial differences, sometimes that can have a really messy effect in itself. Ultimately that's when I realized there were certain genetic themes within ethnicities that I would really be remiss to not discuss.
And in the book, before I even started talking about the phenomenon of black athletes being over-represented in certain sports, there's a certain section where I just discuss whether race has any sort of genetic meaning in the first place. I really felt obligated to have that section, because A., I thought it was really interesting, and B., I was not happy to be told that people were not publishing data that they had because they were worried it would be extrapolated into intellectual differences between races. Which has nothing to do with the work that they're doing! Nothing.
Not to mention the fact that that idea—of, like, a biological teeter-totter, that athleticism is somehow inversely proportional to intelligence—isn't the cause of bigotry, it's a result of it. That notion didn't even exist until it was associated with African-Americans, when they started dominating certain sports.
You wrote a few chapters about the present and the future of sports medicine, specifically about how now we're able to diagnose risk factors before they become deadly. In some cases, at least. And I was shocked at what you discovered about brain injuries in boxers and football players. You discovered something that news agencies had sort of glossed over.
I've written about it three or four times and it never seems to get any traction whatsoever—but it's been known for quite a while now that this gene called ApoE4. First it was discovered as a risk factor for Alzheimer's, and if you have one version of the so-called "wrong" copy, which is ApoE4, you're at increased risk for Alzheimer's, threefold. If you have two copies, about an eightfold risk.
It was first discovered in the mid-1990s that this gene is sort of—I don't want to overstate it too much, but it's like this master brain-injury recovery key. And it's involved in all manner of recovery from any trauma, so people who get in car accidents are more likely to die, or more likely to have brain bleeding and less likely to recover, more likely to have post-injury seizures, if they have a copy of this ApoE4 gene. And now all the data today shows that the same kind of head trauma that's in the news all the time now for sports, people with the ApoE4 gene don't do as well with it.
The very first brain that researchers dissected at BU was a guy who had two copies of the ApoE4 gene. Two percent of the population has that. And one thing that was really lost in the headline when they came out with a study a few years ago about all these brain injuries in boxers and football players with Chronic Traumatic Encephalopathy, was the overrepresentation of the ApoE4 gene. So clearly, now, I think it's indisputable that this gene is overrepresented among people who get brain damage from getting hit in the head.
That's something we talk about at The Atlantic pretty frequently, and often agonize over—the ethics of supporting or promoting sports that can be so deadly later in life. It seems like more attention to this genetic disposition could change that conversation.
It frustrates me that we haven't paid that much attention to this part of it. All the way back in 2002, the chief medical officer of the New York State Athletic Commission was thinking about having all boxers take ApoE4 screenings. They decided not to do that, for a number of reasons—one of them being that there was no protection for genetic information at the time. And still, genetic information isn't perfectly protected, but it's better. He was thinking about doing it back then, and nobody paid attention to it.
I'm not saying that I should be the one to tell anyone what to do, but if somebody knew they were at increased risk of getting dementia playing a contact sport, I would say, "Hey, you know, maybe track is a better option!"
So when I was asking doctors why we aren't offering this to athletes, they said, "Well, basically, the thinking in the genetics community has been twofold. One: It's just predisposition risk. You either have this disease or you don't, and people have difficulty understanding that. And two: There's nothing you can do about it."
I've said, "Well, it's a risk factor. You tell people that smoking is a risk factor," and the doctors' response is, "They can stop smoking. They can't change their DNA."
And I've said, "Yeah, but they can choose not to play football! Or not to be a boxer!"
It frustrated me that the medical community was sort of thinking for the patients. It's true that if you learn that you have the "bad" version, not only are you at a greater risk for brain injury when you get hit in the head, but you're also more predisposed to Alzheimer's. You can't find out one without finding out the other, and some people might not want to find that out. But that should be a choice, I think. When I went around asking athletes, most of them told me, "Absolutely! I wouldn't hide from that information." None of them had any problem understanding that having the gene didn't mean they would get CTE. Only a small number told me no—most said it wouldn't make them retire or anything, but they would rather know than not know.
Now, there's a small number of people they're actually testing. They're offering it to parents and such. But it's still quite rare.
To what extent is that going to become a thing athletes and parents of athletes do routinely in the future? To what extent is that something they already do routinely?
Well, now there are these direct consumer genetic-testing companies, and they've now started to offer testing for that gene. I think that will grow quite a bit—although there are a lot of questions about how to regulate it.
But now these companies and this whole field of genomics are starting to offer testing for predispositions related to things like hearing, tendons, and ligaments, I think that's going to expand, and we should be having those conversations now, because there will be tricky ethical issues there.
The kind of testing that I don't think will expand or be that useful is people testing their kids to see if they'll be some sort of superstar. We're still so far off from knowing how to deal with that knowledge—at this point, you're so much better off just finding out by having them just try to play the sport.
In some cases, at least. For things like hypertrophic cardiomyopathy—the genetic disease that's the most common cause of sudden cardiac death among athletes—that testing is already there now. That just needs more awareness. If someone in your family drops dead, it's usually offered—or if someone goes into cardiac arrest and is revived, you can take their DNA, find out if they have the known mutations, and then you can test the rest of their family for it. Then instead of the rest of the family wondering if they're going to drop dead suddenly, or seeing a cardiologist every couple of months for the rest of their lives, they can know that they have it or they don't.
If they do, they get defibrillators—and that's spreading already now. It could still use more awareness. The only thing worse than losing one child in the family would be losing two—and that happens so often.
You mentioned in the book that there's a soccer player with a defibrillator who had a cardiac arrhythmia during a game a few years ago, keeled over, and then just popped back up! That's amazing.
Is that what the future of sports will look like? What's the next frontier of genetics in sports?
Not to jump too far ahead, but a really interesting expanding area concerns what things you can do to turn certain genes on and off. Broadly, that's called the science of epigenetics—you can't change your genes, but you can sort of affect how they respond to training, to diet, et cetera.
But that's way far into the future. I think genetically tailoring your diet is already a thing; for example, one of the doctors I talked to for the book also does genetic testing for retired athletes. And he knows who'll respond better to fish oil supplements for brain health, things like that. I think we'll start to see more of that.
One other thing that's happened organically is individualized training. Great coaches sometimes do this intuitively: They have these intuitions about genetics, in that they recognize that certain athletes are responding to certain types of training while others aren't. There's one scientist I talked to who actually takes biopsies, so that he has the muscle-fiber types of athletes—percentages of fast-twitch and slow-twitch, stuff like that—and develops training regimens for them accordingly. Most athletes are recreational and won't be getting muscle biopsies, but hopefully we can start paying more attention to what individuals respond to.
We all have different genomes, so for optimal training, we all need a different environment. Coaches are starting to realize that now, that they need to have individualized training if they want to get the very best out of an athlete. That will maybe start to spread down to lower levels—hopefully not to the point where people are pushing kids to get too crazy and stressed out by it. That's bad. But I do think we're going to increasingly see science point to the fact that there's no cookie-cutter training method that works for everyone.