July 25, 2019 — In this week’s episode we’re talking about new research comparing the health of athletes in the National Football League and Major League Baseball. The study looked at 6,000 athletes between the years of 1979 and 2013. During that period, there were 517 deaths among NFL players and 431 deaths among MLB players, translating into a 26% higher mortality rate among football players compared with baseball players. The findings showed that while NFL players died of neurodegenerative diseases at a higher rate than MLB players, both groups of athletes were more likely to die of cardiovascular disease than brain diseases.
The study was led by Marc Weisskopf, Cecil K. and Philip Drinker Professor of Environmental Epidemiology and Physiology at the Harvard Chan School. The research comes amid increasing concerns over the effects of repeated head trauma on the health of NFL players. And while the study did show a difference in death rates, it’s still unclear exactly what’s driving that disparity—and how to address it. We sat down with Weisskopf to discuss the research and the unanswered questions he’s hoping to answer in the future.
You can subscribe to Harvard Chan: This Week in Health by visiting iTunes or Google Play and you can listen to it by following us on Soundcloud, and stream it on the Stitcher app or on Spotify. If you’re a fan, we’d appreciate you leaving a rating and review wherever you listen.
Pro-Athlete Mortality Gap (Harvard Medical School)
NOAH LEAVITT: Coming up on Harvard Chan: This Week in Health…
Pro sports and health.
MARC WEISSKOPF: The first step is really to say, does a certain occupational group have a higher risk of some outcome. So we’re sort of at that stage now where we’re trying to say, if we compare to people who otherwise seem more similar, does it still seem like there’s something different about the football players. And that’s really what our paper gets at now, as it says yes, there is something. If you compare athletes at the same sort of professional level, football players seem to be worse off in this regard.
NOAH LEAVITT: In this week’s episode: A new study finds that pro football players appear to be at an elevated risk of death compared to pro baseball players. We’ll explore the factors that may be driving this.
Hello and welcome to Harvard Chan: This Week in Health. I’m Noah Leavitt.
In this week’s episode we’ll be talking about new research comparing the health of athletes in the National Football League and Major League Baseball.
The study looked at 6,000 athletes between the years of 1979 and 2013. During that period, there were 517 deaths among NFL players and 431 deaths among MLB players, translating into a 26% higher mortality rate among football players compared with baseball players.
The findings showed that while NFL players died of neurodegenerative diseases at a higher rate than MLB players, both groups of athletes were more likely to die of cardiovascular disease than brain diseases.
The study was led by Marc Weisskopf, Cecil K. and Philip Drinker Professor of Environmental Epidemiology and Physiology at the Harvard Chan School.
The research comes amid increasing concerns over the effects of repeated head trauma on the health of NFL players.
And while the study did show a difference in death rates, it’s still unclear exactly what’s driving that disparity—and how to address it.
I had the chance to sit down with Weisskopf to discuss the research and the unanswered questions he’s hoping to answer in the future.
Take a listen.
NOAH LEAVITT: I thought one of the interesting things that you’re doing in the study is that previous research has compared to mortality among NFL players to kind of the general public, but in this study you looked at major league baseball players. So why was that an important comparison to make, to be comparing athletes to other athletes instead of the general public?
MARC WEISSKOPF: Right, right, sure. So yeah, so the prior– the few prior papers have come out suggesting differences between football players and the general population, which is certainly of interest and we need to know that. But the– one important issue when you do that is that the general population is, as one might imagine, very different from the type of person who becomes a football player.
So when you do see differences, it’s a little hard to tell exactly what’s going on. And what is typically found, and in much of what these other papers have seen, is that football players look better off than the general population. Now, that’s really not that surprising, since you obviously have to be a certain type of person to play in the National Football League or whatever professional sports league you are.
Not only healthier probably because of the physical fitness needed and the training you go through for to play those sports, but you may also just from the very beginning be a different kind of person. The general population includes people that are not healthy. And if you are not healthy, you’re never going to get into the NFL. So you’re kind of comparing apples and oranges a bit.
It’s certainly worth knowing that professional athletes, like football players, are at lower risk overall than the general population for, say, cardiovascular mortality. But to then ascribe that to the sport itself is difficult to do. In other settings, it’s known as the healthy worker effect where workers, despite whatever exposures they may have, often appear better off to the general public. But that’s because they have to be healthy to work in the first place.
NOAH LEAVITT: And is one of the challenges too like simply a matter of scale? Like there are so few people who make it to that level of the NFL. Is that– is that also a challenge, because they’re just– scales are different?
MARC WEISSKOPF: That’s– it’s less about the scale, other than that that’s a problem in that you don’t have a whole lot of people in the population you want to look at. So that population is a small one. And it– when you have small numbers, that can affect what kind of statistics you can do. But other than that, that’s OK.
If the general population were, in fact, comparable in other ways, that would be a perfectly fine comparison to do. And it’s, as I say, it still is. I mean, we know that football players are in many ways– they have a lower rate of dying than the general public. Exactly why is a little uncertain.
NOAH LEAVITT: And so was there a particular reason that you chose to compare NFL to MLB for this study, as opposed to, say, soccer players or NHL players?
MARC WEISSKOPF: Sure, yeah. So there were a couple of main considerations. I mean, first is the availability of data. So we need to have a list of these players and be able to track them and know what they die of.
The second is to have– again, this gets to your last point, a large enough comparison. So general population is huge. It’s a great comparison to make, because you have lots of numbers.
If you take a sport like, say, basketball, there aren’t– just aren’t that many players, so it really impedes your ability to look at differences. So MLB baseball players were a good option. One, because there are fantasy leagues and people track these guys like crazy. So we can get data on them, and it’s online even.
But the other is just that it’s a sport that has a lot of players. And it has a long history, so we have a big comparison group. That relates also to one prior paper looked at NFL players, career players versus replacement players during the strike years.
And that also is a better comparison than the general population. But it’s much, much smaller, because there just aren’t– weren’t that many replacement players. So the ability they– for example, they could only look at overall mortality, simply because the numbers were too small. So with Major League Baseball we had a much bigger population and one that was still an elite athlete playing professional sport.
NOAH LEAVITT: Was there anything– consideration too, I guess, I think what’s interesting is the difference between football is brutal, physical, and baseball not typically thought of as a super physical game.
MARC WEISSKOPF: Yeah, yeah. So I will say that is– that was another personal thing. So we’re interested overall in the health differences between these guys.
But in some of my other work I focus on particular diseases, one of which is Lou Gehrig’s disease, or ALS. And certainly with that one there’s some interest in the idea that head trauma may be related. And so I was also having that in mind with exactly what you’re saying, is that baseball players make a good comparison with that regard because they have some of the same physical fitness and other characteristics of football players but much less head injury.
NOAH LEAVITT: And so you just talked a second ago about, for example, like the availability of player data for MLB. So who are the players you study? What years were they playing in their respective leagues?
MARC WEISSKOPF: Yeah. So the beauty of MLB is that we really were able to find available even online literally every baseball player who played prior to the MLB. Even in the 1800s we had some guys, right?
So we could have used whatever we wanted in terms of the baseball players. The football players though, that data was not as available. And so we took advantage of a cohort that was constructed by NIOSH, the National Institute for Occupational Safety and Health.
And so we were restricted to how they constructed that group. And they constructed that group and it was players that had to have played after 1959, I believe. And they had to have had five career seasons. That– because of the vested them in things and that’s how they identified these players.
So in order to– and because of that then, that wound up restricting sort of the birth years of these guys. Because to fit those criteria, it wasn’t everybody. So essentially to be more comparable, we did the same thing for the baseball players. So we actually wound up cutting our baseball player pool way down to match the NFL cohort entry criteria that NIOSH had already set up.
NOAH LEAVITT: How big of a hurdle of this was just finding the data? Because you mentioned for MLB It’s so widely available, but NFL was obviously much of a challenge. So was that one of the just initial barriers, like we need to find the data, then we can do this study?
MARC WEISSKOPF: Absolutely, absolutely. I actually had meetings with Don Fehr at the MLB Players Union, had conversations with the NFL Players Association way, way back. Then discovered online that baseball and fantasy leagues are wonderful, and that I was able to source data that way.
That’s not as available for football. And that was a major hurdle for me. And that’s why it really wasn’t until this, an NFL paper came out using the NIOSH data that I thought maybe we could access that. And in fact, because NIOSH is federal, it can be accessed.
NOAH LEAVITT: What were your key findings, and where did you find significant differences between the NFL and MLB players in terms of death or cause of disease?
MARC WEISSKOPF: Sure. So the first thing is that what we found is NFL players have a higher rate of mortality than baseball players. So they are worse off overall.
Importantly, it looks like those differences are driven by in particular cardiovascular death and neurodegenerative death. And in the prior studies we’d referred to where they were comparing to the general population, they actually found that football players were less likely to die from cardiovascular disease in the general population. So our results kind of flip that on the head.
They also, of course, flip the overall mortality, which for football players is less than the general population but now more than baseball players. The prior papers that compared to the general population also saw a higher risk of neurodegenerative death, and we see that as well compared to the MLB players. So those were the main findings, that higher mortality in the football players largely driven by cardiovascular and neurodegenerative deaths.
NOAH LEAVITT: And so do you have a sense of– or I guess it’s probably hard to say directly what causes each, but you have a sense of what may be driving that those increases in deaths?
MARC WEISSKOPF: Right. So we have theories about that, but this is very similar to sort of standard occupational epidemiology, where the first step is really to say, does a certain occupational group have a higher risk of some outcome. So we’re sort of at that stage now where we’re trying to say, if we compare to people who otherwise seem more similar, does it still seem like there’s something different about the football players.
And that’s really what our paper gets at now, as it says yes, there is something. If you compare athletes at the same sort of professional level, football players seem to be worse off in this regard. Obviously, the next step is to try and figure out what that may be.
Now, we have thoughts about it. Obviously, the cardiovascular factors, football players have a lot of traditional risk factors for cardiovascular disease that we would want to explore more. I also do work with the football players health study, and as part of that we– cohort study we’ve started there, we do see that things like weight changes at different times are related to worse cardiovascular health later, certain injuries seem to predispose too. So we think things like weight, obviously, and BMI is a big one, and the types of injuries football players get at putting them at risk of this.
Obviously with the neurodegenerative, we’re particularly interested in the idea of head injuries. More work needs to be done to show that. This paper can’t say that on its own. But it’s something we’re certainly thinking about.
NOAH LEAVITT: So in other words, what you’re saying is you mean this study seems to be like we want to identify OK, it seems like they’re at this elevated risk. Now it becomes a question of why and then what do we do about it.
MARC WEISSKOPF: Right. And that’s partly because if we see there’s a problem, what do you, yeah, what are you going to do about it. So at the stage where we’ve done now, we can say, OK, well, let’s not play football and that would solve this problem. But that seems maybe a little extreme.
So the idea would be to figure out are there aspects of the game that might be contributing to this that are more amenable to change, that might preserve the game while trying to eliminate some of these problems. And so we can start to look at things in more depth, at like position, I mean, if it’s a weight issue, there’s clearly positions in the NFL vary by their body physique. And so we can look at questions like that. Obviously, the numbers start to get a little smaller, but these are the kinds of things we would want to explore.
NOAH LEAVITT: So that was my next question, would be kind of what would be some of the next questions that you would look to answer. And then how, I guess, what are some of the barriers there. I mean like you said, you might have less data available. So how do you kind of tackle those issues?
MARC WEISSKOPF: Yeah.
NOAH LEAVITT: No pun intended.
MARC WEISSKOPF: Yeah. So, right, exactly. Yeah, I mean, so key barriers are exactly that. As you start to delve into finer details of what’s going on in the sport, the numbers get smaller in terms of who you’re comparing to whom.
From the perspective of this paper and the type of data we have here, there’s only so far you can go, because we’re really only looking at mortality. That’s part of the rationale for doing something like we’re doing with the football players health study, which is to try and get a cohort of former players that we can get other types of outcomes from than just death. The other aspect that also is why the football players health study helps is really on the exposure side.
So there is data out there that we can find on who they played for, what position they played, what years they played. But you’d really like to go even deeper, and say how many snaps, were you a starter, were you not a starter, what kinds of injuries did you have, all sorts of details that you really can’t get from online or things like that. You need to develop a cohort and a questionnaire to get that kind of intricate details. But we can look at things of whether it’s certain patterns change over time because body physique has changed over time, sort of training habits have changed over time. So those are the kinds of things we would try and explore more.
NOAH LEAVITT: As I hear you describing that, I’m wondering is it helpful at all that baseball and football games are televised? So if you had a cohort, you could actually go back and watch the games and say, oh, this player had this many hits to his head, or–
MARC WEISSKOPF: Right, right. No, absolutely. And in fact, again, as part of the football players health study, not this paper we’re talking about today, but part of the football players health study, there’s work going on with some groups that do exactly that. They take detailed video breakdown of games and recreate sort of collisions and injuries and how these things happen to get forces and stuff like that.
So I think it does help. The only problem is that recently that’s been the case. Going way back in time, it’s not quite as extensive, the video we have. But to the extent we can take advantage of things like that, absolutely that’s a good way to go.
NOAH LEAVITT: And do– I mean, I think when a lot of– I think a lot of the public focus on football players health has been focused on head trauma, because that has gotten the most attention. But do you think there needs to be a broader look at health to take into fact cardiovascular disease, other potential health factors?
MARC WEISSKOPF: Oh, absolutely. I mean, right, there are many aspects that could be problematic from a health perspective. And we need to look at them all and understand which ones are the most important and which ones are really creating a problem later for later life health down the line.
I know there’s been a lot of interest in the head injury, and with football this chronic traumatic encephalopathy. And it’s– obviously, that pathology in the brain is very extreme and not something we would like to see. I think it’s important, though, to note, and in fact this comes out of our paper here, that even with this increased risk of rate of mortality from neurodegenerative diseases, including all the neurodegenerative diseases, Parkinson’s, ALS and dementias, still if you– our projections were that if you have a thousand football players and a thousand baseball players and compare them that by age 55 there’s really, what our data suggests from this paper, is only one more death from a degenerative disease by age 55.
Now, one more death is a problem, and we would like to stop it. But it isn’t like everybody is getting this. And even by age 75, the suggestion were, it was, I think, about 11 more neurodegenerative deaths. Which again is a problem, and we want to avoid any excess deaths, but it’s not sort of that everybody will have this, right?
And so I think that’s also an important message we’re getting at here. Along with that is there are other things going on in football than just the head injuries. And we ought to be trying to look at the whole thing.
NOAH LEAVITT: And it’s probably– I guess is it a case of too being that deaths are only part of the story, that it might just be quality of life? I mean, you might not have dementia, but you could have trouble sleeping over the course of your life.
MARC WEISSKOPF: Absolutely.
NOAH LEAVITT: Those types of things.
MARC WEISSKOPF: Absolutely. And that’s the kind of thing that, from the way we did this paper now, we can’t get at those details. That’s the kind of thing the football players health study is trying to get at.
They’re extremely important too. Because especially when it comes to things like dementia, there’s a risk if someone thinks they absolutely are going to get this horrible outcome that they’ve become fatalistic to some extent. But a lot of dementia problems, cognitive problems that occur, are actually from factors that could be modifiable. It might be sleep problems. It could be cardiovascular problems that are treatable.
So part of what we’re trying to do is understand why are people getting these things. And are some of the reasons potentially modifiable? Even later in life, even after you’ve played football, so that we can treat them.
So someone, a football player presents to a doctor, we really want to make sure that the doctor doesn’t just think, oh, they played football, got head injuries, are going to have this neurodegenerative problem. Can’t do anything about it. As opposed to being aware that there may be other issues, like sleep problems and cardiovascular problems that are treatable and might actually impact the quality of life of the player now.
NOAH LEAVITT: So do you see this as just almost like a workplace health issue limited to football players, or are there kind of some broader public health implications? I mean, I’m thinking even of parents who are worried about their kids playing youth football or high school football.
MARC WEISSKOPF: Yeah, I mean, it certainly is. I mean, obviously this is an extreme group and they have a very particular set of exposures. But there I think it is– it could potentially be sentinel in some ways that these are issues for other populations. I would more immediately go to things like military populations that might face some of the same sort of extreme stress, possibly head injury and other types of injuries.
The kid issue is very important. Our work really can’t address that, and it gets problematic if you’re only looking at the professional athletes. Because by definition, to become a professional athlete you probably were OK by the time you got to high school or college. So that may be less reflective of what’s going on with the whole population of kids playing these sports. Hugely important question, but really requires a different approach to study that.
NOAH LEAVITT: So I think the framing of kind of studying this from a workplace health issue is really interesting. And, I guess, are there any other lessons from other kind of workplace health issues that you and kind of other researchers are drawing upon, either in terms of lessons from modifying work conditions or even just designing studies?
MARC WEISSKOPF: Yes. So certainly in the designing studies. I mean, this is basically like an occupational epidemiology study.
And there is an approach to doing that. There are certain pitfalls that you have to be aware of in those types of studies. And so we’re certainly drawing on that.
And as I sort of alluded to before, this paper now is kind of the first step in that process to really nail down that we’ve got a problem going on here. There is a difference. What is that difference?
And now the occupational approach is let’s dig deeper into what are the factors that are leading to that, with obviously the idea that we want to be able to intervene on those, modify them so that we can keep the job, in this case football, without– while at the same time improving the working conditions so we don’t have as much of the health effects. It’s sort of like the days of asbestos or something, that we knew that asbestos mining work, turns out that’s bad for you. Well, are there– is there personal protective equipment?
Is there ways we can change how you go about the mining, or how you handle the rock that can reduce the exposure to the stuff that’s actually producing the problem? So ideally, we would like to do that in this case too. Let’s find things we might be able to modify so that baseball players and football players don’t look any different in their mortality.
NOAH LEAVITT: In doing this work, have you had a chance to speak with any current or former players? If so, what has their reaction been?
MARC WEISSKOPF: So not necessarily specifically with the work related to this paper, but in my work with the football players health study we have an advisory group that includes many former players that they’re very interested in these issues. I mean, I will say they love their game, right? And they want to play their game.
And that’s not uncommon. I mean, even workers that know they’re exposed to very dangerous things, that’s their livelihood. That’s their love. They want to do it. And so our job is then to figure out how can we allow these people to keep doing that and– while making it safer for them. So let’s do the work to figure out if they’re– if there are levers we can pull to protect them.