The Rock n’ Roll marathon kicked off from downtown Tempe earlier this month and, that morning in Tempe, PM2.5 air particulate levels were sitting around 10 micrograms — that’s good.
But, Elvira Fleury, a graduate student at Harvard’s School of Public Health, says even under "acceptable" levels of pollution like that, air particulates can slow you down.
And it was during a winter in which we’ve seen a lot of air quality alerts and warnings about exercising outside — especially for at-risk populations, like older people, or people with asthma.
Fleury studied this phenomenon recently and found that even a little change in air pollution slows down even the fastest marathoners. Fleury joined The Show to discuss what they found — and what that means for the rest of us.
Full conversation
ELVIRA FLEURY: So we looked at nine different marathons that we chose based off of their size, so more than 5000 finishers dating back to 2003 and we included all finish times.
GILGER: So, very broad. How did you measure pollution levels like this? Was actually measuring pollution levels like on the race course. It's very specific. It sounds like.
FLEURY: Yeah, so we used a spatiotemporal machine learning model. It uses information such as satellite aerosol optical depth, so how much sunlight aerosols in a column of the atmosphere scatter and absorb. It uses monitoring station measurements within 500 kilometers and other factors such as topography, meteorology, land use, state, longitude and latitude, and it makes estimates at any point. So, we made estimates at the start line, finish line, and every mile along the course, and took the average of those estimates.
GILGER: OK, so tell us what you found here. It sounds like there was a big difference on high pollution days, even for really, you know, fit and fast runners.
FLEURY: So for a one microgram increase in race day, PM 2.5 we actually found that on average, male runners were 32 seconds slower and women were 25 seconds slower for a one microgram per cubic meter increase. So it doesn't actually sound like that much when you put it that way, but that would translate to something around two minutes slower finish times for a normal change in PM 2.5.
GILGER: OK, so a lot of numbers in there, right? But the difference here is, is quite noticeable, especially for runners like a couple of minutes is a lot.
FLEURY: Exactly.

GILGER: Tell me a little bit about your own interest in this. You are a runner. It sounds like that must have played a role in why you were interested in looking at this impact?
FLEURY: Yeah, absolutely. So I and my mentor, who's also on this paper, are both avid runners. We've both run multiple marathons. I've run the Marine Corps Marathon, which is in this study. So it really came from personal interests. We're both in environmental health, so we just started wondering, how do less appreciated environmental conditions impact performance? And we found that there have been some studies looking at PM 2.5, but they're really among elite runners or NCAA athletes. So we were really interested to use our perspective and you know, our skill set to answer this question.
GILGER: Yeah, talk a little bit about the trends within this like, were there any differences you found in slow versus fast runners, etc?
FLEURY: Yeah, yeah. So actually, PM 2.5 levels overall have actually improved. So, they've gotten lower. They got lower over the course of the study, which is, you know, a little bit of good news. We also looked at different quantiles of performance. So being able to say, how does it affect the median runner, versus how does it affect a very fast runner, versus how does it affect maybe a slower marathoner? And most of those estimates were pretty imprecise, but we did see slightly stronger associations among the median and faster runners, which surprised us honestly. And we think that could be for a number of reasons. The first is that slower runners, their performance might be affected by a greater variety of factors, like walking through aid stations or illness and injury. That would make it harder to detect a signal, but also the faster you're running, the faster, the harder you're breathing, so the particle deposition rate is larger, so that could influence it.
GILGER: And, I mean, what really strikes me about this is that you're looking at, like, super healthy people, and the impact that this kind of pollution level has on, yeah, like, What is this for the rest of us who aren't running marathons? Like, how bad is it for everybody?
FLEURY: That's a great question, and one I like to emphasize, because you are right. This is among a very specific segment of the population that has the ability to even run a marathon, which is a great accomplishment, that has trained so they're cardiovascularly conditioned, right? And even beyond that, these are levels that, for the most part, are below health based levels used to guide states in managing air quality, and its short term exposures. It's just that we only met. We only have exposure during the race.
GILGER: Right.
FLEURY: Right? So for the rest of us, it kind of highlights what we know and what many, many, many studies have shown, which is that PM 2.5 is harmful for health and especially for sensitive populations.
GILGER: Yeah, right.
FLEURY: So people who are ill, or, you know, when you're developing your lungs, so it really adds to this notion that the PM 2.5 is harmful for health.
GILGER: Yeah. So, I mean, there's that level of it, but I wonder, as a runner yourself, looking at the findings that you were able to gather, how do you think runners can use this data to help them improve their own performance or maybe avoid some of these bad consequences? Is there anything they can do?
FLEURY: Yeah. So what I often tell people is, if the air quality is unhealthy, if your weather app or the news is saying, “Don't go out and run”, don't go out and run. Right but barring that, if there's no reason to think that it would be unhealthy to run in these conditions, training for a race takes a lot of time, a lot of energy, and is a huge accomplishment. So what I would say, what I would do myself, is go out and run the race, right but then afterwards, when you're evaluating your performance, you might think about it the way you think about heat. So if it was really hot on the day of your race and you got dehydrated, you might factor that in when you're seeing okay, I ran this fast, but maybe I would have run a little faster if it had been cooler.
GILGER: Sure.
FLEURY: So something similar with PM 2.5 right? Something to consider when, when you're evaluating your performance or your time after the race.
GILGER: Yeah, it's making you work a little bit harder.
FLEURY: Exactly, and you might not even notice it yourself, so it's just something to think about.