Earlier this week, a bunch of outlets reported that a new study in the American Journal of Cardiology showed that running too much was associated with an increased risk of mortality (it crossed our desks at Science of Us and we planned to write about this surprising finding, but then we saw the Cut already had). It was the sort of study that leads to forehead-smacking: you don’t run enough, you’re more likely to die; but you run too much, and you’re about equally likely to die. How can anyone win? Luckily, as Justin Wolfers points out at the Upshot, there is way less here than meets the eye.
Sometimes, unpacking these sorts of studies requires a fair amount of statistical weed-whacking. Not here. Wolfers lays out a really simple case for why this finding is, if not total bunk, then at least completely inconclusive:
• the researchers surveyed a bunch of Danish joggers, “categorizing 878 of them as light, moderate or strenuous joggers“; 40 were identified as “strenuous”
• ten years later, the researchers checked in on how everyone was doing, and found that 17 of the joggers — two of them from the “strenuous” group — had died
• so over the course of ten years, two out of 40 strenuous joggers had died, the highest proportion of deaths in any of the three groups
Here’s the problem: Statistically speaking, the fact that two out of 40 strenuous joggers had died over the course of a decade really tells us nothing about strenuous jogging. It’s just way too small a sample size. Plus, as Wolfers points out, the study doesn’t look into how the runners died and “whether those two deaths were from causes that could plausibly be related to running.” Wolfers goes on to explain how the researchers tried to sidestep these statistical weaknesses with a too-cute-by-half trick, so if you want more details, check out his post.
In the meantime: Yes, if you’re out of shape or have a cardiovascular disease or something, you should check with a doctor before engaging in any sort of exercise regime. But beyond this obvious exception, go ahead and run. As Wolfers writes: “Most Americans need to worry about exercising too little, not too much[.]”
