An important part:<p>> This is an observational study, and as such, can’t establish cause. And the researchers caution that the number of included studies was small and their methods varied considerably, which may have influenced the results.<p>People who are sick or have other health issues will do very little running, so correlation and causation and all that.
My preference is to see what the NHS "Behind the Headlines" sites makes of such studies [1], rather than read press releases.<p>> The findings support general understanding that physical activity is good for health. However, this does not mean that running in itself is necessary as an activity. The comparison groups may have included a varied mix of those who were sedentary and those who took other forms of activity but just did not run.<p>[1] <a href="https://www.nhs.uk/news/lifestyle-and-exercise/running-reduces-risk-early-death-study-reports/" rel="nofollow">https://www.nhs.uk/news/lifestyle-and-exercise/running-reduc...</a>
I was skeptic, so I looked at the related paper: <a href="https://bjsm.bmj.com/content/early/2019/09/25/bjsports-2018-100493" rel="nofollow">https://bjsm.bmj.com/content/early/2019/09/25/bjsports-2018-...</a><p>It looks like a fairly good systematic review and meta-analysis. Their "Assessment of study and evidence quality" seems quite comprehensive, and they filtered out several studies affected by selection bias.<p>They mention the causation possibility:<p>> the association between running and mortality may partially be explained by assuming that sick participants (who are more likely to die) were less likely to participate in running.<p>Still, even if "running" isn't a direct cause for increase longevity, we can't rule out that more promotion of running would have positive effects (placebo effect, have unhealthy people use running as a keystone habit to build other healthy habits, etc).<p>What I'd love to see is more meta-analysis like this one for other sports. I believe that running is known for its orthopedic stress risks, and other sports like swimming might be seen to deliver the same positive effects as running, without the negative impacts.
This podcast has an episode on that study, highlighting the constraints: <a href="https://open.spotify.com/episode/67MXt1VL6BoXt0PW5i6pA0?si=GWdupcT7QYSyfN5WAE9Y2g" rel="nofollow">https://open.spotify.com/episode/67MXt1VL6BoXt0PW5i6pA0?si=G...</a>
Need multiple regression analysis to see how individual factor contributes. If running is there many others may be excluded but it is the excluded one might explain more.