> On the one hand, physical activity is clearly one of the best interventions for preventing physical disease and mental suffering
As far as I've read, it's decidedly unclear that general physical activity as an intervention is proven for much of anything; as Dr. Ashley mentions, most of the data we have is from observational studies. I think it's fair to say that for the vast majority of people, a moderate amount of physical activity is almost certainly better overall than being a total couch potato. But it's also not as simple as "active good, sedentary bad". In particular, I've seen a few studies suggesting that physical activity at one's job might not have the same disease-prevention benefits as physical activity for leisure (e.g. [1]), which casts some doubt on the idea that the benefits are primarily coming from physical activity per se.