I didn't have look on the studies but I would not be surprised if a decent amount of participants were completely healthy individuals. And maybe (more from random sampling) some unsuspicious mildly overweight without other problems. Especially in the earlier cohorts of testing.

Right, you should read it though, we're in the weeds over here, it's not just sort of free-assocating chat, we're picking apart specific things about the article. One of them, as I mentioned 4 up, is that the study with the 65% # is confounded because the groupings involve type 1 diabetes, and also, the number rebounds higher than the # who stopped