> My understanding is that even for those who receive positive results, such results are highly anecdotal and individually specific. Probably even coincidental

> My understanding is that even for those who receive positive results, such results are highly anecdotal and individually specific. Probably even coincidental. People don't try one thing at a time - they try multiple things at the same time (the diet, the doctor's prescription, more/less exercise, some natural remedies, etc). They're in no position to tell which one worked, or if only multiple changes worked, together only, but not individually, or even just random chance. As far as Survivorship bias goes, following the 100/10/1 rule, for every 100 readers of a forum, 10 may post comments and 1 may contribute original content. I mean, even if every single posting on a forum is positive, there's potentially 90 failures who did not care to create an account to post at all.