Here's what it would take to really prove the causative effects claimed in the article: pick a bunch of random people and then force them to either drink or not

Here's what it would take to really prove the causative effects claimed in the article: pick a bunch of random people and then force them to either drink or not, exercise or not, smoke or not.. etc. There's no way anyone is ever going to do this experiment, and I just find the claims made in the article somewhat dubious as a result. On the other hand, they are likely to be right (yes?) and probably it's better to make such claims than not to. But hey, if you are a sedentary type, if you smoke, if you drink, what is causing those behaviours? Just forcing some exercise doesn't change those deeper causes (what ever bio/emoto/chemico causes), and may very well make the person absolutely miserable or worse.