Some premises of this article are at least sound, at least, fishy. Easy language, lot of promises, and a lot of plausible things but not necessarily with causal relation.
It starts with "We now know that depression is associated with [...]", where AFAIK we are far from pinning down etiology of depression (there may be multiple mechanisms, etc). And it is one of reasons why antidepressant have relatively low effectiveness. The abstract ends with a longer list of things that make it flashy.
Moreover, things like smoking are sometimes considered actually having mildly anti-depressive effect (though it is disputed).
I don't say that the paper is wrong. Just a paper not listing doubts and promising a lot should be less believed that one making very humble observations (but the later one would not make HN frontpage, hence the bias).
It starts with "We now know that depression is associated with [...]", where AFAIK we are far from pinning down etiology of depression (there may be multiple mechanisms, etc). And it is one of reasons why antidepressant have relatively low effectiveness. The abstract ends with a longer list of things that make it flashy.
Moreover, things like smoking are sometimes considered actually having mildly anti-depressive effect (though it is disputed).
I don't say that the paper is wrong. Just a paper not listing doubts and promising a lot should be less believed that one making very humble observations (but the later one would not make HN frontpage, hence the bias).