It's incredibly frustrating seeing people who acknowledge they have no specialized knowledge in a field make sweeping statements about it, encouraging psuedo-scientific beliefs about psychiatry (which is, make no mistake about it, more of an 'art' than a science, but so was the beginning of many now-scientific fields). Nobody tries to tell a cancer patient how to treat their cancer, yet so many people feel obliged to opine about mental illness.
What this article is discussing is the (unquantified) subset of people who experience serious side-effects from anti-psychotics. The key takeaway from the article:
> “For some people, remaining on medication long-term might impede a full return to wellness,” Insel wrote. “For others, discontinuing medication can be disastrous.”
> The problem is that we do not know who is in which group.
> Some people can learn to talk themselves down from delusional thoughts. And some people might choose hearing voices over being 30 pounds overweight and tired all of the time. The point is that this is not a choice I should be making for my patients; it is a choice I need to make with them.
Of course, if somebody is having a psychotic break or mentally unstable, you can't even begin to have that conversation about their available options because they will be unable to focus or make rational choices. You pretty much need to start with the medication, then work from there to figure out which group they're in, and adapt their treatment accordingly (i.e. stick with meds or drop them but help them learn how to manage it).
I found the article to be a long-winded way of saying "psychiatrists ought to seriously consider dose reduction in their treatment regime", which is something that I doubt anybody would argue against. The devil is in the details (for some people, dose reductions can be incredibly dangerous), and the article does not delve into what a sensible strategy would look like.