Hi! I've written a blog post about disadvantages of the categorical diagnostic systems (DSM and ICD) behind psychiatry and how it affects, for example, how we see mental health disorders and the start of getting treatment. Perhaps there (and additional reading in footnotes) is something to this topic as well! I'm a psychology student, and in our studies transdiagnostic and dimensional approaches are strongly present there.
G'day Marissa! I'm admittedly not the best-versed in psychiatry specifically, since I've focused more on psychotherapy in the past. My general vibe from reading & research I've done is that (for pharmacotherapy only, can't speak to crisis care):
My personal theory is that drugs are good for preventing people from doing harm to themselves for a short period, and in many cases the causes of the underlying depression go away on their own. But they probably shouldn't be used to permanently improve someone's mood, at which point we should focus on improving their environmental conditions and retraining their learned responses to stimuli.
But in a more general sense, I haven't come across a lot of general reviews assessing the effectiveness one way or the other in a deliberately unbiased way, but I haven't looked hard. I think it's likely that the role split between psychologists and psychiatrists, and the industrial split in funding between the two, is likely to make this research very hard. Anecdotally, I liked Johann Hari's Lost Connections, which begins with a pop-science assessment of the evidence against psychiatry while remaining balanced enough to describe when it's valuable, but I wouldn't call it unbiased.