That is not, in fact, what they find, whatever they claim. Science has no objective means to diagnose depression, or more importantly to distinguish among many

That is not, in fact, what they find, whatever they claim. Science has no objective means to diagnose depression, or more importantly to distinguish among many fundamentally different kinds, other than by noting what responds to different medications. It should be easy to understand why designing a valid RCT of an antidepressant is practically impossible. But there is a continuous demand for RCTs, so people run, instead, invalid RCTs. Running invalid RCTs is pointless. Reporting their results as if valid is, at best, misleading. At worst, dishonest.