> (often against best current practice guidance) This could be one incremental angle to push on, which could reduce the harm from over-prescription without req

> (often against best current practice guidance) This could be one incremental angle to push on, which could reduce the harm from over-prescription without requiring a radical overturning of approaches to mental health. There are a number of cases where drugs are prescribed despite any decent evidence that they actually improve outcomes in those cases, and those really should not be happening. Part of it is that, at least in the US, there is an odd step function in drug approval. A drug has to be shown both safe and effective for treating some illness to be approved. In this approval process, things like placebo effects are controlled for, representative groups are sought, etc. If a drug isn't approved for any uses yet, it cannot be sold. But once it's approved as effective for any illness, it suddenly becomes legal to prescribe for anything at all, even "off-label" uses unrelated to the approved indications and regardless of any evidence of effectiveness for those alternative uses. Physicians tend to develop idiosyncratic ideas about how they think drugs work in these alternative cases, because they are not doing any of what's done in proper clinical trials, like controlling for placebo effects, or even really solid data collection. I'm not sure if actually banning off-label prescriptions would be the right solution, but off-label prescription without convincing evidence-based support for the decision should be resorted to cautiously. There still seem to be too many psychiatrists who approach it as some kind of seat-of-the-pants endeavor, rather than following statistically validated best-practice guidelines.