You are correct that how any of the drugs work is currently poorly understood. Presumably, there are individual biochemical differences in either drug receptor

You are correct that how any of the drugs work is currently poorly understood. Presumably, there are individual biochemical differences in either drug receptor sites on nerve cells or in metabolism of the drug after ingestion that make some kind of treatment difference, but there are not convenient medical tests yet to predict which kind of patient will respond to which drug. When I advise friends about pondering different medical treatments for most disease, my pickling in the research on human behavior genetics in the weekly journal club I attend suggests that you look to what works for your close relatives. If a particular medicine seems to be helpful with few side effects for a first-degree relative (parent, sibling, or child), give it a try. If a particular drug has been useless for someone closely related to you, for whatever condition you are concerned about, start your search with some other drug. And so on. Eventually we MAY get to a deeper genetic understanding of individual differences in response to prescribed medicines, but we are a long way from that so far.