Yeah I can't blame you for not seeing a solution---the key I think is that there is in fact not one solution but many---my current theory is that any psychosis is highly heterogeneous but being treated as a single condition. So take cancer for example----everybody is like when are we going to find the cure for cancer but there is no one cure, its different for every type of cancer, specifically tailored. Lets say you decide that cancer can be cured with mastectomy----yeah will that will take care of a percent and keep it from coming back but if you have colon cancer and get a mastectomy well that would just be stupid right----that's what I feel like the pdocs are doing now----they only have one treatment APs which is like mastectomy and they'll give it to you whether its specific to your subtype or not. You could be a man and they'd be like well we've found mastectomy works in like 25% of cancer cases so you're getting it (even though the incidence of breast cancer in men is very low but this is assuming no knowledge of that). I know it sounds stupid but that's what they are actually doing...because they don't know anything about psychosis subtypes yet. Anyway my current theory is some of the meds they've developed in the past probably work but only on specific subtypes so they don't do well in clinical trials but they may work if we start genetically screening for subtypes. So I think there is hope that there is some stuff that already made it to phase I or II trials might be able to be dusted off and tried again in a specific sub population of affected individuals. Part of the problem is it seems to be polygenic as well so even in one person its not like its any one particular gene and that makes the puzzle a lot harder---the type of math needed to figure this out might just be impossible I'm not sure---I wish I could talk some math dudes into looking at the GWAS data.