[markroseman]

While there is some work done on matching people up with specific drugs, this is still very much in the research phase.

Still, for many people, it's more trial and error than it needs to be. Many family doctors still don't match up symptoms with neurotransmitters, and then put people on the wrong class of drugs altogether (e.g. something like Wellbutrin which works on norepinephrine and dopamine instead of an SSRI for anxiety which is more associated with serotonin).

Part of the reason is that new drugs keep coming out so it's difficult to keep up, and eventually they all start to blur together.

The best practical predictor of success on a given medication today is if you have a close relative who's had success on that medication.

[numinary1]

That certainly makes sense since the problem has a higher likelihood of stemming from the same root cause, but also speaks to how little is known.

So include the social graphs of people who are on antidepressant discussion boards and pick up on friends, family, socioeconomic features ...

I so want to use ML tools on medical treatment data, but it's so hard to come by. Have toyed with a couple of startup ideas based on the tendency of people who share a condition to establish ties. Crohn's disease, cutaneous lymphoma. But the groups are too small. Mood disorders on the other hand...