[trustmath]

> I suppose if one holds the view that men and women are mentally identical on average, it would lead one to the conclusion that the gender disparity in diagnosis rates is due to social factors (or some kind of medical discrimination).

It might be easier to think about this on a different level of granularity. The problem has a self referential issue (requires social to evaluate social) if discrimination is the largest factor considered. Aggregate social factors over time yield discrimination - can't be corrected if the solution defines the problem.