I'm willing to go along with this, but there's lots of evidence to suggest the contrary. There's also evidence suggesting more severe forms of mental health pro

I'm willing to go along with this, but there's lots of evidence to suggest the contrary. There's also evidence suggesting more severe forms of mental health problems are more common in larger more population dense areas, and that migration from less dense to more dense areas in particular is associated with increased risks of problems later. For what it's worth, the datasets being used seem a little odd to me, in that there are many many datasets that could have also been used and were not. It seems odd to say this, but it has a cherry-picked feel to it given the number of other datasets that could have been used. It's interesting though; despite the problems I think there's lots of unknowns in this area. I would have liked to see it extended to longitudinal data in a wider variety of datasets that are available, and other forms of psychopathology.