They should do a study on Religious Fanatics and mental health.

Everything I've read on the subject suggests that religion is much more of a societal phenomenon than an individual phenomenon. The way to test its health effects would not be to compare groups of individuals against other groups of individuals, but to compare the health of predominantly atheistic societies against predominantly religious societies. The difficulty in doing that is that there are not and never have been any predominantly atheistic societies. Probably because the ones who tried it could not hold together a functional society.

This researcher is clearly operating with a motive. Could it have been only an accident that the two most potent aspects of religious participation - social support and personality - just happened to not be addressed by the survey he used?

When Dr. Speed finds ANY group, let alone an entire society, whose effectiveness at social support, social cohesion, and personality stabilization rival that of any major world religion, I'll be all ears.