[strken]

It's almost impossible to run a good UV exposure association study because going outside correlates with health. Everything clever you can do to get around that (UV-B areas are what the article chooses) runs into issues with introducing even more uncontrolled variables.

I'm getting a bit jaded with vitamin D and UV studies and it's possible that one will come along and prove me wrong, but thus far they've generated a lot of hype but failed when experimentally tested.