As another medical fella (hi, Serenitude), it might be useful for me to chime in to confirm that, in medical research, simple statistical association will (quite rightly) get you precisely nowhere when you're trying to interest a funding body in your research.
One of the standard references in these matters is Sackett & Tugwell's Clinical Epidemiology, and they give nine criteria by which to judge possible causation, as opposed to simple statistical linkage.

1) Is there evidence from true experiments in humans?
2) Is the association strong?
3) Is the association consistent?
4) Is the temporal relationship correct?
5) Is there a dose-response relationship?
6) Does the association make epidemiological sense?
7) Does the association make biological sense?
8) Is the association specific?
9) Is the association analogous to a previously proven causal assocation?

The association between uranium and diabetes fails the majority of these tests, so in terms of "medical science" it doesn't get off the ground.

Grant Hutchison