Those of us interested in attempts to identify the cause of cancers and the long term adverse impacts of exposure to drugs would be well aware of the difficulties in determining whether or not there's a causative relationship, even when there's a strong correlation. This excellent and very entertaining article by Jon Borwein, Laureate Professor of Mathematics and Michael Rose, PhD Candidate, School of Mathematical and Physical Sciences at University of Newcastle, Australia will hopefully clarify what can be a very confusing subject:
I well remember the concern years back (only recently debunked) as to whether higher incidences of childhood leukaemia were due to the children living under high voltage power lines. As one researcher commented, the apparent correlation could be due to poorer diets provided to children living in the lower socio-economic neighbourhoods associated with high voltage power distribution and nothing to do with the electric and magnetic fields from the power lines.
Here's a summary (from 2005) on the association between magnetic fields from electric power and cancer from the National Cancer Institute:
Photo: There was definitely correlation between this not so sleepy lizard hissing at me and me waving my hat at him (causation)...