I asked Dr. Bill Koslovsky, the "Wireless-Doc," about a Swedish study that concluded there was an elevated risk of a rare form of cancer in cell phone users: His response is that the study seems good -- he needs to read it in more depth -- but the method they chose leads to the least predictive accuracy. It's a useful tool to look at a small group of people with a particular disease and tease out causes, but it's no smoking gun. Better, a large sample over a large time with no particular outcome being looked for: take 1,000 people over 10 years and track many environmental variables, including cell phone use, and see if you can identify more disease in sub-groups.
Take a look, by comparison, at the phenomenal Nurses' Health Study, which started in 1976 with over 121,000 female registered nurses. They've been followed in fits and starts with large participation for decades, and the results have revealed more specific information about human--and specifically the much ignored female side of human--health than anything else.
I'd like to see a study start now on cell phones involving such a large group. They could track individual cell phone models, among other factors. But I think the only folks with the resources to fund such a study is the cell telephone industry. And you've got to be kidding.
The science hasn't appeared to change much since this mollifying Clinton-era (Nov.-Dec. 2000) article in an FDA publication even with this latest study.