Chicago Reader calls out "missing" mobile phone health study: European papers apparently gave this study that indicated no increased risk of glioma based on a retrospective (i.e., "what do you remember about your use") study on cell phone use across five countries. The study found that for people with gliomas and more than 10 years' cell phone use, the side that they remembered having used more frequently to speak on a mobile phone had an increased correlation with the side that the glioma was located on. From the abstract, I can't see if they removed side bias: there's a potential for someone with a brain tumor to remember incorrectly which side they normally talked on.
The conclusion is, "our results overall do not indicate an increased risk of glioma in relation to mobile phone use, the possible risk in the most heavily exposed part of the brain with long-term use needs to be explored further before firm conclusions can be drawn." Which is fine: It means more animal and other forms of study are needed, since this study didn't provide a firm conclusion.
There's a small cohort in this study: 1,522 glioma patients (a form of brain cancer that has sidedness, and thus some notion that it might reveal increased cancer risk derived directly from mobile phone radiation) and 3,301 control patients. A previous study from 2005 by the same authors about another form of cancer that could be predictably caused also found no increased risk of acoustic neuroma's, but concluded that more study of users with 10 years or more of use was needed.
The Danish study of 425,000 mobile phone users based on their phone records--rather than their recollection--showed no elevated health risks, and received wide play. The Danish study was retrospective, but relied on actual data of hours of use and phone types.