A Report on Non-Ionizing Radiation

Where Is Interphone?

October 9, 2007

Why is the Interphone study not finished yet? "The interpretation is not straightforward," Elisabeth Cardis told Microwave News in an interview from her office at IARC in Lyon, France. The data are "very difficult to interpret because of the potential problems of recall and selection bias," she explained.

Cardis is leading the Interphone project, an international epidemiological study investigating the possible link between the use of cell phones and brain tumors. The study was originally slated to be completed years ago.

Asked when the paper with the combined data from all 13 countries would be out, Cardis replied, "Soon, I hope."

IARC has posted an update on the study on its Web site. It includes a table with the key results from individual countries as well as those from two combined analyses. The update also provides a list of the more than 30 papers published by various members of the Interphone team.

The update addresses the most widely discussed —and provocative— findings: The significantly increased risks of gliomas and acoustic neuromas, by up to 40% and 80%, respectively, among those who used mobile phones for ten years or more on the side of the head the tumors developed. These results came out of pooled analyses of data from five northern European countries (see our January 22 post). Cardis noted that these elevated risks "might either reflect a true causal association or be artifactual, related to recall bias among the cases."

One unexpected finding that has emerged from a number of Interphone study groups is that using a cell phone appears to protect against developing a tumor. (See, for example, the letters from Sam Milham, a Washington state epidemiologist, to the American Journal of Epidemiology and the British Journal of Cancer.) This result, Cardis explained, may be due to selection bias, that is, the controls and/or the cases may not be completely representative. "It's hard to believe the use of a phone for a few minutes a month could be protective against brain cancer," Cardis said. If selection bias is in fact at work, it may mean that the published results underestimate the true cancer risks.

The analysis of the Interphone data continues and may provide some clarification, especially with respect to long-term users. "Manuscripts presenting results of the international analyses, based on much larger numbers of long-term and heavy users, are in preparation," the IARC update states, adding that, "More detailed analyses are also underway, focusing on more precise localization of tumors using 3-dimensional radiological images, and on the analysis of the effect of RF exposure at the location of the tumor."