Hardell L, Mild KH, Carlberg M (2002 b)
This paper is based on the same study as the one published in the European Journal of Cancer Prevention (2002;11:377-386).
Here the authors have concentrated on malignant tumours of the brain, whereas the earlier paper looked at both benign and malignant tumours. As before, the data on exposure are based on a questionnaire given to survivors or a family member. Information was ascertained about phone use (type, duration and extent of use). Cases of brain tumours diagnosed between 1 January 1997 and June 30 2000 were identified from cancer registries in four regions of Sweden. In total, 1111 cases with a malignant brain tumour were reported in this time period. However 393 cases that had died were not included, nor another 69 who were excluded for various reasons. The analyses were based on the remaining 649 cases and an equal number of controls, matched on age and sex.
The results were essentially the same as in the first paper. Overall, there was no increased risk from exposure to various types of phone. The risk for analogue phones, digital phones, and for cordless phones was 1.13 in each case, although the results were not statistically significant.
Some subgroup analyses did show some statistically significant results. For instance ipsilateral use of an analogue phone (i.e. use of the phone on the same side as the location of the tumour) had a risk of 1.85 (CIs 1.16 - 2.96), and digital phones had a risk of 1.69 (1.05 -2.41). While the risk for cordless phones of 1.46 was not quite statistically significant, the risk for these phones in people who had more than 319 hours of cumulative use was 2.15 (1.13 - 4.09). It should be noted that there was no other evidence of a dose-response trend. In addition, in most cases the risk on the contralateral side was reduced, although none of the results was statistically significant. This trend for reduced risks on the contralateral side suggests a protective effect from the use of a phone on the other side of the head - something difficult to explain. Similar results were seen when cases of astrocytoma were analyzed separately.
These authors continue to have results at variance from other researchers, who have not identified an increased risk with ipsilateral use. Cordless phone use has never been implicated as a risk factor for brain tumours. It is noteworthy that the cases were not "incident" cases - that is, they were not identified immediately after the diagnosis, but were those cases diagnosed over a 3 year period and still alive at the time of the study. This makes reporting bias more likely.