Hardell L, Carlberg M, Mild KH (2006c)

The authors present analyses from two case-control studies, carried out in 1997-2000 and 2000-2003. There were 3,279 patients initially reported from the cancer registry. This included benign and malignant tumours. Because of a number of exclusions 2,437 (65%) were finally included. Of these, 1,008 had a malignant brain tumour. In this paper they discuss the results for malignant brain tumours. A total of 905 cases and 2,162 controls were included. It is not clear why 905 cases are discussed in this paper, rather than 1,008. The cases were initially matched to 1 control, but in this analysis, rather than use a matched analysis, they have included all controls, including those recruited for cases of benign brain tumour. The design of the study has been discussed in other papers. Briefly, subjects were asked to fill in a mailed questionnaire, sent to their home about two months after the diagnosis.

Numerous results are presented, although there does not appear to have been any statistical correction for multiple analyses. Statistically significant risks were seen for analogue, digital, and cordless phones, and the risk increased in those with greatest cumulative use, or with more than 10 years of use. Increased risk is also seen with ipsilateral use. Multivariate analysis reduced the calculated risk, although for analogue phones those with more than 10 years of use had an OR of 1.9 (CI 1.4-2.6). Digital users in the 5-10 years group had an OR of 1.3 (1.03-1.7) and for >10 years use their OR was 1.9 (0.98-3.8).Cordless phone users had an OR of 1.3 that was not statistically significant.

The authors in the discussion section review the discrepancy between their findings and those of other authors.

Home             Links              Sitemap               Contact Us
© McLaughlin Centre for Population Health Risk Assessment