Schoemaker MJ and Swerdlow AJ (2009) Risk of pituitary tumors in cellular phone users: a case-control study. Epidemiology 20: 348-354.
Although several studies have looked at intracranial tumours and mobile phone use, few have examined a link with pituitary gland tumours, despite their intracranial location and a lack of known risk factors.
This was a population-based case-control study of pituitary tumors, including 291 cases (63% response rate) and 630 controls (43%), based in Southeast England. The study formed an extension of the ongoing INTERPHONE study. Patients had to be 18–59 years of age. Pituitary tumor diagnoses were made between December 2000 and February 2005. Control subjects were selected from the patient lists of general practices in the study region. The control series was frequency matched on the sex, age, and health-authority distribution of the entire cases series. Face-to-face interviews were conducted by trained research nurses. Subjects were asked whether they had “ever” used a cellular phone prior to the interview, and more exposure history was collected from those who reported an average of 1 or more calls a week for at least 6 months.
Phone use less than 1 year prior to diagnosis was ignored for the statistical analyses, because this period is unlikely to be etiologically relevant. All estimates were adjusted for sex, age, reference date, Townsend deprivation score (a socioeconomic index), and geographic region.
The authors write that the risk of a pituitary tumors was not increased in regular phone users overall, with the odds ratio (OR) estimated as 0.9 (95% confidence interval, 0.7– 1.3). Other ORs were also close to 1.0 for nearly all exposure indices. Odds ratios marginally above 1.0 were noted in groups with cumulative phone use of more than 10 years. Estimates were not materially affected by adjusting for hands-free phone use. Finally, subjects who had more than the median hours of use in the population (51 hours), accrued 10 or more years prior to diagnosis, had an OR of 1.6 (0.8 –3.6). Using slightly different measures for their statistical modeling (the lower or upper value of reported ranges instead of the midrange value) did affect estimates of cumulative use, but the authors did not believe these differences to be of material importance.
Discussion and Conclusion
The authors summarize their results as supporting the view that risk of developing a pituitary tumor was not associated with regular phone use overall, nor with duration of use, time since first use, or cumulative number of calls or hours of use. The authors conclude with a discussion of the possible influences of errors in recalling historical phone use and potential selection bias from low response rates.