Handheld Cellular Telephone Use and Risk of Brain Cancer
A new study in the northeastern United States has evaluated the use of mobile phones among 469 people with brain cancer and 422 matched controls. Overall no association was found between the use of mobile phones – most of which were analogue – and the risk of brain cancer during the few years subsequently. Further research would be desirable to look for any longer-term effects, based on larger numbers of people and also specifically for the use of digital phones.
This study was based on patients at five academic medical centres. The cases were 469 men and women aged between 18 and 80 years of age with primary brain cancer. The controls were in- patients from the same hospitals, most of whom had been admitted for non-cancerous conditions, and who were matched to the cases according to factors such as age and sex. Using a structured questionnaire, trained interviewers obtained information on the patients’ use of mobile phones, as well as – for example – on their occupation and medical history.
Relatively few of the patients reported having used mobile phones regularly – 14% of cases and 18% of controls. The odds ratio (relative risk) for ever having used a mobile phone regularly was 0.85 (95% confidence interval 0.6-1.2); an odds ratio less than 1 corresponds to a decreased risk, while a value greater than 1 corresponds to an increased risk. The mean duration of mobile phone use was 2.7 years for cases and 2.8 years for controls, and there were no increasing trends in risk with increasing frequency of use or estimated cumulative use. There were no indications of raised risks for tumours in specific sites of the brain. There was a slight suggestion of an association between the side of the head on which the phone was held and the side of the head on which the tumour occurred, but this did not hold for tumours in the region of the brain that might receive the highest radiofrequency exposures from mobile phone use. Brain cancers of the common histological types did not show an association with mobile phone use; the odds ratio for one uncommon type of tumour (neuroepitheliomatous) was elevated although not statistically significant.
This was a well-conducted study. Nevertheless, some caveats may be attached to the interpretation of its findings. The proportion of mobile phone users and their average duration of use are both relatively low, reflecting the fact that most of this use took place in the early and mid-1990s. Consequently, the patterns of exposure are somewhat different from those arising nowadays. Furthermore, most of the mobile phones considered in the US study were analogue and operated at frequencies of around 800 to 900 MHz, whereas mobile phones in the UK operate at either 900 or 1800 MHz.
Since information on mobile phone use was collected directly from the patients in most instances, there might be a concern about whether the brain cancer cases and controls responded in a similar way, given that the severity of illness tended to be greater among cases than controls. In addition, any differences in referral patterns between cases and controls could potentially lead to artefactual findings, although the patterns of mobile phone use among controls appeared to be similar to those in the general US population. A further point concerns possible latency effects, in that some types of tumours may take many years to develop. Studies that account for long induction periods would be desirable.
Further research on mobile phone use and brain tumours is in progress. Results from another study in the United States are due to be published within the next year. Also, data collection is commencing for an international study, which is being co-ordinated by the International Agency for Research on Cancer. This study will include data from 13 countries, including the UK, with first results expected in 2004.
NRPB Response Statement R5/00
21 December 2000
Last updated 21st December 2000