Monday May 17 2010
The study did not find a plausible link with risk of brain tumours
“Half an hour of mobile use a day ‘increases brain cancer risk’”, The Daily Telegraph has claimed. It says a landmark study into the health risks of mobile use has found that 30 minutes a day over 10 years increases the risk of tumours.
The research in question was a well-conducted analysis of several international studies that actually found no plausible evidence of a link between cancer and mobile phone use. Some newspapers have selectively quoted a few results in this research that suggest a significant link, but this is misleading in the context of the overall results. The researchers themselves explain these few anomalous results, and conclude that there are no conclusive signs of an increased risk of brain tumours.
Overall, this study does not provide evidence that mobile phones cause cancer, a finding echoed by the majority of studies on the matter, although sadly not by most newspapers.
Where did the story come from?
The study was carried out by an international group of hundreds of researchers known as the INTERPHONE Study Group, supported by the International Agency for Research on Cancer (IARC) at the World Health Organization. The IARC is conducting ongoing research and analyses into the possible health effects of low-level exposure to the radio frequency electromagnetic waves used by mobile phones. Numerous different sources provided funding for each of the international research centres.
The researchers also declare that mobile phone companies provided part of the funding for this study. However, an agreement allowed them to maintain complete scientific independence. Technical support was provided by the Canadian Wireless Telecommunications Association, which had no involvement in the design or conduct of the study. A travel grant for one of the researchers was supported by the Australian Centre for Radiofrequency Bioeffects Research, and some of the researchers own shares in Telstra Australia, which is a mobile phone provider.
The study was published in the peer-reviewed medical journal, The International Journal of Epidemiology.
Newspapers have featured a confusing mix of reports on the implications of this research: The Daily Telegraph suggests that half an hour a day can increase brain cancer risk, while the Daily Mail says “long conversations” and “prolonged use over many years” pose a potential risk. BBC News says that the study is inconclusive. A number of these reports appeared before the publication of the research paper itself, and may have been influenced by a series of alleged internet leaks that selectively used data taken out of its correct scientific context.
What kind of research was this?
This study was a series of international case-control studies designed to determine whether exposure to radio frequency from mobile phones is associated with cancer risk, specifically tumours of the brain, acoustic nerve and parotid gland (the largest salivary gland). The researchers say that much of the research into a supposed link between mobile use and cancer is to address public concern rather than any particular biological principle: the frequency of radio waves used in mobile phones does not break DNA strands, and therefore cannot cause cancer in this way.
The researchers report that this is the largest case-control study of mobile phones and brain tumours conducted to date. Generally, case-control studies involve comparing a group of people with a disease with those who do not have the disease, and seeing which characteristics or exposures are significantly different between them. As a study design, case-control studies have some shortcomings. Most importantly, they cannot prove that one thing causes another, only that they are associated.
An alternative way to research the relationship between an exposure and a disease might be a prospective study, which follows a population over time and waits for cases to develop. However, brain tumours are rare and take a long time to develop, so the very long follow-up and large number of participants needed to do this may make this type of study less appropriate.
What did the research involve?
Sixteen study centres from 13 countries participated in this study, and shared a common protocol to encourage similar study methods. The studies were pooled for this analysis to allow a large, single analysis of cancer cases and controls.
Cases were adults aged 30 to 59 years with a glioma or meningioma brain tumour diagnosed between 2000 and 2004. For each of the cases, a control person was selected and was matched against them in terms of age (within five years), sex and the region where they lived. There were small differences in the way countries ran this part of the study. For example, Germany chose two controls per case, while Israel also matched participants for ethnicity.
The researchers identified only 3,115 meningiomas and 4,301 gliomas across all the study centres, along with 14,354 controls. Not all potential candidates completed their interviews or were matched with controls, leaving 2,409 meningioma cases, 2,662 glioma cases and 5,634 matched controls to be included in the analyses. The majority of meningioma cases were in women (76%) and the majority of glioma cases were in men (60%), reflecting the known epidemiology of these cancer types.
Cases were interviewed shortly after their diagnosis, and their matched control was interviewed at around the same time. A trained interviewer applied a computer-assisted questionnaire to collect information on the use of mobile phones and potential confounding factors (that may be linked with either mobile phone use or cancer outcomes), including social and demographic status, medical history, smoking, and potential exposure to electromagnetic fields or ionising radiation at work or through other sources. Details about tumours were also collected from the cases.
The results from 14 participating centres were analysed separately and pooled in an analysis, which assessed whether there was an association between cancer and mobile phone use. The results from UK North and UK South were not pooled because of the large numbers. The researchers were interested in whether:
- regular users (an average of at least one call per week for a period of six months) had a different risk to those who had never been regular users
- the length of time as a regular caller had any effect
- the cumulative number of calls had any effect
- the duration of calls had any effect.
When they were analysing call duration etc, the researchers compared the cases with a group of people who had some mobile phone but less than the average of one call a week over six or more months. Cases were also compared with people who had never used a mobile phone. The researchers decided upfront only to adjust their main analyses for factors that showed a particular strength of relationship with exposure or outcome. They adjusted for education level as a proxy indicator of social and economic status.
Different analyses were done to account for the location of the tumours and the side of the head that a person reported placing their phone against most often. The researchers undertook separate analyses to assess whether a number of methodological issues had any effect on the results
What were the basic results?
For both meningioma and glioma, the study found no increased cancer risk with mobile phone use. In fact, it found the risk of cancer was lower in those who had regular mobile phone use in the past one or more years (21% and 19% respectively).
When analysing cumulative call time, the researchers split cumulative call time into 10 levels. In the lowest nine ranges (from less than five hours and up to 1,640 hours) there was no increased rate of either type of brain tumour. There was a small increase in the number of cases of glioma in those who had used their phone for 1,640 hours (the highest level of use) or more, i.e. 1.4 times increased risk.
However, the researchers say that there were “implausible values of reported use in this group”, i.e. some users with brain tumours estimated that they spent an unrealistic 12 hours or more each day on their mobile phone. The researchers suggest that there may be some data quality issues within this group, given that the cost of mobile phone calls at this time would make this prohibitive and that there may be impaired recollection for some people.
In the analysis of the link between preferred phone ear and tumour location, the only significant result was for the group of people who reported 1,640 hours or more of lifetime usage of their device on the same side of their head as their glioma tumour. As above, there may be issues concerning data quality with this group of individuals.
How did the researchers interpret the results?
The researchers note that there are several ways to interpret the largely negative associations between mobile phone use and cancer risk. On balance, they conclude that ‘INTERPHONE finds no signs of an increased risk of meningioma among users of mobile phones’. For glioma, they note that although they found one or two significant increases in risk in the highest users, the overall results are inconclusive as there are likely to be errors in this data.
Overall, the researchers say that they have “no certain explanation for the overall reduced risk of brain cancer among mobile phone users in this study”, although they do not think it is likely that mobile phones have a protective effect.
This study has not found conclusive evidence to support a link between mobile phone use and brain tumours. According to the researchers, it is the biggest case-control study on the subject to date, making the findings particularly important.
While there is a need for further research into longer-term mobile phone use, this study certainly does not support the clear-cut claims of some newspapers that “talking for 30 minutes a day” increases the risk of brain tumours.
While there are a few spikes in results, these individual results should be interpreted in the context of the data as a whole. In their paper, the researchers themselves provide plausible explanations for these results. They clearly conclude that there is no evidence of an increased risk of meningioma among users of mobile phones, and that for glioma, the overall results are inconclusive.
Alongside the general shortcomings of case-control studies, the following points should be considered when interpreting these results:
- In the main, this study actually found an apparent reduced risk of brain tumours with mobile phone use, but the researchers dismiss this as being a real association and give possible explanations for these findings. These include sampling differences in the participating centres, missed cases or misdiagnosis.
- Many people declined entry into the study, so participation was also quite low - 78% among meningioma cases, 64% among glioma cases and 53% among controls. There were also some differences between those who responded and those who did not.
- As with all case-control studies, this one cannot prove causation, i.e. it cannot prove that mobile phone use or lack of it was having an effect on levels of cancer and not the other way round. They say, for example, that having early symptoms of a brain tumour may dissuade people from using mobile phones – although this is unlikely to account for all the patterns seen in these data.
- The researchers acknowledge that their adjustment for education is not a perfect adjustment for socioeconomic status.
- They explain the possible reasons for the few significant results they found. The small positive links they found between the highest levels of cumulative call time and risk of glioma have been discussed.
- One disadvantage of case-control studies is that they do not give any indication of the absolute risks of disease. Brain cancers are rare. In 2006, the incidence (i.e. number of new cases) of brain or central nervous system cancers diagnosed in the UK was about seven in every 100,000 people. Across 13 countries, only 3,115 meningiomas and 4,301 gliomas were identified over the study period (four years). The vast majority of people do not develop these diseases.
- Cancers may take a long time to develop, and ongoing analyses are important.
Overall, the emphasis some newspapers have placed on selected results of this research is misleading. This study does not provide evidence that mobile phones cause cancer. More research will follow and over time, as data gathers, the longer-term effects of mobile use can be assessed.