A few weeks ago the classification of radiofrequency electromagnetic fields (RF-EMF, the radiation from cell phones) as a possible human carcinogen by the International Agency for Research on Cancer (IARC) made headlines. At the time, my position was that it is hard to draw conclusions from a press release, but that given the criteria for classification as a possible human carcinogen, the information in the press release and the data available in the scientific literature I didn’t see a need to panic.
Now more technical information about the IARC evaluation is available, published in the Lancet; while it still isn’t the full IARC technical monograph (which we probably won’t see for several months at least) it does provide a bit more insight into the basis for the IARC classification. At virtually the same time, another analysis of the same data by the International Commission for Non-Ionizing Radiation Protection (ICNIRP) Standing Committee on Epidemiology has been published, which concluded that “the accumulating evidence is increasingly against the hypothesis that mobile phone use can cause brain tumours in adults.” Which of these two apparently conflicting conclusions should we believe?
First a look at the IARC evaluation, as summarized in the Lancet. As expected, the evaluation relies primarily on the INTERPHONE case-control study, which looked at cell phone use (based on recall) in 2708 people with glioma and 2972 controls, and found that for most groups there was no increased risk of glioma (in fact reduced risk in most cases – though as discussed below this is likely a study design artifact), except for the group with the highest cell phone use (>1640 hours of use) which showed an increased risk of glioma, with some suggestion of an increased risk on the side of the head and in the part of the brain where exposure would be highest. Similar results were reported in a Swedish pooled analysis.
The IARC Working Group acknowledged that both of these studies are susceptible to bias, but overall believed that the results “could not be dismissed as reflecting bias alone, and that a causal interpretation between mobile phone RF-EMF exposure and glioma is possible.” On this basis they found there is “limited evidence in humans” for carcinogenicity, although this was not unanimous and some members of the Working Group found the data to be inadequate. They also reviewed animal studies; the available chronic animal studies did not show an increased cancer risk from RF-EMF; a few studies showed an increased cancer risk from a combined exposure to RF-EMF and a known carcinogen but the meaning of that isn’t clear. They also concluded there was only “weak” evidence of a mechanism for RF-EMF to cause cancer. The combined result was classification of RF-EMF as “possibly carcinogenic to humans”, which as I have previously noted would not normally result in regulatory agencies treating it as a carcinogen.
The ICNIRP evaluation also looked at the INTERPHONE study. They found several issues in the study design, including evidence that people’s memory of their cell phone usage was generally biased, and people with lower cell phone use who didn’t have glioma were more likely to refuse to participate in the study; these biases are believed to be the cause for the reported lower glioma risks in low/moderate cell phone users than people who didn’t use cell phones at all. They found data suggesting that people had better recall of the number of calls made than the amount of time they spent on their cell phones, and found that if number of calls was used instead of total hours spent on the phone there was no indication of increased risk of glioma. There was also no increased risk observed for those who had used cell phones the longest. They also noted that the INTERPHONE study showed that individuals who had a very high usage over the past 1 to 4 years had a higher risk than those with similar total usage and more than 10 years of use, which is the opposite of the trend that would be expected for most carcinogens (though a possible alternative explanation is that cancer risk is more a function of the intensity of exposure than cumulative dose).
They found data on where in the brain the tumours were reported to also be subject to bias and to be very inconsistent.
The ICNIRP evaluation also considered the data on brain cancer incidence in several countries with reliable records, including Nordic Countries, Switzerland and the US; in all cases there has been no indication of an increase in brain tumour incidence (as I discussed previously, the trend is the same in Canada).
Overall ICNIRP concluded that the combined evidence suggests that there is no significant increase in adult brain tumours within 10 to 15 years of cell phone use, but noted that there are no data on the risk of childhood tumours.
These two evaluations do not really contradict each other as much as it might seem. Neither give a clear black and white conclusion – both show that the available studies have significant flaws and biases that make it difficult to draw solid conclusions. The two evaluations were done using a different approach; IARC looked at the evidence within a specific framework and set of rules for comparison with other evaluations, while ICNIRP looked at the overall weight of evidence and tried to make general conclusions.
Both evaluations acknowledge some major shortcomings in the data. In particular, cell phone use has only really been widespread for about 10 years, and some cancers may take longer than that to develop. There also aren’t any good data for risks to children, who might be expected to be more sensitive due to thinner skulls and still-developing brains.
Overall, I stand by my original conclusion that there is no reason to panic unless brain cancer incidence rates start to increase, but at the same time if you’re an extremely heavy user of cell phones it probably wouldn’t hurt to take at least minimal precautions. Also of note, the IARC evaluation mentions that the electromagnetic field exposure from newer 3G phones (or from Bluetooth headsets) is about 100 times lower than for traditional cell phones, which means that even if the amount of cell phone usage continues to increase, exposure will likely decrease.
Baan, R., Grosse, Y., Lauby-Secretan, B., El Ghissassi, F., Bouvard, V., Benbrahim-Tallaa, L., Guha, N., Islami, F., Galichet, L., & Straif, K. (2011). Carcinogenicity of radiofrequency electromagnetic fields The Lancet Oncology, 12 (7), 624-626 DOI: 10.1016/S1470-2045(11)70147-4
Swerdlow, A.J., Feychting, M., Green, A.C., Kheifets, L., & Savitz, D.A. (2011). Mobile Phones, Brain Tumours and the Interphone Study: Where Are We Now? Environmnetal Health Perspectives DOI: 10.1289/ehp.1103693