Back in the news – mobile phones and cancer

Headlines yesterday suggesting heavy mobile phone use is ‘likely’ to cause cancer are misleading, and don’t reflect what the evidence really shows.

We’ve discussed the problems with a lot of the evidence around mobile phones and cancer before. Overall, this evidence shows that mobiles are unlikely to cause brain tumours (nor any other type of cancer).

It’s important to consider this whole body of evidence, rather than individual studies, particularly since lots of these have come to different conclusions. And many of the studies have issues, meaning that their findings may not be completely reliable.

Yesterday’s study is no exception. Let’s have a look at what it found, why we’re not convinced by its findings, and how it fits into the bigger picture.

New research

The latest study, carried out in France and published in the journal Occupational and Environmental Medicine, is a type of study called a case-control study. These studies compared people with cancer to similar healthy people, along with information about their lives to find patterns.

The new study looked at two specific types of brain tumour – glioma and meningioma. Crucially, the scientists found no link overall between whether people used a mobile phone and whether they developed either of these types of tumour. But they did see a tentative link among a small group of people who said they’d used their phone the most. Hence the headlines.

But before we think about where this finding sits in relation to existing research, it’s important to note that there were a number of problems with the study:

• People were asked to remember detailed information about their mobile phone use in the past. The ‘controls’ – healthy people on the study chosen because they were broadly similar (e.g. in age and sex) to the patients involved – had to remember back more than a year longer than people with tumours. And the patients’ memories could have been affected by their beliefs about whether mobile phones caused their tumour, or even by the tumour itself.
• For nearly a fifth of the people diagnosed with tumours – those who were very sick or who had died, a simplified questionnaire was filled out on their behalf by their relatives.
• The study is quite small: the ‘highest use’ group contained only around 20 glioma cases and 10 meningioma cases. Although some of the results are ‘statistically significant’ (in other words, unlikely to be explained solely by chance), the researchers don’t seem to have corrected for doing several tests (this is important, since the more statistical tests you do during a study, the more likely one of them is to seem ‘significant’ by chance).
• The study was carried out several years ago – the tumours were diagnosed between 2004-2006. Brain tumours generally take a long time to develop. So the types of phones and the way they worked back then mean the results may not hold for more modern technology.

What does other research show?

In 2011, the International Agency for Research on Cancer (IARC) concluded that the evidence on mobile phones and gliomas was ‘limited’, as it was for another type of brain tumour (acoustic neuroma), and that for other cancer types there wasn’t enough good evidence to come to a conclusion at all (Though it’s worth noting that, as you might expect with such a controversial subject, a minority of the expert panel felt the evidence around glioma and acoustic neuroma was also inadequate to reach a conclusion.

How did they arrive at this conclusion? IARC’s decision was dominated by two studies. One called INTERPHONE (a very large, international study), and a separate large Swedish study. We’ve summarised these findings in this previous blog post, so we won’t rehash that here. But it’s worth noting both of these studies followed the same (case-control) design as this new French study – looking back at what people remember doing. A much more reliable method is known as a ‘cohort’ design, which asks people about their current phone use, then follows them over time to see what happens to them.

Since IARC’s report, two more notable cohort studies have been published. The first, a Danish study published just too late to be included in the IARC discussions, found no link with any type of brain tumour at all, nor with glioma or meningioma in particular. This study was based on subscriber data about people’s mobile use, though, which could theoretically introduce errors if people shared mobiles or used a phone in someone else’s name or a company phone.

Then last summer the biggest study so far – the UK Million Women Study – found no link between mobile phones and any kind of cancer, any kind of brain tumour, nor glioma or meningioma in particular.

There’s a caveat – in the women who used phones the most, the research did show a small increased risk of developing a type of rare brain tumour called an acoustic neuroma.

But an update to this research, including even more recent data meant that this link was no longer seen, nor any link to any other type of brain tumour. The Million Women Study is large and well-conducted, but does still have some limitations – not least that it is restricted to women and there was limited opportunity to update women’s phone use after the start of the study.

But the Danish cohort study found no increased risk of acoustic neuromas, even in people with the highest levels of use. And neither did a cohort study conducted in Sweden published earlier this year. This Swedish study interestingly found evidence to suggest that using a mobile phone may make people who have an acoustic neuroma more likely to have it diagnosed, which could help to explain the conflicting results seen so far.

What next?

As it currently stands, it seems unlikely that using a mobile phone can cause brain tumours, particularly as lab research hasn’t shown a biological way this could happen. And rates of the brain tumours in question haven’t seen an increase over time, despite mobile phone use booming.

But there isn’t enough good evidence to say with absolute confidence that no risk exists.

It’s a well-worn cliché that more research is needed, but in this case it’s the only way to resolve the question and be sure. Some key members of the IARC expert panel recently published their take on where we go next, including recommendations for research that needs to be done.

By focusing on high-quality research, learning from the limitations of past studies, and the ability to gather more long-term data as time goes on, we will eventually come to a clear understanding of any cancer risks posed by using a mobile phone.

It’s important to remember the overall risk of developing brain tumours is small – fewer than 10,000 people are diagnosed with any form of brain or central nervous system tumour each year in the UK, out of a population of more than 63 million.

But when the media report small increases in these already-small chances as meaning people who do a particular, common, thing are “particularly likely” to develop a disease – as one newspaper did – it not only is a complete misrepresentation of what a study says or means. It also scares people.

This may sell newspapers, but it’s extremely unhelpful when trying to help the public understand what they can – and can’t – do to help reduce their chances of cancer.

Sarah

Reference

Coureau G., Bouvier G., Lebailly P., Fabbro-Peray P., Gruber A., Leffondre K., Guillamo J.S., Loiseau H., Mathoulin-Pelissier S. & Salamon R. & Mobile phone use and brain tumours in the CERENAT case-control study, Occupational and Environmental Medicine, DOI: 10.1136/oemed-2013-101754

Image via Wikimedia Commons