COSMOS - Mobile Phones and Health

Chris - With mobile communications playing an increasingly major part in our lives, many people are worried, quite justifiably, about the potential health impacts of this technology.  For this reason, a large study called COSMOS has been set up to track the ongoing health of 250,000 mobile phone users over a 30-year period.  The leader of the study is Professor Paul Elliott.  He's from Imperial College in London and he's with us today. Hello, Paul.

Paul -   Hi, Chris.

Chris -   Every good study has got to have a fantastic acronym to represent it, COSMOS being no exception.  What does it stand for?

Paul -   It's a COhort Study on MObile phoneS in health, and the key there is the 'cohort' which means, as you say, a prospective study over many years following the health of people in relation to their mobile phone usage.

Chris -   It sounds fairly trivial, but I'm sure it's not, to recruit a huge number of people and then try to get data from them, so how have you gone about this?  Who were the study subjects and how are you following them?

Paul -   We identified people through mobile phone companies who sent out a large number of letters to subscribers, and the idea was to get people at different levels of usage - low, medium, and high, so that we can get a broad range of use.  All these letters went out to subscribers in the UK and then we invited people who received those letters to log in to our website, answer the questionnaire, and give consent to take part in the study.

Chris -   How often are you following the people up?

Paul -   They give consent to us getting copies of their usage of mobile phones through the operators every year.  So we get three months usage every year, and then the idea is to get back to them on regular intervals with questionnaires.  Regular being over a 4-year period so we plan to go back to them with a questionnaire.

Chris -   What about keeping in touch with them?  One problem that big longitudinal studies like yours often face is people who get lost to follow up for various reasons.  So, how are you going to make sure you keep people in the loop?

Paul -   Clearly working on mobile phones is good.  We can track their phone either directly through the participant or through the mobile phone operators.  The participants also give consent for us to track their health through routine registries such as the cancer registry, the mortality register, and also hospital admission registers.  So we can track people and their health even if we can't find the people themselves.

Chris -   It's quite neat that you can get access to how often they spend on the phone which enables you to establish the dose that they're getting, because obviously when one is looking for causation there's got to be a dose dependent relationship, hasn't there?  The more you use something, the more of an outcome we should see.

Paul -   Exactly.  The key is to get exposure variation, as you say.  The proxy for this will be how often people use the phone, and by stratifying the cohort of outset into low, medium, and high users, we've actually got a very good distribution of usage at baseline.  Then we need to follow up, through continually looking at the mobile phone records, how much people are using.  Then we look for a dose effect in terms of potential health outcomes.

Chris -   One possible confounder here could be that not everyone uses their phone with it glued to their head.  Some people use their phone with it on the table in front of them and they're reading some texts or a Twitter feed.  Can you discriminate between active calls where someone is dosing their brain with radiation at very close measures and more distant operations like that?

Paul -   Yes.  We can differentiate between data use, SMS, and active phone use.  But of course even with active phone use some people will use a hands-free device which will reduce the exposure to the head.  So we do ask about use of hands-free devices as well.

Chris -   What about children though?  One thing I've heard people say a lot is that as children are growing up their bones are less dense, therefore their skull is less dense, therefore the attenuation of any microwaves coming out of the phone going into the brain will be lower in a child.  So their relative dose to potentially sensitive tissue inside could be higher.  Have you got them represented in your cohort too?

Paul -   No.  Our study is a study of adults, but you raise an important question.  Clearly children are exposed.  From what you said earlier in your programme it sounds like children have more phones that they have books!  So this is clearly an issue and it is a question for active research.  Does use in children affect, for example, their concentration, their sleep?  And this is something that, for example, the WHO has called for research on because children are currently under researched.

Chris -   It sounds like the study has been well put together and you're going to follow these people up, and we're going to see if there are any health outcomes.  What do we already know about the risks?  Are there any preliminary data to suggest that there might be an effect?  In other words, what do you think might come up?  If there is an effect, what are you expecting to see?

Paul -   There have been a number of so-called 'case-control' studies.  These are a different design where you find cases of a disease, in particular people looked at different forms of brain cancer, and then a set of control participants who don't have the disease, and ask about their use of mobile phones going backward in time.  So this is based on subjective reporting.  Those studies have been reassuring up to about 10 years of usage.  So that's good news, but what we don't know about is longer term use.  That's where the COSMOS study will be very important because we'll be able to follow people up over many years of usage.

Chris -   Indeed.  It takes 50-60 years of dedicated subscription to cigarettes and the smoking habit to get lung cancer.  So it may be that just looking in the near term, which many of these studies have, is just too short for us to see enough dose for a small or marginal increase in, say, brain cancer rates, if there is a risk of that.

Paul -   That's absolutely correct.

Chris -   So, where do you see this playing out to now?  Are you going to stick to this cohort or are you going to build in additional people who will come along behind this initial group in order to address some of the other questions that you've raised with us today?

Paul -   The cohort is a multinational study, so the UK is one component and we have colleagues also working with us in various other countries: Denmark, Sweden, Finland, and the Netherlands.  The idea is, as you say, to get to a cohort of around 250,000 people.  The study is open to other countries to take part in as long as they can follow the protocol.  I think that clearly the larger the study the more power it will have to detect what might be quite small effects, if there is indeed any effect of long term mobile phone use on health.

Chris -   Paul, thank you very much.