December 2, 2013

Don’t be scared of WiFi

From TV One Breakfast this morning “WiFi detrimental to health, study suggests“. (I didn’t see this live; the Science Media Centre contacted me for comment)

The guest, Mr Kasper from Safe Wireless Technology NZ, said

Overseas research has shown that a person who uses a mobile phone for a year increases their chances of getting brain cancer by 70%, according to the SWTNZ.

The ‘overseas research’ appears to be this, a case-control study of acoustic neuroma in Scandinavia. The first thing to note is that “acoustic neuroma” isn’t the same thing as “brain cancer”. Acoustic neuroma is a rare, benign brain tumour (‘benign’ means it doesn’t spread metastatically), which is usually treatable, though often with long-term effects.  The researchers didn’t suggest that their results applied to other brain tumours; in fact, they assumed the opposite and used people with a different brain tumour, meningioma, as one set of controls for their comparisons.

The story  also says

“There’s so much research and there’s so much scientific evidence now that does more than just suggest that there is a real problem, and people are getting these problems,” Mr Kasper said.

The National Cancer Institute has a good summary of the scientific evidence, and they are not at all convinced. It certainly isn’t the case that there’s a strong association with brain cancer overall.

The new research appears to be better conducted than a lot of the past claims of associations between radio waves and health. It’s working against a strong burden of proof both from animal studies and from the fact that radio waves can’t damage DNA. I don’t think it manages that level of proof, but I think reasonable people could disagree. However, even if we assume that the association specifically with acoustic neuroma is real and causal, it doesn’t really support any concern over WiFi. Cellphones are pressed up against the ear, and so provide higher dose of radio-frequency energy to that ear. WiFi transmitters are typically not pressed up against the ear, and each doubling of the distance reduces the energy by a factor of four. And, since they don’t have to reach as far, WiFi signals are less powerful to begin with.

The story ends with

“We do want the Government to put some money into some independent research.”

I’m generally in favour of the Government putting money into research, but on this particular topic  there’s no real advantage to the research being done in New Zealand, and we have too small a population to contribute much. There are large international studies ongoing; we don’t need a small local one.

If you are worried about cellphones and acoustic neuroma, use headphones with your cellphone. If you are worried about WiFi and brain cancer, then relax.

 

avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »

Comments

  • avatar
    Martin Connelly

    There was also a recent story suggesting that cordless phones were more of a problem than cellphones, at least for young people.
    http://www.stuff.co.nz/technology/gadgets/9309643/Pre-teens-at-risk-from-cordless-phones

    10 years ago

    • avatar
      Thomas Lumley

      Though the new paper says the opposite.

      10 years ago

  • avatar

    Thanks for your comment on this. It’s interesting to hear these conflicting studies, it makes me wonder if there is any place for individual experiential ‘evidence’? Not that I’m necessarily saying there is direct causation, but do you think I should completely ignore the fact that exactly when the brain tumour I had last year was growing, I spent three months in Togo talking on my cellphone (international calls) for at least an hour a day…and doctors have no other explanation for it? Cheers.

    10 years ago

    • avatar
      Thomas Lumley

      The problem with individual experience is that nearly all tumours are unexplained, and nearly all heavy cellphone users don’t get tumours, so it’s very hard to distinguish causation from coincidence. That’s why the research involves pooling the individual experiences of thousands of people with cancer.

      The place where individual experience is most valuable is with immediate, repeatable things. For example, if someone believes they get headaches from WiFi, it would be easy to do a simple bllnded test, to see if they can tell whether a WiFi transmitter is switched on. With twenty or thirty repetitions we could get reasonably convincing evidence one way or the other.

      10 years ago

      • avatar

        Thanks for that – I completely agree. It’s certainly made me interested in the best research methods concerning past (remembered) experiences that are inevitably coloured by what has happened since..

        10 years ago

  • avatar

    All RF EMF was categorised as a class 2B carcinogen in 2011 by the IARC. A team of World renown scientists in the field of Bioelectromagnetics reviewed the larger studies and found sufficient scientific evidence for RF EMF to “possibly” cause cancer and categorised it as a class 2B carcinogen.

    The International Agency for Research on Cancer (IARC) is a part of the World Health Organisation (WHO).

    Cheers
    TGreening
    BTech Information Engineering
    BEMS.ORG Professional Member

    10 years ago

    • avatar
      Thomas Lumley

      To quote the IARC definition:

      Group 2B: “Possibly carcinogenic to humans” There is some evidence that it can cause cancer in humans but at present it is far from conclusive.

      I’d also note that IARC has only ever classified one thing as “not a carcinogen”.

      10 years ago

      • avatar
        Thomas Lumley

        And also, as I pointed out when Group 1 carcinogens came up earlier this year, the IARC definition deliberately has nothing to do with risk at usual levels of exposure.

        10 years ago