December 18, 2009
Spin, Spin, Spin
Pity those who are trying to follow the cell phone–brain tumor story. Their sense of the cancer risk is most likely a reflection of the last thing they read or saw on TV —It all depends on whose sound bite they happen to catch.
Take, for example, a paper published earlier this month in the Journal of the National Cancer Institute (JNCI) by a team of Scandinavian epidemiologists, under a rather bland title — "Time Trends in Brain Tumor Incidence Rates in Denmark, Finland, Norway and Sweden, 1974–2003." But its message is anything but: Because there has been no increase in brain tumors between 1998 and 2003, a period when the use of cell phones "increased sharply," cell phones are cancer safe.
The raw data had already been reported earlier this year in an obscure EC report but had escaped notice by all except the most attentive observers. This time around, JNCI issued a press release with cell phones in the headline. The paper immediately became part of the 24-hour news cycle. "Study Finds No Brain Tumor Link with Mobile Phones," reported Reuters. "Mobile Phones 'Have Not Increased Brain Cancers'," said BBC News. "Four Country Study No Cancer Link to Cell Phone Usage," ran the headline in USA Today.
No one was more emphatic than Brian Williams, the anchor at NBC Nightly News. Here's what he told his six million viewers:
"There's an intriguing headline tonight out of Scandinavia about cell phones and cancer, specifically brain tumors. There's been speculation for years that there has been a link between them, but tonight the results of a very large, very long study of just about everybody in Scandinavia found no link." (Watch the video.)In fact, the new study does not tell us anything new. As was pointed out in that obscure EC report, these results do nothing to reduce the "level of uncertainty" about long-term tumor risks. The Scandinavian researchers concede this in their JNCI abstract: the new data only applies to those who have been using cell phones for five to ten years. For quite a while now, it has been clear that, except for some groups of very heavy phone users, there is no apparent tumor risk until after at least 10 years. The cell phone–brain tumor controversy is about what might happen after that, quite likely after 20-30 years. That's where the uncertainty lies. Anything that does not address this question is not news. It simply provides false reassurances.
Most of the news stories do say that follow-up studies are needed, usually somewhere near the end. We long ago became inured to such boilerplate. When was the last time, some health report didn't end with "more research is needed"? In this case, the main message —that cell phones are safe— must have gotten through long before the reader got to the call for more studies.
It would be easy to blame JNCI and its editor-in-chief, Barnett Kramer, for promoting this paper. This is a replay of how they tried to bury the cell phone story back in 2001. At the time, the journal tapped Robert Park, a physicist, to make the case that some Danish epi data were sufficient to put an end to public concerns about cancer. It didn't work then; we doubt it'll work now. (Disclosure: In that editorial, Park called Microwave News a "fear merchant" for our unwillingness to accept the establishment's party line that the cell phone story is a non-issue.)
But some of the responsibility must lie with the Scandinavian researchers themselves. They left out what most serious epidemiologists would consider a key benchmark: How small a risk could their study detect? There's not a word about this in their paper, perhaps because their study did not have the power to tease out anything but a very large effect.
"It's complete nonsense that an increased risk could be seen in this type of incidence data," Michael Kundi, the head of the Institute of Environmental Health at the Medical University of Vienna, told Microwave News. "One must be very simple-minded not to see this immediately." When asked what kind of risk it could have detected, Kundi replied that the answer depended on a number of assumptions, but that it would be somewhere between a 7-fold and 15-fold increase among long-term users. In other words, the study that is being touted as an all clear could be masking an effect that increases the tumor risk by a factor or ten or perhaps a lot more.
Another aspect of the JNCI paper that went unmentioned in the news stories is that all but one of its six authors are members of the Interphone study project and that they have previously published papers showing statistically significant risks of two types of tumors —gliomas and acoustic neuromas— after ten years of use. So the same people who have shown that there might be a long-term risk are now saying cell phones are safe without ever referring to their own earlier, troubling findings.
It's no secret that many of the authors of the new paper are members of the "There-Is-No-Tumor-Risk-Whatever-the-Data-May-Show" bloc within the Interphone project. They insist that the elevated risks that they themselves, as well as others, have reported are spurious —due to recall bias— and should not be trusted. The deadlock within Interphone between this group and those who do not wear the same rose-colored glasses is the main reason why the Interphone results are still not published more than four years after a draft was first circulated.
Coincidentally —or maybe not— the day before JNCI sent out its press release, the New Scientist, which has a few skeletons in its own EMF closet, published a piece trashing Interphone by Mike Repacholi, the former head of the WHO EMF project. Repacholi is himself a master of the spin game. Here, he argued that "Interphone can't end the debate" over tumor risks because it is riddled with "major flaws." He played the same semantic games as the Scandinavian epidemiologists, writing that the Interphone results published so far have been "mostly negative." His only concession was that they "could not rule out a possible elevated risk of glioma or acoustic neuroma for more than 10 years." He too could not bring himself to say that the 10-year data do point to such tumor risks.
To understand why the spinmeisters are working so hard to pacify the public, you need only to read an advisory released by the EC Research Information Center last summer. It stated, in part: "Although the conditions in which mobile phones are harmful to public health are not clearly established, one can reasonably doubt that the friendly mobile is totally innocent." (This warning was based, in part, on the long-term tumor risks seen in the Scandinavians' epidemiological studies of glioma and acoustic neuroma; the same ones that they left out of their new paper in JNCI.) And just a few days ago, an article in the Financial Times called for consumers to be able to select cell phones on the basis of their SARs —a proposition that has been anathema to Motorola, Nokia and the other manufacturers since the controversy began in the early 1990s. "If the industry really is confident about its products' safety," the FT stated, "it should be more willing to display radiation ratings."
To be sure, the activist community has been doing its own spinning. Last summer, when some —in hindsight, naively— thought that the Interphone paper would soon be released, a team led by Lloyd Morgan, a brain tumor survivor, posted its own position paper, Cell Phones and Brain Tumors: 15 Reasons for Concern, Science, Spin and the Truth Behind Interphone. (Spin meets spin!) And earlier this month, Morgan, together with David Carpenter, Devra Davis and Joel Moskowitz, held a "Media Teleclass on Cell Phones and Brain Tumors."
Nor is the Interphone paper the only one that is having trouble getting into print. At a Congressional hearing in September 2008, Ron Herberman, then the director of the University of Pittsburgh Cancer Institute, publicized some intriguing new U.S. data that show an uptick in brain cancer among 20-29 year olds between 1995 and 2005. These statistics would tell a different story than the latest Scandinavian trend data. Robert Hoover of the NCI testified that he was unaware of any such increase. At the time, Davis, who is collaborating with Herberman, said that a paper would be soon submitted for publication. (The hearing was called by Rep. Dennis Kucinich (D-OH); see: "Are Brain Cancer Rates Rising Among Young Adults?," September 30, 2008.)
So far, the paper has failed to clear peer review. In an interview, Devra Davis said that reviewers at two different journals had found it "too speculative." Nevertheless, she maintains that they do see an increase among the young. "The paper has been submitted to a third journal," Davis said.
As for Interphone, two of the principal investigators have recently told us that it would finally appear in the "next couple of months." (Elisabeth Cardis, the head of the project, did not respond to requests for comment.) We've given up guessing on the timing, but we would make a big bet that when Interphone does go public, the spinning from all sides would make a whirling dervish dizzy.