HHS's call for phone-harm stories selects on the outcome it means to measure

Every account that arrives will come from someone who has already concluded the device was to blame, because that is who the request asks for ^[2]. Reports gathered that way can be counted, but they cannot be turned into a rate. There is no group of people with the same headaches or sleeplessness who attribute them to something else, and no measurement of what radiofrequency exposure anyone in the pile actually received. Kenneth Foster, professor emeritus of bioengineering at the University of Pennsylvania, expects the call to produce a lot of alarming personal anecdotes that shed little light on whether connected devices carry hidden harms ^[3].

"The way that any health agency would manage this," Foster said, "is to have a systematic, careful review looking at all the evidence, considering the strengths and weaknesses, and trying to put together some kind of consensus or holistic picture" ^[4]. "That process is very different than the process of soliciting opinions from various people about what they think happened to them," he said ^[5].

At the same conference Kennedy said HHS will investigate chemtrails, the lines of water condensation from airplanes that some people believe, without evidence, spread secret biological or chemical agents ^[7]. Scientific American separates the two inquiries: chemtrails are conspiracy, while electromagnetic fields and radiofrequency exposure have been studied for decades ^[8].

The band under review, 0 to 300 GHz, includes the exposure a cellphone delivers ^[9]. Radiation at those energies is not strong enough to break a DNA bond, which is how X-rays and other high-energy radiation cause cancer ^[10]. That does not rule out other routes, and the reviews looked for them: results on oxidative stress and fertility in lab animals came back highly variable ^[11].

ARPANSA, the Australian Radiation Protection and Nuclear Safety Agency, whose professor Ken Karapidis worked on the WHO-sponsored reviews ^[14], responded through a spokesperson. It said that "numerous epidemiological studies, including large cohort and case-control studies, have not found evidence that mobile phone use increases the risk of brain cancer or other cancers" ^[13]. The spokesperson also pointed to Karapidis's ecological work, which found that wireless technology use rose massively in Australia over 20 years with no rise in brain cancer incidence ^[15]. An ecological comparison cannot resolve one person's risk from their own handset, though it does test whether a large population-level effect showed up while the exposure spread. The spokesperson's summary was that "the body of evidence accumulated over several decades provides reassurance that exposures from wireless technologies, when they comply with established exposure limits, do not pose a risk to human health" ^[16].

The same reviews were explicit about the quality of what they had to work with. Their authors wrote that their conclusions could be made only with so much confidence, given a lack of high-quality evidence ^[20], and Scientific American reports a continuing need for high-quality confirmatory research ^[21]. Foster's account of the annual output explains why the public record looks less settled than the reviews. About 100 studies come out a year, roughly half of which report biological effects ^[18]. That works out to some 50 papers a year that a worried person can find ^[1]. Most, Foster says, are small exploratory efforts prone to bias ^[17]. "These studies are generally not very persuasive, but John Q. Public has access to the Internet, and they can find all these studies, and they get scared," Foster said ^[19].