Rico's rants: Debunking

Daniel Engber has a Slate article about wearable tech:

I debunk today with a heavy heart, one that is filled with sorrow. Another silly science piece has been emitted to the readers of The New York Times, and zapped into their social media accounts. The Disruptions column in a recent Thursday Style section lays out the health risks of wearable technology. Could that Apple Watch or Google Glass pressed against our skin be dangerous: could cancer-causing radiation be their killer app?
To be honest, I feel a little bad. The column, by tech writer Nick Bilton, is so absurd in its analysis, so specious in its stats, and so annoying in its analogies that I wonder if we shouldn’t just avert our eyes and pretend it never happened. Bilton’s argument, that our wireless devices may be carcinogenic, runs so contrary to popular opinion that it may not be worth attacking. Billions of people use mobile phones without a trace of worry, and just a tiny group will give the Times column’s warnings any more than a moment’s thought. Must we really blast it from the sky?
Alas, facts are facts. The Times column is so weird, and so bad, and so weirdly bad, that this magazine has made the unusual choice of publishing twin takedowns, a rare double debunk. (The other one, from Phil Plait, is here.) I know we’re shooting fish in a barrel with both sides of a coach gun. But get a load of all those fish!
Bilton starts out by comparing the potential danger of wearing an Apple Watch to the health risks of smoking cigarettes. (One can think about wearable tech, he says, “in a similar vein” to tobacco.) This is now a standard move in health-scare journalism: remember when we learned that drinking soda ages you as much as a pack-a-day habit? Remember when the NFL was likened to Big Tobacco, hinting that concussions might be as bad as cigarettes? Remember when people said the same about the sugar lobby, and hinted that cavities might be as bad as cigarettes? Remember when people wondered which was worse, the ravages of smoking, or those of overeating?
Yeah, no. The relative risk of early death from smoking at least a pack of cigarettes per day is about three or four. That means that heavy users are three or four times more likely to die younger than similar people who never smoke. An estimated half million Americans perish every year from tobacco use, despite the fact that only eighteen percent of us are smokers. Here’s how many people have died from wearing Apple Watches: zero.
Bilton knows there’s no evidence of harm from wearable tech (he points out the Apple Watch isn’t yet on store shelves) but his column alludes to serious concerns about the health effects of Google Glass, which was first announced in 2012. What kind of serious concerns, exactly? The answer is buried in a link, on the words serious concerns, to a column in the UK’s Daily Mail. Never mind that it’s preposterous, by definition, to refer to anything published in the Daily Mail as a serious concern; even taken on its own terms, the linked article doesn’t offer any real grounds for worry. (I know we’re shooting fish in a barrel with both sides of a coach gun. But get a load of all those fish!)
Then Bilton turns to the main focus of his piece, cellphones. Since there’s been no definitive research on Google Glass, he says, why not “hypothesize a bit” based on what we know about other forms of mobile technology? That means embarking on a speculative résumé of the dangers of our iPhones, “which give off low levels of radiation, could lead to brain tumors, cancer, disturbed blood rhythms, and other health problems if held too close to the body for extended periods.”
The main citation for this claim is a Swedish oncologist named Lennart Hardell, who found in one study that talking on a mobile or cordless phone could triple the risk of “a certain kind of brain cancer,” Bilton says. (For the record, the linked research paper hints at a doubling of risk, not a tripling, but whatever.) In fact, Hardell— a one-man correlation engine— has published dozens of studies claiming links between wireless phones and cancer, going all the way back to 1999, when cellphones were still made of wood and spackle. In that time, he’s found suggestive evidence that a mobile habit leads to brain cancers, as well as non-Hodgkin’s lymphoma, testicular cancer, basal cell carcinoma, cheek and ear melanoma, and vestibular schwannoma. (He did find negative results for salivary gland tumors and meningiomas.) Oh yeah, and cellphone use might also cause stress, fatigue, and headache in teenagers, especially if they live in the sticks, where phones use more power.
As science journalist George Johnson points out in his superb book The Cancer Chronicles, most epidemiology doesn’t really support this notion. More people are using cellphones than ever before, yet the incidence of malignant brain tumors— a very rare condition— has been steadily decreasing over the last decade.
That may be why the Interphone study, a study of 6,420 brain tumor patients (as compared with more than 7,500 healthy controls), found no consistent relationship between cellphone use and cancer. The World Health Organization notes that animal studies show no increased cancer risk with long-term exposure to radio-frequency fields. “A large number of studies have been performed over the last two decades to assess whether mobile phones pose a potential health risk,” its fact sheet continues. “To date, no adverse health effects have been established as being caused by mobile phone use.”
Bilton mentions the Interphone study, but describes its negative results with a hedge: it “did not find strong links,” he says. That’s true only insofar as the study didn’t find any kinds of links at all. The study concluded that there isn’t any dose-response from cellphone radiation. In fact, for most subsets of analysis, the correlation appeared slightly negative. For the heaviest users, there was an association, but it seemed to be a product of statistical noise.
The Times piece, though, focuses instead on a ruling by the WHO’s International Agency for Research on Cancer, which, in 2011, declared cellphone use a Group 2B factor, meaning that it’s “possibly carcinogenic”. Bilton notes that the IARC’s expansive list of Group 2B chemicals includes those used in dry cleaning and pesticides. He neglects to mention many other, less scary-sounding ones, including aloe vera extract, caffeic acid (present in coffee), and— this may disturb readers in certain boroughs of the nation— the entire practice of woodworking. (“The epidemiological data available suggest that there may be a carcinogenic risk connected with employment as a carpenter or joiner,” the group said.)
To make things worse, so much worse, Bilton turns to an expert in alternative healing named Joseph Mercola. The herbal products magnate has, in the past, promoted using eggplant cream to cure skin cancer and walking barefoot on soil to absorb the Earth’s energy. (Even the scam-adjacent Dr. Oz has described Mercola’s views as “controversial” in the world of batshit-crazy medicine.) Anyway, when reached on his cellphone, Mercola tells Bilton that we should be cautious. Maybe he’s right to take precautions, since, you know, cellphones are on the list of Group 2B carcinogens and everything …
I went over to Mercola.com to see how the doctor views other Group 2B substances. Naturally, he’s flexible. Caffeic acid, the component in coffee that has been designated as “possibly carcinogenic” just like cellphones, turns out to be, per Mercola, a “valuable phytochemical” present in pea shoots. (You can get your daily fill from Dr. Mercola’s $60 Sprout Doctor Starter Kit.) What about aloe vera extract, another possible carcinogen? Last year Mercola called it an immune-booster that can be helpful for curing acid reflux.
With Mercola and Hennart to guide him, Bilton reaches the entirely inevitable conclusion that cellphones might be dangerous, and Apple Watches, too. “I’ll still buy one,” he writes, “but I won’t let it go anywhere near my head. And I won’t let any children I know play with it for extended periods of time.”
That’s a pretty easy stand to take, at least if you don’t have kids. It seems a little silly, though. If cellphone radiation does cause cancer, it would not be in the normal way. In The Cancer Chronicles, Johnson explains that only the highest-frequency radiation— UV light, X-rays, gamma rays— can penetrate our cells and tear up DNA. Microwaves can heat or vibrate tissue, but cellphone fields don’t do even that. If mobile technology does cause cancer, Johnson says, it would have to be through some subtler pathway, perhaps by inducing gentle oscillations in our cells’ metabolism. No one knows how, exactly, that would work.
In Johnson’s view, the giant COSMOS study, which will follow three hundred thousand cellphone users over multiple decades, could offer more conclusive data. But even that won’t be enough to really prove that phones are safe. Such definitive answers are few and far between in public health. We don’t even know for sure, I mean with a hundred percent certainty, that regular old power lines are fully safe. (Some have argued that they increase rates of childhood leukemia. In light of this inherent uncertainty, Johnson calls the cell phone scare a “metastatic meme”, like other “hard, impenetrable kernels of folk science” that get “passed from mind to mind with little deliberation”.
I think the more pressing question is how this particular kernel of folk science got passed into The New York Times. The newspaper has, of course, published excellent and balanced reviews of the very same research, yet here we have this stinky fish flopping into print. The problem may be one of entryways: some research goes straight into the paper’s rigorous, slow-moving Science section, while other studies sneak in through the Style desk or into Op-Ed. It’s in these other sections, and on blogs, that you’ll find the Times’ rare scientific mistakes. That’s not to say these stories should be held for Tuesdays, and ghettoized in Section D. Science has a place in every news vertical. But I think it’s fair to say its visits should be supervised.