To get a sense of the total, complete, and utter mess that is research on the health effects of cell phones, look no further than a study of whether the ubiquitous gadgets raise the risk of brain tumors. “Interphone,” organized by the World Health Organization’s International Agency for Research on Cancer, was the largest (10,751 subjects, ages 30 to 59, in 13 countries), longest (10 years), most expensive (as much as $30 million), and most labor-intensive (48 scientists) study of its kind. That boded well for producing credible conclusions. Instead, Interphone found that using a cell phone decreased the risk of glioma (primary brain cancer) by 19 percent. Even in people who had used cell phones for more than 10 years there was no increased risk of brain tumors, with the exception of those who said they had yakked away for more than 1,640 hours. And the 40 percent increased risk of glioma in this group came with a caveat that is emblematic of this field: this elevated risk, the scientists warned, may be an artifact of “biases and error,” not real. Things got so acrimonious among Interphone scientists that they delayed announcing the results, finally released in May, for four years.
There are many, many ways to screw up experiments on the biological effects of cell-phone radiation, and in 20 years of studies scientists seem to have used every one. The result is a confused public and nearly incoherent government policies that careen back and forth like a drunk after last call. In April, Maine legislators voted against requiring warning labels on cell phones. In May, San Francisco mandated them. A bill to be introduced in Congress would require warning labels nationwide and create a research program—but the last time the government called for studies that would “finally” answer whether cell phones pose a risk of cancer was in 1999, and since then all that’s been accomplished are studies on how to do the studies. Society has never been good at making decisions in the face of scientific uncertainty (what do we do about possibly carcinogenic pesticides? About climate change?), but with cell phones the situation is even worse: it may be impossible to get definitive answers in a reasonable time about whether the radio-frequency radiation the devices emit will kill any of the 4.6 billion people who now use them.
The first big uh-oh experiment, done in Australia and published in 1997, exposed mice to the radiation typical of cell phones (about 800 megahertz to more than 2 gigahertz; this study used 900 MHz) for one hour a day for 18 months. The mice got lymphoma at 2.4 times the rate that unexposed mice did. The alarming finding set off a stampede of research. Two studies in Texas, in 1998, exposed mice to 2,450-MHz radiation for 20 hours a day, every day, for 78 weeks, finding no extra breast cancers compared with mice that weren’t zapped. A 2002 study in Germany, exposing mice to 900 MHz, found no increase in breast cancer. A 2002 Australian study—900 MHz, an hour a day, five days a week, for two years—looked for an increase in lymphomas: nothing. The biggest set of animal tests—called Perform-A, it took eight years, cost $10 million, was organized by the European Commission, and announced results in 2007—found no evidence that cell radiation induces or promotes cancer in exposed mice or rats.
Tests of whether the radiation kills, slows, or otherwise harms sperm are also enough to give you whiplash. In 2003: “no evidence suggesting an adverse effect of cell phone exposure on…testicular function or structure”; 2005: “no support to suggestions of adverse effect on spermatogenesis”; 2005: “a significant genotoxic effect on…spermatozoa”; 2007: “significantly higher incidence of sperm cell death” suggesting “that carrying cell phones near reproductive organs could negatively affect male fertility”; 2009: “significantly reduced percentage of motile sperm.”
Scientists who believe cell phones are dangerous have been throwing out hypotheses to explain away the negative results. Maybe something about the unexposed animals raised their rates of cancer or sperm problems, so by comparison the exposed animals didn’t seem to be harmed. Maybe the studies should have used pulsed, on-off radiation rather than a continuous beam, the better to mimic the way we actually use mobile phones, suggests epidemiologist Devra Davis, whose book Disconnect: The Truth About Cell Phone Radiation, What the Industry Has Done to Hide It, and How to Protect Your Family will be published next month. Maybe it matters whether the lab animals are zapped while in a device like a Ferris wheel or while running around in cages. On the other hand, if these details do matter, maybe that in itself is significant. “One could say that if [cell phones’ inducing cancer] is a robust effect, then it should show up regardless of the setup,” says Louis Slesin, who as editor of Microwave News has been following this issue since 1981. “But since we don’t know the mechanism [by which cell-phone radiation might cause cancer], it’s possible that by changing something we’ve eliminated the effect.”
Scientists who dismiss claims that cell-phone radiation is causing an epidemic of brain cancer argue that there isn’t any mechanism. According to textbook biophysics, only radiation that has enough energy to ionize molecules—that is, knock off electrons—can trigger cancer. (X-rays are ionizing radiation, with some 15 million times the energy of cell phones.) Cell phones don’t emit energy great enough to ionize molecules in living cells. Their radiation is “far below the cancer energy threshold,” notes physicist Robert Park of the University of Maryland, who often battles junk science. But whenever he makes that point in his What’s New e-newsletter, he gets flooded with angry responses insisting there are other ways low-energy radiation can do harm. “I don’t like cell phones and I don’t like writing about cell phones,” says Park, “but the damned issue just won’t go away.”
Studies of the effects of cell-phone radiation on cells growing in lab dishes are also all over the map. A 1997 experiment found that mast cells exposed to 835-MHz radiation proliferated in an unusual way, but another experiment that same year found no increase in cell proliferation under exposure to 837 MHz. A 2004 study found no evidence of DNA damage in rat brain cells when they were exposed to on-off cell-phone radiation, but a 2005 study found breaks in DNA strands in human and rat cells when they were exposed to 1,800 MHz. A 2008 study at the Medical University of Vienna found that cell-phone radiation caused numerous DNA-strand breaks. But a university investigation found that its data had been fabricated. “After literally hundreds of studies, there is nothing approaching a smoking gun,” says Peter Sandman, a risk-communication expert in Princeton, N.J. “Most experts look at this pattern [of no effect or a weak effect] and conclude the risk is either small or nonexistent. If it were big it wouldn’t keep disappearing and reappearing from one study to the next.”
Even scientists who believe we are conducting a vast and dangerous experiment on our brains by using cell phones concede that the balance of evidence, from both in vitro and in vivo experiments, indicates that cell-phone radiation does not alter DNA or chromosomes in a way likely to initiate tumors. “There is no denying that the majority of published studies on radio frequency radiation and the brain do not show any impact,” writes Devra Davis. “The preponderance of the evidence shows that cell phone radiation has little biological impact.” She attributes this to the ways scientists can cook the books to avoid finding an effect.
Studying humans hasn’t clarified things. The first big epidemiology study, in Denmark, used government cancer registries and company records of cell-phone charges. It found no increased rate of brain cancer, the scientists reported in 2001. But almost none of the subjects had used a cell for more than 10 years; 92 percent used one for less than five years. Since the latency period between exposure to a carcinogen and the appearance of brain cancer could well be 30 or 40 years, the negative finding isn’t completely reassuring. Also, the scientists lumped together all cell users, which could hide an effect that appears only at high exposures, and included no children or teens, who might be especially vulnerable.
The Interphone study, which studied people with brain cancers diagnosed between 2000 and 2004, was supposed to do it right. But its finding that using a cell phone decreased the risk of glioma suggests the study was somehow flawed, because there is no plausible biological mechanism by which using a cell phone can be protective. The most likely glitch: something skewed the statistics in a way that reduced the calculated risk of using a cell phone by about 5 to 15 percent, according to Jonathan Samet of the University of Southern California, who analyzed the study for the International Journal of Epidemiology. If the same methodological flaws (which have to do with the control group, the Interphone scientists concede) occurred in the calculation of risk for the heaviest cell users, then their increased risk is both real and greater than reported. The top 10 percent of users—about half an hour a day for 10 years—would have an 82 percent higher risk of glioma. With about seven new cases per 100,000 people per year, that translates into increasing the annual risk from 0.007 percent to about 0.013 percent.
On the other hand—and when it comes to cell phones and cancer, there is always another hand—the study might exaggerate the true risk. If people with brain cancer unconsciously inflate their usage (“I have brain cancer: I must have used my cell hours and hours every week”), it would create a false positive, linking brain tumors to heavy cell use. Indeed, heavy users did overestimate, the Interphone scientists found, with some subjects saying they used their cell five and even 12 hours a day. With these methodological and statistical questions unresolved, says Samet, “the question as to whether mobile-phone use increases risk for brain cancers remains open. We simply do not know the answer.”
It’s not clear whether anyone has the stomach for another years-long study like Interphone to try to find the answer. To do it right would require documenting phone use precisely with company records, and cross-linking those data with cancer registries. Another obstacle is that brain cancer is rare. You’d need to study some 9.6 million people to pick up an elevated incidence of even 20 percent. As for animal tests, this month the National Toxicology Program (NTP), part of the National Institutes of Health, is finally beginning the ones the Food and Drug Administration requested a decade ago. The NTP expects to launch the main studies—exposing animals to radiation for two years—in 2011, and to report results in 2014.
War, Peace and the Weather: Floods in Pakistan and Implications for Afghanistan; America's Number ... 11?; Net Neutrality; Cellphone Fears; Eat, Pray, See?; From the Archives: Health Reform, Psychology and Politics
In the meantime, we will keep using our cell phones. It is striking how little the suspected danger resonates with users. Sure, some don corded headsets or use Bluetooth, which reduce radiation exposure (if you switch off the Bluetooth between calls). Some have given up talking in favor of texting, which by keeping the phone at a distance reduces radiation exposure. A few users even heed manufacturers’ warnings (buried in the instructions) about how far to keep a phone from the body: an inch for the BlackBerry 8300, for instance, and five eighths of an inch for the iPhone. It will be interesting to see whether labeling affects consumer behavior. The relevant spec is SAR, or specific absorption rate, a measurement (in watts per kilogram) of how much radiation is absorbed by biological tissue; this is what San Francisco will require be posted at the point of sale. The legal maximum SAR in the United States is 1.6, but SARs vary widely, with the Motorola V195 at 1.6, but the Samsung Freeform at only 0.48.
For now, most users aren’t letting the warnings dull their cellular infatuation. “People don’t need to read up on the studies to get the accurate impression that the experts simply aren’t sure,” says Peter Sandman. “When people are inclined not to worry [because we love our cell phones] this uncertainty becomes a reason for calm: ‘Even the experts don’t know whether it’s dangerous or not, so why sweat it!’ We really, really don’t want to learn that there are health reasons to restrict our use of this miraculous invention. Uncertainty gives us a reason to stay unconcerned.” And since the scientific uncertainty may never go away, the vast experiment we are embarked on won’t either.