I hate to pick on a fellow news outlet, but this was a breathtaking example of irresponsible science hype from Mother Jones:
What actually happened is that scientists released the results of a big — but not yet peer-reviewed — study in which they exposed rats to the sorts of radio frequency radiation emitted by cellphones and found higher incidences of two types of tumors in male rats, compared with a control group. (Though, oddly, the exposed rats also lived longer.) It's the first study to show non-ionizing radiation causing cancer.
But there are major caveats here: This is just one study (we shouldn't dismiss it, but it's quite possible the results were simply due to chance). The effects were only found in rats (and may not translate at all to humans). And this needs to be weighed against other evidence that cellphones aren't a big risk for people (we've been using phones for decades now with no uptick in brain cancer). This is an important bit of research and deserves careful scrutiny and follow-up. But it's not an occasion for fear-mongering.
Look: Science moves slowly. Individual studies are often wrong, and it's rare for one paper to completely upend everything we know about a topic. There are very few genuinely "game-changing" studies. And reporters need to do a better job of putting this incrementalism in context — rather than preying on people's fears for clicks.
(I'll note that there were quite a few careful, nuanced stories on the cellphone study, including this from Rachel Feltman of the Washington Post and this from Megan Thielking and Dylan Scott of Stat News.)
What the new study on cellphones, rats, and cancer found
The US National Toxicology Program, which is part of the National Institutes of Health (NIH), released the results of a big, two-year study in which researchers exposed rats to the types of radio frequency radiation that cellphones emit and compared them with a control group.
You can read the paper here. Note that the findings have been reviewed by other experts, but they haven't yet gone through a formal journal peer review, which may lead to further edits. An NIH spokesperson told Stat News that the agency is also reviewing the work.
A careful summary of the results was done by Aaron Carroll, Professor of Pediatrics and Associate Dean for Research Mentoring at Indiana University School of Medicine. I'll turn the mic over to him:
They exposed pregnant rats to whole body CDMA- and GSM-modulated radiofrequency radiation, for 9 hours a day, 7 days a week. Then they exposed 90 pups of each sex to three levels of each type of radiation for up to 2 years ...
At the end of the study, survival was lower in the control group of males than in all the exposed males. Survival was lower in the control group of females for two of the three exposed groups. Yet no headlines blared that cell phones extend life. Nor will mine. No statistics are presented on whether this is significant.
Now let’s get to brain cancer. There were no significant differences in the incidences of lesions in exposed male rats compared to controls. There was a "statistically significant positive trend in the incidence of malignant glioma (p < 0.05) 16 for CDMA-modulated RFR exposures." Not GSM, though. No differences were seen in the female rats at all.
The cardiac schwannomas [a type of heart tumor] were more compelling, but again, only for males. No differences for females.
I didn’t see any sample size calculation, nor any discussion of what they expected to see. One of the reviewers did a power calculation ... and found that based on 90 rats per group, the power was about 14%. This means that false positives are very likely. The cancer difference was only seen in females, not males. The incidence of brain cancer in the exposed groups was well within the historical range. There’s no clear dose response. ... And one more thing — the survival of male rats in the control group was relatively low, and if these tumors developed later in life, this could be the whole reason for the difference.
So: The researchers found an increased incidence of two types of tumors in male rats — malignant gliomas in the brain and schwannomas in the heart — after heavy exposure to cellphone radiation. But we can't rule out false positives. And the dangers aren't entirely clear-cut: note that the exposed rats actually lived longer than the unexposed rats.
That's the study. Now keep in mind these big caveats:
1) Single studies, even well-designed studies like this one, don't always hold up. That's why replication is so crucial in science. Other researchers need to dig into this study and try to reproduce its findings.
2) This study was done on rats. Rats have some similarities to humans, and can be good test subjects for medical research, but they are not identical to us. Science is littered with animal studies whose results don't carry over to humans.
3) The researchers treated the rats to very heavy amounts of radiation — nine hours a day, seven days a week — which is far more than most people spend holding their cellphones to their heads. (Many people nowadays rarely hold their cellphones up to their heads at all.)
4) There's a fair bit of countervailing evidence that cellphones aren't a major cancer risk. Here's Carroll again: "Cell phones are UBIQUITOUS in the United States. If they were causing cancer, we would expect to see rates of cancer going up, right? That’s not what we’re seeing. They’ve been decreasing since the late 1980s." Other population-wide studies have shown little evidence of cancer risk in humans from cellphone use.
That's not to dismiss this research, but it does suggest caution is warranted.
Outside experts were much more cautious about the study
Malcolm Sperrin, director of medical physics at Royal Berkshire Hospital in the United Kingdom, thought the study was "interesting and well-designed" but merited careful follow-up:
This is a very interesting and well-designed study of the incidence of two types of tumour in rats. The conclusions suggest that there is a small additional risk from exposure to RF [radiofrequency radiation] similar to that used in mobile phones, although the increase is very small. ...
However, the exposures given to the rats is high, the rats are clearly not human equivalent and perversely, the life expectancy of the rats exposed to the RF is marginally greater than the control. This report is a forerunner of a more comprehensive study but does lend credibility to the cautions necessary in very long RF exposures. ...
It should be borne in mind that the hypothesis of cancer genesis is not postulated and certainly does not reveal a link between normal domestic use of RF devices and health detriment, but does suggest that further studies are warranted.
Kevin McConway, emeritus professor of applied statistics at the Open University in Britain, pointed out that the uptick seen in rat tumors could well be due to chance. He noted that he doesn't intend to stop using his cellphone just yet:
The study is statistically underpowered. That means that not enough animals were used to allow the researchers to have a good chance of detecting a risk from radiofrequency radiation of the size one might plausibly expect, on the basis of previous findings. ...
The concern here is that risks and effects that are detected in underpowered experiments often turn out to be exaggerated — they may or may not still be real effects, but further study shows them to be smaller than the original research indicated. It’s too soon to tell if that will be the case here.
So here we have a study that found fairly weak evidence of small effects of mobile phone radiation on tumours in rats, where it’s plausible that the effects are even smaller than what was found, and where it’s not (yet) clear how far any such results are applicable to humans. I’m not going to stop using my mobile phone in the light of this.
Rodney Croft, director of the Australian Centre for Electromagnetic Bioeffects Research, raised several key questions about the study and noted that it "does not provide reason to move from the current scientific consensus that mobile phone-like exposure does not impact health":
Of particular note is that the rats treated with RF lived longer than the controls (which is counter intuitive given that the increased tumour rates normally lead to reduced lifespan), the controls did not have ‘any’ tumours (which is also not what is normally found), and the lack of clear dose-response relationships raises the possibility that the results may merely be ‘false positives’ (particularly given the large number of statistical comparisons, the one significant result would appear consistent with chance).
It is also noteworthy that the results do not appear consistent with the cancer rates within the human population, nor with the majority of other experimental research, even at the very high exposure levels, which are many times higher than humans are exposed to.
The NTP study will thus need to be fully evaluated once further details become available, and considered within the context of RF emissions science as a whole. At present though, and particularly given a range of uncertainties regarding its results, the NTP report does not provide reason to move from the current scientific consensus that mobile phone-like exposure does not impact health.
This is how scientists usually think about any important new study. They take the results seriously, they point to questions that demand further investigation, but they remain skeptical that this is the last word. The press should strive to follow suit.