Nov 02 2018

That Rat Cellphone Study – I’m Still Not Impressed

In May I reported on the preliminary report of an extensive study of cancer incidence in mice and rats exposed to radio-frequencies. Yesterday the researchers released the final report, with essentially the same conclusions. This is prompting another round of the media reporting that a study links cell phone use to cancer – but the data does not show that, and what it does show is still not impressive.

To summarize the results, and the points I made in my previous review of this study – there were actually two studies, one in mice and one in rats. The researchers exposed some of the animals to intermittent (10 minutes on, 10 minutes off) radio frequencies in varying strengths, whole body, from the womb until death. They found a statistically significant increase in some types of brain and heart tumor in male rats only exposed to the radio frequencies, but not in female rats, and not in male or female mice.

Right there, this is a strange result. Why only male rats? If this effect does not even extend to female rats, or to mice, why should we suspect it extends to humans? Further, whenever study results are quirky like this, in my opinion that calls into question the relevance, and even the reality, of the claimed effect. It smacks of random noise.

Another red flag is the large number of comparisons being made in the study. For example, the researchers looked for a particular type of tumor, schwannoma, in various tissues. They found a statistically significant increase in the heart, but not in other tissues, and not when you look at all schwannomas – only when you consider the heart separately. This looks suspicious for not controlling for multiple comparisons. You have to make a statistical adjustment for doing so, and I see no mention that they did.

Further, the absolute numbers are fairly low, with affected rats all being in the single digits. This makes subtle confounding effects and also random quirky effects more likely.

There was also some indication of possible confounding or random effects. For example, the treated male rats had a lower birth weight than the controls, but returned to normal weight as they aged. Also, the treated rats lived longer than the control rats. The authors speculate that the radio frequency exposure reduced the incidence of kidney disease that is a frequent cause of death in these rats.

The authors concede that behavioral effects on the mothers may have resulted in the low perinatal weight, and of course that can have lasting health effects on the rats. This is a huge potential confounding factor.

These kinds of studies are best considered as relating to hazard, not risk. Hazard is the potential for harm in the right setting, while risk relates to the probability of harm in the real world, or in specific settings. A good analogy is that a shark in an aquarium may represent a hazard, but visiting an aquarium does not present a significant risk of shark attack (unless you jump into the tank with the shark).

So at most this study shows that radio frequency exposure (whole body, whole life, starting in utero) is a potential hazard. It does not indicate that cell phone use is a risk to humans, however, and there are many good reasons to suspect that it isn’t. Epidemiological data in humans has not show a clear correlation between cell phone use and cancer, and brain cancers have not been increasing since the rapid and almost ubiquitous use of cell phones.

But still I am not even convinced this study shows a real hazard. The fact that the data was negative in female rats, in male and female mice, and for most tumor types is important. It limits the applicability of the results, and suggest they may just be random noise or due to some confounding factor.

Unfortunately most people will not read beyond the misleading headlines. I will say, however, I don’t think the headline will have much of a negative effect. I don’t think people are ready to give up their cell phones because of a study in rats.

No responses yet