Opinionista Ivo Vegter 19 October 2017

The campaign against killer radio waves revisited

Some people still believe that radio waves from cellphones make us sick. And they’re offended that I think they’re wrong.

I am not Karl Muller’s favourite sceptic. In fact, Mr Muller, from Sibebe, Swaziland, takes grave issue with my recent column on why radio waves don’t cause headaches, fatigue, nausea, insomnia and, of course, cancer. Allow me to quote his original comment, which he says he was not able to post:

“‘There is no conceivable mechanism by which it can harm you. It’s literally physically impossible (as in prohibited by the laws of physics).’

Not only are there ‘conceivable’ mechanisms. There are mechanisms that have been proven and validated and reproduced in hundreds of biochemistry labs around the world. Take a look at Friedman et al. (2007): just two minutes of low-level phone radiation causes living human cells to be flooded with free radicals, causing severe oxidative stress. This in turn causes genetic, metabolic, and neurological disorders. This study has been quoted over 250 times, only to confirm and amplify its findings. The ERK biochemical cascade Friedman et al. identified is specifically found in the *glial cells* of the brain. The National Toxicology Programme found with low-level whole-body radiation of rats, 3% got gliomas, *brain tumours*. The brain is especially sensitive to this radiation. You need to wake up, Ivo Vegter.”

This turned into a small flood of email correspondence, into which I won’t go here. Suffice to say that he says my column insulted those who are sick, that I was “deliberately directly offensive as well as being totally uninformed”, that I should “do some proper research before you open your snide mouth”. He demanded a prompt response, complained to my editor when I had not replied within 24 hours during a recent long weekend, and threatened to take his case to the ombudsman.

I’ve taken the liberty to consider Muller’s comment to be representative of the correspondence sent to me or Daily Maverick on the subject, since it actually cites two formal studies which we can examine more closely.

Let’s consider the statement he alleges I made about radio-frequency (RF) electromagnetic radiation (EMF): “There is no conceivable mechanism by which it can harm you. It’s literally physically impossible (as in prohibited by the laws of physics).”

I agree: if I made this statement, I should withdraw it. But I did not make this statement, so I cannot withdraw it. Nowhere in the column linked above, nor in the draft I submitted for publication, could I find any occurrences of “conceivable”, “literally”, “physically”, “impossible”, “prohibited” or “laws of physics”.

What I did say is: “…the science is clear: electromagnetic radiation of frequencies lower than visible light cannot plausibly cause significant harm to the human body.”

I’m not just nitpicking. Muller did not just trivially paraphrase what I wrote. He claims to have an honours degree in applied mathematics, and to be in correspondence with scientists concerning EMF. He also claims to edit academic papers, albeit mostly in statistics and econometrics, not physics, biochemistry or medicine. On another occasion, he claimed to be a freelance journalist and long-time newspaper sub-editor. So he should know the significance of the word “significant”.

Very many everyday things affect the human body and do insignificant harm. Our body gets harmed all the time by stresses imposed upon it by the body’s own metabolic activity, by what we eat or drink, by deliberate actions such as cold or sun exposure, by accidental injuries and infections, and by factors in the environment. Most of the routine harm to which the body is exposed is not significant because the body is able to heal itself and because we ordinarily try to avoid excessive amounts of harm.

Having established that he significantly misquoted me, let’s move on to the Friedman study Muller cites. In his wording, radio frequencies, by way of oxidative stress, “causes genetic, metabolic, and neurological disorders”. However, no such cause is established in the study he cites.

The Friedman study – which is hard going if you’re not a biochemist, as neither I nor Muller are – describes a signalling process activated in cells cultured in laboratory conditions by radiation at mobile phone frequencies. This process was found to produce oxidative stress in the cell, and could, in principle, act upon the cell’s DNA transcription. However, the study does not describe any actual harm that might befall a living human being as a result.

A commentary on this study, from the same issue of the Biochemical Journal in which it was published, is instructive. First, it reconfirms that “radio waves have been reported to be unable to directly cause DNA damage, and studies on the effect of radio waves on tumour formation in rodents have failed to show any consistent association between exposure to radio waves and tumours”.

The question then is whether radio waves can indirectly cause DNA damage, via the signalling process Friedman described. It speculates that it could, but that this process “occurs frequently in cells in response to a huge variety of signals, and is an essential component of many aspects of cellular physiology”. It says other studies find no evidence of the potentially cancer-causing oxidative stress Friedman reports, and in any case, “[oxidative stress is] commonly induced in the body as a result of many processes, and organisms have many ways of limiting [its] potential damage”.

It concludes by saying, “Further work will be required to assess the potential pathological role of these findings.”

Ergo, no potential pathological role of these findings has been determined. The Friedman study describes a potential mechanism that could cause harm of some unspecified degree. It offers no evidence whatsoever that such harm would be significant, or that it actually occurs in people in real-world conditions. This is a far cry from Muller’s apparent certainty that cellphone radiation, via oxidative stress, “causes genetic, metabolic, and neurological disorders”.

Muller then says: “The National Toxicology Programme found with low-level whole-body radiation of rats, 3% got gliomas, *brain tumours*.”

He refers to partial findings of a study that will only be fully released in 2018. These results are very weak, however. It exposed rats for nine hours every day to whole-body doses of cellphone radiation equivalent to a specific absorption rate (SAR) of 0, 1.5, 3 and 6 Watts per kilogram of body mass. Even the strongest cellphone signals, however, produce an SAR of only 1.54 W/kg, and then only when actively broadcasting. So by any measure, the rats were not exposed to “low-level” radiation, as Muller claims, but by relatively high levels.

Here’s what the study says about two different kinds of cellphone radiation, CDMA (which we do not use in South Africa) and GSM (which we do): “There was not a statistically significant difference between the incidences of lesions in exposed male rats compared to control males for any of the GSM- or CDMA-modulated RFR (radiofrequency radiation) groups. However, there was a statistically significant positive trend in the incidence of malignant glioma (p < 0.05) for CDMA-modulated RFR exposures.”

First, note that the result was not observed in female rats, although equal numbers of both sexes were used in the study, so the overall incidence rate of glioma is lower than those reported for male rats.

The “positive trend” was only observed for CDMA exposure, and then only for the 900MHz frequency, because that’s the frequency the rats in the study were exposed to (unlike the mice, which got 1,900MHz, but weren’t included in these results). Now, the incidence of glioma was nil for all groups up to 3W/kg exposure, and 3% only for the 6W/kg group, which is hardly a good correlation. In the case of glial cell hyperplasia, 1.5W/kg and 6W/kg produced positive results, but 3W/kg did not.

For GSM exposure, the group with the highest exposure showed a lower incidence of malignant glioma than either of the two lower-exposure groups. For glial cell hyperplasia, only the lowest and highest-exposure groups saw a positive result, but the middle group did not.

The results were entirely inconsistent with one of the most basic rules of pharmacology, the dose-response relationship. Worse, the 3% figure showed up in just one of 32 results categories: male rats (not female) tested for glioma (not hyperplasia) after exposure to 6W/kg (not 0, 1.5 or 3) of CDMA (not GSM) radiation. If you do enough tests, a few are going to be positive purely by random chance, which is exactly what happened here.

If the 3% rate were valid, would it be exceptional? It compares to a 2% average rate in historical control groups, which fall within a range of between 0% and 8%. So no, it would be nothing to write a paper about.

To throw even more cold water on the results, the probability of survival of the rats after about 75 days was higher for the rats exposed to radio frequency radiation, especially for the males.

So the claim that 3% of rats get malignant glioma after being exposed to low-level full-body radio frequencies is patently false.

There are studies that indicate with low levels of certainty that long-term, excessive exposure to high-power radio frequencies could cause some harm, but there are just as many, or more, that conclude just the opposite. The results to date have been inconsistent and inconclusive, especially in studies involving moderate exposure to actual humans in real-world conditions. And the point I made in the original column stands: to reduce exposure, demand more cellphone towers in your area, to limit the power your phone’s transmitter needs to emit.

In his final email to me, Muller invokes the “precautionary principle”, which says: “If an action or policy has a suspected risk of causing harm to the public, or to the environment, in the absence of scientific consensus, the burden of proof that it is not harmful falls on those taking that action.”

First, one cannot prove a negative. It is impossible to prove that something does not cause harm. That is why the World Health Organisation’s International Agency for Research on Cancer (IARC) has only found one substance of the thousand it has tested to be non-carcinogenic (and that was arguably a mistake). The burden of proof demanded by the precautionary principle is impossible to attain.

Second, there is no substantial evidence that radio-frequency signals, at levels ordinarily encountered, cause significant harm to the human body. Until such evidence emerges, that is the scientific consensus.

Third, if government were to prohibit or restrict everything that could possibly cause even a small amount of harm to people, we would destroy innovation. We would never be allowed to take the risks upon which the edifice of scientific and technological progress is founded. We would never be allowed to use the products of this progress, since everything under the sun, both natural and man-made, can possibly cause us harm to some degree.

Finally, applying the precautionary principle could do more harm than good, by denying the world life-saving medicine or poverty-alleviating technology. So the precautionary principle says that the burden of proof that applying the precautionary principle does not do more harm than good lies with those advocating its application. In short, the precautionary principle is self-contradictory and precludes its own application.

Would somebody please think of the children,” to quote Helen Lovejoy from the Simpsons, is always the last resort, and indeed, Muller asks: “Is there sufficient evidence of possible harm, that one should at least give warnings that small children should not be exposed?”

No, I do not think there is sufficient evidence that the radio frequencies used by mobile phones do us harm. Could they, with high enough power and long enough exposure? Perhaps, though it seems unlikely even then. And since breathing too fast or drinking too much water can also kill you, I’d hold off on calling for more regulation or warning labels. Who has time to read them anyway, these days? DM