The (unmentioned) Risk of Bias (RoB) of the Systematic Review on RF-EMF and Cancer

Correction: this sentence was incorrect and is removed with my sincere apology: “This systematic review is a part of the ongoing WHO review of the science of RF-EMF and health.”

Yesterday, been has published a systematic review ‘In vivo Studies on Radiofrequency (100 kHz–300 GHz) Electromagnetic Field Exposure and Cancer: A Systematic Review’ authored by Pinto R, Ardoino L, Villani P, Marino C. Int. J. Environ. Res. Public Health 2023, 20, 2071.

The article presents in a very detailed way how the authors have selected and evaluated the to-date published pertinent scientific studies. Detailed track of evaluation, including all ‘++’ and ‘+’ as well as all ‘–‘ and ‘-‘ is clearly presented and can be examined by readers. Certainly, there will be some disagreements, and those will be published in due time, but the opinions of the authors are clearly visible to anyone who wishes to evaluate them.

Generally, systematic reviews that employ all necessary processes of evaluation of studies, like Cochrane Collaboration, the National Toxicology Program-Office of Health Assessment and Translation (NTP-OHAT), and the guidelines “Preferred Reporting Items for Systematic Reviews and Meta-Analyzes” (PRISMA-P) with protocol registered on PROSPERO, are considered as a gold standard of evaluation of science and also are considered, by many, to be unbiased or as unbiased as only humanly possible.

It certainly might be so, in some cases. However, let’s not forget that systematic reviews are prepared and written by humans who might be, by themselves, biased evaluators. Therefore, while all the data produced by the evaluation is available to readers there is always the possibility of disagreements on the interpretation of the quality of the original studies. Those disagreeing with the final conclusions of any particular systematic review are free to publish their own opinion but the already published systematic review, no matter what the disagreeing opinion says, this original opinion will stand as peer-reviewed published systematic review.

It is expected that a systematic review will be a calm and unemotional review of the science, free of personal opinions and scientific gossip. This is not the case here.

The US NTP study examining RF-EMF carcinogenicity in rodents is the largest and the most comprehensive study executed to date. Its results have been considered clear evidence of the carcinogenicity of RF-EMF. Not everyone agrees with this opinion. But writers of the systematic review are supposed to be above such agree/disagree opinions and they should calmly follow the data as per the protocol of the particular systematic review. This is not the case.

The following sentence suggests that the NTP study has been reviewed with some biased attitudes in the systematic review (emphasis added):

The technical reports of the National Toxicology Program (NTP) [31, 32] on carcinogenesis effects of RF exposure on rats and mice, respectively, were included, even if not published yet; these reports, released in 2018, although controversial, are considered among the most complete studies currently available on the impact of RF exposure on carcinogenesis [21].

Pointing out, that the NTP study reports are ‘considered controversial’ is out of place in the systematic review. It demonstrated the preconceived bias towards this study by the authors of the systematic review.

The authors of the systematic review are also perpetuating incorrect opinion, a myth already, that the NTP study was not peer-reviewed and, hence, is of lesser value than the others, peer-reviewed studies. The NTP study was in fact peer reviewed and the names of peer reviewers are well known:

Rick Adler, Lydia Andrews-Jones, Frank Barnes, J. Mark Cline, George Corcoran, Susan Felter, Jack Harkema, Wolfgang Kaufmann, Asimina Kiourti, James Lin, Tyler Malys, Matthias Rinke, and Laurence Whiteley, and Ms. Kamala Pant.

Therefore, it is false to claim that the NTP study was not peer-reviewed.

The authors of the systematic review were very attentive to the issue of the ‘Risk of Bias’ (RoB). They diligently pointed out how the studies funded by the industry were considered potentially biased. They should be thanked for this approach.

However, the authors have ‘forgotten’ their own ‘Risk of Bias’. Namely, the senior author of the systematic review, Carmela Marino, is a member of the Main Commission of ICNIRP. Coincidentally (?), also ICNIRP considers the NTP study as controversial and also ICNIRP perpetuates the false notion that the NTP study was not peer-reviewed. The membership of ICNIRP is not mentioned in the conflict of interest statement provided by the authors. What is also important to mention is that this particular team of scientists has got the job of preparing a systematic review on cancer thanks to Carmela Marino’s involvement in ICNIRP. So, the ‘ICNIRP-connection’ should have been specified in the conflict of interest statement.

The unanswered question is, how much, ‘a priori’, biased was the scientific team that prepared this systematic review? Was the bias directed solely toward the NTP study or was there a bias toward some other studies too?

The remedy for avoiding this potential ‘a priori’ bias was to have a diverse team of scientists with opinions that agree and that disagree with ICNIRP.

Already reading the abstract of this systematic review one gets an uneasy feeling that there might be something off…

The three sentences in the abstract make this ‘off-feeling’:

Sentence 1. “A significant association between exposure to RF and the increased/decreased risk of cancer does not result from the meta-analysis in most of considered tissues.”

OK, most of the tissues and organs were not affected. Correct.

Sentence 2. “A significant increased/decreased risk can be numerically observed only in heart, CNS/brain, and intestine for malignant tumors.”

OK, some important organs, like heart, brain and intestine, might be affected. Correct

Sentence 3. “Nevertheless, the assessment of the body of evidence attributes low or inadequate evidence for an association between RF exposure and the onset of neoplasm in all tissues.”

What? Does it mean that the lack of effects in most of tissues/organs nullifies the effects observed on only a few tissues/organs, even if they are the significant ones, like the heart, brain, and intestine? Even with the explanations that weaken the significance of the meta-analysis, this is still valid evidence pointing toward the link between RF-EMF exposures and cancers in certain physiologically important tissues and organs.

Finally, the authors did not specify at all something that at least I would like to know. The in vivo animal evidence is never used as stand-alone evidence of carcinogenicity but it is always supplementary evidence that strengthens/weakens the evidence obtained in human studies. The authors did not specify whether the outcome of their systematic review strengthens or weakens the human evidence from epidemiological studies. That is a significant omission.

Conclusion: systematic review is, by itself, not an automatic guarantee of unbiased evaluation of science if it doesn’t consider the potential bias of the team of scientists that prepare it.


4 thoughts on “The (unmentioned) Risk of Bias (RoB) of the Systematic Review on RF-EMF and Cancer

  1. Dear Dariusz, what is your take on the following three papers, especially the 3rd one?

    1) Cellular autofluorescence is magnetic field sensitive, PNAS 2021 Vol. 118 No. 3 e2018043118
    2) Long-term exposure to a hypomagnetic field attenuates adult hippocampal neurogenesis and cognition Nature Communication, 2021, 12:1174
    3) The effect of radiofrequency electromagnetic fields (RF-EMF) on biomarkers of oxidative stress in vivo and in vitro: A protocol for a systematic review, Environment International 158 (2022) 106932

  2. Comment on publication Jan. 2023: In Vivo Studies on Radiofrequency (100 kHz–300 GHz) Electromagnetic Field Exposure and Cancer: A Systematic Review.

    We did the same but distinguished the differences between EMF-patterns that can be related to healthy states versus unhealthy states. Many people have worked on models, see the known models of Fröhlich and Illinger and our proposal: A predictive quantum model that reveals a causal relation between exposures to non-thermal electromagnetic waves and biological effects. When there is evidence and a model, we can improve our current technology.

    Kind regards, Hans Geesink.

  3. ‘This systematic review is a part of the ongoing WHO review of the science of RF-EMF and health’.

    Not sure if this is right? The review was funded by INAIL (National Institute for Insurance against Accidents at Work, Italy).

  4. Thank you very much for sharing this article on ‘biased reviews’, let’s hope that it falls on ‘fertile ears’.

    Kind regards Steven

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.