How safe is safe?

For Radio-frequency radiation, maybe not as SAFE as our masters would have us believe

It is 2 years since HART published an article by Gillian Jamieson on the safety or otherwise of electromagnetic radiation. Her article entitled “Ministers remain in ignorant bliss with regard to 5G health risks” arose in part from a London conference organised by the International Commission on the Biological Effects of Electromagnetic Fields (ICBE-EMF).

Since that time, further concerns have been raised about potential health risks, particularly given the ubiquitous nature of mobile phones and other ‘smart devices’. The WHO proceeded to commission a series of systematic reviews, aiming to allay fears. The reviews largely agreed with the previous official line that everything is fine, but this view is certainly not shared by the ICBE-EMF team. They have written and published a scathing review and held a press conference earlier this month to highlight their concerns.

Gillian attended the conference and has written a timely review of the situation.

WHO commissioned reviews – are they fit for purpose?

Gillian Jamieson

In an effort to define its position on the health effects of electromagnetic fields (EMFs) including radio-frequency radiation (RFR) emitted by mobile phones, phone masts, WIFI and various “smart” devices, the World Health Organisation commissioned 12 systematic reviews (SRs) to summarise the science on this topic. By May this year, they were complete.

The reviews in question had been commissioned by the WHO’s EMF Project. Following an international survey, the health effects to be covered were cancer, male fertility, pregnancy outcomes, cognitive impairment, oxidative stress, electromagnetic hypersensitivity and heat-related effects. The review on heating effects was not concluded, which is strange indeed, as many of the world’s safety exposure limits, set by the International Commission for Non-ionising Radiation Protection (ICNIRP) are based on ICNIRP’s belief that harm only occurs if tissue is heated.

These reviews have, however, been under close scrutiny by a multi-disciplinary team of international scientists at the International Commission on Biological Effects of Electromagnetic Fields (ICBE-EMF). Earlier this month, their detailed critique of the WHO reviews was published, concluding that “due to serious flaws …the WHO-commissioned reviews cannot be used as proof of safety of cell phones or other wireless communication devices and should not be relied upon for the forthcoming WHO Environmental Health Criteria (EHC) monograph.” The 12 reviews are listed in the ICBE-EMF critique in Table 1.

Four of the ICBE-EMF authors, Dr Melnick, Dr Moskowitz, Professor Frank and Dr Mallery-Blythe presented their main arguments in an 60-minute online press-conference on 7^th October 2025 and this is now available in video format. Their press release summarised their position, with the detailed review published in the journal, Environmental Health. Essentially, the conclusions drawn by most of the WHO review teams, that harm caused by RFR is insignificant, cannot be trusted, although two of the reviews did show a definite link between RFR and cancer in animals (SR2), male fertility (SR4A) and pregnancy and birth outcomes (SR4B) and could inform public policy.

For me the question arises, how can SR1 (Karipidis et al, 2024) conclude that a link between RFR and cancer in humans is unlikely, when another major review (Moon et al, 2024) came to the opposite conclusion using almost exactly the same studies? This issue is dealt with in detail by Hardell and Nilsson in their 2025 article as well as by the ICBE-EMF in a letter to the editor. Overarching issues with the WHO reviews were presented as two major problems during the press conference, with more specific details being described in the ICBE-EMF critique.

One major problem identified in the reviews was their misuse of meta-analysis which may often conclude a systematic review. After an initial search of the scientific literature and evaluation of the primary studies, meta-analysis, which uses statistical analysis to give a clearer picture of the findings, may follow. There are, however, sophisticated conditions for its use. These conditions, in line with the Cochrane Collaboration, were not met in the WHO reviews. In this case, says the ICBE-EMF, narrative reviews, describing the findings in words rather than numerical analysis, should have been used, Tellingly, only one WHO review, by Mevissen et al, 2025 (SR2), used narrative review and this was the one that found high certainty of certain cancers in experimental animals.

Why exactly was meta-analysis unsuitable for these reviews? Briefly, there were too few primary studies and too much variability between them so that, for example, studies with vastly different exposure conditions were inappropriately combined. Doing this kind of thing might hide important differences or give a misleading average effect.

Professor Frank suggested during the press conference that the review teams did not have the necessary mixed expertise including epidemiological and biostatistical experts as well as substantive experts in the specific scientific field being reviewed. Problems created by deficits in the make-up of the teams should have been caught in the peer-review process.

A fascinating insight into why peer-review can be ineffective is given by Dr Moskowitz in his blog entry for 24^th April 2025 (please scroll down), when he describes his extremely frustrating experience of participating as a peer-reviewer in one of these WHO systematic reviews. He said,

“By design, this SR excluded most of the relevant research; hence, its conclusion is very limited in scope. Although the authors complied with some of the reviewers’ suggestions, the authors refused to deviate from several problematic decisions made in their protocol paper which was published several years earlier. After two revisions, the special issue editor allowed a critically flawed SR paper to be published in the journal.”

Returning to the 12 systematic reviews, other defects included, “exclusion of relevant studies, weaknesses in many of the included primary studies and lack of a framework for analysing complex processes such as those involved in cognitive functions.”

I’d like to illustrate this with one particularly weak study included in SR1, since it was heavily relied upon by the authors, despite having been dismissed by the International Agency for Research on Cancer (IARC) as uninformative regarding cancer risks due to “considerable misclassification in exposure assessment.”

The study in question is the Danish cohort study. It is extensively discussed by Hardell and Nilsson in an article earlier this year and is also mentioned in the ICBE-EMF critique. The division of exposed and unexposed control groups seems nonsensical. Only mobile phone private subscribers in Denmark between 1982 and 1995 were included in the exposed group, excluding the group most likely to be exposed, which consisted of 200,507 corporate mobile phone users. Users with mobile phone subscriptions after 1995 were not included in the exposed group and were treated as unexposed; all users of cordless (DECT) phones were treated as unexposed although they were exposed to RFR similar to that of mobile phones. In addition, actual exposure data was unknown and no analysis of laterality (the side where the phone was held in relation to the position of the tumour) was performed.

The errors in this study were so serious that others reviewers commented, “After reviewing the four publications on the Danish cohort study, one might rightly wonder whether this cohort was initially set up to show no increased risk.” And Professor Michael Kundi of the Medical University of Vienna described the Danish cohort study as “the most severely biased study among all studies published so far,” as quoted in an article named, “The Danish Cohort Study: The Politics and Economics of Bias.”

Apart from misuse of meta-analysis and other flaws, the ICBE-EMF’s other main concern is that the WHO-commissioned reviews are unduly influenced by ICNIRP members. At least one current or former member of ICNIRP was on each of the 12 review groups. Martin Röösli, a member until last year, co-authored four of the reviews, while Ken Karipidis the current vice-chair of ICNIRP, co-authored three, one of which was the defective SR1 mentioned above. Maria Feychting, who served on ICNIRP for 12 years and is known to be sceptical of cancer risks, helped design the study protocol of six of the reviews, including SR1. This table adapted by Microwave News, gives full details of ICNIRP involvement in each review. Further details of conflicts of interest can be found in Supplementary File 1 of the critique.

“What is wrong with ICNIRP?” was one question following the press conference presentations. The answer was that ICNIRP, whose safety exposure limits are followed world-wide, is a self-selecting group, all of whom believe that biological harm cannot occur below a certain heating (of tissue) threshold, despite the fact that there are hundreds of papers, showing effects below the supposed threshold dose and that numerous effects are shown below the safety limit itself. The limit lacks scientific basis, in that it was based on behavioural studies in the 1980’s which involved 40 to 60-minute exposures of 5 monkeys & 8 rats. When the researchers identified what they considered an apparent threshold they applied arbitrary safety factors. This would never be accepted in today’s world. Long-term effects were not considered e.g. cancer has a latency period, but this is not built into ICNIRP limits. ICBE-EMF were concerned that this mind-set would introduce bias into their evaluations of other effects of RFR.

Another concern is ICNIRP’s ties to major economic industries (telecommunications and military) that have financial interests in exposure standards for RFR. These ties have been described by members of the European Parliament in their “Conflicts of Interest, Corporate Capture and the Push for 5G” report. As the ICBE-EMF states, “Numerous analyses have found that sources of funding can impact the design and conclusions of health effects studies, including those on the effects of non-ionizing EMF.”

The ICBE-EMF considers that all but one of those 12 systematic reviews is inconclusive and that more high-quality research is needed. Professor Frank was concerned that the fundamental scientific errors in all but one of the reviews has led some lay observers and even certain scientists to conclude that present RFR exposures are “nothing to worry about” and even, that these reviews should re-assure us there is no valid scientific basis for questioning the current international safety exposure limits. Such highly misleading suggestions were recently made in a Guardian article, in which Ken Karipidis the lead author of SR1 even misquoted his own review’s conclusions, making them appear more certain than they were. In the light of ICBE-EMF’s detailed critique, these suggestions now appear ludicrous.

The danger is that the WHO uses these sub-standard reviews as a basis for their Environmental Health Monograph, that could serve as the WHO position on the health effects of RFR for many years to come - many years, because the previous update was more than 30 years ago.

However, the ICBE-EMF critique said that SR2, SR4A and SR4B which did show adverse health effects and included quantitative information could be used to alter exposure limits:

The animal cancer SR, which was rated as “high certainty of evidence” for heart schwannomas and “moderate certainty of evidence” for brain gliomas, provided quantitative information that could be used to set exposure limits based on reducing cancer risk. The multiple and significant dose-related adverse effects found in the SRs on male fertility and pregnancy and birth outcomes should also serve as the basis for policy decisions to lower exposure limits and reduce human reproductive risks.”

Any doubts as to whether animal cancer studies were relevant to human health were laid to rest following a question posed at the press conference. Dr. Melnick answered,

“The processes of cancer induction are similar in humans and rats…the International Agency for Research on Cancer (IARC) has said that every known human carcinogen is carcinogenic in experimental animals when adequately tested… Well-conducted animal studies can eliminate the need to wait for the availability of sufficient human cancer data before implementing public health protective strategies.”

ICBE-EMF therefore calls for a thorough and more independent review of this body of evidence to correct some of the scientific weaknesses found across these reviews. Until that is done, they strongly urge the public and the regulatory authorities internationally to consider the currently recommended safety exposure limits to be potentially too high to fully protect the public and the environment. They urge them to do everything possible to reduce RFR EMF exposures in public places, especially where there are particularly susceptible persons such as pregnant women, children or those with chronic disease and disability of any age. Telecoms companies could improve the situation by using engineering solutions suggested in a recent article, without waiting for governments to act.

Dr. Erica Mallery-Blythe, who specialises in electromagnetic hypersensitivity (EHS), and whose presentation described serious weaknesses in the two reviews relating to EHS, said that this was essentially a consent issue and that it was very important that people should be able to give informed consent, or not, to exposures. She would like to see moves to make low EMF spaces in residential areas, in work spaces and in schools to accommodate those who are acutely unwell and also those who simply do not give consent to exposure. She thought that that was a reasonable interim measure while there was an indisputable level of risk.

In answer to a question about what governments could do now, it was stated that governments should acknowledge that there is a risk, though it has not been fully quantified. In the meantime, people should be warned to be careful. People could use wired connections instead of WIFI or at the very least turn WIFI off at night to promote good quality sleep. Admitting to the public that there is some risk is exactly what Michael Mansfield KC had argued for in the judicial review on 5g in 2023, but this was denied him by the judge, who felt that the populace could not deal with nuance.

As Dr Louis Slesin from Microwave News has summarised, “For close to 15 years, the World Health Organization has been struggling to set out its views on the health effects of RF radiation. It hasn’t been going well, and it just got worse.”

My question is, will they actually go back to the drawing board as ICBE-EMF has recommended, so that we can begin to get closer to the truth or is the WHO a lost cause in terms of corporate capture?