The WHO-commissioned systematic reviews on health effects of radiofrequency radiation provide no assurance of safety.

As part of its Electromagnetic Field (EMF) Project, the World Health Organization (WHO) commissioned 12 systematic reviews and meta-analyses to clarify the health effects of radiofrequency radiation (0–300 GHz). Modern wireless communication technologies, such as mobile phones, Wi-Fi routers, and base stations, operate within this frequency range. The reviews cover topics including cancer risks in humans and laboratory animals, fertility issues, cognitive deficits, electromagnetic hypersensitivity (EHS), and oxidative stress. The reviews were published between 2023 and 2025. This paper by Melnick et al. critically evaluates the methodological quality, selection criteria, risk and evidence assessment, and interpretation of the results by the WHO-commissioned teams of authors.

Instead of conducting a primary study, Melnick et al. performed a systematic and methodological critique of twelve systematic reviews and meta-analyses that were commissioned by the WHO. The authors analyzed 6 areas: (1) possible bias on the part of the review authors, (2) inclusion and exclusion criteria for primary studies, (3) adequacy of the design of the included studies, (4) approach used in the reviews, (5) meta-analysis, and (6) risk and evidence assessment.

First, Melnick et al. criticize the overrepresentation of ICNIRP members among the authors of the WHO-commissioned studies. Each working group included at least one current or former ICNIRP member, and some included several. Some individuals appeared in different review teams multiple times. Because the ICNIRP has historically believed that thermal effects are the only health consequences of radiofrequency electromagnetic fields, Melnick et al. argue that the authors are at high risk for systematic bias ("groupthink bias"). Former ICNIRP member James C. Lin (Lin, 2021) came to the same conclusion. In addition, the connections between ICNIRP and industry, particularly in the telecommunications and military sectors, are concerning because these sectors have a major economic interest in exposure limits. Melnick et al. also criticize the glaring deficiencies in the inclusion and exclusion criteria of primary studies in various reviews. These deficiencies resulted in the exclusion of relevant, well-conducted studies while studies of insufficient quality received disproportionate consideration. This is particularly evident in the example of RF effects on oxidative stress markers. Of the 897 original articles that Meyer et al. deemed suitable for their review, only 52 were included in their meta-analysis. This corresponds to less than 6%. Meyer et al. excluded 360 primary studies because the WHO reviewers deemed the biomarkers used in these studies unsuitable for evaluating oxidative stress. However, adequate sham-exposed controls invalidate this argument, demonstrating that the examined biomarkers reliably indicate oxidative stress. In addition, primary studies with low exposure levels (despite their relevance; editor's note) were excluded due to selected field strength thresholds (e.g. E > 1 V/m). Meyer et al. failed to mention the exclusion of protective antioxidant enzyme activity when describing the inclusion and exclusion criteria. These enzymes have long been used as biomarkers for oxidative stress in many studies. Therefore, the Meyer et al. review does not reflect the current state of research. Not only were relevant studies on oxidative stress markers omitted, but also methodologically inadequate studies were given significant weight. For example, the effects of RF radiation on cancer in humans relied heavily on the results of the Danish cohort study despite its serious methodological shortcomings and classification as inconclusive by the International Agency for Research on Cancer (IARC). A second review of "less well-researched" human cancers included only primary studies from 1988 to 2019, excluding the UK Biobank cohort study. With over 400,000 participants, the UK Biobank cohort study is one of the largest and best-conducted studies in this field. The study found statistically significant increases in the incidence of various types of cancer, including overall cancer incidence, in connection with mobile phone use. Another example of inadequacies is the evaluation of electromagnetic hypersensitivity (EHS) (Bosch-Capblanch et al., 2024). In assessing sleep disorders, not a single study involving participants with EHS was included. Furthermore, only power density and frequency were considered relevant exposure criteria. Variables closely related to EHS, such as magnetic fields, extremely low frequency (ELF) pulse modulation, and data transmission, were ignored. Melnick et al. identified weaknesses in nearly all methodologies of the meta-analyses, except for the review by Mevissen et al. (Mevissen et al., 2025). However, Mevissen et al. did not perform a meta-analysis due to the heterogeneity of the primary studies. Despite high heterogeneity, meta-analyses were performed on other topics. In addition, a series of meta-analyses were conducted with too few primary studies (often fewer than 5 and sometimes as few as 2) and with studies that had different designs (e.g. case-control and cohort) and/or endpoints. The resulting data suggest a precision of results that does not exist due to the aforementioned methodological shortcomings. Melnick et al. raise another point of criticism regarding the use of the risk-of-bias tool and the GRADE system. They found that downgrading the strength of evidence was inconsistent. In addition, documentation of assessment decisions and traceability of GRADE steps were often insufficient. Based on their data, 8 of the 12 WHO-funded studies concluded that RF-EMF exposure probably does not cause adverse health effects. However, these conclusions are based on data that the authors classified as having low or very low strength of evidence. This is a logical error because the evidence that RF-EMF exposure is safe is also low or very low.

Melnick et al. identified recurring and systematic methodological flaws in nearly all WHO reviews. These flaws include overly restrictive inclusion criteria, flawed exposure definitions, inappropriate meta-analyses based on a small number of studies, selective weighting of results, inconsistent application of assessment tools, and potential conflicts of interest among authors. Rather than providing a thorough evaluation of the scientific evidence on the health risks of RF-EMF, these reviews create a false sense of security that endangers the health of the general population. In addition, consistent evidence of harmful effects has been found in reviews of cancer and reproduction in animal studies. Melnick's working group argues that the FCC and ICNIRP exposure limits are outdated and based on invalid assumptions (thermal dogma). New, scientifically sound guidelines that adequately protect human health and the environment are urgently needed. Due to their shortcomings, most reviews and meta-analyses commissioned by the WHO do not provide sufficient evidence that mobile phones and other mobile devices are safe.

Editor's note:

This unusually harsh criticism is based on documented methodological processes. The criticisms raised are all valid and understandable. In light of the methodological shortcomings and the presentation of the findings as robust despite this not being the case (keyword: meta-analysis), one must question whether the WHO-reviewers' intentions and methodology were designed to downplay the risks posed by wireless communication technologies. (RH)

Bosch-Capblanch X, Esu E, Oringanje CM, Dongus S, Jalilian H, Eyers J et al. (2024). The effects of radiofrequency electromagnetic fields exposure on human self-reported symptoms: A systematic review of human experimental studies. Environment International, 187: 108612. https://doi.org/10.1016/j.envint.2024.108612

Lin JC (2021). Science, politics, and groupthink [Health Matters]. IEEE Microwave Magazine, 22(5), 24–26. https://doi.org/10.1109/MMM.2021.3056975

Mevissen M, Ducray A, Ward JM, Kopp-Schneider A, McNamee JP, Wood AW et al. (2025). Effects of radiofrequency electromagnetic field exposure on cancer in laboratory animal studies, a systematic review. Environment International, 199: 109482. https://doi.org/10.1016/j.envint.2025.109482

Meyer F, Bitsch A, Forman HJ, Fragoulis A, Ghezzi P, Henschenmacher B et al. (2024). The effects of radiofrequency electromagnetic field exposure on biomarkers of oxidative stress in vivo and in vitro: A systematic review of experimental studies. Environment International, 194: 108940. https://doi.org/10.1016/j.envint.2024.108940