Radiation literature survey

Article publication date

January 2026

ARPANSA review date

25 February 2026

Summary

These two large studies, simultaneously but independently conducted in Korea (Kim et al., 2026) and Japan (Imaida et al., 2026), investigated whether long‑term exposure to radiofrequency electromagnetic fields (RF-EMF) can cause tumours or genetic damage in rats. These studies were particularly designed to substantiate earlier findings from the U.S. National Toxicology Program’s (NTP) study, which had reported increased rates of rare tumours, particularly brain gliomas and heart schwannomas, in male rats exposed to RF-EMF at very high exposures (given as the specific absorption rate (SAR)). To test whether these results could be reproduced and validated, both countries used the same protocol, rat strain, exposure system and RF-EMF signal (900 MHz, CDMA) at a whole‑body SAR level of 4 W/kg. Both studies followed strict Good Laboratory Practice and OECD guidelines (e.g., TG451). Exposure began before birth and continued for two years, with rats receiving RF-EMF exposure for 18 hours and 20 minutes per day in 10 minute on/off cycles. The study included three groups of rats of 70 animals each, one group being those exposed to RF-EMF, one sham exposed and one group of cage controls. The incidence of tumours was assessed by histopathology which underwent international peer review. 

The findings of these studies were remarkably consistent and in contrast to the NTP study as there were no statistically significant changes in tumour incidence in either study. In the Japanese study, the RF-EMF exposed group lived longer than the sham exposed group, although still within normal variation. This was attributed to slightly lower food consumption and weight in the exposed group and was accounted for statistically. Similarly, the studies reported no evidence of DNA damage or chromosomal aberrations in exposed rats. The results of this Japan–Korea collaborative research do not substantiate the results previously reported by the NTP study. 

Published in

Toxicological Sciences

Link to study

Link to Korean and Japanese studies

Commentary by ARPANSA

These two studies are among the largest to investigate tumour outcomes in rats exposed to RF‑EMF, with each including 210 animals. Both studies followed a high standard of methodological rigour, including adequate blinding during exposure and analysis, good exposure characterisation and appropriate statistical methods. Together, these features minimise the potential for bias and position these two studies to have a significant influence on the over-arching scientific literature in this domain and any future meta-analyses. The concerted effort by an international research collaboration to simultaneously and independently conduct two large, long-term animal studies is also commendable and highlights the importance of these results. 

These partial replication studies used a single exposure level of 4 W/kg, compared to the original NTP study that had groups of animals that were exposed at either 1.5 W/kg, 3 W/kg or 6 W/kg. The maximum exposure level of 6 W/kg has been the subject of criticism by the International Commission on Non-Ionizing Radiation Protection (ICNIRP), the German federal office for radiation protection (BfS) and ARPANSA. These criticisms argue that a core temperature rise at this exposure magnitude would exceed 1°C and the methodology used to evaluate body temperature in the NTP pilot study likely underestimated temperature rise. Where the NTP pilot study measured subcutaneous temperature only after exposure ended, contemporary core temperature monitoring studies have used superior surgically implanted monitors that continuously record core temperature throughout exposure (Bala et al., 2025). The rationale for using high doses is that such doses enhance the experiment’s sensitivity to small or rare outcomes  (Haseman & Lockhart, 1994). Without careful control, this can risk tumours occurring  secondary to other  harms caused by exposure (Bucher, 2000). That adds to the challenge of interpreting the results in relation to humans as the cause of any observed effect becomes unclear. It is therefore justified for the Japanese and Korean replications to use 4 W/kg. This is also the putative level for health effects that the whole body SAR restriction is conservatively based off in the ICNIRP guidelines (2020) and the Australian radiofrequency standard RPS-S1 (2021). 

Neither the Korean or Japanese study reported an increased incidence of tumours. These findings align with a recent systematic review and meta-analysis by Pinto et al., (2023), which concluded that evidence for any association between RF‑EMF exposure and tumour development is low or inadequate. In contrast, a systematic review and narrative synthesis by Mevissen et al., (2025) reported high‑certainty evidence for an increased incidence of brain tumours (gliomas) and heart tumours (malignant schwannomas) following RF‑EMF exposure, along with moderate evidence for several other tumour types in rats. ARPANSA has previously provided commentary on this review as well as other national health bodies (BfS, Swedish Radiation Safety Authority). 

The differing conclusions between these reviews originates from their approaches to evaluating the evidence. Mevissen et al. placed substantial weight on the outcomes of two chronic cancer bioassays, particularly the National Toxicology Program (NTP) rat study (NTP,  2018), without fully accounting for its limitations or integrating the findings of other relevant studies. The authors of the Mevissen review justify this methodology by stating that the available research was not similar enough to enable an encompassing synthesis of the evidence. Although the validity of this justification is poor, such reasoning can never be applied to these two new chronic cancer bioassays from Japan and Korea.  As such, even if the critically flawed methodology of the Mevissen et al. review were followed, the conclusions would have to yield to accommodate the findings of this new research. Similarly, the findings reported by Pinto et al. would be further strengthened as the data offered by the two new studies is in alignment with their conclusions. The genotoxicity results are also consistent with a recent systematic review (Romeo et al., 2024) and the results likewise strengthen those conclusions.

Another important issue that should be considered when interpreting these and the results of the NTP studies is the proportion of rats that survived to the end of the experiment. Longer survival rates increase the likelihood of tumour development. In the NTP study, survival among exposed male rats was substantially higher (50–62%) than in the control group (28%). A similar pattern was observed in the Japanese study, although the difference was less pronounced, with 65% survival in exposed rats compared with 43% in sham controls. In contrast, the Korean study reported no meaningful difference in survival rates between exposed and control animals. When exposed animals live significantly longer than controls, as seen in the NTP and Japanese studies, any increase in tumour detection may simply reflect extended lifespan rather than an effect of RF‑EMF. The Japanese study accounted for these differences in mortality through their statistical methods.

For some health endpoints, there is limited relevance of rat cancer studies for extrapolation to human cancer risk. Notably for glioma, some of the molecular features seen in human glioma are not seen in rat glioma, indicating  a fundamental biological difference (Sahu et al., 2022). These types of fundamental differences between human and animal biology speak to the importance of the hierarchy of scientific evidence where the most significant evidence comes from studies on humans. This is because studies on humans can provide more direct and relevant information about human health and disease. Human evidence of the association between RF-EMF on cancer has recently been reviewed in the WHO commissioned systematic reviews looking at observational studies in humans (Karipidis et al., 2024, Karipidis et al., 2025). These studies did not find an association between RF-EMF and any cancer type, including glioma. 

Although the OECD Test Guideline 116 recommends approximately 50 rats per group, and these studies used 70 animals per group, such numbers may still be insufficient to reliably detect rare cancers. This is particularly relevant given that the background incidence of glioma in NTP control groups is typically only 0–4% (NTP, 2025), meaning that these may still have limited power to identify small increases in rare cancers like glioma.

In Australia, devices that emit RF-EMF must comply with the limits prescribed in the radiofrequency standard RPS-S1. It is important to note that under this standard, the maximum permissible whole body SAR for the general public is 0.08 W/kg, one fiftieth of the exposure level used in these animal carcinogenesis studies. Overall, these new animal studies provide further evidence that RF-EMF is not associated with cancer in animals, supporting previous safety assessments (ICNIRP, 2020; SCHEER, 2022). It is similarly supportive of ARPANSA’s assessment that there is no substantiated scientific evidence of adverse health effects from RF-EMF exposure at levels below those prescribed in the standard.

Article publication date

September 2025 

ARPANSA review date

March 2026

Summary

This Spanish case-control study investigated the association between indoor radon and inflammatory bowel disease (IBD). The study included 178 cases and 178 controls that were matched by age and sex. Residential radon levels were measured using passive detectors for three months in each subject’s home. The study investigated if higher residential radon exposure was associated with a high incidence of IBD and whether it increased the incidence of IBD disease flares (followed for one year) among the cases. The study found no association between increased rates of IBD and high indoor radon, even in the highest exposure category of >299 Bq/m³ (odds radio (OR) 0.5, 95% confidence interval, CI, 0.3-0.9). The study also found no association between increased occurrences of flares and the highest category of indoor radon levels (OR 1.5, 95% CI 0.5–4.5). Overall, the authors concluded that higher residential radon exposure was not associated with IBD. 

Published in

Therapeutic advances in gastroenterology

Link to study

Indoor radon concentration and risk and severity of inflammatory bowel diseases: a case-control study - PubMed

Commentary by ARPANSA

This study contributes to the emerging evidence on residential radon exposure and IBD. A previous ecological study in Spain also found no association between higher indoor radon and Crohn's disease or other types of IBD (Mauriz-Barreiro et al 2022). Previous studies of radon exposure have mostly focused on the association between radon and lung cancer. However, a recent systematic review and meta-analysis of 129 studies (Henyoh et al 2024) investigated the association between radon and diseases other than lung cancer, and found no association between radon and these outcomes.

Radon is a naturally occurring radioactive gas that can build up in poorly ventilated areas, particularly in caves and mines, however, it can also build up in home that do not have airflow. Homes that are well ventilated, made of timber or built on stumps have lower radon levels compared to homes on concrete slabs with brick walls. More information can be found on our factsheet Radon exposure and health | ARPANSA. An Australian Radon measurement survey of more than 3300 Australian homes found that that the average concentration of radon is much lower (10 Bq/m³) than global average (40 Bq/m³) (Radon in homes survey and indoor radon map | ARPANSA). This low residential radon level is unlikely to pose any health risk as it’s far below the recommended reference action limit of 200 Bq/m³. 

Article publication date

February 2026

ARPANSA review date

10 February 2026

Summary

This population-based case-control study examined the association between occupational exposure to extremely low frequency electromagnetic fields (ELF-EMF) and postmenopausal breast cancer among women in Canada. The study included 663 breast cancer cases occurring in women aged between 47 and 75 years and 592 controls in the same age cohort. Information on breast cancer cases was gathered from histopathological reports. Occupational ELF-EMF exposure was estimated by using a job exposure matrix (JEM). Overall, the study indicated no elevated risk of breast cancers as a result of workplace ELF-EMF exposure. 

Published in

Journal of Occupational and Environmental Medicine

Link to study

Canadian study indicates no association between occupational ELF-EMF exposure and postmenopausal breast cancer risk

Commentary by ARPANSA

The study found no overall risk of postmenopausal breast cancer and occupational ELF-EMF exposure. Similar findings have been consistently reported previously (e.g., Koeman et al., 2014; Labrèche et al., 2003). The present study advances the evidence base for assessing breast cancer risks associated with occupational ELF‑EMF exposure. However, several limitations should be acknowledged. First, the JEM inherently assigned exposure solely based on job title without considering within‑job variability in exposure levels. This results in non-differential misclassification of exposure biasing association towards null. Second, the study had a relatively low and differential response rate (54% among cases and 41% among controls), which may have introduced selection bias. Third, potential co‑exposures to other occupational carcinogens, such as chemical agents or ionizing radiation, were not assessed. It is ARPANSA’s assessment that there is no substantiated scientific evidence that  ELF-EMF exposure below the limits recommended by international guidelines (e.g., ICNIRP) poses a health risk. More information about exposure to ELF-EMF can be found on the ARPANSA factsheet Electricity and health | ARPANSA.

Article publication date

October 2025

ARPANSA review date

January 2026

Summary

The French Agency for Food, Environmental and Occupational Health & Safety (ANSES) has updated its appraisal of the carcinogenicity of radiofrequency electromagnetic fields (RF-EMF). This was a large undertaking that involved collating and synthesising information from a wide variety of evidence streams into a cohesive assessment for each organ or physiological system. The evidence assessed in the report was published between 2013 and 2024, building on previous ANSES expert appraisals published in 2013 and 2016.  ANSES used a hierarchical assessment where both the type and strength of evidence were used to determine the overall level of evidence. Their structure follows the generalised hierarchy of evidence where epidemiological studies on humans are valued above studies on animals which are, in turn, valued more than studies on cell lines or mechanisms.

For the brain, central nervous system, blood, plasma, immune system, cardiovascular system, liver and the reproductive system ANSES assessed that it was not possible to conclude from the available evidence whether RF-EMF was carcinogenic. This is the lowest possible classification of risk in ANSES’ classification structure. For all other biological systems there was insufficient information to form substantial lines of evidence.

Commentary by ARPANSA

The formulation of ANSES’ conclusory statements arising from the lowest classification of risk in their classification structure can give the impression that there are large uncertainties remaining in the scientific evidence, regardless of the size and quality of the evidence used to arrive at that classification. While some uncertainties remain for less studied cancers, for well-studied cancers their statements do not provide an accurate accounting of the state of the scientific evidence. This can be problematic for readers without further understanding of the underlying scientific evidence, as it is challenging to interpret the meaning of their conclusions accurately without this knowledge. Better formulated conclusory statements that include consideration for both the certainty and direction of effect, such as those employed in GRADE assessments, could provide clarity to an uninitiated reader.

This assessment was requested by the French Director General of Health in response to preliminary reports from the National Toxicology Program’s (NTP) study into the effect of RF-EMF on cancer incidence in laboratory animals (NTP, 2018a, 2018b). These preliminary reports and subsequent full publications indicated a potential effect of RF-EMF on specific cancers at very high exposure levels. However, numerous issues with the NTP studies have been documented by a variety of public health agencies (ARPANSA, ICNIRP, FDA) and so the approach of ANSES to consider those results in the context of the entire scientific body of evidence, including using a hierarchical approach, is justified. ARPANSA similarly uses a hierarchical approach to evaluating evidence which is detailed on the corresponding ARPANSA webpage. ANSES collated all the appropriate research that existed during their assessment however, there have also been notable recent publications that partially replicate the NTP study (Kim, H. et al., 2026;  Imaida, K. et al., 2026). These studies do not confirm the findings of the original study and conclude that there is no carcinogenic effect of RF-EMF.

With regard to the epidemiological evidence, the organ that has been researched the most is the brain due to the assumption of comparatively high near field exposures from mobile phone use. The most comprehensive assessment of this line of evidence to date (Karipidis, K. et al., 2024) did not show an association between RF-EMF exposure and brain cancers and that has been shown further since in high quality prospective cohort studies (Feychting, M. et al., 2024). Other organs have been studied much less in epidemiological studies but still show no association between RF-EMF exposure and cancer (Karipidis, K. et al., 2025). As this type of evidence is the highest in the hierarchy of evidence for long-term health effects such as cancer, it is appropriate that ANSES’ assessment of corresponding organs arrived at their lowest classification of risk.

A similar effort has been undertaken by the World Health Organization in a project to assess the relationship between RF-EMF exposure and a variety of health endpoints, including cancer. This project involved commissioning a series of systematic reviews and will ultimately result in the publication of an environmental health criterion monograph for RF-EMF. Each of the systematic reviews have been published and the evidence is broadly in agreement with ANSES’ conclusions.

Article publication date

5 November, 2025

ARPANSA review date

17 December, 2025

Summary

This case-control study is a partial replication of the international MOBI-kids study (Castano-Vinyals, G. et al., 2022) and investigates the association between mobile phone use and brain tumour incidence in Japanese youth aged between 10 and 29 years. The study examined 120 brain tumour patients (cases) and 360 controls. The analysis examined potential associations with both the duration and intensity (cumulative number of calls and call time) of mobile phone use while adjusting for age and sex. Exposure was also adjusted based on the variant output power of different generations of mobile phone technology and their historical prevalence in Japan. 

An odds ratio and 95% confidence interval was computed for each exposure classification; based on either regular use [0.92 (0.48-1.77)], years of use [0.94 (0.30-2.92)], cumulative number of calls [0.80 (0.32-2.01)] or cumulative call time [0.58 (0.22-1.52)]. In each case there was no association between mobile phone use and brain tumours. These same categories also did not show an association when the exposure was adjusted for mobile phone generation.

Link to

Brain Tumor and Mobile Phone Risk Among Young People: Analysis of Japanese People Using the MOBI-Kids International Case-Control Study

Published in

Bioelectromagnetics

ARPANSA commentary

The original international MOBI-kids study assessed the association between brain tumour incidence and mobile phone use among children from fourteen different countries and it did not find any association. The current study expands on the Japanese subset of that study with key differences being study size, matching of controls and the adjustment of exposure characterisation by mobile phone generation. The adjustment of exposure for mobile phone generation represents an improvement in exposure characterisation as modern mobile phone technology has lower exposure to radiofrequency electromagnetic fields, especially compared to the older 2G networks (Iyare, R. et al., 2021; van Wel, L. et al., 2021).

In addition to confirming results from the international MOBI-kids study, this study is in agreement with a systematic review of all the evidence that shows no association between mobile phone use and brain tumours (Karipidis, K. et al., 2024). In Australia, emissions from mobile phones must comply with the limits prescribed in the radiofrequency standard RPS-S1. This study supports ARPANSA’s assessment that there is no substantiated scientific evidence of adverse health effects at levels below those prescribed in the standard. 

Article publication date

December 2025

ARPANSA review date

December 2025

Summary

This Swiss cohort study evaluated the association between different occupational exposures, including radiofrequency (RF) electromagnetic fields, extremely low frequency magnetic fields, ionising radiation and ultra-violet (UV) radiation, and the incidence of melanoma and squamous cell carcinoma (SCC). The study population was identified from the Swiss national census (an estimated coverage of 98.6% of the population in 2000), which collected data on occupation, migration, and demographics. The census data was linked to cancer registries data from six Swiss regions. The study included 1,077,487 adults aged 20 to 65 years. Occupational exposure of the subjects was assessed using the Canadian Job Exposure Matrix (CANJEM). 

The study found a statistically significant association between UV and melanoma (hazard ratio (HR): 1.23, 95% confidence interval (CI):1.02-1.50). No association was observed between UV exposure and SCC (HR: 1.08, 95% CI: 0.84-1.40). RF, magnetic fields and ionising radiation were not associated with melanoma or SCC. 

Published in

The Science of the Total Environment

Link to study

Occupational exposures and skin cancer incidence in six Swiss cantons

ARPANSA commentary 

The study reports that occupational exposure to UV was associated with melanoma, which aligns with the findings of the WHO systematic review and meta-analysis. However, the WHO systematic review also found an association with non-melanoma skin cancers (NMSC), which would include SCCs. The current study did not find an association between UV exposure and SCC, inconsistent with results from the WHO systematic review. 

The absence of an association between ionising radiation and melanoma or SCC is consistent with previous research, including the recent review by Caramenti et al (2024). Similarly, the study found no association between RF or magnetic fields and either melanoma or SCC. Evidence on these exposures remains limited and generally of low quality. For example, the UK Biobank study reported no association between RF exposure and melanoma but observed a small association with NMSC. In contrast, a Danish cohort study by Poulsen et al (2013) found no link between RF exposure and SCC, and Khan et al (2021) reported no overall increased risk of skin cancer associated with magnetic field exposure among individuals living near power transformers. Overall, the evidence suggests that associations between RF or magnetic fields and skin cancer are weak, and no plausible biological mechanism has been identified to explain such an association.

Article publication date

November 2025

ARPANSA review date

December 2025

Summary

This case-control study examined the association between mobile phone use and screen time and invasive breast cancer. The study included 226 women (77 cases, 97 controls and 52 unconfirmed cases) recruited from diagnostic, mammography, and radiotherapy centres across Iran. Mobile phone and screen use was assessed by structured questionnaires asking about mobile phone call duration, screen time, and phone placement. The study reported that women who spent more than 60 min per day in mobile phone conversations had higher odds of confirmed breast cancer (odds ratio (OR): 3.49, 95% confidence interval (CI): 1.02–11.97). 

Published in

Journal of Research in Medical Sciences

Link to study

Journal of Research in Medical Sciences

Commentary by ARPANSA

The study contains a significant error. In Table 2, the reported mobile phone use for cases and controls shows 94 cases and 77 controls; the number of cases and controls is inconsistent for other characteristics listed in Table 2 as well. However, the authors state elsewhere in the paper that they have 77 cases and 97 controls. It is unclear whether this discrepancy indicates that the case-control status of participants has been misclassified in the analysis or if this is a typological error. This may invalidate the results of the study.  

All mobile phone use in the case-control study was self-reported, introducing recall bias as a source of uncontrolled error. This bias is possible, as cases may remember their phone use differently than controls following diagnosis. Additionally, selection bias may be present, as the authors provide no information regarding loss of subjects or participant exclusion, this means we are unable to assess how attrition may have influenced the outcome of the study.

One of the most comprehensive studies examining the association between mobile phone use and breast cancer is the UK Million Women Study (Benson et al 2013). This prospective cohort study collected data on mobile phone use from 489,769 women and found no evidence of an association between mobile phone use and breast cancer risk. Similarly, a recent systematic review commissioned by the World Health Organization (Karipidis et al., 2025) which evaluated all epidemiological evidence from 1988 to 2019 on mobile phones and cancer, also concluded that there is no association with breast cancer. Therefore, ARPANSA’s assessment is that the scientific evidence in its totality does not support an association between mobile phone use and breast cancer.

Article publication date

October 2025

ARPANSA review date

December 2025

Summary

This prospective cohort study assessed the association between light exposure (day and night) and incidence of cardiovascular diseases in human populations. A total of 88,905 individuals (mean age, 62 years) were included in the study. The data on light exposure from the participants (nearly 13 million hours) were collected from using wrist-worn light sensors. The information on the incidence of cardiovascular diseases (e.g., coronary artery disease, myocardial infarction, heart failure, etc. after light tracking) were gathered from UK National Health Service records. The strength of the association was reported in hazard ratios (HRs) with a 95% confidence interval (CI) adjusting for potential confounders (e.g., sociodemographic and lifestyle, sleep, physical activity, diet and genetic susceptibility). The results [HR (95% CI)] showed that exposure to brighter light at night was associated with higher risks of coronary artery disease [1.3 (1.2-1.5)], myocardial infarction [1.5 (1.3-1.7)], heart failure [1.6 (1.3-1.8], atrial fibrillation [1.3 (1.2-1.5)], and stroke [1.3 (1.1-1.5)]. The study indicated that light at night exposure is a risk factor for cardiovascular disease in adults over 40 years old.

Published in

JAMA Network Open

Link to study

Light Exposure at Night and Cardiovascular Disease Incidence

ARPANSA commentary

The study reports that brighter night light is related to up to nearly 60% higher risk of cardiovascular disease among the people aged 40 years and older. These risk estimates are comparable to those reported in previous smaller studies (e.g., Obayashi et al., 2015; Kim et al., 2023). The strength of the current the study lies in rigorous adjustment for confounding variables and the inclusion of personal light exposure data. Notably, this was the first study on light exposure to account for key cardiovascular risk factors (BMI, sleep, diet, hypertension, cholesterol, and diabetes) thereby disentangling the effects of these factors from those of light exposure.

One of the key limitations of the study is that it predominantly included white (97%) populations and people with higher education levels, higher income, women (57%), and healthier individuals. This eventually flags the study findings while generalising them for other human populations. However, the findings suggest that limiting overexposure of night light is another factor that could help reduce cardiovascular disease risk associated with it. There are some publications which provide useful recommendations for visible light exposure, such as from light emitting diodes, and potential health effects (e.g. ANSES, 2019; Brown et al., 2022). Similarly, the International Commission on Non-ionizing Radiation Protection (ICNIRP) has published a statement on short wavelength light (SWL) exposure from indoor artificial sources and human health acknowledging no scientific consensus on whether night light per se causes health effects. ARPANSA currently does not have specific health advice regarding night light exposure and health. However, it will continue to scrutinise emerging evidence on health risk associated with night light exposure.

Article publication date

September 2025

ARPANSA review date

21 November 2025

Summary

This study updates previous estimates of the proportion of cutaneous melanoma incidence attributable to ultraviolet radiation (UVR) by examining data from 2022. Data on national estimates of cutaneous melanoma cases from 154 countries were extracted from the World Health Organization’s GLOBOCAN 2022 database. Population attributable fractions (PAFs) related to UVR exposure were calculated by sex, age, and country using a minimally exposed Nordic 1930 birth cohort reference population for comparison.The estimates were adjusted for acral lentiginous melanoma, which is not associated with UVR exposure and accounts for about half the cutaneous melanoma cases in dark-skinned populations. The study showed that most of the global cutaneous melanoma burden in 2022 (n=267,353 cases; 57% of them in males) was UVR-attributable. The PAF estimates increased with age; 76% among people aged 30–49 versus 86% among people aged 70 or more years. In Australia/New Zealand, Northern Europe, and North America, more than 95% of cutaneous melanoma cases attributable to UVR exposure. The highest attributable age-standardized rates (per 100,000) were found in regions with populations of lighter skin colour: Australia/New Zealand (76), North Europe (37), and North America (34). The study concluded that excess UVR exposure accounts for more than four-fifths of the global cutaneous melanoma incidence.

Published in

International Journal of Cancer

Link to study

Global burden of cutaneous melanoma incidence attributable to ultraviolet radiation in 2022

ARPANSA commentary

This study provides an updated global estimate of cutaneous melanoma burden for countries and regions categorized by sex and age. Further, the study also improved the earlier methods of estimating PAFs of UVR (Arnold et al., 2018) by adjusting cutaneous melanoma rates to the relative proportion of acral lentiginous melanoma for darker-skinned populations. Regional trends of cutaneous melanoma rates remain similar to earlier estimations (Arnold et al., 2018).  Very high rates in Australia and New Zealand are attributable to high ambient levels of UVR exposure (Xiang et al., 2014; Olsen et al., 2010). As highlighted in the study, most of the global cutaneous melanoma burden is UVR-attributable and hence the role of sun protection remains important in reducing cutaneous melanoma burden. 

The Australian Radiation Protection and Nuclear Safety Agency (ARPANSA) measures the UV index in various locations in Australia and publishes this data in real-time. In view of preventing skin cancer occurrences, Australia promotes the world leading SunSmart program, which recommends people to adopt a combination of five sun protection measures whenever the UV index is 3 or above. The sun protection measure includes Slip (on clothing), Slop (on SPF30 or higher), Slap (on a hat), Seek (a shade), Slide (on sunglasses). ARPANSA also provides evidence-based public health messages in relation to UV protection measures, including sun protection factsheets. The SunSmart Global UV app provides real time sun protection advice for Australian and major international cities to inform people about sun protection measures.

Article publication date

October 2025

ARPANSA review date

November  2025

Summary

This commentary discusses the recently published set of systematic reviews that were commissioned by the World Health Organization (WHO) to evaluate various health endpoints in relation to radiofrequency electromagnetic field (RF-EMF) exposure. The commentary presents criticisms of some aspects of each review with a particular focus on those relating to human observational studies (SR1 A & B, SR3A & B, SR5), oxidative stress (SR9) and non-specific symptoms (SR7, SR8). The commentary touches on study selection criteria, purported authorship bias, the use of meta-analysis and other direct criticisms of included studies. The conduct of risk of bias (RoB) and certainty of evidence (CoE) assessments were also the subject of criticisms as well as some additional semantic commentary around phrasing used in GRADE CoE statements. 

Published In

Melnick et al. Environmental Health

Link to study

https://pubmed.ncbi.nlm.nih.gov/41034851/

Commentary by ARPANSA

Cancer in human observational studies (SR1 A and B)
SR1 A and B remain the most comprehensive and highest-quality systematic reviews on human observational studies investigating RF-EMF and cancer; they conclude that RF-EMF does not cause cancer with varying degrees of certainty for different cancers based on the availability of evidence. The primary criticisms of SR1 by Melnick et al., are due to the inclusion of the Danish cohort study which bases exposure on mobile phone subscription. The exposure assessment in the Danish cohort study has been validated, showing a clear distinction between subscribers and non-subscribers, with only 16% of non-subscribers actually using a mobile phone (Schuz & Johansen, 2007), meaning its exposure assessment can be rated very well and far better than previous studies. The conclusions of SR1 on brain cancer, in particular, are supported by the Cosmos study, (Freychting et al., 2024), which is the most comprehensive observational study to date, that includes a large cohort of participants and assesses exposure via questionnaires on mobile phone use and operator data. The Cosmos study found no association between mobile use and brain cancer, but this result was not included in SR1 because it was published after the cut-off date for inclusion.  Melnick et al., also criticised SR1 for how it assessed exposure comparisons saying it was too basic and only used exposure metrics like“ever versus never mobile phone use”. However, SR1 goes beyond simple exposure proxies by analysing duration of use and dose-related metrics such as cumulative call time and number of calls.

Cancer in experimental animals (SR2) 
In contrast to their criticisms of the other systematic reviews, the commentary praised the systematic review on the effect of RF-EMF on cancer in experimental animals despite the significant flaws in the narrative synthesis of SR2. An extensive critique of SR2 can be found on the ARPANSA website. 

Adverse reproductive outcomes (human observational studies) (SR-3 A and B)
The reviews (SR3A and SR3B) provide state-of-the art methods on investigating whether RF-EMF exposure is related to reproductive outcomes in human populations. Melnick et al, comments that the use of surrogate measures of expo­sure (hours of mobile phone use), do not provide reliable information on exposure to the genitalia or the developing foetus. However, this is a valid and widely accepted approach in epidemiological studies when objective exposure data are unavailable (Teschke, 2003). Another criticism of Melnick et al. is that evidence on the effects of RF-RMF on female reproduc­tive outcomes was made without consider­ation of the extensive literature on oxidative stress due to RF-EMF exposures. Although several experimental studies suggest that RF-EMF exposure may induce oxidative stress, the validity of these findings is undermined by heterogeneity, and other methodological limitations that have been described in  the systematic review on oxidative stress (SR9) (Meyer et al., 2024). 

Male fertility in both in-vivo and in-vitro studies (SR4)
The main criticism for SR4 by Melnick et al. was that they believe that for some of the outcomes where effects were found, the certainty of the evidence should not have been downgraded. They argue that heterogeneity in the included papers does not impact the certainty of results. This is counter to standard systematic review methodology, such as  outlined in the GRADE assessment, which typically considers methodological heterogeneity as a source of inconsistency that reduces confidence in the overall body of evidence. By dismissing these variations as inconsequential, the authors risk overlooking important sources of bias or effect modification that could influence the interpretation of results. Melnick et al., suggests that the conclusions for SR4 should be changed because of some negative results, particularly those reported for laboratory animals and human sperm in vitro. However, the only negative outcome with a high level of certainty in the evidence, had an average exposure across the studies of 23.87 W/kg. This average is over 28,000 times the public limit for whole body average exposure set in the ARPANSA safety standard (RPS S-1)

Effects on cognition in human observational studies (SR5)
Melnick et al. had various criticisms of SR5, one of the mains ones being that a study by Grigoriev et al. (2018) should not have been excluded. However, this study was excluded due to significant methodological flaws, particularly, the lack of a description of exposure assessment, follow-up protocol and participant details.  For instance, the study fails to account for mobile phone use changing between the age of 7 to 17 years, which is critical given that previous research (Thomas et al., 2010; Bhatt et al., 2017) consistently shows increased mobile phone use as children age. Therefore, its exclusion from SR5 is well-founded. Regarding the exclusion of other studies, namely, cross-sectional studies were excluded as they are unable to establish causality, and this reason is clearly described in SR5 and its protocol.

Effects on cognitive performance in human experimental studies (SR6)
Melnick et al.’s main criticism of SR6 is the use of the neuropsychological assessment classification system for cognitive domains (Lezak et al., 2012) and suggest that the incorrectly cited Cattell-Horn-Carrol (CHC) taxonomy should have been used. However, the cognitive domains used for grouping the systematic review outcomes are in fact based on CHC taxonomy and Lezak et al. (2012) only describes ways these different domains can be tested for. This comment by the authors is therefore invalid. Melnick et al., further criticise the heterogeneity of the included studies stating this would reduce the ability of the meta-analyses to detect a small effect. Statistical power is an issue that was discussed in the systematic review and remains an issue in the literature, particularly in singular one off studies. However, the pooling data for use in a meta-analysis, as done in this systematic review, is how the lack of statistical power in the literature can be alleviated.  

Symptoms (human observational studies) (SR-7)
Melnick et al. purport that the health outcomes (tinnitus, migraines, and sleep disturbances) included in SR7 should not have been evaluated. This is contrasting to literature which clearly shows individuals report both of these as short-term and long-term effects (Medic et al., 2017; Lipton et sl., 2001; Zeleznik et al., 2024). The authors also criticised one of the key included studies, the COSMOS study (Auvinen et al., 2019), allegedly citing its inappropriate study methodology, such as exclusions and follow-up time. The COSMOS study excluded the subjects with a history of tinnitus or weekly headaches at baseline to avoid potential reverse causation. This is a common practice in epidemiological studies when the goal is to test the relationships between environmental exposures and health outcomes (Rezende et al., 2022). This approach helps isolate the effect of preexisting conditions and those that align with outcomes after exposure (Poorolajal, 2025). The four-year follow-up period in the COSMOS longitudinal cohort study is a strong design choice for evaluating health symptoms (Kamal et al., 2025). The methodological approaches adopted in the COSMOS study, represent the most robust design and are well supported by evidence.

Human experimental non-specific symptoms (SR-8)
With respect to the systematic review on non-specific symptoms in human experimental studies (SR-8) Melnick et al. assert that, because studies on people without idiopathic environmental intolerance attributed to EMF (IEI-EMF, also known as electromagnetic hypersensitivity or EHS) were included in the review, it cannot effectively review studies on people with IEI-EMF. However, in SR8, the analyses were subdivided by IEI-EMF and non IEI-EMF populations. The commentary also questions why human provocation studies using EMF frequencies outside of the RF section of electromagnetic spectrum are excluded from the review and suggests that their exclusion prevents consideration of the results of studies using RF-EMF exposure. SR-8’s protocol, and the entire set of systematic reviews more broadly, clearly define the exposure that is to be studied and so a large departure of the review from this paradigm would be inappropriate. The inclusion criteria of SR-8 adequately cover the topic under investigation (RF-EMF).

Oxidative Stress (SR-9)
The primary criticism Melnick et al. ascribe to the systematic review on the effect of RF-EMF on biomarkers of oxidative stress (SR-9) is the inclusion criteria. Melnick et al. take issue with the fact that numerous studies were excluded from SR-9 for using an unreliable method of outcome assessment and that studies were excluded for not adequately characterising their exposure system. However, both reasons are valid causes for exclusion regardless of the number of studies that fall under that category. The secondary criticism Melnick et al. have of SR-9 is the subdivision of the meta-analyses into biomarker and biological system pairs which they assert may dilute an overall effect. However, combining outcomes across different organs and markers would create a lack of specificity thus reducing the usability of the results for directing further research and drawing conclusions relevant to human health outcomes. Combining biomarkers and biological systems into a net category only enables vague discussions of miscellaneous oxidative stress. 

ARPANSA has written brief evaluations of each systematic review and published them as part of our radiation literature survey program. They are available here: SR1A & 1B, SR2, SR3A & B, SR4, SR5, SR6, SR7, SR8, SR9.