Hindsight check: LL2-21 (Ch 21 Mobile phone use and brain tumour risk: early warnings, early actions?)#

Source section: EEA, Late lessons from early warnings: science, precaution, innovation (EEA Report No 1/2013), Ch 21, by Lennart Hardell, Michael Carlberg and David Gee (report pp. 509–529; PDF pp. 511–531). There are no panels or responses. The authors are protagonists: Hardell’s group produced one of the two main bodies of evidence the chapter defends, and the EEA narrates its own 2007–2011 warnings (Box 21.2, p. 515). Check window: publication (2013) to late September 2026. The latest event in the chapter is the Italian Supreme Court ruling of 12 October 2012 (pp. 524–525). Checked: 25–26 September 2026.

Method note. - Search. General web search was unavailable for this pass because the session’s search budget was exhausted. I retrieved sources directly from: - the Europe PMC REST API (abstracts, plus open-access full texts where available); - the IARC Monographs website: the 2019 and 2024 priorities reports, the July 2026 newsletter, the list of volumes, and the Cancer in Humans chapter of Volume 102; - ICNIRP (the 2020 guidelines PDF and the commission membership page); - the European Commission (the SCHEER 2023 final opinion); - US FDA (the 2020 literature review) and the NTP website; - the D.C. Circuit (the EHT v FCC opinion); - CourtListener (US case records); - Légifrance and the French Sénat (the 2015 French law); - California Department of Public Health (CDPH); - and, where no primary source was reachable, The Nation, Microwave News and Phonegate Alert (an advocacy group). I flag these where used. - Access gaps. - WHO’s EMF fact-sheet URLs returned “page not found”, and WHO’s IRIS repository did not respond to scripted queries. I rely on WHO’s live Q&A on 5G and on IARC’s 2024 report for WHO’s position and plans. I could not confirm whether WHO’s promised Environmental Health Criteria monograph on radiofrequency fields has been published. - The full text of Karipidis et al. 2024 was blocked (HTTP 403), so I rely on its long structured abstract. I therefore cannot name the “one influential study” or the “five studies reporting implausible effect sizes” it identifies. - The D.C. Courts PDF of Murray v. Motorola (2025) was blocked. Its holding is taken from CourtListener’s indexed text (search excerpts), not a full reading. - The Italian court judgments themselves were not retrieved. The 2017 Ivrea and 2019/2020 Turin rulings are known here from an Italian occupational-medicine review abstract (Polichetti 2020) and an advocacy group’s summary. - The ANSES and ANFR websites (France’s health-safety agency and radio-frequency regulator) could not be read by script. I could not verify whether France’s dedicated research funding on radiofrequency fields comes from a levy on network operators. - The Légifrance text of the 2015 French law was read through a fetch tool that summarised it. Only one sentence (Art. 7) is quoted verbatim. - Annex 3 does not apply. Mobile phones are a new 2013 case with no 2001 counterpart. - Scope. I read only the section digest, the section’s source extract and web sources. Page numbers are report pages of the 2013 volume. - Conflicts of role (relevant because the chapter’s own argument is about independence). - Chapter authors in later work. Hardell co-authored the main later critique of the WHO-commissioned reviews [S4] and a critique of MOBI-Kids [S19]. He and Carlberg also published their own pooled analyses and a causal assessment [S14, S15]. - Authors of the WHO-commissioned human review [S1]: - Ken Karipidis has been an ICNIRP Main Commission member since May 2020 and Vice-Chair since July 2024 [S45]; - Maria Blettner is the scientist the chapter names as the lone dissenter in the 2011 IARC vote (p. 521); - Martin Röösli co-authored CEFALO, which the chapter attacks [S20]; - Susanna Lagorio co-authored the 2011 critique of the first Italian compensation ruling [S57]. - Other ICNIRP overlaps. - Jens Kuhne (ICNIRP member) co-authored the 2026 commentary that downgraded the WHO animal review [S34, S45]. - Young Hwan Ahn (ICNIRP member) co-authored both Japan–Korea animal replication studies [S35, S36, S45]. - Anke Huss, an ICNIRP member on the current commission page, co-authored COSMOS [S11, S45]. I did not check her membership dates. - Industry-linked data. Persson et al. 2012 [S52] is an industry study based on operator network statistics.

Overview#

1. The chapter’s central epidemiological conclusions have been substantially weakened. They were that long-term use raises the risk of glioma and acoustic neuroma, including from cordless phones and especially in adolescents (p. 524). Since 2013, the larger and better-designed studies have been largely null: - Cohorts: the UK Million Women Study update [S10] and the five-country COSMOS prospective cohort [S11]. - Young people: the 14-country MOBI-Kids case-control study [S16]. - Incidence data in the Nordic countries, Australia, Canada, New Zealand, Korea and the US [S23–S28]. - The WHO-commissioned systematic review (Karipidis et al. 2024) concluded, at “moderate certainty”, that mobile phone use “likely does not increase” the risk of glioma, meningioma or acoustic neuroma in adults, or of paediatric brain tumours [S1]. For cordless phones the certainty was only “low” [S1].

2. Not every signal has gone away. - CERENAT (France, 2014) found glioma OR 2.89 in the heaviest users [S9]. - A bias-adjusted reanalysis of Canadian Interphone data kept an OR of about 2 in the top quartile of use [S12]. - IARC’s 2024 advisory group still called the human evidence “mixed” [S40]. - A small, organised dissent continues, including ICBE-EMF (International Commission on the Biological Effects of Electromagnetic Fields) and Hardell [S4, S15, S19]. Some of these scientists (Kundi and Hutter [S8]) now propose that phones speed up existing tumours rather than start them, which would reconcile flat incidence data with modest odds ratios. This hypothesis is untested.

3. The laboratory evidence has moved partly in the chapter’s direction, then been contested. - In 2018 the US NTP (National Toxicology Program) found “clear evidence” of heart schwannomas and “some evidence” of gliomas in male rats [S31]. - In 2025 a WHO-commissioned animal review rated both at “high” certainty [S32, S33]. - In 2026 purpose-built Japanese and Korean replications found nothing [S35, S36]. A quantitative reanalysis, whose authors include the ICNIRP commissioner Jens Kuhne, downgraded the certainty to moderate (heart) and low (brain) [S34]. - ICBE-EMF disputes the replications’ statistical power [S37]. - So biological plausibility is more open than in 2013, even as the human evidence has weakened.

4. IARC has not re-evaluated. - The 2011 Group 2B classification still stands. - IARC’s advisory groups rated radiofrequency fields “high priority” in 2019 and again in 2024 [S39, S40]. The 2024 group suggested an evaluation “in the latter half” of 2025–2029 and said a change in classification “is uncertain” [S40]. - No radiofrequency meeting appears in IARC’s schedule through Meeting 144 (October 2027) [S41].

5. Governments behaved much as the chapter complained, with limited exceptions. - Thermal basis kept. Exposure limits remain designed to limit tissue heating (ICNIRP 2020 [S44]). The EU’s scientific committee SCHEER (2023) found “uncertain to weak” evidence on cancer and advised only a technical revision [S46]. The US FDA (2020) found “no quantifiable causal link” [S47]. A US federal appeals court upheld the FCC’s cancer reasoning in 2021 while faulting it on other health effects [S48]. - Precautionary exceptions: - France’s 2015 law, which bans Wi-Fi in spaces for children under three [S49]; - Berkeley’s point-of-sale notice, upheld in 2019 [S50]; - California’s 2017 advice on reducing exposure [S51].

6. Compensation and litigation split. - Italian courts awarded further workers’ compensation (Ivrea 2017, upheld by Turin 2019–2020), but other Italian courts refused [S56, S58]. - The main US brain-tumour litigation ended in July 2025 with all the plaintiffs’ experts excluded and judgment for the defendants affirmed [S59, S60].

7. What holds up best is the chapter’s institutional and communication diagnosis, in qualified form. - Holds up: - A “possible” classification is read in opposite ways. - Disputes over independence shape which evidence counts. - Latency and exposure assessment really do limit early studies. - Needs qualifying: the lesson that “non-positive is not negative” (p. 511). Later experience shows that several independent null lines, pursued for long enough, can put a tight ceiling on large risks. Cohorts, trends in national incidence and bias analyses are examples. What remains honestly open is narrower: - very long latencies (>15–20 years); - rare tumour subtypes; - the heaviest users; - childhood exposure under current patterns of use [S5, S16].

8. For the project’s lens, this is the report’s clearest candidate for an early warning that has, so far, largely not been borne out. It remains unresolved rather than refuted. The case also shows exposure falling for reasons unrelated to health policy (lower-power networks, texting, data use; see Claim 7). That makes it a useful counter-case to the volume’s dominant “warnings ignored, harm confirmed” pattern.

Claim-by-claim#

Claim 1: Long-term mobile phone use increases the risk of glioma and acoustic neuroma, highest for ipsilateral (same-side) use; no consistent pattern for meningioma (p. 524; also pp. 509, 513, Table 21.1 p. 516)#

What the chapter says. - “Results from the Hardell-group as well as from the Interphone group show an increased risk for glioma and acoustic neuroma associated with long term mobile phone use… The risk is highest for ipsilateral exposure… For meningioma there is no consistent pattern” (p. 524). - Because results differed by tumour type in the same studies, the chapter argues that this “strongly argues against systematic bias” (p. 524).

Subsequent developments.

Evidence consistent with the chapter: - CERENAT (France; published May 2014). - 253 gliomas, 194 meningiomas and 892 matched controls (2004–2006). Regular use was not associated with glioma: OR 1.24 (0.86–1.77). - The heaviest users (≥896 lifetime hours) had glioma OR 2.89 (1.41–5.93) and meningioma OR 2.57 (1.02–6.44). The authors say the data “support previous findings concerning a possible association between heavy mobile phone use and brain tumours” [S9]. - The meningioma finding cuts against the chapter’s “no consistent pattern” for meningioma. - Canadian Interphone reanalysis (Momoli et al., Am J Epidemiol 2017). Glioma OR 2.0 (1.2–3.4) in the top quartile (>558 h), and 2.2 (1.3–4.1) after probabilistic adjustment for recall and selection bias. There was little evidence for meningioma or acoustic neuroma [S12]. - Hardell & Carlberg pooled analysis (2015). 1,498 cases and 3,530 controls. - Glioma OR 1.3 (1.1–1.6) for mobile phones, rising to 3.0 (1.7–5.2) at >25 years’ latency. Ipsilateral OR 1.8. Cordless OR 1.4 [S14]. - In 2017 Carlberg and Hardell concluded that radiofrequency radiation “should be regarded as a human carcinogen causing glioma” [S15]. - Choi, Moskowitz et al. meta-analysis (2020). Cumulative call time >1,000 h “statistically significantly increased the risk of tumors” [S6].

Evidence against: - UK Million Women Study update (Schüz et al., JNCI 2022). - 776,156 women, 3,268 incident brain tumours over 14 years. Glioma RR 0.89 (0.80–0.99) for ever vs never use. - No association with daily use or ≥10 years’ use. Relative risks for temporal and parietal gliomas were “slightly below 1.0” [S10]. - IARC’s 2024 advisory report notes that the earlier signal for acoustic neuroma (10+ years’ use, RR 2.46, 1.07–5.64; Benson et al. 2013) was attenuated to 1.32 (0.89–1.96). It also notes that “the exposure assessment was crude” [S40]. - COSMOS prospective cohort (Feychting et al., Environment International, March 2024). - 264,574 participants in five countries; median follow-up 7.12 years. Call time was regression-calibrated against operator records. - Hazard ratio (HR) per 100 cumulative hours: glioma 1.00 (0.98–1.02), meningioma 1.01, acoustic neuroma 1.02 (0.99–1.06). - Glioma HR 1.07 (0.62–1.86) for ≥1,908 h, and 0.97 (0.62–1.52) for >15 years’ use [S11]. - Critics note the short follow-up (letters and replies in Environment International 2024; see [S11] and linked responses). - Interphone bias simulation (Bouaoun et al., Epidemiology 2024; IARC and ISS authors). Interphone’s J-shaped curve was reproduced under no true effect. Case–control differences in recall error alone produced a spurious OR of 1.91 in the heaviest users. The authors say this makes a causal glioma risk from heavy use “less likely” [S13]. It also shows that bias can produce effects that differ by level of use. That weakens the chapter’s argument that different results by tumour type exclude systematic bias (p. 524). - WHO-commissioned systematic review (Karipidis et al., Environment International, online 30 Aug 2024). - 63 aetiological articles (1994–2022), 22 countries. - Meta-estimates for ever/regular use: glioma mRR 1.01 (0.89–1.13); meningioma 0.92 (0.82–1.02); acoustic neuroma 1.03 (0.85–1.24). No increase with time since first use, cumulative call time or number of calls. - In low-risk-of-bias studies of long-term (10+ years) users: glioma 0.95 (0.85–1.05); acoustic neuroma 1.00 (0.78–1.29). - Simulation studies of incidence trends showed RR >1.5 with a 10+ year induction period to be “definitely implausible”. Excluding five studies with such implausible estimates cut heterogeneity from 77% to 3.6% [S1]. - Conclusion: “moderate certainty evidence that it likely does not increase the risk of glioma, meningioma, acoustic neuroma, pituitary tumours, and salivary gland tumours in adults” [S1]. - Regulatory syntheses reached the same view: - ICNIRP 2020 said that the Hardell results “are not consistent with trends in brain cancer incidence rates” and that no cohort reports a higher risk [S44]; - SCHEER 2023: “uncertain to weak weight of evidence” for neoplastic diseases [S46]; - FDA 2020: “no quantifiable causal link”, noting that some studies of “heavy” users suggest a possible link but show “no clear and consistent pattern” [S47].

Dissent and where the mainstream stands. - ICBE-EMF critique (Melnick et al., Environmental Health, Oct 2025; co-authors include Hardell). It argues that the WHO-commissioned reviews “cannot be used as proof of safety” [S4]. The review authors replied to a separate letter in January 2025 [S3]. - Kundi and Hutter (2026) show by modelling that a growth-promoting effect (a shift of tumours to younger ages) could reproduce the meta-analytic OR of 1.22 and the null cohort results. They acknowledge that this is untested [S8]. - Where the mainstream stands. It now treats a substantial risk as unlikely. The residual uncertainty lies in very long latency, rare subtypes and the heaviest use [S5, S40].

Verdict: weakened. - The large-risk claim, as the chapter framed it (driven by Hardell’s odds ratios of 2–3 and ipsilateral excess), is not supported by the cohort, incidence and systematic-review evidence. - A heavy-use signal persists in some case-control data (CERENAT, Canadian Interphone), and IARC still calls the human evidence “mixed” [S40]. So “refuted” would overstate it.

Implications for weight. - The chapter’s epidemiological conclusion should not be used as an example of a confirmed late lesson. - Its method-level points need two qualifications: - later cohorts partly solved the exposure-assessment problem (regression calibration against operator records); - bias analysis showed that patterns the chapter read as signs of causation can arise from recall error.

Claim 2: Adolescents and children are at higher risk (first use before age 20: glioma OR 3.1, acoustic neuroma OR 5.0); children absorb more energy, have developing brains and longer lives ahead (pp. 514, 524)#

Subsequent developments. - MOBI-Kids (Castaño-Vinyals et al., Environment International, online 30 Dec 2021). - 14 countries; 899 brain-tumour cases aged 10–24 and 1,910 appendicitis controls (2010–2015); 22% overall were long-term (>10 years) users. - Odds ratios for neuroepithelial tumours fell with increasing time since first use, number of calls, call time and estimated dose at the tumour site. The authors attribute these values below 1 partly to proxy recall and pre-diagnosis symptoms. - Conclusion: “no evidence of a causal association between wireless phone use and brain tumours in young people”. They add that the biases “prevent us from ruling out a small increased risk” [S16]. - MOBI-Kids recall validation (van Wel et al. 2024). Among 702 participants, self-reports compared with billing records found “no evidence of differential recall error between cases and controls” [S21]. - Country analyses. Japan and Korea found no association (glioma OR 0.70 in the top tertile) [S17]. A Japanese analysis with wider age limits, published in 2025, also found none [S18]. - WHO-commissioned review. Paediatric brain tumours mRR 1.06 (0.74–1.51). Moderate certainty of no increase [S1]. - Dissent. Hardell and Moskowitz (2022) argue that MOBI-Kids is “uninterpretable”. Their reasons: appendicitis controls, exclusion of centrally located tumours, and lagging that treats recent use as unexposed [S19]. - What remains open. Röösli et al. (2019) list use “during childhood” as a remaining uncertainty [S5]. The dosimetric premise (greater absorption in children’s heads) was not the subject of this check and is not contradicted by anything found.

Verdict: weakened. - The claim of a higher age-specific risk rested on small numbers in one group’s data (e.g. acoustic neuroma OR 5.0 with CI 1.5–16, p. 514). - The purpose-built international study and the WHO-commissioned review both found no increase. Neither can exclude a small risk.

Implications for weight. - The chapter’s argument for extra precaution for children remains defensible as a decision rule under uncertainty: a cheap measure that protects a group with more years of exposure ahead of it. - It should not be presented as resting on demonstrated higher risk.

Claim 3: Cordless (DECT) phone use increases brain tumour risk “when properly assessed and analysed” (p. 524)#

Subsequent developments. - Hardell group. Its pooled data continue to report cordless-phone glioma OR 1.4 (1.1–1.7), rising to 1.7 at >15–20 years’ latency [S14]. - WHO-commissioned review. - Cordless phone use was “not significantly associated” with glioma (mRR 1.04, 0.74–1.46; I² = 74%), meningioma (0.91) or acoustic neuroma (1.16, 0.83–1.61). - It rated this only “low certainty evidence that it may not increase the risk” [S1]. - Studies outside the Hardell group. Few assess cordless phones separately. MOBI-Kids analysed “wireless phones” (mobile and cordless together) and found no association [S16]. COSMOS-France now records cordless use (66% of participants) but has not yet reported on tumours [S53].

Verdict: weakened, with low certainty on both sides. The positive claim still rests almost entirely on one research group. The best independent synthesis finds no significant association, while acknowledging that the evidence base is thin and heterogeneous.

Implications for weight. - The chapter’s methodological point stands on its own terms: counting cordless users as “unexposed” biases mobile-phone estimates towards the null (p. 514). - The substantive claim that cordless phones cause tumours should carry little weight.

Claim 4: National brain tumour incidence data are “of limited value” for testing the case-control associations (pp. 509, 523)#

What the chapter says. - Incidence can be affected by other, unknown risk factors, and the mechanism is unclear (p. 523). - The chapter questions Swedish registry quality and cites rising Danish rates, including an “almost 4-fold increase” in glioblastoma reported in a Danish news release (p. 523). - It dismisses Little et al. 2012 as going “far beyond scientific evidence” (p. 523).

Subsequent developments. Incidence comparison became one of the main lines of evidence against large risks. - United States. - Little et al. 2012: glioma rates were flat in 1992–2008 (−0.02%/yr). Hardell-based projections should have been “at least 40% higher” than observed by 2008, whereas projections from Interphone’s small highly exposed group “could be consistent” [S22]. - Zhang and Muscat (2025), SEER 2000–2021: malignant tumours −0.6%/yr; temporal-lobe tumours −0.06%/yr (not significant); acoustic neuroma +0.09%/yr (not significant) [S28]. - Nordic countries (Deltour et al., Environment International, 2022; 18,232 gliomas, 1979–2016). - Among men aged 40–59, incidence rose only 0.1%/yr. The trends are incompatible with RR ≥1.08 at a 10-year lag, ≥1.2 at 15 years and ≥1.5 at 20 years. - For men aged 60–69, RR ≥1.4, ≥2 and ≥2.5 could be rejected at the same lags. - Denmark is included [S24]. - Australia (Karipidis et al. 2018). No increase in any histological type or in temporal-lobe glioma during 2003–2013. An earlier rise in glioblastoma (1993–2002) is attributed to MRI [S23]. - Canada (Villeneuve et al. 2021; co-authors include Canadian Interphone investigators). Rates were stable in 1992–2015. Applying the Swedish pooled, 13-country Interphone and Canadian Interphone risk estimates overestimated 2015 cases by 50%, 86% and 63% respectively [S25]. - New Zealand (Elwood et al. 2022). A small decrease at ages 10–69 in 1995–2020 [S26]. - Korea (Choi et al. 2021). Glioma and glioblastoma rates rose (2.6–3.9%/yr), but “no association was observed” with subscriber numbers [S27].

Evidence the other way: - England (Philips et al. 2018). Glioblastoma incidence more than doubled (age-standardised rate 2.4 to 5.0 per 100,000, 1995–2015). The authors infer a possible “environmental or lifestyle factor” but explicitly do “not provide additional evidence for the role of any particular risk factor” [S29]. - Sweden (Hardell and Carlberg 2017). Inpatient-register codes for brain tumours of unknown type rose 4.24%/yr in 2007–2015. They argue that the Cancer Register under-reports cases [S30].

How evidence reviews now treat it. - The WHO-commissioned review used incidence simulations as a “credibility benchmark” [S1]. - ICNIRP 2020 and Röösli et al. 2019 describe incidence data as “informative for this specific topic” because adoption was steep, registration nearly complete and competing risk factors few [S5, S44].

Verdict: weakened. - The mainstream now treats incidence data as strong evidence against the large risks reported by some case-control studies, including the chapter’s authors’ own. - The chapter’s caveat survives in a narrow form. Incidence data cannot rule out small relative risks, effects with very long latency, or growth-promotion effects that shift diagnosis earlier [S8]. Rising glioblastoma registrations in some countries remain unexplained and contested.

Implications for weight. The chapter shows an asymmetry the digest already flagged: it dismisses incidence data when flat and cites them when rising (p. 523). That asymmetry is itself a transferable lesson about motivated treatment of evidence. The claim that descriptive data are “of limited value” should carry little weight.

Claim 5: Hardell group and Interphone glioma results are “similar” when the same inclusion and exclusion criteria are used (p. 520)#

Subsequent developments. - Little et al. 2012 (the half the chapter omits). The chapter reports only that the Hardell-based projections were incompatible with US data (p. 523). The same paper found Interphone’s modest excess among heavy users “could be consistent” with them [S22]. - Choi et al. 2020, subgroup meta-analysis by research group, for regular use: - Hardell studies OR 1.15 (1.00–1.33); - Interphone-related studies OR 0.81 (0.75–0.89); - others null [S6]. - Villeneuve et al. 2021. Both sets of estimates overestimated Canadian incidence [S25]. - ICNIRP 2020. Hardell’s studies differ in reporting significantly raised risks “already after less than five years” of use and at “quite low levels of cumulative call time” [S44]. - Bouaoun et al. 2024. Interphone’s heavy-user excess can be produced by bias alone [S13]. - Karipidis et al. 2024. A single “influential” study drove much of the glioma heterogeneity, and five studies had implausible effect sizes. The abstract does not name them [S1].

Verdict: weakened. Independent analyses consistently treat the two bodies of evidence as different. Hardell’s results are the outlier, and the Interphone heavy-user excess is itself now more plausibly attributed to bias.

Implications for weight. The chapter’s framing of the two sets of studies as “complementary and generally mutually supportive” (p. 509) should not be relied on. The episode is a good example of authors assessing their own evidence (digest insight 11).

Claim 6: The IARC 2B classification did not have “any significant impact on governments’ perceptions of their responsibilities” (pp. 509, 524)#

Subsequent developments. - IARC itself. - The Volume 102 monograph was published in 2013 [S38]. - The 2019 advisory group recommended radiofrequency fields as “High priority (and ready for evaluation within 5 years)”. It cited new case-control results (Hardell 2013, 2015; CERENAT) and the NTP and Ramazzini animal studies [S39]. - The 2024 advisory group repeated the rating, judged the human evidence “mixed” and the animal evidence new. It suggested an evaluation “in the latter half of the next 5 years” to await ongoing bioassays, adding that “a change in the current classification… is uncertain” [S40]. - As of IARC’s July 2026 newsletter, the scheduled meetings are 143 (cannabis smoking, November 2026) and 144 (lead and cadmium, October 2027). Volumes 139–142 cover other agents. No re-evaluation has occurred or been scheduled [S41]. - Exposure standards. - ICNIRP 2020. Restrictions are set to limit “temperature rise”. The guidelines state: “There is no evidence that additional precautionary measures will result in a benefit to the health of the population” [S44]. ICNIRP says it considered “low-level” and “non-thermal” evidence and found no substantiated effect. - EU (SCHEER 2023). No “moderate or strong level of evidence for adverse health effects” below EU limits; only a technical revision for new dosimetric quantities [S46]. - United States. FDA 2020 found “no quantifiable causal link” [S47]. In EHT v FCC (D.C. Cir., 13 Aug 2021), the court held that the FCC “offered an adequate explanation” on cancer. It remanded for failing to explain why its limits protect against “negative health effects unrelated to cancer”, including effects on children and long-term exposure [S48]. I did not check the FCC’s response on remand. - WHO. Its current Q&A still says “no adverse health effect has been causally linked with exposure to wireless technologies” and that “tissue heating is the main mechanism”. This is close to the 2011 wording the chapter criticises (p. 520). The Q&A also still promises a risk assessment “by 2022” [S42]. IARC’s 2024 report says it will be an Environmental Health Criteria monograph [S40]; I could not confirm publication. - Precautionary measures after 2011: - France, Loi n° 2015-136 (9 February 2015). It bans fixed wireless terminals in spaces for children under three (“L’installation d’un équipement terminal fixe équipé d’un accès sans fil à internet est interdite dans les espaces dédiés à l’accueil, au repos et aux activités des enfants de moins de trois ans”, Art. 7). It also: - requires Wi-Fi in primary schools to be switched off when not used for teaching; - mandates a public information policy on “responsible” phone use (Art. 6); - assigns radiofrequency monitoring and research programmes to a health agency (Art. 3) [S49]. - Berkeley, California. A point-of-sale notice ordinance (2015) was upheld by the Ninth Circuit (CTIA v. City of Berkeley, 928 F.3d 832, 2 July 2019). The Supreme Court denied certiorari on 9 December 2019 [S50]. - California Department of Public Health. Guidance of December 2017 on How to Reduce Exposure to Radiofrequency Energy from Cell Phones (speakerphone or headset; text rather than call) [S51].

Verdict: partly held up. - As a description, the chapter was largely right, and the pattern persisted. The 2B classification did not change the thermal basis of limits in ICNIRP, the EU or the US, and IARC has not re-evaluated in 15 years. - National and local precautionary steps did follow in some places. - The chapter implies this inaction was a failure. Given the later evidence (Claims 1–4), mainstream bodies would say the evidence did not warrant more. The case against inaction rests on low cost and residual uncertainty rather than on demonstrated harm.

Implications for weight. - The chapter supports a lesson about institutional inertia after a hazard classification. - It is a weak example of harmful inertia. The same facts fit a regulator correctly declining to act on a signal that later weakened. - Both readings should be kept.

Claim 7: Recommendations. Reduce head exposure (texting, hands-free, better design), especially for young people; reconsider the thermal-based standards; label phones and warn; fund independent research via a levy on phones and masts (Box 21.2, p. 515; p. 524)#

Subsequent developments, by recommendation. 1. Reduce head exposure. - Largely happened, but mostly through technology and use rather than health policy, as the chapter’s own footnote 9 anticipated (p. 515). - An industry study using Swedish operator network statistics found average 3G handset output power during voice calls “below 1 mW”, “<1% of the maximum available” [S52]. It predates the chapter but shows why network changes lowered exposure. - Official advice to use speakerphones, headsets and texting appeared in some jurisdictions (California 2017 [S51]; France’s 2015 information duty [S49]). - Behaviour change was incomplete. In COSMOS-France (2017–2019), most users reported short weekly call times (the modal category was 5–29 min/week) but also held the phone to the head, not using loudspeakers or hands-free kits [S53]. - I did not retrieve national statistics on voice minutes. 2. Reconsider thermal standards: not adopted. ICNIRP 2020 kept a heating-based rationale [S44]; SCHEER 2023 advised only a technical update [S46]; the FCC retained its limits, and the court upheld it on cancer [S48]. ICBE-EMF continues to argue that limits “lack scientific credibility” [S4]. 3. Labelling and warnings: patchy, local. - Berkeley’s notice survived constitutional challenge [S50]. - France’s 2015 law strengthened information duties [S49]. - I found no EU-wide or US federal cancer-warning labelling. 4. Research levy administered independently: not verified as adopted in the form proposed. - Major post-2013 evidence was publicly funded. The WHO-commissioned review was funded by WHO, New Zealand’s health ministry, Italy’s ISS and Australia’s ARPANSA [S1]. France’s ANSES funds a dedicated radiofrequency research line, with project codes such as 2020-CRD-RF20-01 and EST-2016-2RF-04 [S53, S54]. - I could not confirm from a primary source that this line is financed by a charge on network operators. - The chapter’s companion concern, “funding bias” (fn 10, p. 515), got mixed support. A 2017 systematic review found study quality associated with higher risk estimates but no significant relation between funding source and results [S7].

Verdict: partly held up. Exposure to the head probably fell, largely through unplanned technical change. The thermal-standard reform was not adopted, labelling appeared only locally, and a levy is unverified.

Implications for weight. The case supports a technology-neutral lesson: exposure can fall for reasons unrelated to precautionary policy, through design changes made for other purposes (battery life, network efficiency). This complicates any judgement about whether precaution “worked”. It gives little support to the claim that standard-setters were wrong to keep the thermal basis.

Claim 8: Evidence is increasing that heavy long-term users who develop glioma or acoustic neuroma should be compensated; the Italian Supreme Court ruling of 12 October 2012 was the first case in the world (pp. 509, 524–525)#

Subsequent developments. - Marcolini case. Microwave News (23 October 2012) confirms the facts: the Brescia Court of Appeal ruled in December 2009, and the Court of Cassation affirmed on 12 October 2012 [S55]. An Italian occupational-medicine review (Polichetti 2020) notes that the Cassation decides on legitimacy, not merits. So the fact that its decision came after the IARC classification “is irrelevant” to its scientific basis [S56]. Lagorio and Vecchia (2011) had criticised the first-instance expert testimony as “seriously flawed” [S57]. - Later Italian cases. - Polichetti (2020) reports that “some Italian courts” recognised an occupational origin in the last decade. The Ivrea proceeding’s court-appointed experts reasoned that “the rarity of the circumstance is indicative of a causal association” [S56]. - Two courts, Cremona and Milan, refused. Their reasoning: a “possibly carcinogenic” (2B) agent cannot meet the “more likely than not” standard [S56]. The review notes that the media “virtually ignored” the negative judgments [S56]. - Turin Court of Appeal. It upheld the 2017 Ivrea ruling in Romeo v. INAIL (judgment 904/2019 of 3 December 2019, published 13 January 2020) [S58]. This is from an advocacy group’s account, not the judgment itself. According to that account, the court relied on court-appointed experts and gave less weight to studies by authors linked to ICNIRP, SCENIHR (the EU’s former scientific committee on emerging health risks) or industry funding [S58]. - United States. - In Motorola v. Murray (D.C. Court of Appeals, en banc, 20 October 2016), the court adopted Federal Rule of Evidence 702 for expert evidence [S59]. - On remand, the trial court excluded “the proffered opinion testimony of all of appellants’” experts and granted summary judgment. The Court of Appeals affirmed on 17 July 2025 (“we affirm the judgment of the trial court”) [S60]. This is from CourtListener’s indexed text; I did not read the full opinion.

Verdict: weakened. - In Italy the chapter’s expectation partly came true: at least one further compensation award was upheld on appeal. - Elsewhere it did not. Italian courts were themselves divided, the major US litigation failed on scientific admissibility, and the science the chapter cited as “increasing” evidence has since weakened (Claims 1–4).

Implications for weight. - The episode supports digest insight 10 only in a descriptive form: legal systems can act on individual cases before science settles. - It shows the results depend heavily on the standard of proof and on how courts vet experts. Italy’s civil “more likely than not” test with court-appointed experts produced different answers from a US Rule 702 gatekeeping regime. - It is not evidence that the underlying risk is real.

Claim 9: The IARC decision started a worldwide “spinning machine” similar to the tobacco industry’s campaign against IARC’s passive-smoking work (p. 521, fn 16)#

What the chapter offers. - As evidence it quotes industry and trade-association statements after the decision (the manufacturers’ forum MMF, the operators’ association GSMA, and the Federation of Finnish Technology Industries; pp. 520–521), plus blogs and Microwave News. - It cites no internal industry documents for mobile phones. The tobacco parallel (Ong and Glanz 2000) rests on Philip Morris documents.

Subsequent developments. - Documentary evidence. The one internal document I found cited later is a 1994 Motorola memo saying the company had “war gamed” the science. It predates IARC 2011 by 17 years and is reported in a 2018 investigation in The Nation [S61]. That article also reports, without primary documents, that: - three industry trade associations had observer status at the 2011 Lyon meeting; - “two industry-funded experts” sat on the working group; - the industry “mounted a campaign to discredit” Hardell [S61]. - Public-record industry action. The industry (CTIA) litigated to block Berkeley’s point-of-sale notice and lost [S50]. - What was not found. No later disclosure (litigation discovery, archives, investigative release) shows a coordinated post-2011 strategy against IARC comparable to the Philip Morris programme. The Murray litigation ended on expert admissibility [S60]. - Evidence cutting the other way. - The main later null studies were publicly funded (Million Women, COSMOS, MOBI-Kids, the WHO-commissioned review [S1, S10, S11, S16]). - A systematic review found no significant association between funding source and results [S7]. - Independence disputes persist in structural form. ICNIRP commissioners led or co-authored several key later reassuring or downgrading analyses [S1, S34, S35, S36, S45]. Critics and one Italian court treat this as a conflict of interest [S4, S58]. That is an independence concern about expert networks, not documentary evidence of an industry campaign.

Verdict: unclear. The chapter’s claim was asserted rather than evidenced, and nothing found since supplies the missing documents for the post-2011 period. Industry positioning and litigation against disclosure are documented. Overlaps in personnel between standard-setters and evidence reviewers are real and contested.

Implications for weight. - The “manufactured doubt” analogy should not be used as an established part of this case. - The more defensible and technology-neutral lesson is structural. When the same small expert network sets standards, reviews the evidence and runs the replication studies, its conclusions will be distrusted whatever their merit. Independence has to be designed in and visible (digest insight 6).

Claim 10: CEFALO, had it been available to IARC, would have supported the 2B classification (operator-recorded OR 2.15 at >2.8 years since first subscription) (p. 514)#

Subsequent developments. - The CEFALO authors’ own interpretation (Aydin et al., JNCI, July 2011). - 352 cases and 646 controls aged 7–19. Regular use OR 1.36 (0.92–2.02); ≥5 years since first use OR 1.26 (0.70–2.28). - In the operator-record subset, risk “was related to the time elapsed since the mobile phone subscription was started but not to amount of use”. There was no excess in the brain areas most exposed. - Conclusion: “The absence of an exposure-response relationship either in terms of the amount of mobile phone use or by localization of the brain tumor argues against a causal association” [S20]. - IARC’s published monograph (2013). Its Cancer in Humans chapter does not cite Aydin et al. 2011 (my text search of the IARC PDF), so the published evaluation did not rely on CEFALO either way [S38]. - Later child and adolescent evidence (MOBI-Kids [S16]; the Japanese and Korean analyses [S17, S18]; the WHO-commissioned paediatric meta-estimate [S1]) found no association.

Verdict: weakened. - The chapter selected one subgroup trend (time since subscription) and set aside the authors’ main reason for scepticism (no dose-response by amount or by tumour location). Later, larger studies of young people did not support an association. - The chapter’s narrower criticism holds: CEFALO was small, with short latency, and could not by itself be “reassuring evidence” (p. 514). MOBI-Kids’ own authors likewise say a small risk cannot be excluded [S16].

Implications for weight. This is a clear instance of selective reading of a study (digest insight 11). Use it as an example of how contested evidence is argued over, not as support for the 2B classification.

Implications for the digest’s transferable insights#

Sources#

Human epidemiology: systematic reviews, pooled analyses and commentary - [S1] Karipidis, K., Baaken, D., Loney, T., Blettner, M., et al. The effect of exposure to radiofrequency fields on cancer risk in the general and working population: a systematic review of human observational studies – Part I: Most researched outcomes. Environment International 191:108983 (online 30 Aug 2024). https://doi.org/10.1016/j.envint.2024.108983 (abstract read via Europe PMC; full text blocked) - [S2] Karipidis, K., et al. … Part II: Less researched outcomes. Environment International (online 11 Jan 2025). https://doi.org/10.1016/j.envint.2025.109274 (title and authors only) - [S3] Karipidis, K., et al. Response to the letter from Di Ciaula et al. Environment International (online 31 Jan 2025). https://doi.org/10.1016/j.envint.2025.109276 (title only) - [S4] Melnick, R. L., Moskowitz, J. M., Héroux, P., …, Hardell, L., … (ICBE-EMF). The WHO-commissioned systematic reviews on health effects of radiofrequency radiation provide no assurance of safety. Environmental Health (2 Oct 2025). https://doi.org/10.1186/s12940-025-01220-4 - [S5] Röösli, M., Lagorio, S., Schoemaker, M. J., Schüz, J., Feychting, M. Brain and salivary gland tumors and mobile phone use: evaluating the evidence from various epidemiological study designs. Annual Review of Public Health 2019. https://doi.org/10.1146/annurev-publhealth-040218-044037 - [S6] Choi, Y. J., Moskowitz, J. M., Myung, S. K., Lee, Y. R., Hong, Y. C. Cellular phone use and risk of tumors: systematic review and meta-analysis. IJERPH (Nov 2020). https://doi.org/10.3390/ijerph17218079 - [S7] Prasad, M., et al. Mobile phone use and risk of brain tumours: a systematic review of association between study quality, source of funding, and research outcomes. Neurological Sciences (Feb 2017). https://doi.org/10.1007/s10072-017-2850-8 - [S8] Kundi, M., Hutter, H. P. Interpretation of epidemiological studies on the relationship between mobile phone use and cancer. Epidemiologia (June 2026). https://doi.org/10.3390/epidemiologia7030086

Human epidemiology: adult studies - [S9] Coureau, G., et al. Mobile phone use and brain tumours in the CERENAT case-control study. Occupational and Environmental Medicine (May 2014). https://doi.org/10.1136/oemed-2013-101754 - [S10] Schüz, J., Pirie, K., Reeves, G. K., Floud, S., Beral, V. Cellular telephone use and the risk of brain tumors: update of the UK Million Women Study. JNCI 114 (2022). https://doi.org/10.1093/jnci/djac042 - [S11] Feychting, M., Schüz, J., Toledano, M. B., et al. Mobile phone use and brain tumour risk – COSMOS, a prospective cohort study. Environment International (online 2 Mar 2024). https://doi.org/10.1016/j.envint.2024.108552 (letters and replies: https://doi.org/10.1016/j.envint.2024.108665 ; https://doi.org/10.1016/j.envint.2024.108664 ; https://doi.org/10.1016/j.envint.2024.108808) - [S12] Momoli, F., et al. Probabilistic multiple-bias modeling applied to the Canadian data from the Interphone study. American Journal of Epidemiology 186 (2017). https://doi.org/10.1093/aje/kwx157 - [S13] Bouaoun, L., Byrnes, G., Lagorio, S., Feychting, M., et al. Effects of recall and selection biases on modeling cancer risk from mobile phone use: results from a case-control simulation study. Epidemiology (May 2024). https://doi.org/10.1097/EDE.0000000000001749 - [S14] Hardell, L., Carlberg, M. Mobile phone and cordless phone use and the risk for glioma – analysis of pooled case-control studies in Sweden, 1997–2003 and 2007–2009. Pathophysiology (online Oct 2014; 2015). https://doi.org/10.1016/j.pathophys.2014.10.001 - [S15] Carlberg, M., Hardell, L. Evaluation of mobile phone and cordless phone use and glioma risk using the Bradford Hill viewpoints. BioMed Research International (Mar 2017). https://doi.org/10.1155/2017/9218486

Human epidemiology: children and young people - [S16] Castaño-Vinyals, G., Sadetzki, S., Vermeulen, R., et al. Wireless phone use in childhood and adolescence and neuroepithelial brain tumours: results from the international MOBI-Kids study. Environment International 160 (online 30 Dec 2021; 2022). https://doi.org/10.1016/j.envint.2021.107069 - [S17] Kojimahara, N., et al. Impact of radiofrequency exposure from mobile phones on the risk of developing brain tumors in Korean and Japanese adolescents: a MOBI-Kids case-control study. Journal of Epidemiology (2023/2024). https://doi.org/10.2188/jea.JE20230005 - [S18] Kojimahara, N., et al. Brain tumor and mobile phone risk among young people: analysis of Japanese people using the MOBI-Kids international case-control study. Bioelectromagnetics (Dec 2025). https://doi.org/10.1002/bem.70032 - [S19] Hardell, L., Moskowitz, J. M. A critical analysis of the MOBI-Kids study of wireless phone use in childhood and adolescence and brain tumor risk. Reviews on Environmental Health (online May 2022). https://doi.org/10.1515/reveh-2022-0040 - [S20] Aydin, D., Feychting, M., Schüz, J., et al. Mobile phone use and brain tumors in children and adolescents: a multicenter case-control study (CEFALO). JNCI 103 (July 2011). https://doi.org/10.1093/jnci/djr244 - [S21] van Wel, L., et al. Validation of mobile phone use recall in the multinational MOBI-Kids study. Bioelectromagnetics (May 2024). https://doi.org/10.1002/bem.22507

Incidence-trend studies - [S22] Little, M. P., Rajaraman, P., Curtis, R. E., et al. Mobile phone use and glioma risk: comparison of epidemiological study results with incidence trends in the United States. BMJ 344:e1147 (Mar 2012). https://doi.org/10.1136/bmj.e1147 - [S23] Karipidis, K., Elwood, M., Benke, G., et al. Mobile phone use and incidence of brain tumour histological types, grading or anatomical location: a population-based ecological study. BMJ Open (Dec 2018). https://doi.org/10.1136/bmjopen-2018-024489 - [S24] Deltour, I., Poulsen, A. H., Johansen, C., Feychting, M., Johannesen, T. B., Auvinen, A., Schüz, J. Time trends in mobile phone use and glioma incidence among males in the Nordic countries, 1979–2016. Environment International (online 24 Aug 2022). https://doi.org/10.1016/j.envint.2022.107487 - [S25] Villeneuve, P. J., Momoli, F., Parent, M.-É., Siemiatycki, J., Turner, M. C., Krewski, D. Cell phone use and the risk of glioma: are case-control study findings consistent with Canadian time trends in cancer incidence? Environmental Research (May 2021). https://doi.org/10.1016/j.envres.2021.111283 - [S26] Elwood, J. M., Win, S. S., Aye, P. S., Sanagou, M. Trends in brain cancers (glioma) in New Zealand from 1995 to 2020, with reference to mobile phone use. Cancer Epidemiology (Aug 2022). https://doi.org/10.1016/j.canep.2022.102234 - [S27] Choi, K. H., et al. Mobile phone use and time trend of brain cancer incidence rate in Korea. Bioelectromagnetics (Sept 2021). https://doi.org/10.1002/bem.22373 - [S28] Zhang, L., Muscat, J. E. Trends in malignant and benign brain tumor incidence and mobile phone use in the U.S. (2000–2021): a SEER-based study. IJERPH (June 2025). https://doi.org/10.3390/ijerph22060933 - [S29] Philips, A., Henshaw, D. L., Lamburn, G., O’Carroll, M. J. Brain tumours: rise in glioblastoma multiforme incidence in England 1995–2015 suggests an adverse environmental or lifestyle factor. Journal of Environmental and Public Health (June 2018). https://doi.org/10.1155/2018/7910754 - [S30] Hardell, L., Carlberg, M. Mobile phones, cordless phones and rates of brain tumors in different age groups in the Swedish National Inpatient Register and the Swedish Cancer Register during 1998–2015. PLoS One (Oct 2017). https://doi.org/10.1371/journal.pone.0185461

Experimental animal evidence - [S31] US National Toxicology Program. Cell phone radio frequency radiation (study page; Technical Reports published November 2018). https://ntp.niehs.nih.gov/whatwestudy/topics/cellphones (accessed 26 Sep 2026) - [S32] Mevissen, M., Ducray, A., Ward, J. M., et al. Effects of radiofrequency electromagnetic field exposure on cancer in laboratory animal studies, a systematic review. Environment International 199 (online 25 Apr 2025). https://doi.org/10.1016/j.envint.2025.109482 - [S33] Mevissen, M., et al. Corrigendum to the above. Environment International (online 14 July 2026). https://doi.org/10.1016/j.envint.2026.110368 - [S34] Belenki, D., Baaken, D., Meyer, F., Kuhne, J. Commentary on the systematic review of radiofrequency field exposure and animal cancer by Mevissen et al. (2025). Environment International (online 9 Mar 2026). https://doi.org/10.1016/j.envint.2026.110154 - [S35] Imaida, K., et al. The International Collaborative Animal Study of mobile phone radiofrequency radiation carcinogenicity and genotoxicity: the Japanese study. Toxicological Sciences (online 12 Jan 2026). https://doi.org/10.1093/toxsci/kfag002 - [S36] Kim, H. S., et al. The International Collaborative Animal Study …: the Korean study. Toxicological Sciences (Mar 2026). https://doi.org/10.1093/toxsci/kfag001 - [S37] Melnick, R. L., Moskowitz, J. M., Héroux, P. (ICBE-EMF). Conflicting views in experimental carcinogenesis: a commentary on design and methodological deficiencies in the follow-up validation studies on radiofrequency radiation. Environmental Health (29 July 2026). https://doi.org/10.1186/s12940-026-01324-5

IARC and WHO - [S38] IARC. Non-ionizing Radiation, Part 2: Radiofrequency Electromagnetic Fields. IARC Monographs Vol. 102 (2013); Cancer in Humans chapter searched. https://publications.iarc.who.int/126 ; https://publications.iarc.who.int/download/mono102-007-CancerinHumans.pdf - [S39] IARC. Report of the Advisory Group to Recommend Priorities for the IARC Monographs during 2020–2024 (2019), pp. 148–149. https://monographs.iarc.who.int/wp-content/uploads/2019/10/IARCMonographs-AGReport-Priorities_2020-2024.pdf - [S40] IARC. Report of the Advisory Group to Recommend Priorities for the IARC Monographs during 2025–2029 (2024), agent 099, pp. 171–174. https://monographs.iarc.who.int/wp-content/uploads/2024/11/AGP_Report_2025-2029.pdf ; priorities poster (April 2024): https://monographs.iarc.who.int/wp-content/uploads/2024/04/Priorities-poster-streams-April2024.v5.pdf ; summary: Berrington de González, A., et al. Lancet Oncology (Apr 2024). https://doi.org/10.1016/S1470-2045(24)00208-0 - [S41] IARC Monographs Newsletter, Issue No. 15 (July 2026), upcoming meetings. https://monographs.iarc.who.int/wp-content/uploads/2026/07/Newsletter-IssueNo15-forweb.pdf ; IARC List of Volumes (current, through Vol. 142). https://monographs.iarc.who.int/wp-content/uploads/2018/07/List-of-Volumes.pdf - [S42] WHO. Questions and answers: Radiation: 5G mobile networks and health (undated; live 26 Sep 2026). https://www.who.int/news-room/questions-and-answers/item/radiation-5g-mobile-networks-and-health - [S43] Baan, R., et al. Carcinogenicity of radiofrequency electromagnetic fields. Lancet Oncology 12 (July 2011) (context only). https://doi.org/10.1016/S1470-2045(11)70147-4

Standards, regulation and official guidance - [S44] ICNIRP. Guidelines for limiting exposure to electromagnetic fields (100 kHz to 300 GHz). Health Physics 118(5) (May 2020). https://doi.org/10.1097/HP.0000000000001210 ; PDF: https://www.icnirp.org/cms/upload/publications/ICNIRPrfgdl2020.pdf - [S45] ICNIRP. Commission membership page (accessed 26 Sep 2026). https://www.icnirp.org/en/about-icnirp/commission/index.html - [S46] SCHEER. Final Opinion on the need of a revision of the annexes in Council Recommendation 1999/519/EC and Directive 2013/35/EU, in view of the latest scientific evidence available with regard to radiofrequency (100 kHz–300 GHz). Adopted 18 April 2023. https://health.ec.europa.eu/document/download/aaf66644-cfb2-4f9b-8a73-df7ef2b3a2a4_en?filename=scheer_o_044.pdf ; consultation page: https://health.ec.europa.eu/consultations/scheer-public-consultation-preliminary-opinion-scientific-evidence-radiofrequency_en - [S47] US FDA, Center for Devices and Radiological Health. Review of Published Literature between 2008 and 2018 of Relevance to Radiofrequency Radiation and Cancer (February 2020). https://www.fda.gov/media/135043/download - [S48] Environmental Health Trust v. FCC, No. 20-1025 (D.C. Cir., decided 13 Aug 2021). https://media.cadc.uscourts.gov/opinions/docs/2021/08/20-1025-1910111.pdf - [S49] France. Loi n° 2015-136 du 9 février 2015 relative à la sobriété, à la transparence, à l’information et à la concertation en matière d’exposition aux ondes électromagnétiques. https://www.legifrance.gouv.fr/loda/id/JORFTEXT000030212642/ (read via a fetch tool’s summary; Art. 7 quoted) ; Sénat legislative dossier: https://www.senat.fr/dossier-legislatif/ppl13-310.html - [S50] CTIA – The Wireless Ass’n v. City of Berkeley, 928 F.3d 832 (9th Cir., 2 July 2019). https://www.courtlistener.com/opinion/4636043/ctia-the-wireless-assn-v-city-of-berkeley/ ; certiorari denied, 140 S. Ct. 658 (9 Dec 2019). https://www.courtlistener.com/opinion/9364366/ctia-the-wireless-assn-v-city-of-berkeley/ - [S51] California Department of Public Health. How to Reduce Exposure to Radiofrequency Energy from Cell Phones (December 2017; date from document metadata). https://www.cdph.ca.gov/Programs/CCDPHP/DEODC/EHIB/CDPH%20Document%20Library/Cell-Phone-Guidance.pdf

Exposure, use patterns and research funding - [S52] Persson, T., Törnevik, C., Larsson, L. E., Lovén, J. Output power distributions of terminals in a 3G mobile communication network. Bioelectromagnetics 33 (online Oct 2011; 2012). https://doi.org/10.1002/bem.20710 - [S53] Deltour, I., Guida, F., Ribet, C., Zins, M., Goldberg, M., Schüz, J. Use of mobile phones and radiofrequency-emitting devices in the COSMOS-France cohort. IJERPH (Nov 2024), including the ANSES funding statement. https://doi.org/10.3390/ijerph21111514 - [S54] Regrain, C., et al. Design of an integrated platform for mapping residential exposure to RF-EMF sources. IJERPH (July 2020), with ANSES funding statement EST-2016-2RF-04. https://doi.org/10.3390/ijerph17155339

Courts and compensation - [S55] Microwave News. Italian Supreme Court affirms tumor risk from long-term use of a cell phone (23 Oct 2012). https://microwavenews.com/news-center/italian-supreme-court-affirms-tumor-risk - [S56] Polichetti, A. [Jurisprudence on occupational radiofrequency diseases]. Giornale Italiano di Medicina del Lavoro ed Ergonomia (Dec 2020), PMID 33600662 (abstract). https://europepmc.org/article/MED/33600662 - [S57] Lagorio, S., Vecchia, P. [An Italian Court recognizes the occupational origin of a trigeminal neuroma in a mobile telephone user…]. La Medicina del Lavoro (2011), PMID 21485052 (abstract). https://europepmc.org/article/MED/21485052 - [S58] Phonegate Alert (advocacy organisation). The Court of Appeal of Turin confirms the link between a head tumour and mobile phone use (15 Jan 2020), reporting judgment 904/2019 of 3 Dec 2019, Romeo v. INAIL. https://www.phonegatealert.org/en/the-court-of-appeal-of-turin-confirms-the-link-between-a-head-tumour-and-mobile-phone-use/ (secondary and partisan; the judgment itself was not retrieved) - [S59] Motorola Inc. v. Murray, 147 A.3d 751 (D.C., en banc, 20 Oct 2016). https://www.courtlistener.com/opinion/4313668/motorola-inc-v-michael-patrick-murray/ - [S60] Murray v. Motorola, Inc., No. 23-CV-0700 (D.C. Court of Appeals, 17 July 2025), read via CourtListener index excerpts. https://www.courtlistener.com/opinion/10635088/murray-v-motorola-inc/

Industry conduct and independence - [S61] Hertsgaard, M., Dowie, M. How big wireless made us think that cell phones are safe: a special investigation. The Nation (29 Mar 2018). https://www.thenation.com/article/archive/how-big-wireless-made-us-think-that-cell-phones-are-safe-a-special-investigation/ (investigative journalism; used only where no primary source was available)

Consulted, not relied on for specific claims - Nordhagen, E. K., Flydal, E. Self-referencing authorships behind the ICNIRP 2020 radiation protection guidelines. Reviews on Environmental Health (2022). https://doi.org/10.1515/reveh-2022-0037 - Hardell, L., Carlberg, M. Health risks from radiofrequency radiation, including 5G, should be assessed by experts with no conflicts of interest. Oncology Letters (2020). https://doi.org/10.3892/ol.2020.11876 - Redmayne, M. International policy and advisory response regarding children’s exposure to RF-EMF. Electromagnetic Biology and Medicine (2016). https://doi.org/10.3109/15368378.2015.1038832 - Lin, J. C. Health and safety practices and policies concerning human exposure to RF/microwave radiation. Frontiers in Public Health (2025). https://doi.org/10.3389/fpubh.2025.1619781 - Frank, J. W. Epidemiological criteria for causation applied to human health harms from RF-EMF exposure: Bradford Hill revisited. Frontiers in Public Health (2025). https://doi.org/10.3389/fpubh.2025.1559868