The Law Says “Shall.” The Science Says “Act.”

Restart NTP’s RF research and modernize U.S. RF‑safety standards—now.

Regulatory delay is not scientific uncertainty. Over the last 18 months, the weight of evidence on radio‑frequency electromagnetic fields (RF‑EMF) has shifted, while U.S. rules remain stuck in the 1990s. The result is a widening gap between what the science now signals and what our institutions still say. That gap puts families at risk—and it violates federal law.

What the newest science says (2024–2025)

• Animal cancers (WHO‑commissioned, 2025):
  A World Health Organization project published in Environment International finds high‑certainty evidence that RF‑EMF exposure increases malignant heart schwannomas and gliomas in male rats. Benchmark‑dose modeling reported BMD = 0.177 W/kg and 1.92 W/kg for heart schwannomas (two studies) and BMD = 4.25 W/kg for gliomas; the authors also note these positives did not show a monotonic dose‑response versus sham—something that must be understood mechanistically, not waved away. This is a hazard signal at the core of cancer risk assessment. PubMed

• Male fertility (WHO‑commissioned, 2024 + 2025 corrigendum):
  The WHO review of experimental mammal studies originally graded reduced pregnancy rate as moderate‑certainty. In 2025, a corrigendum revised the pooled effect to OR = 1.68 (95% CI 1.06–2.65) and explicitly described a detrimental effect. That’s stronger language, and it matters for policy. PubMed+2PubMed+2

• Human pregnancy outcomes (2025 cohort):
  A prospective cohort in Yazd (BMC Pregnancy & Childbirth, n = 1,666) reports that longer cell‑phone call duration during pregnancy correlates with higher risks of miscarriage and abnormal infant weight and length after adjustment. Associations per time increment are small but directionally consistent with animal signals on reproduction. BioMed Central+1

• Mechanism (why non‑thermal exposure matters):
  A 2025 mechanistic review (Frontiers in Public Health) details how pulsed/modulated RF can perturb voltage‑gated ion channels, dysregulate intracellular Ca²⁺, and drive oxidative stress—biologically plausible pathways at power levels below heating thresholds. This is exactly where 21st‑century wireless operates. PMC+1

Note: Germany’s radiation office (BfS) has urged caution about calling the animal‑cancer evidence “high‑certainty,” arguing the review’s integration may overweight a few positive studies. That’s a policy debate, not a negation of the WHO review’s own conclusion. Either way, no responsible reading of 2025 can say “no evidence.” Doris

What our regulators still haven’t done

• The FCC still relies on exposure limits rooted in the mid‑1990s (SAR 1.6 W/kg peak for the general public). In 2021, the D.C. Circuit ruled the FCC’s 2019 decision to keep those limits “arbitrary and capricious” and remanded for a reasoned explanation on non‑cancer harms. Four years later, the public is still waiting. eCFR+2Federal Communications Commission+2

• NTP’s RF program was shut down on February 2, 2024, stopping the most important U.S. government effort to clarify mechanisms and replicate cancer findings. That decision left a vacuum right when hazard signals were strengthening. Microwave News+1

The law isn’t a suggestion: Public Law 90‑602 (1968) uses “shall.”

Congress wrote this in plain English. Under the Radiation Control for Health and Safety Act, the HHS Secretary “shall establish and carry out an electronic product radiation control program,” including (1) developing and administering performance standards and (2) planning, conducting, coordinating, and supporting research to minimize harmful exposures. Shall means must. Congress.gov

Today, the HHS Secretary is Robert F. Kennedy Jr. That’s not speculation; it’s the official record. And he personally has history here—the D.C. Circuit’s 2021 Environmental Health Trust v. FCC remand followed litigation that included plaintiffs he has worked with. He now holds the legal pen the statute envisioned. HHS.gov+1

What “doing the job” looks like—practical steps HHS can take this year

1. Restart and expand NTP’s RF research immediately.
   Fund replications of NTP and Ramazzini cancer bioassays; prioritize prenatal and early‑life exposures; include real‑world pulsed/modulated signals; pre‑register protocols; publish all raw data. Microwave News+1

2. Stand up an HHS‑led RF Hazard Program under PL 90‑602.
   Use the Act’s authority to develop and administer performance standards for electronic products that emit RF (that includes consumer devices), focusing on duty‑cycle, peak modulation, body‑contact and sleep‑proximity controls—while coordinating with FCC on transmitter rules. The statute explicitly instructs the Secretary to plan and support research and to develop standards. Congress.gov

3. Coordinate with FCC on an updated national exposure framework.
   The D.C. Circuit remand is still open. HHS should supply the health science record the court said was missing, so FCC can modernize 47 CFR § 1.1310 (or HHS can act where FCC cannot—on product performance). Update OET‑65 guidance at the same time to reflect contemporary science and use patterns. Federal Communications Commission+2eCFR+2

4. Issue precautionary guidance for pregnancy and childhood settings now.
   On the strength of the Yazd cohort and the animal fertility evidence, advise shorter call duration, device‑to‑body distance, night‑time airplane mode, and router placement outside bedrooms—pending full standards revision. Low cost, high upside. BioMed Central+1

5. Commit to full transparency and conflict‑of‑interest firewalls.
   Public registries, open data, and explicit COI policies are the antidote to “industry said / industry paid” arguments—whichever industry is in the room.

Why this is not “alarmism”—it’s alignment

• Hazard signals in animals (two tumor types, high‑certainty per WHO project) + reproductive detriment in experimental mammals (OR ≈ 1.68 after corrigendum) + human pregnancy cohort associations + plausible non‑thermal mechanisms ≠ “no evidence.” It equals “Act now while you keep learning.” PMC+4PubMed+4PubMed+4

• IARC has treated RF‑EMF as Group 2B (“possibly carcinogenic”) since 2011 and is prioritizing re‑evaluation in 2025–2029. If anything, momentum is toward stronger hazard classification, not weaker. IARC+1

• Regulatory stagnation (1990s limits, 2021 remand unaddressed, a shuttered NTP program) is a process failure, not proof of safety. The law anticipated this: that’s why it says “shall.” Justia+1

A plain‑language promise to the public

Until the federal standards catch up, here’s what families can do today (no drama, just physics): keep devices off the body, prefer speaker or wired audio, avoid long calls in low‑signal areas, airplane mode near the pillow, put routers away from sleeping spaces. These steps reduce exposure without changing your life. They’re good practice, period.

Pull‑quote you can use anywhere

When Congress wrote “shall,” it meant it. HHS must restart RF research and help set modern safety standards—because families can’t live inside 1990s limits while 2025 science says act. Congress.gov+1

Key sources (recent & load‑bearing)

• WHO‑commissioned animal‑cancer review (2025)—high‑certainty evidence for heart schwannomas and gliomas; BMDs include 0.177 W/kg and 4.25 W/kg; Environ. Int. (open abstract). PubMed

• Male fertility review (WHO project)—2025 corrigendum: pooled OR = 1.68 detrimental effect on pregnancy rate. PubMed+1

• Human cohort (2025, Yazd)—longer call duration in pregnancy linked to miscarriage and abnormal birth measures. BioMed Central

• Mechanism (2025)—pulsed/modulated RF → ion channels/Ca²⁺ → oxidative stress (non‑thermal). PMC

• NTP program termination (2024)—reporting and commentary. Microwave News+1

• Legal duty (PL 90‑602)—HHS “shall” run an electronic product radiation control program incl. standards and research. Congress.gov

• FCC remand (2021)—court found 2019 decision to keep 1996 limits arbitrary and capricious re non‑cancer effects. Federal Communications Commission

• Context note—BfS critique of Mevissen approach (method‑integration debate). Doris