QuantaDose Press Releases

RFK Jr. Was Right to Remove FDA’s Blanket “Cell Phone Radiation Is Safe” Assurances

Why removing categorical FDA safety assurances was the scientifically responsible move: high‑certainty animal cancer evidence, non‑thermal biological effects, and a court‑recognized failure of thermal‑only assumptions.

RFK Jr. Was Right to Remove FDA’s Blanket Assurances on Cell Phone Radiation Safety

In January 2026, the U.S. Department of Health and Human Services (HHS), led by Secretary Robert F. Kennedy Jr., confirmed it would initiate a new study on cellphone radiation and health. In the same reporting window, HHS also explained that the Food and Drug Administration removed webpages that had presented older, categorical assurances that cellphones are “not dangerous,” specifically so that updated research into electromagnetic radiation and health could proceed without being undercut by outdated certainty statements.

That removal was not “anti‑science.” It was the opposite: a correction to public‑facing messaging that had drifted beyond what the evidence can honestly support.

This RF Safe editorial explains why.

1) The central issue is not “human causation” — it is whether blanket assurances are scientifically defensible

Blanket safety assurances implicitly claim that the existing exposure limits are biologically protective across realistic conditions of use, across vulnerable populations (including children), and across relevant endpoints (including cancer, neurological effects, reproductive effects, vascular effects, and more).

That is a much stronger claim than “the evidence is complicated” or “there is no definitive proof of harm in humans.”

Even the U.S. Court of Appeals for the D.C. Circuit has addressed this mismatch between conclusory reassurance and reasoned scientific explanation. In Environmental Health Trust v. FCC (2021), the court criticized FDA‑relied‑upon statements as conclusory when they asserted that exposure below current limits does not cause harmful health effects without articulating the factual bases. The court specifically flagged the possibility that a major predicate of the thermal‑only framework — the asserted non‑existence of non‑thermal biological effects — “may no longer be accurate.”

The court did not “prove harm.” It did something more fundamental for public health: it recognized that categorical reassurance is not the same thing as scientific accountability.

Removing categorical assurances aligns public messaging with that legal and scientific reality.

2) Precautionary public‑health standards are built on animal toxicology — because waiting for human proof turns people into the experiment

Across environmental health and toxicology, animal carcinogenicity bioassays have long been used to identify hazards before irreversible outcomes accumulate in humans. That is not controversial; it is foundational.

The World Health Organization’s commissioned animal‑cancer systematic review in Environment International (2025) explicitly situates chronic animal cancer bioassays as “commonly used to identify potential human carcinogens,” and it elevates those bioassays as the most directly applicable evidence stream for carcinogenicity assessment.

This is precisely why categorical consumer‑facing assurances (“cellphones are not dangerous”) were scientifically inappropriate: the most precaution‑relevant evidence stream has strengthened, not weakened.

3) WHO‑commissioned 2025 animal cancer review: “High certainty” signals for the two tumor types that matter most

The 2025 WHO‑commissioned systematic review of RF‑EMF exposure and cancer in laboratory animals (Mevissen et al., Environment International) is not a blog post. It is a structured evidence assessment that applied risk‑of‑bias methods and graded certainty using GRADE‑style approaches refined for toxicology.

Its core finding is straightforward:

  • High certainty of evidence for increased risk of glioma in animals (brain tumors derived from glial cells).

  • High certainty of evidence for increased risk of heart schwannomas in male rats.

Importantly, the review also highlights why simplistic “dose → heat → harm” thinking does not resolve RF risk:

It explicitly raises uncertainty about whether a monotonic dose‑response holds, whether SAR is even the right dose metric for RF‑EMF adverse effects, and which exposure metric best translates risk (whole‑body vs localized, intensity vs cumulative exposure).

That is not an argument for complacency; it is an argument for retiring blanket assurances.

4) Convergence matters: NTP and Ramazzini observed the same tumor type — across vastly different exposure regimes

When two large rodent programs report rare tumor findings in the same tissue type, the correct scientific response is not to wave it away as an artifact. The correct response is to stop telling the public “safe,” and to investigate.

The U.S. National Toxicology Program (NTP)

NTP’s two‑year rat study reported malignant schwannomas in the heart with no malignant heart schwannomas in sham controls, and malignancies appearing across exposed male groups (and at least one exposed female group), along with related preneoplastic changes.

For the GSM‑modulated exposure condition, the NTP’s own tabulated data show 0% in male sham controls and malignant heart schwannomas observed in exposed groups (with a statistically significant trend test in that table).

The Ramazzini Institute (Italy) far‑field “base station‑like” exposures

ICNIRP’s own evaluation summarizes the Ramazzini exposure levels as whole‑body SARs of 0.001, 0.03, and 0.1 W/kg, with a statistically significant increase in heart schwannomas in male rats reported at the highest level (0.1 W/kg).

Why this convergence directly undermines “it’s only high dose” dismissals

A common dismissal is that NTP is “high dose.” That framing misses the more important scientific fact: the same tumor type appears in a second program at orders‑of‑magnitude lower whole‑body SAR.

When a hazard signal shows up across distinct exposure regimes, the narrative that “only heating matters” becomes increasingly indefensible as a public‑facing assurance.

5) “Non‑thermal effects” are not fringe — even a federal court record acknowledges the premise is eroding

The words “thermal” and “non‑thermal” are often used rhetorically. The more useful framing is this:

  • Thermal effects are harms caused by heating tissue.

  • Non‑thermal biological effects are changes in biological systems that occur at exposure levels not explained by meaningful tissue heating.

The D.C. Circuit opinion lays out that exposure to lower levels of RF radiation might cause “non‑thermal” biological effects, notes the debate, and — most consequentially for policy — states that petitioners’ evidence implies the factual premise of non‑existence of non‑thermal effects underlying RF guidelines “may no longer be accurate.”

That is exactly why categorical public assurances are misleading. They treat a contested and eroding premise as settled.

6) A real‑time, testable human physiology signal exists — without requiring a “smoking gun” cancer claim

RF safety debates often get trapped in a false binary: either “proven human cancer” or “safe.”

But biology does not work that way. Sub‑clinical physiological disruptions can be measurable, repeatable, and relevant — even when downstream population‑level disease signals are difficult to isolate in noisy real‑world epidemiology.

A 2025 peer‑reviewed Frontiers/PMC‑archived paper reported ultrasound imaging of a healthy volunteer’s popliteal vein before and after 5 minutes of direct skin exposure to an active smartphone (idle but with antennas on). Post‑exposure imaging showed a dramatic change in the vessel’s acoustic appearance consistent with rouleaux formation (red blood cell aggregation) and sluggish flow, with follow‑up imaging showing persistence and a later repeat session suggesting reproducibility in that same subject.

This is not, by itself, a population‑level risk estimate. It is something different — and crucial:

  • It is a plausible example of a non‑thermal biological effect observable in real time.

  • It illustrates that the question “does RF do anything biologically at non‑thermal levels?” is not hypothetical; it is experimentally testable.

Once non‑thermal biological effects are plausibly observable, categorical “safe” assurances are not scientifically responsible.

7) Why studies often cited as “reassuring” cannot justify blanket safety messaging

This RF Safe analysis is not about litigating every epidemiology paper. The point is narrower and more important:

Even if epidemiology were perfectly null, it still would not justify categorical assurance when animal cancer bioassays show high‑certainty signals and mechanisms/non‑thermal endpoints remain unresolved.

That said, a major reason blanket assurances are inappropriate is that “null” human evidence is not the same as “proof of safety.” Well‑known design limitations can dilute real effects:

The Danish subscription‑cohort problem: misclassification that dilutes risk

Swerdlow’s 2011 commentary highlights that subscription‑based cohort studies have a weakness not present in interview case‑control studies: subscription data exclude corporate subscriptions and the named subscriber may not be the true user — both of which can dilute a true association.

Consistent with that, commentary around the Danish cohort points out that corporate subscribers were excluded from the “exposed” group in cohort construction, pushing likely heavier users out of the exposure category and into comparison categories.

A study design that can misclassify heavy users as “unexposed” cannot be used as the basis for telling the public “safe.”

INTERPHONE: the largest case‑control study did not equal “no signal”

IARC’s own INTERPHONE press release reports that in the highest decile of recalled cumulative call time (≥1,640 hours), the odds ratio for glioma was 1.40 (95% CI 1.03–1.89), while also noting issues such as implausible self‑reported values.

The key policy implication is not that INTERPHONE “proves” causation. It is that even the flagship human study did not justify categorical “no danger” messaging — especially when today’s exposures and use patterns differ substantially from the study era.

Biases can push results toward “no effect” — even when effects exist

Modern analyses of recall and selection biases in INTERPHONE‑like designs show how major biases can affect risk estimation and interpretation.

Again: this is not an argument that epidemiology “proves” harm. It is an argument that epidemiology cannot ethically be weaponized as a safety guarantee.

8) Nonlinear and non‑monotonic behavior is not a rhetorical escape hatch — it is explicitly part of the modern scientific uncertainty

The “nonlinear exposure‑response” point is not a flourish. It is directly acknowledged as an unresolved issue in the WHO‑commissioned animal cancer review, which explicitly raises whether a monotonic dose‑response holds and whether SAR is the right metric.

That one sentence alone is enough to disqualify blanket, consumer‑facing certainty statements. A system that might not behave monotonically cannot be safely summarized as “below limit = safe,” because the meaning of “dose” and “threshold” is precisely what is in question.

9) Why removing the assurances was the correct, evidence‑based move

With the evidence landscape above, the decision to remove categorical assurances can be defended on straightforward scientific‑integrity grounds:

  1. High‑certainty animal carcinogenicity signals exist for glioma and heart schwannoma in rodents per the WHO‑commissioned review.

  2. Two major animal programs converge on the same tumor type (heart schwannoma), and the exposure regimes differ by orders of magnitude.

  3. A key factual predicate of current RF guidelines is under dispute — including in federal court language referencing non‑thermal biological effects.

  4. Human physiology endpoints can be observed on short timescales in at least preliminary ultrasound work, underscoring that “non‑thermal effects” are testable rather than imaginary.

  5. Null or mixed epidemiology cannot be treated as proof of safety given exposure misclassification and other limitations that dilute associations.

  6. HHS itself framed the FDA removals as clearing out old conclusions while updated research proceeds. That is the correct posture for public agencies when the evidence is evolving and the prior messaging was categorical.

In short: removing the assurances was not an overreaction. It was a necessary correction — because there is no scientific basis for absolute, blanket reassurance.

10) What public health communication should say instead of “safe”

A scientifically honest public‑facing statement would look more like this:

  • Current U.S. exposure limits were developed primarily to prevent thermal injury.

  • Evidence about non‑thermal biological effects and long‑term endpoints continues to evolve.

  • Animal cancer bioassays and systematic reviews warrant ongoing evaluation and precaution.

  • Practical exposure‑reduction steps are reasonable, especially for children, without requiring panic or abandonment of modern communications.

That kind of language is the opposite of alarmism. It is responsible risk communication.

11) Practical exposure‑reduction steps that do not require fear — only competence

RF Safe readers looking for low‑controversy, low‑cost risk reduction can focus on distance, duration, and wired substitution:

  • Use speaker mode or wired audio rather than holding a phone to the head.

  • Avoid carrying an active phone directly against the body for long periods.

  • Prefer wired connections for fixed devices when feasible.

  • Keep high‑duty wireless transmitters (routers/access points) away from sleeping areas where practical.

These are not “extreme” steps; they are basic hygiene measures in an evidence‑uncertain environment.

FAQ

Did the FDA “admit” cellphones are dangerous?

Public reporting indicates FDA webpages with older, categorical conclusions were removed while HHS undertakes updated research. That is best interpreted as a correction in messaging certainty, not a single definitive admission of harm.

Why isn’t “no proven harm in humans” the end of the discussion?

Because precautionary standards are designed to prevent harm before definitive human outcomes accumulate, and because epidemiology has known exposure‑measurement limitations.

What is the strongest scientific reason categorical assurances were inappropriate?

The WHO‑commissioned animal cancer systematic review graded certainty as high for glioma and male‑rat heart schwannoma, and it explicitly raised whether monotonic dose‑response and SAR‑based metrics are even valid for RF risk extrapolation.

Primary keyword: cell phone radiation safety
Secondary keywords: RFK Jr HHS, FDA removed cell phone safety pages, non‑thermal RF effects, NTP cell phone study, Ramazzini Institute study, WHO 2025 RF animal cancer review, FCC exposure limits, SAR guidelines