Non‑Thermal EMF Harm Signals (Moderate Evidence): Reproductive DNA Damage, Pregnancy Risk, Tumor Relevance, and Ecological Disruption

Executive Summary

This thread (Harm • Moderate evidence • All years) contains 13 studies spanning RF-EMF, ELF/power-line fields, and real-world wireless use proxies. The pattern is not “one disputed endpoint.” It is multi-domain biological disruption—including DNA integrity in sperm, pregnancy loss and growth outcomes, tumor translational relevance, cardiovascular risk signals, and ecosystem-level impacts on pollination.

Bottom line for policy: when adverse biological effects are reported without relying on heating—and when the exposures are common in daily life—a safety framework that only prevents thermal injury is incomplete. This report synthesizes why that matters and why precaution is warranted.

High-impact findings in this packet

• Reproductive DNA damage under 5G-relevant RF conditions: Butković et al. (2024) found significantly increased sperm DNA fragmentation after 2 hours of exposure at 700/2500/3500 MHz (10 V/m) in vitro.
• Pregnancy risk signal with call-time dose pattern: The Yazd cohort study (2025) associated longer cell-phone call duration with higher miscarriage risk and abnormal infant growth metrics.
• Tumor biology relevance from lifetime RFR bioassay samples: Brooks et al. (2024) reported that mutations in rat RFR-associated tumors show partial overlap with human cancer gene alterations, supporting translational relevance.
• Ecological harm: Molina-Montenegro et al. (2023) linked power-line EMF to honeybee stress responses and reduced pollination/seed set.
• Large-cohort cardiovascular signal: UK Biobank (2023) reported a dose–response between weekly call time and incident hypertension.

What This Report Does — and Does Not — Claim

What it does

• Synthesizes evidence that biological effects occur below, and outside the logic of, thermal injury.
• Connects mechanistic plausibility (e.g., oxidative stress/ion-channel disruption) to reproductive, developmental, cardiovascular, tumor, and ecological endpoints.
• Explains why thermal-only RF guidelines are not designed to detect or prevent these effects.

What it does not

• It does not require definitive proof of disease-specific human causation for every endpoint before recommending precaution.
• It does not treat “regulatory compliance” as proof of biological safety.
• It does not mislabel behavioral/screen-time studies as RF-dosimetry studies; where exposure is primarily behavioral/light-mediated, that is stated explicitly.

Why Thermal-Only Standards Are Inadequate

Thermal-only standards assume that if RF exposure does not measurably heat tissue, it is biologically inert. The studies in this thread challenge that assumption in multiple ways:

• DNA integrity and oxidative stress pathways can be altered without requiring bulk heating—especially in highly sensitive cells (e.g., sperm) and during developmental windows.
• Signal characteristics (frequency, modulation/pulsing, intermittency) can matter biologically even when average power is low—consistent with non-linear and non-monotonic responses in living systems.
• Population vulnerability is not uniform: pregnancy, fertility, and genetic susceptibility (as suggested in UK Biobank 2023) imply that a single “safe for all” thermal threshold is a poor public-health model.

Evidence of Non-Thermal Biological Effects (by evidence cluster)

1) Reproductive and genetic integrity signals

Butković et al. (2024) exposed boar semen in vitro to continuous 5G-band RF (700, 2500, 3500 MHz) at 10 V/m for 2 hours and reported significantly increased sperm DNA fragmentation across frequencies, with notable effects at 2500 MHz and 3500 MHz. This is a direct harm signal because DNA fragmentation is not a “comfort symptom”—it is a biologically consequential endpoint relevant to fertility and potential downstream developmental risk.

Why this matters for standards:

• Thermal limits are not designed around gamete DNA integrity.
• Reproductive cells can be vulnerable to oxidative and membrane-mediated stress at exposures that do not produce meaningful heating.

2) Pregnancy and developmental outcomes

The Yazd women cohort (2025) followed 1,666 pregnancies and reported that greater cell-phone call duration during pregnancy was associated with higher miscarriage risk (reported as an incremental RR per minute) and with abnormal birth weight and abnormal infant height.

This is policy-relevant even with imperfect exposure measurement because:

• Real-world public health decisions often rely on usage proxies when dosimetry is unavailable.
• Pregnancy is a high-stakes vulnerability window; modest risk shifts can translate into substantial population burden.

3) Tumor relevance and translational support from lifetime exposure bioassay samples

Brooks et al. (2024) performed targeted next-generation sequencing on rat gliomas and cardiac schwannomas arising in a lifetime low-dose RFR bioassay context and reported that a subset of tumor variants overlapped with homologous alterations in human cancer genes (COSMIC), while also noting differences (e.g., IDH hotspot-homologous mutations largely absent).

This paper does not re-litigate incidence; it strengthens the argument that:

• RFR-associated tumors can show cancer-relevant molecular alterations, supporting biological plausibility and translational concern.
• Dismissing animal tumor findings as “non-human” is scientifically lazy when molecular features show partial convergence.

4) Cardiovascular risk signal in a large prospective cohort

UK Biobank (2023) (212,046 participants; median 12 years) reported a small but significant higher risk of incident hypertension among mobile phone call users, and a dose–response with weekly call time (highest category >6 h/week showing higher risk than minimal call time). The study also reported higher risk in those with high genetic susceptibility plus longer call time.

Why this matters:

• A dose–response pattern in a large cohort is a meaningful harm signal even when exposure is self-reported.
• Genetic susceptibility is consistent with a biologically literate view of risk: not everyone responds the same way, which undermines one-size-fits-all thermal thresholds.

5) Environmental/ecological disruption (ELF/power-line EMF)

Molina-Montenegro et al. (2023) combined field observations near high-voltage towers with experimental manipulation and reported:

• Stress-related molecular signatures in honeybees (heat-shock proteins; antioxidant pathway genes)
• Reduced pollinator visitation and reduced seed production in a honeybee-pollinated plant near EMF

This matters because it demonstrates that EMF can:

• Act as a biological stressor in non-human organisms
• Scale up to ecosystem service disruption (pollination), which is a public-interest harm category often ignored in human-only safety frameworks.

6) Symptom clusters near base stations (exposure proxied by proximity)

J Biomed Phys Eng (2026) analyzed 699 adults living near base stations and used machine-learning models to predict symptom patterns, with distance to base station among key predictors. While model performance is not proof of causation, the study reflects a recurring real-world issue: people report symptom clusters in proximity to RF infrastructure, and current standards offer little in the way of biologically grounded investigation pathways.

Mechanistic Plausibility (why non-thermal effects are scientifically coherent)

Panagopoulos et al. (2025) (mechanistic narrative review) argues that wireless communication fields—described as microwave carriers with ELF/ULF modulation/pulsing—can plausibly drive biological effects via voltage-gated ion channel (VGIC) dysregulation and downstream reactive oxygen species (ROS)/oxidative stress, with potential indirect DNA damage.

Even as a review (not new data), it is important because it:

• Provides a coherent framework linking signal properties to cellular stress pathways.
• Aligns with endpoints in this packet where oxidative stress and DNA integrity are central concerns (e.g., sperm DNA fragmentation).

Smartphone “Use” Studies: Important Harms, but Not RF-Dosimetry Evidence

Several studies here address harms associated with smartphone use:

• Sleep impairment and next-day performance in an RCT crossover in elite athletes (2026)
• Sleep quality associations with smartphone addiction (2026)
• Working memory and altered fNIRS activation/connectivity with high screen time (2025)
• Disengagement cycles in a daily diary design (2026)

These are legitimate public-health harms, but they are primarily driven by behavioral factors, screen light, arousal, and time displacement, not quantified RF exposure. They should be used to support precaution in device habits, while RF policy arguments should lean more heavily on the reproductive, developmental, mechanistic, animal, and infrastructure-related evidence.

Regulatory Failure and Policy Implications

A thermal-only compliance model fails the public in at least four ways highlighted by this thread:

• It ignores endpoints that are not heat injuries (DNA fragmentation, oxidative stress signaling, developmental vulnerability).
• It underweights reproductive and prenatal protection, where the cost of being wrong is generational.
• It treats variability as noise, despite evidence consistent with susceptibility (genetics, developmental windows, signal-specific sensitivity).
• It is not ecologically literate, despite evidence that EMF can disrupt pollinators and plant reproduction.

Precautionary Principle: Who needs protection most?

• Children and adolescents: longer lifetime exposure, developing nervous systems, and behavior patterns that increase cumulative exposure.
• Pregnancy: early developmental windows can be uniquely sensitive; the Yazd cohort (2025) underscores why “wait for certainty” is not an ethical default.
• Fertility and future generations: sperm DNA integrity findings (Butković et al., 2024) elevate the stakes beyond transient symptoms.

Conclusion

Across this moderate-evidence harm thread, the scientific record shows biological effects that are not captured by a heating-only paradigm—including sperm DNA damage under 5G-relevant RF exposure conditions, pregnancy loss and growth associations with call-time, tumor molecular relevance in lifetime RFR bioassay samples, dose–response cardiovascular risk signals, and ecological disruption of pollination.

The policy conclusion is straightforward: if standards only prevent heating, they are not designed to prevent the kinds of biological effects repeatedly reported here. That is not a reason for complacency—it is a reason to modernize RF health policy around biology, vulnerability, and precaution.

---

Footnotes (full study links)

1. Problematic smartphone use and disengagement in first-year college students (2026). https://pubmed.ncbi.nlm.nih.gov/41633115/

2. Smartphone addiction and sleep quality among college students (2026). https://pubmed.ncbi.nlm.nih.gov/41678505/

3. Evening smartphone exposure impairs sleep and performance (2026). https://pubmed.ncbi.nlm.nih.gov/41668954/

4. Decision support system for symptoms near base stations (2026). https://jbpe.sums.ac.ir/article_49803.html

5. Smartphone use and working memory (fNIRS) (2025). https://pubmed.ncbi.nlm.nih.gov/41668776/

6. EMW exposure and pregnancy/birth outcomes in Yazd women (2025). https://bmcpregnancychildbirth.biomedcentral.com/articles/10.1186/s12884-025-07512-4

7. Mechanism review: ELF & wireless EMF biological effects (Panagopoulos et al., 2025). https://pubmed.ncbi.nlm.nih.gov/40547468/

8. Air pollution exposure and birth weight in ECHO cohort (2025) (thread mismatch: not EMF). https://pubmed.ncbi.nlm.nih.gov/41632155/

9. Genetic profiling of rat gliomas and schwannomas from lifetime RFR study (Brooks et al., 2024). https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0296699

10. Personal light exposure patterns and incident type 2 diabetes (Windred et al., 2024) (thread mismatch: light, not EMF). https://www.thelancet.com/pdfs/journals/lanepe/PIIS2666-7762(24)00110-8.pdf

11. 5G RF effects on boar semen DNA integrity (Butković et al., 2024). https://pubmed.ncbi.nlm.nih.gov/39342826/

12. EMF disrupt pollination service by honeybees (Molina-Montenegro et al., 2023). https://www.science.org/doi/full/10.1126/sciadv.adh1455

13. Mobile phone calls and new-onset hypertension (UK Biobank) (2023). https://academic.oup.com/ehjdh/advance-article/doi/10.1093/ehjdh/ztad024/7131479

Included studies

• [Problematic smartphone use and disengagement in first-year college students: A daily diary study of between- and within-person differences. (2026)](/mel/paper.php?id=4333)
• [The influence of smartphone addiction on sleep quality among college students: The parallel mediating roles of perceived stress and health-promoting lifestyle. (2026)](/mel/paper.php?id=4329)
• [Evening smartphone exposure impairs sleep quality and next-day performance in elite soccer players: a randomized controlled trial. (2026)](/mel/paper.php?id=4322)
• [A Decision Support System for Managing Health Symptoms of Living Near Mobile Phone Base Stations (2026)](/mel/paper.php?id=2690)
• [The impact of smartphone use on working memory in college students: a functional near-infrared spectroscopy study. (2025)](/mel/paper.php?id=4323)
• [The association of widely used electromagnetic waves exposure and pregnancy and birth outcomes in Yazd women: a cohort study (2025)](/mel/paper.php?id=2660)
• [A comprehensive mechanism of biological and health effects of anthropogenic extremely low frequency and wireless communication electromagnetic fields (2025)](/mel/paper.php?id=2627)
• [Air Pollution Exposure and Birth Weight in the ECHO Cohort. (2025)](/mel/paper.php?id=17)
• [Genetic profiling of rat gliomas and cardiac schwannomas from life-time radiofrequency radiation exposure study using a targeted next-generation sequencing gene panel (2024)](/mel/paper.php?id=237)
• [Personal light exposure patterns and incidence of type 2 diabetes: analysis of 13 million hours of light (2024)](/mel/paper.php?id=141)
• [Effects of 5G radiofrequency electromagnetic radiation on indicators of vitality and DNA integrity of in vitro exposed boar semen (2024)](/mel/paper.php?id=78)
• [Electromagnetic fields disrupt the pollination service by honeybees (2023)](/mel/paper.php?id=477)
• [Mobile phone calls, genetic susceptibility, and new-onset hypertension: results from 212 046 UK Biobank participants (2023)](/mel/paper.php?id=461)