# RF-033 — Prospective cognitive-test association with modeled brain dose

Why this matters
This study followed adolescents and measured changes in memory rather than asking only how much they used phones. A negative association appeared in the subgroup with operator records, while the full-sample estimate was not significant. Developmental function matters; this qualified pattern deserves independent scrutiny.

Id: RF-033

Author: Foerster et al.

Full Authors: Milena Foerster; Arno Thielens; Wout Joseph; Marloes Eeftens; Martin Röösli

Year: 2018

Title: A Prospective Cohort Study of Adolescents' Memory Performance and Individual Brain Dose of Microwave Radiation from Wireless Communication

Journal: Environmental Health Perspectives

Citation: Foerster M, Thielens A, Joseph W, Eeftens M, Röösli M. A Prospective Cohort Study of Adolescents' Memory Performance and Individual Brain Dose of Microwave Radiation from Wireless Communication. Environ Health Perspect. 2018;126(7):077007. doi:10.1289/EHP2427.

Doi: 10.1289/EHP2427

Pmid: 30044230

Topic: Human cognition; adolescents

Verification Group: Original text reviewed

Record Role: Original text reviewed

Featured: True

Exposure: Individually modeled RF brain dose combining device/network use and absorption estimates; operator records calibrate call time; personal measurements/geospatial modeling for far-field sources. Dose is modeled, not directly measured absorbed energy; typical use dominated by GSM calling.

Design: HERMES Swiss school cohort, two sampling waves 2012-2016; 843 followed; 670 completed figural tests, 669 in full adjusted model. Standardized computerized memory testing approximately 1 year apart; regression adjusted for demographic, lifestyle and baseline-score factors.

Finding: Adjusted figural-memory change per IQR brain dose: full sample n=669, -0.22 points (95% CI -0.47 to 0.03; IQR 953 mJ/kg/day), not statistically significant. Operator-data subgroup n=274: -0.26 (-0.42 to -0.10; IQR 341 mJ/kg/day). Right-side users n=532: -0.39 (-0.67 to -0.10).

Boundary: Observational association, residual confounding and exposure-model uncertainty. Full-sample estimate crosses zero. No clinical injury or causal inference established. This expands/reanalyzes the first wave reported by Schoeni 2015 plus newly recruited wave; not wholly independent replication.

Related: Swiss TPH/Basel with Ghent/IMEC dosimetry; separate from Zurich sleep experiments. Schoeni 2015 first-wave DOI 10.1016/j.envint.2015.09.025 is not an independent cohort.

Protection Question: Could developmental cognition be associated with exposure? The operator-data subgroup finding deserves scrutiny alongside the nonsignificant full-sample estimate, cohort overlap and exposure-model uncertainty; it does not establish a harmful dose.

Access: Original indexed body read for methods/results/limitations, primary publisher searchable record and original journal PDF checked for math/funding/COI

Locator: Methods; Table 3 and Figure 2; laterality Figure 3; Discussion 'Comparison with Previous Analysis'; Acknowledgments; PDF first page conflicts

Notice: Scite metadata checked 20 Oct 2026; no editorial notice returned. Primary bibliographic records and targeted notices were checked where retrievable; no retraction or expression of concern located. This is not certification that no notice exists.

Funding: Swiss National Science Foundation project 138190; European Community FP7 GERONIMO grant 603794.

Conflicts: Authors declared no actual or potential competing financial interests.

Source Url: https://doi.org/10.1289/EHP2427

Retrieval Links: https://doi.org/10.1289/EHP2427 | https://pubmed.ncbi.nlm.nih.gov/30044230/ | https://www.emmind.net/openpapers_repos/Applied_Fields-Hazads/Microwave_Effects/MW_Various/2018_A_Prospective_Cohort_Study_of_Adolescents%E2%80%99_Memory_Performance_and_Individual_Brain_Dose_of_Microwave_Radiation_from_Wireless_Communication.pdf

Verification Date: 2026-10-20