RF Safe

Relationship between radiofrequency radiation from cell phones and brain tumor incidence trends in South Korea

Environmental research · 2023 · Mixed · Low evidence

← Research stream
MixedLow evidencePaper 516

Using nationwide South Korean data, the study reports statistically significant positive correlations between cell phone subscription rate (lagged by 10 years) and ASIR for three benign brain tumor categories (D32, D33, D32.0) and three malignant categories (…

Study snapshot

PublicationEnvironmental research · 2023
Study typeecological
Reported directionMixed
Evidence ratingLow

Abstract

Relationship between radiofrequency radiation from cell phones and brain tumor incidence trends in South Korea Moon J. The relationship between radiofrequency-electromagnetic radiation from cell phones and brain tumor: The brain tumor incidence trends in South Korea. Environmental Research (2023). doi: 10.1016/j.envres.2023.115657. Highlights • Positive correlation for malignant neoplasm of cerebrum, except lobes and ventricles/the frontal lobe/the temporal lobe. • Positive correlation for benign neoplasm of meninges/brain and other parts of the central nervous system/supratentorial brain. • The highest correlation coefficient: malignant neoplasm of the frontal lobe/the temporal lobe. • Cell phones: RF-EMR exposure through the frontotemporal side of the brain. • Four reasons for statistically insignificant results of recent studies. Abstract Introduction The aim of this study is to investigate the relationship between the nationwide cell phone subscription rate and the nationwide incidence of brain tumors in South Korea. The nationwide cell phone subscription rate was used as a proxy for the RF-EMR exposure assessment. Methods The data for cell phone subscriptions per 100 persons from 1985 to 2019 were found in the Statistics, International Telecom Union (ITU). The brain tumor incidence data from 1999 to 2018 provided by the South Korea Central Cancer Registry operated by the National Cancer Center were used. Results In South Korea, the subscription rate increased from 0 per 100 persons in 1991 to 57 per 100 persons in 2000. The subscription rate became 97 per 100 persons in 2009 and 135 per 100 persons in 2019. For the correlation coefficient between cell phone subscription rate before 10 years and ASIR [Age-Standardized Incidence Rate] per 100,000, a positive correlation coefficient with a statistical significance was reported in 3 benign brain tumors (International Classification of Diseases, ICD-10 code, D32, D33, and D32.0) and in 3 malignant brain tumors (ICD-10 code, C71.0, C71.1, and C71.2). Positive correlation coefficients with a statistical significance in malignant brain tumors ranged from 0.75 (95% CI 0.46–0.90) for C71.0 to 0.85 (95% CI 0.63–0.93) for C71.1. Discussion In consideration of the fact that the main route for RF-EMR exposure has been through the frontotemporal side of the brain (the location of both ears), the positive correlation coefficient with a statistical significance in the frontal lobe (C71.1) and temporal lobe (C71.2) can be understood. Statistically insignificant results from recent cohort and large population international studies and contrasting results from many previous case-control studies could indicate a difficulty in identifying a factor as a determinant of a disease in ecological study design. pubmed.ncbi.nlm.nih.gov Excerpts Numerous systematic reviews and meta-analyses on the association between Radiofrequency-Electromagnetic Radiation (RF-EMR) exposure from cell phones and the incidence of brain tumors have been published until the present time (Belpomme et al., 2018; Elwood, 2003; Morgan et al., 2015; Myung et al., 2009; Wang et al., 2018). The conclusion of each study varied depending on the magnitude of exposure, the location of brain tumors, the histologic type of brain tumors, and a number of confounding factors such as research team and funding source (Belpomme et al., 2018; Hardell et al., 2008a; Morgan et al., 2015; Wang et al., 2018).... C70 Malignant neoplasm of meninges C71 Malignant neoplasm of brain C72 Malignant neoplasm of the spinal cord, cranial nerves, and other parts of the central nervous system C70.0 Malignant neoplasm of cerebral meninges C70.9 Malignant neoplasm of meninges, unspecified C71.0 Malignant neoplasm of cerebrum, except lobes and ventricles C71.1 Malignant neoplasm of the frontal lobe C71.2 Malignant neoplasm of the temporal lobe C71.9 Malignant neoplasm of brain, unspecified C72.0 Malignant neoplasm of spinal cord D32 Benign neoplasm of meninges D33 Benign neoplasm of brain and other parts of the central nervous system D32.0 Benign neoplasm of brain The AAPC [Average Annual Percentage Change] for C72, C71.0, C71.1, C71.2, C72.0, D32, D33, and D32.0 was positive with statistical significance The AAPC for C70, C70.0, and C71.9 was negative with statistical significance. For the correlation coefficient between cell phone subscription rate before 10 years and ASIR per 100,000, a positive correlation coefficient with a statistical significance was reported in 3 benign brain tumors (D32, D33, and D32.0) and in 3 malignant brain tumors (C71.0, C71.1 and C71.2). A negative correlation coefficient with statistical significance was reported in 3 malignant brain tumors: C70, C70.0, and C71.9 [i.e., malignant meninges, cerebral meninges, and brain, unspecified]. Conclusion In South Korea, for the correlation coefficient between cell phone subscription rate before 10 years and ASIR per 100,000, a positive correlation coefficient with a statistical significance was reported in 3 malignant brain tumors (C71.0, C71.1, and C71.2) [i.e., cerebrum except lobes and ventricles, frontal lobe and temporal lobe] and 3 benign brain tumors (D32, D33, and D32.0) [i.e. nonmalignant meninges, brain and other CNS, and brain, supratentorial]. Among these results, that for malignant brain tumors in the frontal lobe (C71.1) and in the temporal lobe (C71.2) is suspicious of the association with RF-EMR emitted from cellular phones. The other 4 possible hypotheses do not fit well with the observed phenomena. Statistically insignificant results from recent cohort or large population international studies and contrasting results from many previous case-control studies might come from several issues. A statistically significant increased risk can be found if (ⅰ) a more accurate exposure assessment such as site-specific, time-integral of SAR for each individual is applied or (ⅱ) massive populations over 100,000 are studied. sciencedirect.com

Machine Enhanced Logic

AI evidence extraction

Structured fields below were generated from the paper’s abstract and metadata to make the evidence easier to inspect. Automated extraction may be incomplete or incorrect; verify important details against the original paper.

Study typeecological
Effect directionmixed
PopulationNationwide population of South Korea (registry-based incidence trends)
Sample sizeNot supplied
ExposureRF · Cell phone subscription rate per 100 persons (1985–2019) used as proxy; correlations assessed with a 10-year lag versus incidence (1999–2018).
Evidence strengthLowConfidence: 74% · Peer reviewed: yes

Main findings

Using nationwide South Korean data, the study reports statistically significant positive correlations between cell phone subscription rate (lagged by 10 years) and ASIR for three benign brain tumor categories (D32, D33, D32.0) and three malignant categories (C71.0, C71.1, C71.2). Reported positive correlation coefficients for malignant categories ranged from 0.75 (95% CI 0.46–0.90) for C71.0 to 0.85 (95% CI 0.63–0.93) for C71.1. Statistically significant negative correlations were also reported for some malignant categories (C70, C70.0, C71.9).

Outcomes measured

  • Brain tumor incidence trends (ASIR per 100,000) by ICD-10 subtype
  • Correlation between cell phone subscription rate (10-year lag) and ASIR for specific benign and malignant brain tumor categories
  • Average Annual Percentage Change (AAPC) for selected ICD-10 brain tumor categories

Limitations

  • Ecological study design using subscription rate as a proxy for RF-EMR exposure (potential ecological fallacy and exposure misclassification).
  • Potential confounding and time-trend effects are not resolved by correlation analyses alone.
  • No individual-level phone use, laterality, or SAR-based exposure metrics are used (authors note need for more accurate exposure assessment).
  • Registry and diagnostic/practice changes over time could influence incidence trends (not quantified in the provided text).
View raw extracted JSON
{
  "publication_year": 2023,
  "study_type": "ecological",
  "exposure": {
    "band": "RF",
    "source": "cell-phones",
    "frequency_mhz": null,
    "sar_wkg": null,
    "duration": "Cell phone subscription rate per 100 persons (1985–2019) used as proxy; correlations assessed with a 10-year lag versus incidence (1999–2018)."
  },
  "population": "Nationwide population of South Korea (registry-based incidence trends)",
  "sample_size": null,
  "outcomes": [
    "Brain tumor incidence trends (ASIR per 100,000) by ICD-10 subtype",
    "Correlation between cell phone subscription rate (10-year lag) and ASIR for specific benign and malignant brain tumor categories",
    "Average Annual Percentage Change (AAPC) for selected ICD-10 brain tumor categories"
  ],
  "main_findings": "Using nationwide South Korean data, the study reports statistically significant positive correlations between cell phone subscription rate (lagged by 10 years) and ASIR for three benign brain tumor categories (D32, D33, D32.0) and three malignant categories (C71.0, C71.1, C71.2). Reported positive correlation coefficients for malignant categories ranged from 0.75 (95% CI 0.46–0.90) for C71.0 to 0.85 (95% CI 0.63–0.93) for C71.1. Statistically significant negative correlations were also reported for some malignant categories (C70, C70.0, C71.9).",
  "effect_direction": "mixed",
  "limitations": [
    "Ecological study design using subscription rate as a proxy for RF-EMR exposure (potential ecological fallacy and exposure misclassification).",
    "Potential confounding and time-trend effects are not resolved by correlation analyses alone.",
    "No individual-level phone use, laterality, or SAR-based exposure metrics are used (authors note need for more accurate exposure assessment).",
    "Registry and diagnostic/practice changes over time could influence incidence trends (not quantified in the provided text)."
  ],
  "evidence_strength": "low",
  "confidence": 0.74,
  "peer_reviewed_likely": "yes",
  "stance": "concern",
  "stance_confidence": 0.66,
  "summary": "This ecological study examined whether South Korea’s nationwide cell phone subscription rate (used as a proxy for RF-EMR exposure) correlates with nationwide brain tumor incidence trends. With a 10-year lag, statistically significant positive correlations were reported for several benign (D32, D33, D32.0) and malignant (C71.0, C71.1, C71.2) brain tumor subtypes, including frontal and temporal lobe malignancies. The study also reports statistically significant negative correlations for some malignant categories (C70, C70.0, C71.9), and notes the difficulty of identifying determinants using ecological designs.",
  "key_points": [
    "Exposure was assessed indirectly using national cell phone subscriptions per 100 persons as a proxy for RF-EMR exposure.",
    "Brain tumor outcomes were based on national registry ASIR data by ICD-10 subtype.",
    "With a 10-year lag, statistically significant positive correlations were reported for D32, D33, D32.0 and for C71.0, C71.1, C71.2.",
    "For malignant subtypes, reported positive correlations ranged from 0.75 (C71.0) to 0.85 (C71.1) with stated 95% CIs.",
    "Statistically significant negative correlations were reported for C70, C70.0, and C71.9.",
    "The discussion links frontal/temporal findings to typical phone placement near the frontotemporal region.",
    "Authors emphasize that ecological design and exposure assessment limitations may contribute to inconsistent findings across study types."
  ],
  "categories": [
    "Cell Phones",
    "RF Exposure",
    "Cancer Epidemiology",
    "Brain Tumors",
    "Population Trends"
  ],
  "tags": [
    "Ecological Study",
    "Cell Phone Subscriptions",
    "RF-EMR Proxy Exposure",
    "Brain Tumor Incidence",
    "Age-Standardized Incidence Rate",
    "South Korea Cancer Registry",
    "ICD-10 C71",
    "Frontal Lobe Tumors",
    "Temporal Lobe Tumors",
    "Benign Meningioma",
    "Time-Lag Analysis",
    "Correlation Coefficient"
  ],
  "keywords": [
    "radiofrequency",
    "cell phone",
    "subscription rate",
    "ecological study",
    "brain tumor",
    "incidence trends",
    "ASIR",
    "South Korea",
    "ICD-10",
    "C71.1",
    "C71.2",
    "D32"
  ],
  "suggested_hubs": [
    {
      "slug": "cell-phones",
      "weight": 0.95,
      "reason": "Study evaluates cell phone-related RF-EMR exposure proxy and brain tumor incidence trends."
    }
  ],
  "social": {
    "tweet": "Ecological study (South Korea) used cell phone subscription rates as a proxy for RF-EMR exposure and compared them with brain tumor incidence trends. With a 10-year lag, significant positive correlations were reported for some benign (D32/D33) and malignant (C71.0–C71.2) subtypes, with some negative correlations for others.",
    "facebook": "A 2023 ecological analysis in South Korea examined whether national cell phone subscription rates (as a proxy for RF-EMR exposure) track with brain tumor incidence trends. The study reports statistically significant positive correlations (10-year lag) for several benign and malignant subtypes, including frontal and temporal lobe malignancies, while some malignant categories showed negative correlations. As an ecological design, it cannot establish individual-level causation.",
    "linkedin": "Environmental Research (2023): An ecological study in South Korea assessed correlations between nationwide cell phone subscription rates (proxy for RF-EMR exposure) and registry-based brain tumor incidence (ASIR) with a 10-year lag. Significant positive correlations were reported for select benign (D32/D33) and malignant (C71.0–C71.2) subtypes, alongside negative correlations for some malignant categories, highlighting both signal and limitations of trend-based analyses."
  }
}
AI can be wrong. Always verify against the paper.

AI-extracted fields are generated from the abstract and metadata and may be incomplete or incorrect. This material is for informational purposes and is not medical advice.

Identifiers and source

DOI: 10.1016/j.envres.2023.115657
PubMed: 36906274

Read the original paper →

Related papers

← Back to the Research Library