RF Safe

Probabilistic Multiple-Bias Modeling Applied to the Canadian Data From the Interphone Study of Mobile Phone Use and Risk of Glioma, Meningioma, Acoustic Neuroma, and Parotid Gland Tumors

pubmed · 2017 · Harm · Moderate evidence

← Research stream
HarmModerate evidencePaper 2258

After adjustment for selection and recall biases, the odds ratio for glioma among the highest quartile of mobile phone users was 2.2 (95% CI: 1.3, 4.1), indicating increased risk. There was little evidence of increased risk for meningioma, acoustic neuroma, o…

Study snapshot

Publicationpubmed · 2017
Study typecase_control
Reported directionHarm
Evidence ratingModerate

Abstract

- Canadian data from INTERPHONE study of mobile phone use and head tumor risk reported for the first time Since the 13-nation Interphone study was published in 2010, several methods papers have been published that reanalyze the data to correct for biases in the original paper. In most instances the glioma risk estimates increased after adjustment for study biases among long term or heavy mobile phone users. The following paper just published in the American Journal of Epidemiology applies statistical adjustments to the Interphone data from Canada. The authors found that the risk estimate for glioma among the highest quartile of cell phone users increased after adjustment. Risk estimates for other types of head tumors did not change. The substantive results reported for glioma risk in Canada should be of interest to Health Canada. The original Canadian Interphone study data which had not been reported previously found a statistically significant doubling of risk for glioma among the highest quartile of cell phone users (over 558 lifetime hours). In contrast, the original 13-nation Interphone study found a 1.4-fold increase in glioma among the highest decile of use (1640 or more lifetime hours). After adjustment for selection and recall bias, the 2.0 odds ratio originally found in Canada increased to 2.2. Among the potential explanations the authors provided for this disparity between Canada and the 13-nation study (which included Canadian data): "real differences in risk related to different communication technologies between Canada and other INTERPHONE countries." Momoli F, Siemiatycki J, McBride ML, Parent MÉ, Richardson L, Bedard D, Platt R, Vrijheid M, Cardis E, Krewski D. Probabilistic multiple-bias modelling applied to the Canadian data from the INTERPHONE study of mobile phone use and risk of glioma, meningioma, acoustic neuroma, and parotid gland tumors. Am J Epidemiol. 2017 May 23. doi: 10.1093/aje/kwx157. Abstract We undertook a re-analysis of the Canadian data from the thirteen-country INTERPHONE case-control study (2001-2004), which evaluated the association between mobile phone use and risk of brain, acoustic neuroma, and parotid gland tumors. The main publication of the multinational INTERPHONE study concluded that "biases and errors prevent a causal interpretation". We applied a probabilistic multiple-bias model to address possible biases simultaneously, using validation data from billing records and non-participant questionnaires as information on recall error and selective participation. Our modelling sought to adjust for these sources of uncertainty and to facilitate interpretation. For glioma, the odds ratio comparing highest quartile of use (over 558 lifetime hours) to non-regular users was 2.0 (95% confidence interval: 1.2, 3.4). The odds ratio was 2.2 (95% confidence interval: 1.3, 4.1) when adjusted for selection and recall biases. There was little evidence of an increase in the risk of meningioma, acoustic neuroma, or parotid gland tumors in relation to mobile phone use. Adjustments for selection and recall biases did not materially affect interpretation in our Canadian results. ncbi.nlm.nih.gov Excerpts The OR of 2.0 for glioma in the highest cumulative exposure category (558+ hours of cumulative call time) in the Canadian study is higher than the value of 1.4 in the highest cumulative exposure category (1640+ hours) in the international study (4). This may simply reflect sampling variability, differential biases between study centers, matching strategies, or real differences in risk related to different communication technologies between Canada and other INTERPHONE countries (see appendix of (27)). Unlike in the Canadian data, the INTERPHONE multinational study found marked decreased risk associated with most measures of phone use and an increased risk only in the highest decile of use. The study Group concluded that "biases and errors prevent a causal interpretation" (4, pg.1). To the extent that the bias model applied in the present re-analysis of the Canadian data is reasonable, conventional modelling of existing data likely resulted in slight underestimation of the magnitude of associations; however, interpretation of bias-adjusted results would not have materially changed from the original Canadian results.

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 typecase_control
Effect directionharm
PopulationCanadian participants in the INTERPHONE study
Sample sizeNot supplied
Exposurehighest quartile of use over 558 lifetime hours
Evidence strengthModerateConfidence: 70% · Peer reviewed: yes

Main findings

After adjustment for selection and recall biases, the odds ratio for glioma among the highest quartile of mobile phone users was 2.2 (95% CI: 1.3, 4.1), indicating increased risk. There was little evidence of increased risk for meningioma, acoustic neuroma, or parotid gland tumors. Adjustments did not materially change interpretation of Canadian results.

Outcomes measured

  • glioma
  • meningioma
  • acoustic neuroma
  • parotid gland tumors

Limitations

  • Potential sampling variability
  • Differential biases between study centers
  • Matching strategies
  • Uncertainty if differences reflect real risk differences or methodological factors
View raw extracted JSON
{
  "study_type": "case_control",
  "exposure": {
    "band": null,
    "source": "mobile phone",
    "frequency_mhz": null,
    "sar_wkg": null,
    "duration": "highest quartile of use over 558 lifetime hours"
  },
  "population": "Canadian participants in the INTERPHONE study",
  "sample_size": null,
  "outcomes": [
    "glioma",
    "meningioma",
    "acoustic neuroma",
    "parotid gland tumors"
  ],
  "main_findings": "After adjustment for selection and recall biases, the odds ratio for glioma among the highest quartile of mobile phone users was 2.2 (95% CI: 1.3, 4.1), indicating increased risk. There was little evidence of increased risk for meningioma, acoustic neuroma, or parotid gland tumors. Adjustments did not materially change interpretation of Canadian results.",
  "effect_direction": "harm",
  "limitations": [
    "Potential sampling variability",
    "Differential biases between study centers",
    "Matching strategies",
    "Uncertainty if differences reflect real risk differences or methodological factors"
  ],
  "evidence_strength": "moderate",
  "confidence": 0.7,
  "peer_reviewed_likely": "yes",
  "keywords": [
    "mobile phone",
    "glioma",
    "head tumors",
    "bias adjustment",
    "case-control study",
    "INTERPHONE",
    "Canada"
  ],
  "suggested_hubs": [
    {
      "slug": "mobile-phone-exposure",
      "weight": 0.9,
      "reason": "Study focuses on mobile phone use and cancer risk."
    }
  ]
}
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: 10.1093/aje/kwx157
PubMed: Not supplied

Read the original paper →

Related papers

← Back to the Research Library