The role of digital device use on the risk of migraine: a univariable and multivariable Mendelian
Abstract
The role of digital device use on the risk of migraine: a univariable and multivariable Mendelian randomization study He Z, Qiu F, Yang J, Zhao M. The role of digital device use on the risk of migraine: a univariable and multivariable Mendelian randomization study. Front Neurol. 2024 Oct 30;15:1462414. doi: 10.3389/fneur.2024.1462414. Abstract Background: The pervasive integration of digital devices into daily life has raised concerns about their potential health impacts. This study aimed to explore the causal relationships between digital device use and the risk of migraine using Mendelian randomization (MR). Methods: Genetic data on digital device use and migraines were sourced from large-scale genome-wide association studies conducted by the UK Biobank, the FinnGen study, and the International Headache Genetics Consortium. Univariable MR (UVMR), meta-analysis, and multivariable MR (MVMR) approaches were conducted to explore and verify the causal effects of digital device use (including mobile phone use, computer use, playing computer games, and watching television) on migraine risk. Sensitivity analyses were conducted using Cochran's Q, MR-Egger intercept test, MR pleiotropy residual sum and outlier, MR Radial, MR Steiger, and leave-one-out methods. Results: UVMR analyses revealed that genetically predicted mobile phone use was significantly associated with an increased risk of overall migraine (odds ratio [OR] = 2.39, p = 9.78e-5) and migraine without aura (MO) (OR = 2.25, p = 0.024). Additionally, there were significant positive associations between genetically predicted television watching and the risk of overall migraine (OR = 1.63, p = 2.12e- 5) and MO (OR = 2.10, p = 4.98e-5). These results were further supported by the meta-analysis and MVMR analysis. Sensitivity analysis indicated no heterogeneity or pleiotropy. Conclusion: This comprehensive MR study provides preliminary evidence for the causal impact of mobile phone use and television watching on the risk of migraines. Further studies are needed to explore these associations across different populations. : frontiersin.org
AI evidence extraction
Main findings
Using Mendelian randomization with GWAS data, genetically predicted mobile phone use was associated with increased risk of overall migraine (OR=2.39, p=9.78e-5) and migraine without aura (OR=2.25, p=0.024). Genetically predicted television watching was also positively associated with overall migraine (OR=1.63, p=2.12e-5) and migraine without aura (OR=2.10, p=4.98e-5); results were supported by meta-analysis and multivariable MR, with sensitivity analyses indicating no heterogeneity or pleiotropy.
Outcomes measured
- overall migraine
- migraine without aura (MO)
Limitations
- Frequency/band/SAR or other EMF exposure metrics were not reported
- exposure was digital device use as a behavioral trait.
- Population characteristics and sample sizes were not provided in the abstract.
- Authors describe the evidence as preliminary and note need for studies in other populations.
View raw extracted JSON
{
"study_type": "other",
"exposure": {
"band": null,
"source": "mobile phone, television, computer, computer games (digital device use)",
"frequency_mhz": null,
"sar_wkg": null,
"duration": null
},
"population": null,
"sample_size": null,
"outcomes": [
"overall migraine",
"migraine without aura (MO)"
],
"main_findings": "Using Mendelian randomization with GWAS data, genetically predicted mobile phone use was associated with increased risk of overall migraine (OR=2.39, p=9.78e-5) and migraine without aura (OR=2.25, p=0.024). Genetically predicted television watching was also positively associated with overall migraine (OR=1.63, p=2.12e-5) and migraine without aura (OR=2.10, p=4.98e-5); results were supported by meta-analysis and multivariable MR, with sensitivity analyses indicating no heterogeneity or pleiotropy.",
"effect_direction": "harm",
"limitations": [
"Frequency/band/SAR or other EMF exposure metrics were not reported; exposure was digital device use as a behavioral trait.",
"Population characteristics and sample sizes were not provided in the abstract.",
"Authors describe the evidence as preliminary and note need for studies in other populations."
],
"evidence_strength": "low",
"confidence": 0.74,
"peer_reviewed_likely": "yes",
"keywords": [
"Mendelian randomization",
"GWAS",
"UK Biobank",
"FinnGen",
"International Headache Genetics Consortium",
"mobile phone use",
"television watching",
"digital device use",
"migraine",
"migraine without aura"
],
"suggested_hubs": []
}AI can be wrong. Always verify against the paper.
Comments
Log in to comment.
No comments yet.
NNEMF synthesis
He and colleagues used Mendelian randomization, a method that treats inherited genetic variants as stand-ins for a behavior, to test whether digital-device habits are causally related to migraine. Genetic data came from the UK Biobank, FinnGen, and the International Headache Genetics Consortium. The exposures were genetically predicted mobile phone use, computer use, playing computer games, and watching television. The outcomes were overall migraine and migraine without aura. No frequency, modulation, intensity, specific absorption rate (the energy tissue absorbs), or heating measurement is reported. The dose is not reported, so it cannot be compared with the 1.6 watts per kilogram phone limit, and this record does not establish that the exposure was below the level that heats tissue. Univariable Mendelian randomization found that genetically predicted mobile phone use was associated with higher odds of overall migraine (odds ratio 2.39, p = 9.78e-5) and migraine without aura (odds ratio 2.25, p = 0.024). Genetically predicted television watching was associated with higher odds of overall migraine (odds ratio 1.63, p = 2.12e-5) and migraine without aura (odds ratio 2.10, p = 4.98e-5). Meta-analysis and multivariable Mendelian randomization supported those associations. Sensitivity tests (Cochran's Q, MR-Egger intercept, MR pleiotropy residual sum and outlier, MR Radial, MR Steiger, and leave-one-out) indicated no heterogeneity or pleiotropy. The abstract does not report a significant association for computer use or playing computer games. That pattern is a null for those two behaviors, not a refutation of a field effect: the instruments capture self-reported device habits, not waveform, pulsing, or absorbed energy, and the study did not test recovery, challenge, or tissue-level coupling. This is a human genetic-epidemiology result, not a measurement of calcium timing, voltage-sensor behavior, mitochondrial redox, or spin chemistry. It does not show that radiofrequency exposure causes migraine. It shows that genetic liability to two device-use behaviors tracks higher migraine odds in these cohorts. Relevance to people is the finding itself; it does not travel through an animal model. What it does not establish is the physical exposure, the mechanism, or whether the association would hold in other populations, which the authors themselves flag as needing further study.