Pulse Brain · Growing Health Evidence Index
Tier 3 — Observational / field trialPeer-reviewed

Assessing the suitability of summary data for two-sample Mendelian randomization analyses using MR-Egger regression: the role of the I2 statistic

Jack Bowden, Fabiola Del Greco M, Cosetta Minelli, George Davey Smith, Nuala A. Sheehan, John R. Thompson

International Journal of Epidemiology · 2016

Read source ↗ All evidence

Summary

This methodological paper proposes IGX2, an adaptation of the meta-analytic I2 statistic, to quantify violations of the NO Measurement Error assumption in two-sample Mendelian randomization using MR-Egger regression. Through simulations and real data application, the authors demonstrate that MR-Egger causal estimates are biased towards the null when this assumption is violated, with dilution proportional to the strength of violation. The paper evaluates simulation extrapolation as a corrective approach and demonstrates its utility in mitigating bias.

UK applicability

This is a statistical methods paper with broad applicability to genetic epidemiology research conducted in the United Kingdom and internationally. UK researchers using summary-level genetic data for causal inference would benefit from adopting the IGX2 metric to assess and potentially correct for measurement error bias in their analyses.

Key measures

IGX2 statistic, F-statistic, causal effect estimates, type I error rates, bias in MR-Egger regression, pleiotropy test performance

Outcomes reported

The study evaluated the suitability of the I2-statistic adaptation (IGX2) for quantifying violations of the NO Measurement Error assumption in Mendelian randomization analyses, and assessed the performance of simulation extrapolation as a corrective method.

Theme
Measurement & metrics
Subject
Measurement methods & nutrient profiling
Study type
Research
Study design
Methodological study with simulation and real data application
Source type
Peer-reviewed study
Status
Published
Geography
International
System type
Laboratory / in vitro
DOI
10.1093/ije/dyw220
Catalogue ID
BFmor3gaas-pg1ave

Topic tags

Pulse AI · ask about this record

Dig deeper with Pulse AI.

Pulse AI has read the whole catalogue. Ask about this record, its theme, or how the findings apply to UK farming and policy — every answer cites the underlying studies.