Summary
This narrative review by Eleanor Sanderson examines multivariable Mendelian randomisation as a methodological extension for conducting mediation analysis using genetic instruments. The paper explains how MVMR enables decomposition of direct and mediated causal pathways whilst avoiding confounding bias, and critically appraises the advantages and limitations of this approach compared to conventional mediation analysis.
Regional applicability
As a methodological review with no UK-specific application, the findings are globally applicable and would inform epidemiological and causal inference practice in the UK and elsewhere. UK biobank data and similar genetic cohorts could employ these methods to investigate mediated causal pathways in health and disease.
Key measures
Causal effect estimates; direct effects; indirect (mediated) effects; bias due to confounding
Outcomes reported
This narrative review examines the application of multivariable Mendelian randomisation (MVMR) to estimate causal effects in mediation analysis. The paper evaluates how MVMR decomposes direct and indirect (mediated) effects whilst retaining the causal inference advantages of genetic instruments.
Topic tags
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.