Summary
This paper extends Mendelian randomization methodology to the multivariable setting, clarifying how to interpret causal effects when multiple exposures are considered simultaneously and developing diagnostic tests for instrument validity in both single-sample and two-sample summary data analyses. The authors use simulations and theoretical reasoning to distinguish between scenarios where secondary exposures act as confounders, mediators, pleiotropic pathways or colliders. The methods are demonstrated using UK Biobank data, establishing MVMR as a robust tool for determining direct causal effects across a range of epidemiological scenarios.
UK applicability
The methodology was developed and validated using UK Biobank data, making the findings directly applicable to UK epidemiological research. The diagnostic tools and interpretation frameworks will be of immediate utility to UK health researchers conducting Mendelian randomization studies with multiple exposures.
Key measures
Instrument strength and validity assessments; causal effect estimates under different causal scenarios (confounding, mediation, pleiotropy, collider bias); applications to education, cognitive ability and BMI in UK Biobank
Outcomes reported
The study developed and validated methods for multivariable Mendelian randomization (MVMR) to estimate direct causal effects of multiple exposures on outcomes whilst accounting for confounding, mediation, pleiotropy and collider bias. The methods were demonstrated using UK Biobank data to estimate effects of education and cognitive ability on body mass index.
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.