Rxivist logo

An examination of multivariable Mendelian randomization in the single sample and two-sample summary data settings.

By Eleanor Sanderson, George Davey Smith, Frank Windmeijer, Jack Bowden

Posted 27 Apr 2018
bioRxiv DOI: 10.1101/306209 (published DOI: 10.1093/ije/dyy262)

Mendelian Randomisation (MR) is a powerful tool in epidemiology which can be used to estimate the causal effect of an exposure on an outcome in the presence of unobserved confounding, by utilising genetic variants that are instrumental variables (IVs) for the exposure. This can be extended to Multivariable MR (MVMR) to estimate the effect of two or more exposures on an outcome. We use simulations and theoretical arguments to clarify the interpretation of estimated effects in a MVMR analysis under a range of underlying scenarios, where a secondary exposure acts variously as a confounder, a pleiotropic pathway, a mediator and a collider. We then describe how instrument strength and validity can be assessed for an MVMR analysis in the single sample setting, and how such tests can be extrapolated to the popular two-sample summary data setting. We illustrate our methods using data from UK biobank to estimate the effect of education and cognitive ability on body mass index. We show that MVMR analysis consistently estimates the effect of an exposure, or exposures, of interest and provides a powerful tool for determining causal effects in a wide range of scenarios with either individual or summary level data.

Download data

  • Downloaded 1,356 times
  • Download rankings, all-time:
    • Site-wide: 6,402 out of 83,642
    • In epidemiology: 56 out of 1,556
  • Year to date:
    • Site-wide: 21,797 out of 83,642
  • Since beginning of last month:
    • Site-wide: 17,093 out of 83,642

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide


Sign up for the Rxivist weekly newsletter! (Click here for more details.)