MRC INTEGRATIVE EPIDEMIOLOGY UNIT (IEU)
SEMINAR SERIES

 SEMINAR

 Thursday, 18^th February, 2016

16.00 – 17.00 - Room OS6 – Oakfield House

 
Dr Stephen Burgess

Sir Henry Welcome Post-Doctoral Fellow

Department of Public Health and Primary Care

School of Clinical Medicine, University of Cambridge

 

“Testing and estimating non-linear effects using Mendelian randomization”

 

Abstract

Instrumental variable methods can estimate the causal effect of an exposure on an outcome using observational data. Many instrumental variable methods assume the exposure--outcome relationship is linear, but in practice this assumption is often in doubt, or perhaps the shape of the relationship is a target for investigation. This issue is investigated in the context of Mendelian randomization, the use of genetic variants as instrumental variables. We demonstrate the performance of a simple linear instrumental variable method when the true shape of the exposure--outcome relationship is not linear. In this case, the linear instrumental variable estimate approximates a population-averaged causal effect. This is the average difference in the outcome if the exposure for every individual in the population is increased by a fixed amount. We also present a novel method for estimating the effect of the exposure on the outcome within strata of the exposure distribution. This enables the estimation of “localized average causal effects” within quantile groups of the exposure. Estimates of localized average causal effects reveal the shape of the exposure--outcome relationship for a variety of models, which can be estimated using a sliding window about, or by smoothing the localized average causal effects (we suggest the use of fractional polynomials). These methods are used to investigate the relationships between body mass index and a range of cardiovascular risk factors.

 Biography

Stephen Burgess completed his BA and MMath (Part III) in Mathematics from the University of Cambridge. He studied for a PhD in the MRC Biostatistics Unit, Cambridge, from 2008-11 working on methods for Mendelian randomization analysis. He joined the Cardiovascular Epidemiology Unit in the Department of Public Health and Primary Care of the University of Cambridge in September 2011. In January 2012, Stephen was awarded a Wellcome Trust Fellowship to continue theoretical and applied work in the field of Mendelian randomization. He organizes the unit's working group on applied Mendelian randomization investigations.

 ALL WELCOME