Developing Mendelian randomization methods for late onset, time-to-event data

PhD project (3/4 yr research project leading to independent research at the doctorate level)

Dr Jack Bowden, 2nd supervisor TBC, 3rd supervisor TBC

Return to list


Mendelian randomization (MR) uses genetic data to probe questions of causality within the instrumental variable framework. Statistical methods for Mendelian randomization (MR) analysis are well developed and understood for continuous and, to at least some degree, for binary outcomes. Far less progress has been made in developing methods for time-to-event outcomes, such as all-cause mortality.

Aims & objectives

1. To develop MR methods that can account for censoring due to dropout.

2. develop models that can assess whether the non-linear association observed between certain health exposures and survival outcomes are

(a) Truly non-linear,
(b) A product of confounding, or
(c) A signal that some of the genetic instruments are invalid instruments


The PhD will investigate the use of: additive hazard, rank-preserving structural failure time, and frailty models to account for censoring; Local Average Treatment Effects to account for non-linearity; and pleiotropy robust methods such as MR-Egger regression and Weighted Median estimation to account for invalid instruments


Tchetgen et al. Instrumental variable estimation in a survival context. Epidemiology 2015

Silverwood RJ et al . Testing for non-linear causal effects using a binary genotype in a Mendelian randomization study: application to alcohol and cardiovascular traits. IJE 2014

Created on July 20, 2017, 10:13 a.m.

Edit this page