4-year PhD: Detecting and adjusting for selection bias within a Mendelian randomization study


   Faculty of Health and Life Sciences

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
Prof Kate Tilling, Prof George Davey-Smith, Dr J Bowden  No more applications being accepted  Funded PhD Project (Students Worldwide)

About the Project

We are seeking a talented statistician or epidemiologist for an exciting methodological research PhD studentship starting in September 2018. The studentship will be based in the MRC Integrative Epidemiology Unit at the University of Bristol, a leading centre for causal methods development.

The successful applicant will be supervised by Prof Kate Tilling, Prof George Davey Smith and Dr Jack Bowden. Full funding is provided by the UK Medical Research Council (see standard MRC funding details).
Background

A major challenge in using Mendelian randomization to analyse data from large cohorts is the issue of non-random selection into the analytical sample. There are several reasons why the analytical sample may not represent the population of interest, including: nonrandom sampling, nonrandom participant refusal, loss to follow-up, subgroups analysis, and missing data. In all these cases there may be both known and unknown factors that influence the selection of participants for analysis. For example, the response rate for recruitment into the UK Biobank study was particularly low (6%) and it is known that individuals with higher than average levels of educational attainment and general health are over-represented in the study [1].

Selection into (or dropout from) a study has the potential to distort the findings of future MR analyses due to collider bias [2]. For example, if the exposure of interest causes dropout then subsequent SNP-exposure estimates used to derive causal effects will also be biased [3]. In standard observational analyses, non-random sampling is usually adjusted for via inverse probability weighting, and dropout from a study by either inverse probability weighting or multiple imputation. Approaches to selection bias within the Mendelian randomization framework are currently under-developed.

Aims
This PhD will develop methods for detecting and adjusting for selection and other associated biases within the framework of an MR study. The methods developed will be applied to Mendelian randomization analyses which have external information on selection bias, either via comparison with routine data or via baseline data on subjects who subsequently drop out of the study.

Methods
Specific methodologies to be explored will include multiple imputation, inverse probability weighting, conditional logistic regression [4] and data augmentation [5]. Simulations will be used to compare the different methods, and the situations under which each might be preferable in practice. The methods developed will be applied to the UK Biobank study, and other studies with plausibly different selection mechanisms. The project would suit someone with a strong background

How to apply
Closing date: 9am, 21st December 2017

Application: Apply online using the University of Bristol PhD application system. `Select the Faculty of Health Sciences and then Population Health Sciences PhD programme and state in the application that you are applying for funding from the “MRC Integrative Epidemiology PhD programme”.
Selection process: Shortlisted applicants will be invited for interview in January 2018. The successful candidate will be offered a studentship to start in September 2018.

References

[1] Fry A, Littlejohns TJ, Sudlow C, Doherty N, Adamska L, Sprosen T, Collins R, Allen NE. Comparison of Sociodemographic and Health-Related Characteristics of UK Biobank Participants With Those of the General Population. American Journal of Epidemiology, 2017.
[2] Munafò M, Tilling K, Taylor A, Evans D, Davey Smith G. Collider Scope: How selection bias can induce spurious associations. bioRxiv. https://doi.org/10.1101/079707. 2016
[3] Hughes R, Davies N, Davey Smith G and Tilling K et al. Selection bias in instrumental variable analysis. BioRxiv 2017 https://www.biorxiv.org/content/biorxiv/early/2017/09/22/192237.full.pdf
[4] Breslow NE, Day NE, Halvorsen KT, Prentice RL, Sabai C (1978). "Estimation of multiple relative risk functions in matched case-control studies.". Am J Epidemiol. 108 (4): 299–307
[5] Clayton D. Conditional Likelihood Inference under Complex Ascertainment Using Data Augmentation Biometrika 2003. 90: 976-981

Where will I study?