In the middle of applying to universities? | SHARE YOUR EXPERIENCE In the middle of applying to universities? | SHARE YOUR EXPERIENCE

4-year PhD Studentship: Development of methods to triangulate multiple sources of evidence in epidemiology

   Faculty of Health Sciences

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

Click here to search for PhD studentship opportunities
  Prof J Higgins, Prof Kate Tilling  No more applications being accepted  Self-Funded PhD Students Only

About the Project

Triangulation in epidemiology is the strategic use of multiple methods to answer a single question [1,2]. Lawlor, Tilling and Davey Smith explained how causal inferences can be strengthened by triangulating results from several approaches with different key sources of potential bias [1]. For example, randomized trials, Mendelian randomization studies and traditional multivariable regression analyses of observational evidence might all tackle a question relating to the same exposure-outcome effect. The studies may produce different effect estimates because they are (i) differentially relevant (i.e. asking subtly different questions, e.g. in relation to the period or patterns of exposure), (ii) differentially biased (i.e. compromised by different types of problems) and/or (iii) subject to chance. The statistical methods for combining the results from multiple sources of evidence within a triangulation framework are underdeveloped. Quantitative triangulation needs to combine the three key issues (differential relevance, differential bias and chance) in a statistical model and assess the extent to which the observed data fit together (i.e. are coherent).

Aims and Objectives

  1. To develop methods for quantitative triangulation of aetiological epidemiological evidence to estimate the effect of an exposure on an outcome.
  2. To illustrate methods through application to case studies.


The PhD will involve the development, illustration and evaluation of quantitative methods for triangulation of evidence. An evidence synthesis approach known as multiparameter evidence synthesis should provide a fruitful framework within which to explore this [3]. The student will examine existing methods for producing these statistical models and for assessing coherence and develop these into novel approaches for drawing conclusions about causal effects of an exposure on an outcome. Bayesian methods will be explored because they are flexible and allow incorporation of external information through prior distributions. In addition to working on novel statistical methods, the student may explore other methodological questions. These include how we should evaluate the risk of bias in studies for which formal frameworks (such as RoB 2 [4] and ROBINS-I [5]) have not been developed, and what sources of information are available about biases to inform prior distributions. Methodology will be illustrated through application to important causal questions in epidemiology, and the successful student will be encouraged to find case studies matching with their own interests.


Triangulation, evidence synthesis, bias, consistency, meta-analysis, epidemiology, aetiology, causal inference

How to apply for this project

This project will be based in Bristol Medical School - Population Health Sciences in the Faculty of Health Sciences at the University of Bristol.

Please visit the Faculty of Health Sciences website for details of how to apply

Funding Notes

This project is open for University of Bristol PGR scholarship applications (closing date 25th February 2022)
The University of Bristol PGR scholarship pays tuition fees and a maintenance stipend (at the minimum UKRI rate) for the duration of a PhD (typically three years but can be up to four years).


1. Lawlor DA, Tilling K, Davey Smith G. Triangulation in aetiological epidemiology. IJE 2016;45:1866-1886
2. Munafò MR, Davey Smith G. Robust research needs many lines of evidence: Replication is not enough. Nature. 2018;553:399-401.
3. Ades AE, Sutton AJ. Multiparameter evidence synthesis in epidemiology and medical decision-making: current approaches. JRSS-A 2006;169:5-35.
4. Sterne JAC, Savović J, [others], Higgins JPT. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898.
5. Sterne JAC, Hernán MA, [others], Higgins JPT. ROBINS-I: a tool for assessing risk of bias in non-randomized studies of interventions. BMJ 2016;355;i4919.
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs