Bayesian Hierarchical Models Combining Different Study Types and Adjusting for Covariate Imbalances: A Simulation Study to Assess Model Performance

C. Elizabeth McCarron,Jean-Eric Tarride,Lehana Thabane,Ron Goeree,Eleanor M. Pullenayegum

doi:10.1371/journal.pone.0025635

Abstract

BackgroundBayesian hierarchical models have been proposed to combine evidence from different types of study designs. However, when combining evidence from randomised and non-randomised controlled studies, imbalances in patient characteristics between study arms may bias the results. The objective of this study was to assess the performance of a proposed Bayesian approach to adjust for imbalances in patient level covariates when combining evidence from both types of study designs.Methodology/Principal FindingsSimulation techniques, in which the truth is known, were used to generate sets of data for randomised and non-randomised studies. Covariate imbalances between study arms were introduced in the non-randomised studies. The performance of the Bayesian hierarchical model adjusted for imbalances was assessed in terms of bias. The data were also modelled using three other Bayesian approaches for synthesising evidence from randomised and non-randomised studies. The simulations considered six scenarios aimed at assessing the sensitivity of the results to changes in the impact of the imbalances and the relative number and size of studies of each type. For all six scenarios considered, the Bayesian hierarchical model adjusted for differences within studies gave results that were unbiased and closest to the true value compared to the other models.Conclusions/SignificanceWhere informed health care decision making requires the synthesis of evidence from randomised and non-randomised study designs, the proposed hierarchical Bayesian method adjusted for differences in patient characteristics between study arms may facilitate the optimal use of all available evidence leading to unbiased results compared to unadjusted analyses.

Highlights

Evidence of the effects of interventions is a critical component of health care decision making as it contributes to the comparison of alternative interventions in terms of their relative costs and effects
The informative prior approach appears to give less biased results than the model adjusted for aggregate study values while bias is roughly equal for both the model adjusted for aggregate values and the unadjusted model
An increase in the study arm size for the non-randomised studies relative to the randomised studies tends to increase the precision of the estimates for all of the models

Summary

Introduction

Evidence of the effects of interventions is a critical component of health care decision making as it contributes to the comparison of alternative interventions in terms of their relative costs and effects. Bayesian hierarchical models have been proposed to combine evidence from different types of study designs such as randomised and nonrandomised studies [1,2] Due to their inherent design, RCTs are more likely to be balanced in terms of patient characteristics between study arms than non-randomised studies, but they are subject to strict inclusion and exclusion criteria which may limit their generalisability. Other sources of bias, both internal (e.g., performance, attrition) and external (e.g., population, intervention) [3], may exist, it is the increased likelihood of imbalances among the non-randomised studies that constitutes the principal difference between randomised and non-randomised studies [4]. The objective of this study was to assess the performance of a proposed Bayesian approach to adjust for imbalances in patient level covariates when combining evidence from both types of study designs

Objectives

Results

Conclusion