Skip to main content
Effective Health Care Program
Home » Products » Bayesian Approaches to Indirect Comparisons » Bayesian Approaches to Indirect Comparisons

Bayesian Approaches to Indirect Comparisons


Topic Abstract

Introduction: Systematic reviews are a primary synthesis of evidence for healthcare decision makers confronted with questions about the comparative effectiveness and safety of interventions. The strength and usefulness of such reports, however, depends largely on the availability, comparability, and consistency of quality studies addressing a particular question of interest. While head-to-head randomized controlled trials (RCTs) are generally considered the gold standard for evaluating the comparative efficacy, effectiveness, and safety of healthcare interventions, many competing interventions have never been compared directly.

Objectives: In this project, we explore Bayesian statistical modeling approaches to this problem, and compare them with established classical approaches for indirect comparisons.

Approaches: In addition to comparing existing classical and Bayesian methods for binary data in the context of various application domains, we will also undertake a preliminary simulation study to compare classical and Bayesian operating characteristics (e.g., mean squared error for estimation or prediction, or Type I error and power for hypothesis testing) of the two approaches under a variety of assumed true states of nature. A second primary focus of the work is to develop, implement, and evaluate novel Bayesian approaches to indirect comparisons in a variety of binary and non-binary model settings. For instance, hierarchical models that simultaneously consider more than one endpoint (say, efficacy and safety), handle varying observation times, and properly assess the commensurability of various data sources of varying quality, all await development.