How to Conduct a Bayesian Network Meta-Analysis

Dapeng Hu; Annette M O'Connor; Chong Wang; Jan M Sargeant; Charlotte B Winder

doi:10.3389/fvets.2020.00271

How to Conduct a Bayesian Network Meta-Analysis

Front Vet Sci. 2020 May 19:7:271. doi: 10.3389/fvets.2020.00271. eCollection 2020.

Authors

Dapeng Hu¹, Annette M O'Connor², Chong Wang^{1

3}, Jan M Sargeant⁴, Charlotte B Winder⁴

Affiliations

¹ Department of Statistics, Iowa State University, Ames, IA, United States.
² Department of Large Animal Clinical Sciences, College of Veterinary Medicine, Michigan State University, East Lansing, MI, United States.
³ Department of Veterinary Diagnostic and Production Animal Medicine, 2203 Lloyd Veterinary Medical Center, Iowa State University, Ames, IA, United States.
⁴ Department of Population Medicine, University of Guelph, Guelph, ON, Canada.

Abstract

Network meta-analysis is a general approach to integrate the results of multiple studies in which multiple treatments are compared, often in a pairwise manner. In this tutorial, we illustrate the procedures for conducting a network meta-analysis for binary outcomes data in the Bayesian framework using example data. Our goal is to describe the workflow of such an analysis and to explain how to generate informative results such as ranking plots and treatment risk posterior distribution plots. The R code used to conduct a network meta-analysis in the Bayesian setting is provided at GitHub.

Keywords: Bayesian; network meta-analysis; systematic review; tutorial; veterinary science.