Network Meta-analysis: Application and Practice using R software

SHIM, SUNG RYUL; KIM, SEONG-JANG; LEE, JONGHOO; Rücker, Gerta

doi:2019.0.0.e2019013

Methods

Epidemiology and Health 2019;e2019013.
DOI: https://doi.org/10.4178/epih.e2019013 [Accepted] Published online April 8, 2019.

Network Meta-analysis: Application and Practice using R software

SUNG RYUL SHIM^1,2

, SEONG-JANG KIM^3,4

, JONGHOO LEE⁵

, Gerta Rücker⁶

¹Department of Preventive Medicine, Korea University College of Medicine, Seoul, Korea
²Urological Biomedicine Research Institute, Soonchunhyang University Hospital, Seoul, Korea
³Department of Nuclear Medicine, College of Medicine, Pusan National University, Yangsan, Korea
⁴BioMedical Research Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Korea
⁵Department of Internal Medicine, Jeju National University Hospital, Jeju National University School of Medicine, Jeju, Korea
⁶Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg, Freiburg, Germany

Correspondence SUNG RYUL SHIM ,Tel: 01035843336, Email: sungryul.shim@gmail.com

Received: March 14, 2019 Accepted after revision: April 8, 2019

Abstract

The objective of this study was to describe general approaches of network meta-analysis that are available for quantitative synthesis of data using R software. We conducted a network meta-analysis using two types of concepts: Bayesian and frequentist approach. The packages of R software were gemtc for Bayesian approach and netmeta for frequentist approach. In estimating a network meta-analysis models in a Bayesian framework, rjags package is one of conformance fitting tools. The package of rjags implemented Markov Chain Monte Carlo simulation with graphical outcomes. The estimated overall effect sizes, test for heterogeneity and moderator effect, and the publication bias were reported using R software. Especially authors stressed two flexible modellings of Bayesian and frequentist for overall effect sizes in network meta-analysis. This study focused on the practical methods of network meta-analysis rather than theoretical concepts for Korean researchers who were non-majored in statistics. Through this study, authors hope that many Korean researchers will use R software to perform a network meta-analysis more easily and that related research will be activated.

Keywords: Meta-analysis; network meta-analysis; Multiple treatments meta-analysis; Mixed treatment comparison; consistency; R software