Introduction to network meta-analysis: understanding what it is, how it is done, and how it can be used for decision-making.

Abstract

Network meta-analysis (NMA), a statistical technique that allows systematic reviewers to simultaneously compare more than 2 alternatives, makes use of indirect evidence from studies comparing interventions of interest to a common comparator. The capacity for multiple simultaneous comparisons makes NMA appealing for evidence-based decision-makers. This article, aimed at users of systematic reviews (SRs) with NMAs and at those who are considering conducting SRs with NMAs, provides an introductory level overview of this topic. We describe the main considerations that those conducting systematic reviews with NMA should bear in mind, including decisions regarding grouping interventions into analysis nodes, and testing the assumptions that assure the validity of NMA. We explain and illustrate how both systematic reviewers and users should draw conclusions from NMA that are appropriate and useful for decision-making. Finally, we provide a list of tools that facilitate the conduct and interpretation of NMAs.