Use of network meta-analysis in systematic reviews: a survey of authors.

Andrew W Lee

doi:10.1186/s13643-015-0174-4

Abstract

BackgroundThe reporting of network meta-analysis in systematic reviews has increased rapidly since 2009. This qualitative study was undertaken to identify authors’ perceptions of the use of these methods and of what standards for conduct and reporting should apply.MethodsThis is a survey of authors of systematic reviews reporting network meta-analysis.ResultsThe response rate was 32 % of the authors contacted, with these authors responsible for 34 % of the fully published systematic reviews identified within the period searched. Almost all authors would use the method again. Elements of reporting standards were proposed. Responses revealed some tensions between the view that use of network meta-analysis should be more easily accessible, particularly in the form of software tools, and concern that there is some inappropriate use of the methods, which wider use and greater accessibility could exacerbate.ConclusionsAuthors demonstrated strong support for adoption of standards for conduct and reporting. The elements of reporting standards proposed are consistent with those included in the 2015 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension statement. Adoption of standards for conduct and reporting will be a significant step towards clarifying what is appropriate use of the methods and what is not. This should be followed by the development of a critical appraisal tool to support end users of systematic reviews reporting network meta-analysis.

Highlights

The reporting of network meta-analysis in systematic reviews has increased rapidly since 2009
The PRISMA statement mentioned meta-analyses that combine direct and indirect comparisons but prior to 2015 did not contain recommendations for reporting that were specific to network meta-analysis (NMA) methodology
A basis for standards of conduct of NMA can be found in the National Institute for Clinical Excellence (NICE) Decision Support Unit’s Evidence Synthesis Technical Support Documents (TSDs) [7] and in reports on the interpretation and conduct of mixed treatment comparison (MTC) published in 2011 by the International Society for Pharmacoeconomics

Summary

Introduction

The reporting of network meta-analysis in systematic reviews has increased rapidly since 2009. Pairwise meta-analysis remains by far the most commonly used method of analysis within systematic reviews of healthcare interventions [1], outnumbering the use of network meta-analysis by more than 20:1, but the use of network meta-analysis (NMA) methods, including mixed treatment comparison (MTC), is increasing rapidly [2]. These methods combine direct and indirect evidence to address the frequent absence of randomised trials that directly compare all the interventions of interest.

Methods

Results

Discussion

Conclusion