A methodological systematic review of 494 published network meta-analyses

ID: 

2

Session: 

Poster session 1

Date: 

Monday 24 October 2016 - 10:30 to 11:00

Location: 

All authors in correct order:

Chaimani A1, Petropoulou M1, Nikolakopoulou A1, Salanti G2
1 Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Greece
2 Institute of Social and Preventive Medicine, University of Bern, Switzerland
Presenting author and contact person

Presenting author:

Anna Chaimani

Contact person:

Abstract text
Background: Previous empirical studies exploring the characteristics of networks of interventions have raised the need for improving the quality of network meta-analyses (NMA) applications. These empirical studies have included networks published up to the end of 2012 at the latest. Since then, several developments have been made in the field of NMA and many tutorial and guidance papers have been published.

Objectives: We aim to present how the methodological and reporting quality of NMA applications has evolved over the years, to monitor the rate of adoption for the new methodological developments and provide an updated overview of the characteristics of published networks.

Methods: We compiled a database of 494 published NMAs published up to April 2015. We updated the collection presented by Nikolakopoulou (1) to include NMAs published after 2012 and we extracted data on additional clinical and methodological network characteristics that had not been previously considered. We performed a descriptive analysis for all the characteristics we extracted from the eligible networks of interventions.

Results and Conclusions: Publication of NMAs has increased exponentially over the years. We found that the prevalence of NMAs that do not evaluate the transitivity or the consistency assumption has decreased considerably, and there is an important drop in using inappropriate methods to evaluate consistency after 2013. There is also an important increase after 2013 in the percentage of networks that compare only pharmacological interventions; this finding potentially indicates a tendency to narrow the inclusion criteria to increase the chances of a consistent network. This apparent improvement in the methodology employed in NMA applications could be the result of a proliferation of tutorials and methodological articles in 2012 and 2013.

1. Nikolakopoulou A, Chaimani A, Veroniki AA, Vasiliadis HS, Schmid CH, Salanti G. Characteristics of Networks of Interventions: A Description of a Database of 186 Published Networks. PLoS ONE. 2014 Jan 22;9(1):e86754.