MetaInsight: An interactive user-friendly “point and click” web application to conduct network meta-analysis

Session Type
Workshop - training
Category
Network meta-analysis
Target audience
Statisticians, Reviewers, Clinicians, Decision makers
Level of difficulty
Basic
Description

Background Network meta-analysis allows the comparison of multiple competing interventions for a condition. Such analyses have become very popular due to their relevance for clinical decision-making. MetaInsight is a freely available user-friendly web-app with a “point and click” interface, including novel graphical displays to aid understanding and facilitate effective communication of the (often complex) results of a network meta-analysis to a range of stakeholders, including patients and decision makers. It is actively used worldwide with approximately 600 user hours per month and continuously being developed according to user needs. Objectives i. To provide an interactive demonstration of the functionality of MetaInsight v4 including user-interactive features and novel graphical visualisations ii. To provide a user’s experience of using MetaInsight for conducting network meta-analyses for NIHR and Cochrane reviews iii. For participants to gain hands-on experience of MetaInsight by completing a worksheet which will guide them through the analysis of a network meta-analysis dataset using the app Description This workshop will provide an introduction to MetaInsight for conducting network meta-analysis, including the perspectives of a current user of the app, and demonstrate the wide array of functions available within the app as well as ‘tips & tricks’ for running an analysis. Participants will be provided with ‘hands-on’ experience of using the app. They will be given an example dataset and worksheet designed to guide them, at their own pace, through the stages of conducting an analysis in MetaInsight. Topics covered will include: loading data; obtaining statistics using both Frequentist and Bayesian analyses; visualising the NMA results, including ranking statistics, using novel visual graphical displays; conducting sensitivity analyses; and a new feature for exporting a report of the analysis including all data inputs, results tables and figures, facilitating analysis transparency. There will be plenty of time provided for questions, discussion and feedback. Finally, there will be a brief discussion of the current limitations of the app and future planned developments. Participants will be required to bring a laptop or tablet, or may share with another participant.