Course Level: Intermediate
Course Limit: 40
Format Requirements: The course will be based on lectures interspaced with simple pen & paper exercises, group discussions and demonstrations of the WinBUGS software, including how to use the recently published WinBUGS code for pairwise and network meta-analysis (available from http://www.nicedsu.org.uk/). The course will follow the material available in the NICE DSU Technical Support Documents in Evidence Synthesis (http://www.nicedsu.org.uk/) and forthcoming series of MDM papers (to be published in July 2013). Attendees should have a basic understanding of meta-analysis and regression models. A calculator will be required for some exercises.
Background: When more than two treatments have been compared in randomised controlled trials for a certain patient population, Network Meta-Analysis (NMA), also termed Mixed Treatment Comparisons, can be used to estimate the effectiveness of each treatment relative to every other in a coherent way, even if certain treatment pairs have not been directly compared. The results of this coherent analysis can then be used to inform decisions, whether based on clinical or cost-effectiveness. The need to correctly propagate parameter uncertainty and correlation structure through to decision uncertainty makes the use of Bayesian methods for NMA particularly attractive in a decision making context. The recent series of Technical Support Documents in Evidence Synthesis (http://www.nicedsu.org.uk/) and MDM papers provide a general framework for NMA, including how to explore and adjust for heterogeneity, how to test for inconsistency, synthesis to inform baseline natural history, embedding evidence synthesis in cost-effectiveness analysis, and a reviewers’ checklist.
Description and Objectives: We will go through the methods for NMA described in the recent series of NICE DSU Technical Support Documents in Evidence Synthesis (http://www.nicedsu.org.uk/) and forthcoming series of MDM papers, describing the concepts and assumptions and demonstrating how to fit NMA models for a variety of different data formats common to medical decision making problems in a Bayesian framework. Interpretation of results, incorporation into cost-effectiveness models and methods for assessing and explaining heterogeneity and inconsistency will also be described.
Using examples we will introduce NMA and the generic Bayesian framework, demonstrate how to fit the NMA models in WinBUGS using available code, discuss the implications of inconsistency and how to check for it, describe how to fit meta-regression models to explain heterogeneity and how to incorporate the NMA results into a cost-effectiveness analysis.
By the end of the course participants will:
- be able to understand the assumptions and principles of NMA;
- have seen a demonstration of NMA in WinBUGS;
- be able to find the resources required to estimate pairwise and network meta-analysis parameters using WinBUGS for a variety of outcome types commonly used to inform decision models;
- be able to interpret the results of NMA for different outcomes;
- be aware of techniques to explore heterogeneity and test for inconsistency;
- be aware of synthesis methods to inform baseline natural history;
- understand how the results from NMA in WinBUGS can be used to inform a cost-effectiveness model;
- be aware of a reviewers’ checklist tool for critical appraisal of published NMAs.