Cochrane Methods Training Event (PART 1)
Statistical considerations in indirect comparisons and network meta-analysis
This training event addressed methods and applications for the comparison of multiple interventions in Cochrane Reviews, including indirect comparisons and network meta-analysis.
Audience: This two-day course was targeted at a statistical audience, and specifically at statistical editors, statistical advisors and statisticians working with Cochrane Review Groups.
Content: Statistical methods for indirect comparisons and network meta-analysis; validity of underlying assumptions; bias and quality assessment; presentation of findings; computer practicals. Examples of previously published meta-analyses were used to illustrate the concepts.
Objectives: by the end of the course, participants should be able:
- to understand the principles, steps and statistical methods involved in indirect comparisons and network meta-analyses;
- to understand the biases that can distort indirect comparisons and network meta-analysis, including conflict among different sources of evidence, and ways to address these issues;
- to be aware of current thinking in presenting findings from indirect comparisons and network meta-analyses, including issues related to risk of bias and quality (within Summary of Findings tables); and
- to support Cochrane editorial bases in their support of review authors undertaking indirect comparisons and network meta-analysis.
Expectations for after the course: We hope that participants will be willing to be part of a network of methodologists who are able to provide support to each other in addressing the topics covered, and that they will continue to engage with the Comparing Multiple Interventions Methods Group.
Introduction and Course Overview
Review of Standard Meta-Analysis Methods and Introduction to Indirect Comparisons
Validity of Indirect Comparisons
Network Meta-Analysis with Meta-Regression
Problems Introduced by Multi-Arm Trials- Full Network Meta-Analysis
Identifying and Addressing Inconsistency in Network Meta-Analysis
Presenting and Evaluating the Evidence from a Network Meta-Analysis