Quebec 2013

Our Workshops at the 21st Cochrane Colloquium - Sept 19-23 2013

 - Comparing multiple treatments 1: Overview or intervention review?  - 

               Click here for slides from this workshop


• To introduce key issues in the use of indirect comparisons in Cochrane reviews, including the identification of an appropriate question, the construction of a network map, and the decision of whether an intervention review or an overview is the better format.

• To discuss methodological considerations when undertaking such reviews.

Description: Indirect comparisons can be useful in the synthesis of evidence on comparative effectiveness, particularly when the evidence from head-to-head trials is lacking or insufficient. A variety of methods are available, ranging from simple narrative methods to formal network meta-analysis. This is the first of two related workshops. It will focus on non-statistical aspects and will draw upon the results of a special meeting on this topic organized by the Comparing Multiple Interventions Methods Group in April 2012. We will introduce the concept of indirect comparison and describe the construction of a network of interventions. We will discuss appropriate questions for a review involving indirect comparisons and present a roadmap for deciding whether an intervention review or overview fits best. Lastly, we will discuss methodological considerations when undertaking such reviews. We will encourage participants to share their experiences and perspectives on this rapidly developing area of methodology.

 - Comparing multiple treatments 2: Statistical methods for network meta-analysis - 

Objectives: To understand the statistical methodology of network meta-analyses and the assumption of consistency

Description: This is the second out of two workshops on how to conduct Cochrane reviews that aim to compare more than two interventions and it is focused on statistical methodology for network meta-analysis. Network meta-analysis is the statistical methodology used to combine evidence in a network of trials that compare more than two interventions. The workshop will provide insight to network meta-analysis models that can be used to derive estimates for the relative effects of all treatments of interest. We will explore the different implementation alternatives through worked examples and we will discuss extensively the underlying assumption of consistency. We will present approaches to check for and incorporate inconsistency in the results and we will present applications of multiple-treatments meta-regression models. Finally, we will discuss concerns regarding the role of bias in network meta-analysis.