Abstract

Systematic reviewers conducting pairwise meta-analyses sometimes encounter multi-arm studies. To include these studies, in order to avoid a unit-of-analysis error, often two or more arms are combined or the control arm is split up. In this tutorial we present five different approaches that can be used. Particularly, we present a novel approach (Method 4) that to the best of our knowledge has not been presented before. We demonstrate their application on three selected data sets, discuss their scope of application and their advantages and limitations, and give recommendations.