Network Meta-Analysis (ADDED FEE) — Professional Development Continuing Education Course
ASA , Health Policy Statistics Section
Comparative effectiveness usually involves evaluation of multiple interventions and may involve multiple outcomes measured at multiple times as well. Meta-analysis, whether of continuous or discrete outcomes, has in the past focused on summarizing the evidence comparing two treatments or classes of treatments. Recently, methods have been developed to integrate comparisons of multiple treatments into coherent models that allow simultaneous comparison of all treatments, combining the direct evidence from head-to-head studies with indirect evidence from trials that involve common comparators. The network models provide estimates of the relative effectiveness or harms of all included treatments, and a ranking with associated probability estimates. These methods depend on a crucial assumption that the direct and indirect evidence are compatible (consistency) and that treatments are mutually exchangeable across studies (transitivity). This course will introduce meta-analysis in the context of evidence-based science and will then outline the basic principles of network meta-analysis and assessment of the validity of its assumptions including the key role that potential effect modifiers play. Examples of its application to different types of outcomes, both efficacy and safety with discussion of incomplete data problems will be discussed.
Instructor(s): Christopher Schmid, Brown University, Thomas Trikalinos, Brown University
