Abstract

Network Meta-analysis (NMA) is a generalization of pairwise meta-analysis that permits multiple comparisons (three or more treatment groups) in a single analysis. It includes both direct comparisons of treatments which have been compared head-to-head and also indirect comparison of treatments that have not been compared head-to-head, but do possess a common comparator thus accounting for a broader evidence base. A pairwise meta-analysis results in a single pooled estimate, whereas a NMA leads to the computation of several such estimates. The present article illustrates the procedure of NMA by including a simple practical example.