Network meta-analysis (NMA) is a relatively recent development, which extends principles of meta-analysis to the evaluation of multiple treatments in a single analysis. This is achieved by combining the direct and indirect evidence. Direct evidence refers to evidence obtained from randomized control trials (RCTs); for example, in a trial comparing treatments A and B, direct evidence is the estimate of relative effects between A and B. Indirect evidence refers to the evidence obtained through one or more common comparators.

in the absence of RCTs that directly evaluate A and B, interventions A and B can be compared indirectly if both have been compared to C in studies (forming an A-B-C “loop” of evidence). The combination of direct evidence and indirect evidence is called mixed evidence.