What is indirect comparison?
Indirect comparison can be used to compare treatments that have not been directly compared with each other in a head-to-head trial. It is often used when there is no evidence or insufficient evidence from head-to-head trials, or when more than two treatments are of interest.
Indirect comparisons are usually conducted using network meta-analysis, an extension of meta-analysis that includes more than two treatments. Network meta-analysis is also referred to as multiple-treatments meta-analysis.
Network meta-analysis includes indirect treatment comparison and mixed treatment comparison, although all of these terms are often used interchangeably.
Like meta-analysis, indirect comparison combines data from different studies (usually randomised controlled trials) in order to produce overall estimates of treatment effects. Basic assumptions required for indirect comparisons include a homogeneity assumption as per standard meta-analysis, a similarity assumption for indirect comparison and a consistency assumption for the combination of direct and indirect evidence. It is essential to fully understand these basic assumptions in order to produce valid indirect comparisons.
Indirect comparison is often part of the systematic review process. The validity of any indirect comparison also depends on the studies on which it is based.
The use of indirect comparison has increased rapidly in recent years, and indirect comparisons are now accepted by many health technology assessment agencies.