When assessing the efficacy of a treatment in any Clinical Trial, it is recommended by the International Council for Harmonisation to select a single meaningful endpoint. However, a single endpoint is often not sufficient to reflect the full clinical benefit of a treatment in multifaceted diseases, which is often the case in rare diseases. Therefore, the use of a combination of several clinical meaningful endpoints is preferred. Combining endpoints on a test statistics level or on the level of p-values, in general, ignores the correlation between the endpoints, while combining information on the subject level in composite endpoints does allow to take correlation into account. Standard methods of analysis for a composite endpoint, however, are limited in a number of ways, not the least in the number and type of endpoints that can be combined as well as its poor small sample properties. The recently proposed class of non-parametric generalized pairwise comparison methods on the other hand allow for any number and type of endpoints and has good small sample properties. Moreover, this very flexible class of methods allows for prioritizing the endpoints by clinical severity, for matched designs and for adding a threshold of clinical relevance.
Our main focus will be on introducing the generalized pairwise comparison ideas and concepts as well as demonstrating its benefit in small sample trials, more particularly in including patient relevant endpoints, such as quality of life (QoL). Knowledge of common statistical inference terminology is essential to successfully follow the course.