Analysis of Multiple Endpoints in Clinical Trials

Most human diseases are characterized by multidimensional etiology and the efficacy of an experimental treatment frequently needs to be assessed on multiple outcome measures — commonly referred to as endpoints. There is a variety of ways in which the contribution of each endpoint can be accounted for in the primary analysis, for example, the trial’s sponsor can treat endpoints as independent entities or as manifestations of a single underlying cause. The following examples illustrate the common approaches to analyzing multiple endpoints.