Composite endpoints in clinical trials

Composite endpoints are often used in clinical trials, especially in the cardiovascular area. Decreases in sample size requirements, ability to assess the net effect of an intervention and to avoid bias in presence of competing risk are the most cited advantages for their use. However, there is a risk of misinterpretation when heterogeneity among components with respect to either importance, number of events, or magnitude of treatment effect exist. In the following review we present a conceptual discussion about the rationale and interpretation of such variables. Also, a user's friendly guide to interpret the results of clinical trials based on composite endpoints is presented. We also present an empirical study that provides evidence of the use of misleading composite endpoints in cardiovascular clinical trials.