Composite endpoints in large randomised cardiovascular outcome trials: use can improve statistical efficiency but may dilute treatment effects and complicate interpretation when driven by less meaningful events - Noticias de Vera Health