The AIRE study--a trial involving over 2,000 patients with clinical heart failure after myocardial infarction--assessed different outcomes. The primary outcome was total mortality. The secondary outcome was time to death, or progression to severe/resistant heart failure, reinfarction or stroke. Tertiary analyses included hospitalization and mode of death. To obtain maximum quality in the validation of outcomes and to optimize the quality of the data in the study, an end-points committee reviewed all data following preset definitions prior to code break. All definitions used are reported here. Experience of the validation process demonstrated that there are wide variations in clinical practice and terminology not only among countries but also among individual investigators. Variation in the clinical handling and treatment of patients makes it imperative to validate outcomes centrally on the basis of available facts.