Patient-Reported Outcomes as End Points in Heart Failure Trials

Heart failure is a growing health-care concern affecting tens of millions of individuals globally. Although traditional therapeutic strategies have focused on reducing the risk for hospitalization and mortality, the importance of patient-reported outcomes (PROs) in patients with heart failure is increasingly being recognized. Regulatory agencies consider PROs part of their evaluation of drugs and devices, and professional society guidelines may recommend interventions that improve PROs. However, for several reasons, the effect of interventions on PROs reported in heart failure trials currently is difficult to interpret. There is no consensus on the timing and frequency of PRO assessments. Moreover, it has been difficult to establish a minimal clinically important difference, that is, the minimal change in a PRO score that is meaningful to a patient. In addition, traditional methods of analyzing and reporting PROs such as comparison of mean differences across groups or responder analysis are prone to statistical artifacts and misinterpretation. This article presents an in-depth discussion of these issues, with the Kansas City Cardiomyopathy Questionnaire used as an example, to facilitate the use of PROs in heart failure research, regulatory, and clinical settings.