Clinical endpoints in pragmatic heart failure trials: From data collection to clinical endpoint classification

Abstract

Clinical endpoint classification (CEC)—that is, evaluation of clinical events using pre-defined criteria—is commonly conducted in clinical trial operations to ensure systematic and consistent assessment of endpoints needed to assess the intervention's safety and efficacy. This is particularly relevant for heart failure (HF) trials given the subjective decision-making around hospitalizations and variation in how worsening HF events are managed (both in hospital and in ambulatory settings). Several CEC strategies have been adopted to address the growing need for pragmatic clinical trials that enhance generalizability and minimize research burden on trial sites and patients. This review summarizes common CEC strategies including the traditional approach, investigator-reported endpoints, CEC using real-world data and CEC utilizing large language models. We summarize CEC strategies used in recent HF pragmatic trials and present challenges and considerations for CEC in HF pragmatic trials from the selection of clinical endpoints and data collection to CEC.