Data monitoring in the Heart Outcomes Prevention Evaluation and the Clopidogrel in Unstable Angina to Prevent Recurrent Ischemic Events trials: Avoiding important information loss

Abstract

Caution should always be exercised in considering early termination for any randomized clinical trial not only to avoid reacting to a temporary trend in the data, but also to avoid the loss of important treatment information concerning secondary outcomes, key subgroups, and patient safety data. For the Heart Outcomes Prevention Evaluation (HOPE) and Clopidogrel in Unstable Angina to Prevent Recurrent Ischemic Events (CURE) trials, their Data and Safety Monitoring Boards (DSMBs) ensured that this information was complete prior to making their decisions. For the HOPE trial, the DSMB did unblind the principal investigator early, who agreed with their assessment and stopped the trial before its scheduled conclusion. For CURE trial, the DSMB decided not to recommend early termination but allowed the trial to continue to collect important safety data to its planned end. DSMBs play a vital role in ensuring that information that effects clinical practice is ultimately obtained. © 2006 Springer-Verlag New York.