Monitoring Rare Events During an Ongoing Clinical Trial

• Activity Number: 83 - SPEED: Survival Analysis
• Type: Contributed
• Date/Time: Sunday, July 29, 2018 : 4:00 PM to 4:45 PM
• Sponsor: Biometrics Section
• Author(s): Haley Hedlin* and Victoria Ding
• Companies: Stanford University and Quantitative Sciences Unit, Stanford School of Medicine
• Keywords: clinical trial; rare event; sequential monitoring

Abstract

We conduct a simulation study to evaluate approaches to monitoring rare events during an ongoing clinical trial. We consider the setting where a data and safety monitoring board or another entity is monitoring a rare event during the course of a two-arm study and desires guidelines around whether the event is occurring more in one arm than the other. We do not consider stopping rules, but instead present guidelines to guide the monitoring in combination with other accumulating information during the trial. Our work is motivated by a real trial where a rare safety event was of interest. Most existing monitoring methods assume sufficiently large sample sizes or number of observed events. We aim to provide useful guidelines to be used in monitoring rare events while information is accumulating during a trial. We evaluate the operating characteristics of the proposed approach over a range of simulated trial settings. Importantly, we focus on settings where the initial sample sizes are relatively small and the proportion experiencing events are very small. We present guidelines with desirable operating characteristics to guide rare event monitoring during an ongoing clinical trial.