|
Abstract:
|
Conventional practice for monitoring safety in a clinical development program monitors the accumulating numbers of adverse events (AEs) reported from trials in the development program. Recent regulatory guidelines require monitoring the cumulative AE report counts to identify possible between-treatment adverse event risk differences even from blinded trials. Conventional statistical methods for assessing between-treatment AE risks cannot be applied when the trials are blinded. However, CUSUM charts can be used to monitor the accumulation of AE occurrences. CUSUM charts for monitoring AE occurrence are based on assumptions about the process generating the AE counts in a trial as expressed by informative prior distributions. This article describes the construction of control charts for monitoring AE occurrence based on statistical models for the processes, characterizes their statistical properties, and describes how to construct useful prior distributions. Application of the approach to two AEs of interest in a real trial gave nearly identical results for binomial and Poisson observed event count likelihoods.
|
