|
Abstract:
|
This presentation provides a practical framework toward Confirmability of studies comparing medical treatments. We show that, in randomized controlled trials (RCTs) with repeated measures, it is possible to separate between subjects biological variability, and within subject measurement variability. What we call the ETZ decomposition in fact further separates between subject variability into a baseline (before treatments) variability component and a trajectory (after treatment) variability component. With this decomposition, it becomes possible to assess separately the magnitude of potential adverse impact of each variability component on the confirmability of a promising Phase 2 study result in a Phase 3 study. A drug developer can then rationally decide whether it is worthwhile to invest in additional resources to reduce the variability of a particularly impactful component, to increase the confirmability of study results.
|
