|
Abstract:
|
Incorporating historical control data to augment the control arm in randomized controlled trials (RCTs) is one way of increasing their efficiency and feasibility when adequate RCTs cannot be conducted. Recent work has focused on propensity score methodology to address prior-data conflict owing to measured covariates. We propose a conditional power prior approach weighted by standardized mortality ratio weights, the inverse of the conditional probability of membership in the historical trial multiplied by the conditional probability of membership in the current trial for each individual based on patient-level covariates, implying standardization of historical controls to the current trial population. A robust version of the proposed prior is also considered using two component mixture priors with the proposed and non-informative priors as components. Simulation results show that the proposed method yields reduced bias in treatment effect estimates, type I error at the nominal level, and improved power under scenarios with varying degrees of covariate imbalance between trials, providing meaningful implications for use of historical controls to facilitate the conduct of adequate RCTs.
|
