All Times EDT
Keywords: Clinical Trial, Patient Accrual, Enrollment Modeling
Patient accrual projection is a topic gaining more attention in the statistics literature in recent years. A number of methods have been proposed in this area. Some approaches are sophisticated but complicated to implement. We aim to implement a simple and robust empiric Bayes model that is suitable for practical use. We assume the underlying enrollment rate at each site constant over time, which is site-specific and comes from a common Gamma distribution. Choice of prior parameters can be data driven. We tested the model in a number of internal oncology trials with various enrollment patterns, which yield satisfactory results. Compared to a flexible nonparametric model (Zhang and Long, 2010), the new model was associated with a narrower credible interval as a result of parametric assumptions. With the caveat that the model prediction can be off when the model assumption was substantially violated. R codes were available upon request.
Reference: Zhang X and Long Q, Stochastic Modeling and Prediction for Accrual in Clinical Trials, Statistics in Medicine 2010, 29(6) 649-658. https://doi.org/10.1002/sim.3847