B0423
Title: Nonparametric Bayesian Q-learning for optimization of dynamic treatment regimes in the presence of partial compliance
Authors: Indrabati Bhattacharya - Florida State University (United States) [presenting]
Ashkan Ertefaie - University of Rochester (United States)
Kevin Lynch - University of Pennsylvania (United States)
James McKay - University of Pennsylvania (United States)
Brent Johnson - University of Rochester (United States)
Abstract: Existing methods for the estimation of dynamic treatment regimes are limited to intention-to-treat analyses, which estimate the effect of randomization on a particular treatment regime without considering the compliance behaviour of patients. A novel nonparametric Bayesian Q-learning approach is proposed to construct optimal sequential treatment regimes that adjust for partial compliance. The popular potential compliance framework is considered, where some potential compliances are latent and need to be imputed. The key challenge is learning the joint distribution of the potential compliances, which is accomplished using a Dirichlet process mixture model. The approach provides two kinds of treatment regimes: (1) conditional regimes that depend on the potential compliance values; and (2) marginal regimes where the potential compliances are marginalized. Extensive simulation studies highlight the usefulness of the method compared to intention-to-treat analyses. The method is applied to the adaptive treatment for alcohol and cocaine dependence study (ENGAGE), where the goal is to construct optimal treatment regimes to engage patients in therapy.
