Characterizing Outcome Distributions of Dynamic Treatment Regimes

• Activity Number: 118 - Emerging Challenges in Precision Medicine
• Type: Topic Contributed
• Date/Time: Monday, July 29, 2019 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Daniel Lizotte*
• Companies: The University of Western Ontario
• Keywords: dynamic treatment regimes; reinforcement learning; prediction intervals; tolerance intervals; multiobjective optimization; sequential decision-making

Abstract

The goal of this work is to better convey the evidence for or against clinically significant differences in patient outcomes induced by different dynamic treatment regimes by examining not only differences in mean outcomes, but differences in outcome distributions. We present a framework for computing and presenting prediction regions and tolerance regions for the outcomes of a dynamic treatment regime operating within a multi-objective Markov decision process (MOMDP). Our framework draws on two bodies of existing work: one in computer science for learning in MOMDPs, and one in statistics for uncertainty quantification. We review the relevant methods from each body of work, present our framework, and illustrate its use in a precision medicine problem. Finally, we discuss potential future directions of this work for supporting sequential decision-making.