A Bayesian Nonparametric Approach for Estimating Individualized Treatment-Response Curves

Abstract

We study the problem of estimating the continuous response over time to interventions using observational time series---a retrospective dataset where the policy by which the data are generated is unknown to the learner. We are motivated by applications where response varies by individuals and therefore, estimating responses at the individual-level is valuable for personalizing decision-making. We refer to this as the problem of estimating individualized treatment response (ITR) curves. In statistics, G-computation formula (Robins, 1986) has been commonly used for estimating treatment responses from observational data containing sequential treatment assignments. However, past studies have focused predominantly on obtaining point-in-time estimates at the population level. We leverage the G-computation formula and develop a novel Bayesian nonparametric (BNP) method that can flexibly model functional data and provide posterior inference over the treatment response curves at both the individual and population level. On a challenging dataset containing time series from patients admitted to a hospital, we estimate responses to treatments used in managing kidney function and show that the resulting fits are more accurate than alternative approaches. Accurate methods for obtaining ITRs from observational data can dramatically accelerate the pace at which personalized treatment plans become possible.

References in corpus (5)

• Doubly Robust Off-policy Value Evaluation for Reinforcement Learning
• A Framework for Individualizing Predictions of Disease Trajectories by Exploiting Multi-Resolution Structure
• Achieving a Hyperlocal Housing Price Index: Overcoming Data Sparsity by Bayesian Dynamical Modeling of Multiple Data Streams
• A Latent Gaussian Process Model with Application to Monitoring Clinical Trials
• Observational-Interventional Priors for Dose-Response Learning

Cited by in corpus (10)

• A Review of Challenges and Opportunities in Machine Learning for Health
• Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes
• Deep Counterfactual Networks with Propensity-Dropout
• Product Kernel Interpolation for Scalable Gaussian Processes
• Treatment-Response Models for Counterfactual Reasoning with Continuous-time, Continuous-valued Interventions
• Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders
• Neural Pharmacodynamic State Space Modeling
• Human-AI Collaboration with Bandit Feedback
• DeepRite: Deep Recurrent Inverse TreatmEnt Weighting for Adjusting Time-varying Confounding in Modern Longitudinal Observational Data
• Deep Bayesian Estimation for Dynamic Treatment Regimes with a Long Follow-up Time