A Bayesian Machine Learning Approach for Optimizing Dynamic Treatment Regimes

Medical therapy often consists of multiple stages, with a treatment chosen by the physician at each stage based on the patient's history of treatments and clinical outcomes. These decisions can be formalized as a dynamic treatment regime. This article describes a new approach for optimizing dyn...

Full description

Saved in:
Bibliographic Details
Published in:Journal of the American Statistical Association Vol. 113; no. 523; pp. 1255 - 1267
Main Authors: Murray, Thomas A., Yuan, Ying, Thall, Peter F.
Format: Journal Article
Language:English
Published: United States Taylor & Francis 03.07.2018
Taylor & Francis Group,LLC
Taylor & Francis Ltd
Subjects:
Algorithms
Approximate dynamic programming
artificial intelligence
Backward induction
Bayesian additive regression trees
Bayesian analysis
Bayesian theory
Clinical outcomes
Cognitive style
Counterfactual thinking
Decision analysis
Decisions
equations
Gibbs sampling
Health services
Inference
Machine learning
Medical decision making
Medical treatment
Optimization
patients
Physicians
Potential outcomes
Regression analysis
Regression models
Sampling
Simulation
Statistical inference
Statistical methods
Statistics
Theory and Methods
therapeutics
Treatment methods
Treatment outcomes
Variables
Bayesian Additive Regression Trees
Gibbs Sampling
Potential Outcomes
Backward Induction
Approximate Dynamic Programming
ISSN:0162-1459, 1537-274X, 1537-274X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Medical therapy often consists of multiple stages, with a treatment chosen by the physician at each stage based on the patient's history of treatments and clinical outcomes. These decisions can be formalized as a dynamic treatment regime. This article describes a new approach for optimizing dynamic treatment regimes, which bridges the gap between Bayesian inference and existing approaches, like Q-learning. The proposed approach fits a series of Bayesian regression models, one for each stage, in reverse sequential order. Each model uses as a response variable the remaining payoff assuming optimal actions are taken at subsequent stages, and as covariates the current history and relevant actions at that stage. The key difficulty is that the optimal decision rules at subsequent stages are unknown, and even if these decision rules were known the relevant response variables may be counterfactual. However, posterior distributions can be derived from the previously fitted regression models for the optimal decision rules and the counterfactual response variables under a particular set of rules. The proposed approach averages over these posterior distributions when fitting each regression model. An efficient sampling algorithm for estimation is presented, along with simulation studies that compare the proposed approach with Q-learning. Supplementary materials for this article are available online.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Yuan’s and Murray’s research was partially supported by Award Number R01-CA154591 from the National Cancer Institute.
The work of the first three authors was partially funded by NIH/NCI grant 5-R01-CA083932.
ISSN:0162-1459
1537-274X
1537-274X
DOI:10.1080/01621459.2017.1340887