Evaluating and Learning Optimal Dynamic Treatment Regimes under Truncation by Death
Machine Learning
2025-10-10 v1 Machine Learning
Methodology
Abstract
Truncation by death, a prevalent challenge in critical care, renders traditional dynamic treatment regime (DTR) evaluation inapplicable due to ill-defined potential outcomes. We introduce a principal stratification-based method, focusing on the always-survivor value function. We derive a semiparametrically efficient, multiply robust estimator for multi-stage DTRs, demonstrating its robustness and efficiency. Empirical validation and an application to electronic health records showcase its utility for personalized treatment optimization.
Cite
@article{arxiv.2510.07501,
title = {Evaluating and Learning Optimal Dynamic Treatment Regimes under Truncation by Death},
author = {Sihyung Park and Wenbin Lu and Shu Yang},
journal= {arXiv preprint arXiv:2510.07501},
year = {2025}
}
Comments
30 pages, 5 figures, 6 tables, The Thirty-Ninth Annual Conference on Neural Information Processing Systems