Skip to content
DraconiticResearch
Search Sources API
226,742 studies31 sourcesUpdated 43 min ago
PreprintarXiv licencearXiv

Risk-Aware Quantile Learning for Personalized Dynamic Treatment Regimes

Abstract

Sequential clinical decision-making often involves more than maximizing average efficacy. Clinicians may need to simultaneously optimize clinically relevant tails of the outcome distribution, control treatment-related risk, and choose among multiple treatment options. Existing quantile dynamic treatment regime (DTR) methods capture distributional features of treatment outcomes but remain largely restricted to efficacy-only objectives and binary treatments. To address these limitations, we propose Risk-Aware Quantile Dynamic Treatment Regimes (RQDTR), a unified framework that optimizes a prespecified quantile of the cumulative potential outcome while explicitly incorporating treatment-related risk. We also develop an angle-based formulation for jointly learning decision rules across multiple treatment categories. Our framework includes three interpretable subclasses: efficacy-only quantile learning, constraint-based learning with population-level risk control, and utility-based learning through a composite benefit-risk utility. Theoretically, we establish identification and oracle equivalence, surrogate and estimator consistency under increasing quadratic penalties, asymptotic risk feasibility, and finite-sample performance error bounds. Extensive simulation studies and applications to All of Us major depressive disorder and MIMIC-III sepsis data demonstrate that RQDTR improves tail-oriented efficacy and achieves more favorable benefit-risk trade-offs than existing quantile DTR methods.

Related

  1. Estimating Optimal Dynamic Treatment Regimes for Personalized Education: A Tutorial and Applications with Machine Learning
    Chenguang Pan, Yuxuan Li, Thomas Harris +1
    arXivPreprintCC BY 4.0Full textStatisticsarXiv
  2. Deformed Q-learning for High-Dimensional Dynamic Treatment Regimes
    Ming Cheng, Zhao Chen, Christina Dan Wang
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  3. Nonparametric Efficient Estimation of Dynamic Treatment Regimes with Competing Risks
    Nitya Shah, Laura D. Carbone, Howard A. Fink +2
    arXivPreprintCC BY 4.0Full textStatisticsarXiv
  4. Sharp Causal Bounds for Dynamic Treatment Regimes
    Alexander Alvarez, Sebastian E. Ferrando
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  5. Comparison of estimated personalized treatment rules under varying interference structures and treatment rates using dynamic weighted ordinary least squares
    Alexandra Mossman, Ranjani Somayaji, Sonya Heltshe +2
    BMC Medical Research MethodologyPeer-reviewedCC BY-NC-ND 4.0Full textStatisticsPMC
BibTeX
@misc{lei2026riskaware,
  title = {{Risk-Aware Quantile Learning for Personalized Dynamic Treatment Regimes}},
  author = {Chunyin Lei and Annie Qu},
  howpublished = {arXiv},
  year = {2026},
  month = nov,
  doi = {10.48550/arxiv.2608.05434},
  eprint = {2608.05434},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.05434},
}
RIS
TY  - UNPB
TI  - Risk-Aware Quantile Learning for Personalized Dynamic Treatment Regimes
AU  - Chunyin Lei
AU  - Annie Qu
PY  - 2026
DA  - 2026/11/13
PB  - arXiv
DO  - 10.48550/arxiv.2608.05434
UR  - https://arxiv.org/abs/2608.05434
AB  - Sequential clinical decision-making often involves more than maximizing average efficacy. Clinicians may need to simultaneously optimize clinically relevant tails of the outcome distribution, control treatment-related risk, and choose among multiple treatment options. Existing quantile dynamic treatment regime (DTR) methods capture distributional features of treatment outcomes but remain largely restricted to efficacy-only objectives and binary treatments. To address these limitations, we propose Risk-Aware Quantile Dynamic Treatment Regimes (RQDTR), a unified framework that optimizes a prespecified quantile of the cumulative potential outcome while explicitly incorporating treatment-related risk. We also develop an angle-based formulation for jointly learning decision rules across multiple treatment categories. Our framework includes three interpretable subclasses: efficacy-only quantile learning, constraint-based learning with population-level risk control, and utility-based learning through a composite benefit-risk utility. Theoretically, we establish identification and oracle equivalence, surrogate and estimator consistency under increasing quadratic penalties, asymptotic risk feasibility, and finite-sample performance error bounds. Extensive simulation studies and applications to All of Us major depressive disorder and MIMIC-III sepsis data demonstrate that RQDTR improves tail-oriented efficacy and achieves more favorable benefit-risk trade-offs than existing quantile DTR methods.
ER  -