Safe Individualized Treatment Rules with Controllable Harm Rates

2025/05/08 by Patricia Wu, Wu, Peng, Qingwu Jiang +5 · 2 citations
Mathematics · #Advanced Causal Inference Techniques #FOS: Computer and information sciences #Methodology (stat.ME) #Statistical Methods and Inference #Statistical Methods in Clinical Trials

paper · pdf · doi:10.48550/arxiv.2505.05308

openalex publication_date 2025/05/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Estimating individualized treatment rules (ITRs) is crucial for tailoring interventions in precision medicine. Typical ITR estimation methods rely on conditional average treatment effects (CATEs) to guide treatment assignments. However, such methods overlook individual-level harm within covariate-specific subpopulations, potentially leading many individuals to experience worse outcomes under CATE-based ITRs. In this article, we aim to estimate ITRs that maximize the reward while ensuring that the harm rate induced by the ITR remains below a pre-specified threshold. We first derive the explicit form of the oracle ITR. However, the oracle ITR is not achievable without strong assumptions, as the harm rate is generally unidentifiable due to its dependence on the joint distribution of potential outcomes. To address this, we propose two strategies for estimating ITRs with a harm rate constraint under partial identification and establish their large-sample properties. By accounting for both reward and harm, our method provides a reliable solution for developing ITRs in high-stakes domains where harm is a critical consideration. Extensive simulations demonstrate the effectiveness of the proposed methods in controlling harm rates. We apply the proposed method to analyze two real-world datasets from a new perspective, assessing the potential reduction in harm rate compared with historical interventions.

Citations

• Cross-World Assumption and Refining Prediction Intervals for Individual Treatment Effects
• Quantifying Individual Risk for Binary Outcome
• Policy Learning with Distributional Welfare
• Safe Policy Learning under Regression Discontinuity Designs with Multiple Cutoffs
• Personalized Decision Making -- A Conceptual Introduction
• What's the Harm? Sharp Bounds on the Fraction Negatively Affected by Treatment
• Orthogonal Policy Learning Under Ambiguity
• Optimal Decision Rules Under Partial Identification
• Explainable Reinforcement Learning for Broad-XAI: A Conceptual Framework and Survey
• Empirical Welfare Maximization with Constraints
• Conformal Inference of Counterfactuals and Individual Treatment Effects
• Towards optimal doubly robust estimation of heterogeneous causal effects
• On Robustness of Individualized Decision Rules
• Nonparametric Causal Effects Based on Incremental Propensity Score Interventions
• Policy Learning With Observational Data
• Generalized Random Forests
• Residual Weighted Learning for Estimating Individualized Treatment Rules
• Fast learning rates for plug-in classifiers under the margin condition
• On the Application of Probability Theory to Agricultural Experiments. Essay on Principles. Section 9
• Estimating causal effects of treatments in randomized and nonrandomized studies.

Cited by

• Pseudo-strata learning via maximizing misclassification reward
• Adaptive Data-Borrowing for Improving Treatment Effect Estimation using External Controls

Related

• A Comprehensive Framework for the Evaluation of Individual Treatment Rules From Observational Data
• Estimation of Treatment Harm Rate via Partitioning
• Doubly Robust Fusion of Many Treatments for Policy Learning
• Treatment Effect Estimation using Invariant Risk Minimization
• Efficient augmentation and relaxation learning for individualized treatment rules using observational data