Safe Individualized Treatment Rules with Controllable Harm Rates

2025-05-08

AI Summary

In precision medicine, conventional individualized treatment rule (ITR) estimation based on the conditional average treatment effect (CATE) neglects individual-level harm, potentially worsening outcomes for some patients. To address this, we propose an optimal ITR estimation framework subject to an upper bound on the harm rate. First, we derive the explicit closed-form optimal ITR under this constraint. Second, we construct a provably consistent plug-in estimator. Third, we develop three robust estimation strategies under partial identification settings. Theoretical analysis establishes consistency and asymptotic normality of the estimator. Simulation studies confirm strict adherence to the prespecified harm-rate threshold. Applied to real-world right-heart catheterization data, our method significantly reduces harm relative to historical interventions while improving clinical safety. This work transcends the CATE paradigm by jointly maximizing efficacy and ensuring controllable, individualized harm—achieving a principled balance between benefit and safety.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Large-scale security measurements

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 derive the explicit form of the optimal ITR, propose a plug-in estimator when both the CATE and harm rate are identifiable, and establish the large-sample properties of the proposed estimator. Given that the harm rate is generally unidentifiable due to its dependence on the joint distribution of potential outcomes, we further propose {three} strategies for estimating ITRs with a controllable harm rate under partial identification. By accounting for both reward and harm, our method offers a reliable solution for developing ITRs in high-stakes domains where harm is crucial. Extensive simulations demonstrate the effectiveness of the proposed methods in controlling harm rates. We apply the method to the Right Heart Catheterization dataset, assessing the potential reduction in harm rate compared to historical interventions.

Problem

• Estimating ITRs that maximize reward while controlling harm rates
• Addressing unidentifiable harm rates in ITR estimation strategies
• Ensuring harm rates remain below pre-specified thresholds in treatments

Innovation

• Optimal ITR form derived for reward maximization
• Plug-in estimator for identifiable CATE and harm
• Three strategies for harm control under partial identification