Dimension-reduced outcome-weighted learning for estimating individualized treatment regimes in observational studies

TL;DR

This paper introduces a dimension reduction method for estimating individualized treatment regimes in observational studies, improving accuracy and interpretability by capturing treatment effect heterogeneity in a low-dimensional space.

Contribution

It proposes a novel sufficient dimension reduction approach combined with kernel-based covariate balancing, enabling more accurate and consistent estimation of optimal treatment regimes in high-dimensional settings.

Findings

Achieves universal consistency under mild conditions.

Demonstrates improved performance in simulations.

Successfully applied to ICU sepsis data.

Abstract

Individualized treatment regimes (ITRs) aim to improve clinical outcomes by assigning treatment based on patient-specific characteristics. However, existing methods often struggle with high-dimensional covariates, limiting accuracy, interpretability, and real-world applicability. We propose a novel sufficient dimension reduction approach that directly targets the contrast between potential outcomes and identifies a low-dimensional subspace of the covariates capturing treatment effect heterogeneity. This reduced representation enables more accurate estimation of optimal ITRs through outcome-weighted learning. To accommodate observational data, our method incorporates kernel-based covariate balancing, allowing treatment assignment to depend on the full covariate set and avoiding the restrictive assumption that the subspace sufficient for modeling heterogeneous treatment effects is also sufficient for confounding adjustment. We show that the proposed method achieves universal consistency, i.e., its risk converges to the Bayes risk, under mild regularity conditions. We demonstrate its finite sample performance through simulations and an analysis of intensive care unit sepsis patient data to determine who should receive transthoracic echocardiography.