🤖 AI Summary
To address the challenges of accuracy, interpretability, and computational efficiency in estimating individualized treatment rules (ITRs) from high-dimensional clinical data, this study systematically evaluates 16 state-of-the-art ITR methods under both randomized and observational study designs. We propose a novel pretreatment covariate screening strategy integrating causal inference principles, Lasso regularization, inverse probability weighting, A-/Q-learning, and model aggregation. Furthermore, we develop a large-scale simulation framework that, for the first time, characterizes method-specific performance boundaries across diverse data-generating mechanisms. Key contributions include: (1) substantial improvements in ITR estimation accuracy and clinical interpretability; (2) the first practical, high-dimensional ITR method selection guideline tailored for real-world clinical research; and (3) open-sourcing of the complete simulation codebase, providing an off-the-shelf decision-support toolkit for clinical investigators.
📝 Abstract
Individualized treatment rules, cornerstones of precision medicine, inform patient treatment decisions with the goal of optimizing patient outcomes. These rules are generally unknown functions of patients' pre-treatment covariates, meaning they must be estimated from clinical or observational study data. Myriad methods have been developed to learn these rules, and these procedures are demonstrably successful in traditional asymptotic settings with moderate number of covariates. The finite-sample performance of these methods in high-dimensional covariate settings, which are increasingly the norm in modern clinical trials, has not been well characterized, however. We perform a comprehensive comparison of state-of-the-art individualized treatment rule estimators, assessing performance on the basis of the estimators' accuracy, interpretability, and computational efficacy. Sixteen data-generating processes with continuous outcomes and binary treatment assignments are considered, reflecting a diversity of randomized and observational studies. We summarize our findings and provide succinct advice to practitioners needing to estimate individualized treatment rules in high dimensions. All code is made publicly available, facilitating modifications and extensions to our simulation study. A novel pre-treatment covariate filtering procedure is also proposed and is shown to improve estimators' accuracy and interpretability.
