Answers are grounded in this paper and cite their sources.
Individualized treatment rules tailor treatment decisions to individual patients' characteristics. These rules can be estimated without bias using data from randomized clinical trials. However, the benefits derived from trials, which often have restrictive inclusion-exclusion criteria, may not reflect the true benefits when applied to a broader target population. To address this challenge, we utilize large real-world data from the target population to propose a consistent estimator to evaluate the benefits of the individualized treatment rules and obtain their inferential properties. Furthermore, we derive two novel composite features. These two features, replacing the potentially high-dimensional original features, preserve the consistency of the evaluation for the estimated treatment rule and yield the smallest variability for inference. Finally, we conduct extensive simulation studies to demonstrate the small-sample performance of the proposed estimators and further illustrate the method via application to tailor treatment of type 2 diabetic patients.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: