Meta-analysis of individualized treatment rules via sign-coherency

TL;DR

This paper introduces a novel meta-analytic approach for learning individualized treatment rules across multiple sites, leveraging sign-coherency to improve model generalizability and adaptively tuning models for heterogeneous data.

Contribution

It develops a new method for multi-site ITR estimation that incorporates feature sign-coherency and adaptive model tuning, extending existing methodologies to handle site heterogeneity.

Findings

Method performs well under varying heterogeneity levels.

Application to electronic health records demonstrates practical utility.

Sign-coherency improves model robustness across sites.

Abstract

Medical treatments tailored to a patient's baseline characteristics hold the potential of improving patient outcomes while reducing negative side effects. Learning individualized treatment rules (ITRs) often requires aggregation of multiple datasets(sites); however, current ITR methodology does not take between-site heterogeneity into account, which can hurt model generalizability when deploying back to each site. To address this problem, we develop a method for individual-level meta-analysis of ITRs, which jointly learns site-specific ITRs while borrowing information about feature sign-coherency via a scientifically-motivated directionality principle. We also develop an adaptive procedure for model tuning, using information criteria tailored to the ITR learning problem. We study the proposed methods through numerical experiments to understand their performance under different levels of between-site heterogeneity and apply the methodology to estimate ITRs in a large multi-center database of electronic health records. This work extends several popular methodologies for estimating ITRs (A-learning, weighted learning) to the multiple-sites setting.