Meta-analysis of individualized treatment rules via sign-coherency
This work addresses the challenge of generalizing ITRs across heterogeneous medical sites, which is crucial for improving patient outcomes and reducing side effects in personalized medicine, but it is incremental as it extends existing methodologies.
The authors tackled the problem of learning individualized treatment rules (ITRs) from multiple datasets with between-site heterogeneity, which can hurt model generalizability, by developing a meta-analysis method that jointly learns site-specific ITRs while borrowing information about feature sign-coherency. They applied the method to a large multi-center electronic health records database and extended existing ITR methodologies like A-learning and weighted learning to the multiple-sites setting.
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.