Learning Explainable Treatment Policies with Clinician-Informed Representations: A Practical Approach

TL;DR

This paper presents a practical method for developing explainable treatment policies in digital health, emphasizing clinician-informed representations to improve efficacy, interpretability, and real-world applicability in remote patient monitoring.

Contribution

It introduces a pipeline that incorporates clinical domain knowledge into treatment policy learning, demonstrating improved performance over black-box approaches in real-world RPM settings.

Findings

Clinician-informed representations lead to more effective policies.

Policies from clinician-informed data are more efficient and interpretable.

Collaboration between ML and clinicians enhances digital health interventions.

Abstract

Digital health interventions (DHIs) and remote patient monitoring (RPM) have shown great potential in improving chronic disease management through personalized care. However, barriers like limited efficacy and workload concerns hinder adoption of existing DHIs; while limited sample sizes and lack of interpretability limit the effectiveness and adoption of purely black-box algorithmic DHIs. In this paper, we address these challenges by developing a pipeline for learning explainable treatment policies for RPM-enabled DHIs. We apply our approach in the real-world setting of RPM using a DHI to improve glycemic control of youth with type 1 diabetes. Our main contribution is to reveal the importance of clinical domain knowledge in developing state and action representations for effective, efficient, and interpretable targeting policies. We observe that policies learned from clinician-informed representations are significantly more efficacious and efficient than policies learned from black-box representations. This work emphasizes the importance of collaboration between ML researchers and clinicians for developing effective DHIs in the real world.