Reinforcement Learning on AYA Dyads to Enhance Medication Adherence
🤖 AI Summary
Adolescent and young adult (AYA) patients and their care partners often exhibit declining medication adherence following hematopoietic stem cell transplantation, posing significant clinical challenges in transitional self-management.
Method: We propose the first multi-agent reinforcement learning (MARL) framework explicitly designed for clinical dyadic interventions. It comprises three domain-informed agents—modeling patient symptom management, caregiver support, and dyadic shared decision-making—optimized via hierarchical, knowledge-guided cooperative policy learning and reward shaping to deliver dynamic, personalized digital interventions. A dyadic simulator, trained on real-world clinical data, was used for validation.
Contribution/Results: Our approach improves simulated adherence rates by approximately 3% over random baselines—a statistically significant gain. This work pioneers the systematic integration of MARL into dyadic clinical intervention design, uniquely accommodating both individual heterogeneity and relational dynamics. It establishes a novel paradigm for precision digital support during the AYA transition to self-management.
Application Category
📝 Abstract
Medication adherence is critical for the recovery of adolescents and young adults (AYAs) who have undergone hematopoietic cell transplantation (HCT). However, maintaining adherence is challenging for AYAs after hospital discharge, who experience both individual (e.g. physical and emotional symptoms) and interpersonal barriers (e.g., relational difficulties with their care partner, who is often involved in medication management). To optimize the effectiveness of a three-component digital intervention targeting both members of the dyad as well as their relationship, we propose a novel Multi-Agent Reinforcement Learning (MARL) approach to personalize the delivery of interventions. By incorporating the domain knowledge, the MARL framework, where each agent is responsible for the delivery of one intervention component, allows for faster learning compared with a flattened agent. Evaluation using a dyadic simulator environment, based on real clinical data, shows a significant improvement in medication adherence (approximately 3%) compared to purely random intervention delivery. The effectiveness of this approach will be further evaluated in an upcoming trial.
Problem
Research questions and friction points this paper is trying to address.
Enhance medication adherence in AYAs post-HCT
Address individual and interpersonal adherence barriers
Personalize interventions using Multi-Agent Reinforcement Learning
Innovation
Methods, ideas, or system contributions that make the work stand out.
Multi-Agent Reinforcement Learning approach
Personalized intervention delivery
Dyadic simulator environment evaluation
🔎 Similar Papers
No similar papers found.
