Paper Number
2356
Paper Type
Completed
Description
Preventive care, including routine check-ups and screenings, aims to avert severe illnesses and champion health equity. However, existing recommendations often neglect the need for personalization and patient convenience, resulting in significant underutilization. This study proposes a multi-objective reinforcement learning framework tailored for optimizing patient-centric diabetes-related preventive care, balancing patient convenience and treatment cost. Based on the electronic health records from over 500,000 patients, we show that the optimal preventive care rate could be fourfold the current rate. Our framework could cut annual patient costs by 1.1%, with more pronounced savings for groups such as young adults, the elderly, males, and diabetic patients. We further validate this method with the Michigan Model for Diabetes, a well-established diabetes progression simulation software. Our study contributes to the design of healthcare decision support systems, spotlighting the significance of personalization and the pressing need for value-based incentives to enhance preventive care adoption among targeted patient groups.
Recommended Citation
Guo, Tianjian; Bardhan, Indranil; and Wen, Wen, "Preventive Care Now or Pay Later? A Personalized Medicine Approach for Healthcare Management" (2023). ICIS 2023 Proceedings. 26.
https://aisel.aisnet.org/icis2023/ishealthcare/ishealthcare/26
Preventive Care Now or Pay Later? A Personalized Medicine Approach for Healthcare Management
Preventive care, including routine check-ups and screenings, aims to avert severe illnesses and champion health equity. However, existing recommendations often neglect the need for personalization and patient convenience, resulting in significant underutilization. This study proposes a multi-objective reinforcement learning framework tailored for optimizing patient-centric diabetes-related preventive care, balancing patient convenience and treatment cost. Based on the electronic health records from over 500,000 patients, we show that the optimal preventive care rate could be fourfold the current rate. Our framework could cut annual patient costs by 1.1%, with more pronounced savings for groups such as young adults, the elderly, males, and diabetic patients. We further validate this method with the Michigan Model for Diabetes, a well-established diabetes progression simulation software. Our study contributes to the design of healthcare decision support systems, spotlighting the significance of personalization and the pressing need for value-based incentives to enhance preventive care adoption among targeted patient groups.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
16-HealthCare