Data-driven Hospital Admission Control: A Learning Approach
Mohammad Zhalechian, Esmaeil Keyvanshokooh, Cong Shi, Mark P. Van Oyen (2023) Data-Driven Hospital Admission Control: A Learning Approach. Operations Research 0(0).
54 Pages Posted: 18 Aug 2020 Last revised: 13 Aug 2023
Date Written: July 16, 2020
Abstract
The choice of care unit upon admission to the hospital is a challenging task due to the wide variety of patient characteristics, uncertain needs of patients, and the limited number of beds in intensive and intermediate care units. The care unit placement decisions involve capturing the trade-off between the benefit of better health outcomes versus the opportunity cost of reserving higher level of care beds for potentially more complex patients arriving in the future. By focusing on reducing the readmission risk of patients, we develop an online algorithm for care unit placement under the presence of limited reusable hospital beds. The algorithm is designed to (i) adaptively learn the readmission risk of patients through batch learning with delayed feedback and (ii) choose the best care unit placement for a patient based on the observed information and the occupancy level of the care units. We prove that our online algorithm admits a Bayesian regret bound. We also investigate and assess the effectiveness of our methodology using hospital system data. Our numerical experiments demonstrate that our methodology outperforms different benchmark policies.
Keywords: online learning, contextual bandit, regret analysis, readmission, data-driven admission control
Suggested Citation: Suggested Citation