Date of Award
2026
Degree Type
Dissertation
Degree Name
Doctor of Philosophy in Business Administration
Specialization
Operations and Supply Chain Management
Department
General Business
First Advisor
Maling Ebrahimpour
Second Advisor
Yuehwern Yih
Abstract
Hospitals operate as complex service systems in which patient flow, bed capacity, staffing, and discharge coordination must be carefully managed. Unplanned 30-day hospital readmissions disrupt these operations by generating unexpected demand for inpatient services, increasing congestion, and reducing the efficiency of care delivery. In the US, hospitals are financially penalized when readmission rates exceed regulatory thresholds, making effective readmission management both an operational and financial priority.
In light of the literature and practice, this study aims to develop personalized prediction models that estimate the probability of 30-day unplanned hospital readmission through continuous-time models and machine learning (ML) models using patient-level and hospital-level data from the Nationwide Readmissions Database (NRD). To achieve this goal, our study consists of three parts. The first part focuses on conducting a systematic literature review that analyzes and synthesize 30-day unplanned readmission estimation models at acute care hospitals (ACHs). The second and third parts aim to develop personalized readmission estimation models. The second part includes a two-state continuous-time model that includes Cox Proportional Hazard model to incorporate patient-level covariates. The third part focuses on implementing concept drift, online learning algorithms, and online feature selection in readmission estimation using patient-level and hospital-level covariates.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Pakdil, Fatma, "ESSAYS ON PREDICTING HOSPITAL READMISSIONS: PERSONALIZED RISK ESTIMATION USING MULTI-STATE MODELS AND MACHINE LEARNING" (2026). Open Access Dissertations. Paper 4551.
https://digitalcommons.uri.edu/oa_diss/4551