Managing emergency care operations through optimal patient routing decisions and predictive demand estimation during public health crisis

This dissertation contributes to managing emergency-care operations. Within healthcare systems, managing emergency-care operations entail relatively higher levels of complexity due to uncertainties of demand, diversity of patient conditions, and complexities due to time and capacity pressures. Effectively managing the demand and the utilization of the capacities in emergency care operations has been a topic of interest in practice and in academic. The three studies use data-driven methods to improve the efficiency of the healthcare system from both the supply and demand sides. Here are the three pieces. First, from the supply-side perspective of managing the delivery of emergency care, we examine fast-track (FT) routing decisions inside the Emergency Department, which are critical components of the healthcare system. Using data from two Canadian hospitals, we find that FT routing decisions are not purely clinical-driven; rather, ED operational status related to congestion is also associated with FT routing decisions. We use an instrumental variable approach to quantify the impact of the FT routing decisions on patient outcomes (i.e., ED length of stay and revisit rate). Second, based on the findings from the first study, we develop a prescriptive decision model for FT routing of ED patients. We propose a multi-class queueing model to derive the optimal routing policy that balances access to care with quality of care. Furthermore, we introduce two heuristic policies that not only promise ease of implementation but also exhibit superior performance compared to the current policy employed in the hospitals under study. Finally, our third study focuses on the critical task of forecasting hospitalization demand during disruptive events that trigger global public health emergencies, such as the COVID-19 pandemic. The abrupt emergence of such epidemics presents significant economic and social challenges that necessitate prompt action from policymakers, with the healthcare system at the forefront. We propose a stochastic discrete-time compartmental model enhanced with a hospitalization compartment, in tandem with a Bayesian framework. Our aim is to predict the demand for hospitalizations more accurately, thereby enabling hospitals to allocate their resources more effectively. This provides insights to improve the management of emergency care capacities, beds, and ventilators, which are all crucial elements in responding to health crises.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Business Administration

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2023

Author dc:creator

• Hao, Shuai

Contributors dc:contributor

• Xu, Yuqian
• Subramanyam, Ramanath
• Anand, Gopesh
• Mukherjee, Ujjal

• Emergency Department
• Bayesian Inference
• Covid-19
• Empirical Healthcare
• Behavioral Operations
• Fast-track Routing
• Queueing

Statement dc:rights

• Copyright 2023 Shuai Hao

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/121229

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F121229