Optimizing Equity in Healthcare Access: An Integer Programming Approach to Emergency Department Closures and Travel Burden Disparities in Ontario

Emergency department (ED) closures in Ontario disproportionately affect rural communities. Patients in rural areas already face inequitable access to care, with many lacking a primary care provider. In many communities, the ED serves as both emergency and primary care. When an ED closes, local patients must still seek care, often traveling long distances to the next closest ED. This extra travel burden varies across Ontario, making ED closures have a disproportionate impact on rural communities in terms of travel burden. This thesis addresses these disparities by filling the gap in ED closure research and proposing an optimization framework as a solution. First, it examines the travel burden caused by closures and the resulting disparities in access between Northern and Southern Ontario. Data on ED closures from January 2022 to December 2024 were gathered from public sources. A network analysis mapped populations at the dissemination area (DA) level to the nearest EDs using the Google Maps API. Travel times to specialized services including computed tomography, magnetic resonance imaging, acute stroke thrombolysis centres, endovascular therapy centres, and intensive care units were also calculated. For historical closure data, travel burden for the closest DAs was calculated. A simulation then hypothetically closed all rural EDs one at a time, calculating the resulting travel burden. To compare regional disparities, closures in Northern and Southern Ontario were analyzed for both historical and simulated data. Post-closure travel burdens were also assessed across dimensions of the Ontario Marginalization Index. Second, this thesis introduces a framework for planning ED closures that minimizes travel burdens while balancing operational constraints. Two Integer Linear Programming (ILP) models were developed within this framework. The Capacitated p-Median Problem (CpMP) model identifies closures that minimize patient travel times, with results aligning to some historical closures (e.g., Durham Memorial and Chesley District), reflecting sensitivity to factors such as ED proximity and urban service density. The new Temporal Geo-constrained CpMP (TG-CpMP) extends this with multi-period planning and geographic equity mechanisms. Compared to ad hoc closures, TG-CpMP reduced affected DAs by 35–76% and lowered maximum travel times.

Department dc:contributor.department

Computing

Year dc:date.issued

2025

Author dc:creator

• Keene, Spencer

Advisor dc:contributor.supervisor

• Choudhury, Salimur

• Emergency department closures
• Rural healthcare access
• Travel burden
• Integer linear programming
• Healthcare equity
• Facility location optimization

Statement dc:rights

• Attribution-NonCommercial-NoDerivatives 4.0 International

Licence dc:rights.uri

• http://creativecommons.org/licenses/by-nc-nd/4.0/

Language dc:language.iso

eng

Handle dc:identifier.uri

https://hdl.handle.net/1974/34778

OAI identifier oai:identifier

oai:queensu.scholaris.ca:1974/34778