Emergency Department Utilization Following a Community Hospital Closure: A Population-Based Comparative Study
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Importance
Hospital closures may alter access to emergency care. However, changes in utilization among residents of affected communities are less well characterized than changes in volume at surrounding hospitals.
Objective
To compare changes in emergency department (ED) utilization among Madera County residents following the closure of Madera Community Hospital (MCH) with changes among Fresno County residents, and to examine geographic differences by pre-closure reliance on MCH.
Design, Setting, and Participants
Retrospective population-based comparative study of annual California hospital Patient Origin/Market Share data, 2015-2024, with an exploratory comparative interrupted time-series analysis. Pre-pandemic years were 2015-2019, pandemic-affected pre-closure years were 2020-2022, and the two complete closure years were 2023-2024. Encounters were classified by county of patient residence, regardless of treating hospital location.
Main Outcomes and Measures
Combined ED encounters (ED-only plus inpatient-from-ED) and each component separately. County-level utilization was expressed as encounters per 1,000 residents using annual population estimates. Geographic comparisons by pre-closure MCH dependence used ZIP-code encounter counts.
Results
From 2022 to 2023, the first complete year following closure, combined ED utilization declined from 398.47 to 333.61 encounters per 1,000 Madera County residents (−16.3%), compared with 399.61 to 398.49 (−0.3%) among Fresno County residents. Madera ED-only rates declined 20.4%, while inpatient-from-ED rates increased 13.4%; however, the adjusted comparative analysis did not identify a consistent additional Madera-specific increase in inpatient-from-ED utilization. Exploratory regression estimated Madera-specific closure-period rate ratios of 0.871 for combined ED utilization and 0.848 for ED-only utilization. Combined ED encounter counts declined 18.4% in high-MCH-dependence ZIP codes versus 5.8% in low-dependence ZIP codes. Statistical inference was limited by the inclusion of only two counties, two complete closure years, and substantial overdispersion.
Conclusions and Relevance
The period following MCH closure was associated with a substantial reduction in ED utilization among Madera County residents relative to Fresno County residents, concentrated in ED-only encounters and communities previously most dependent on MCH. The observational design does not establish causation or determine whether reduced utilization reflected substitution to other care settings, decreased demand, or unmet medical need.
