Health Policy and Health Services Research
Background: Emergency department closure can complicate patients' access to care and place clinical strain on the local health system. Less is known about the downstream effects on operational performance at remaining EDs. The objective of this study was to determine the impact of Memorial Hospital of Rhode Island's (MHRI) closure on operational benchmarks at two neighboring EDs, compared to a geographically distant control. We also aimed to describe ambulance diversion in RI pre- and post-closure.
Methods: We performed an interrupted time-series analysis utilizing data obtained from the electronic health records of a large health system. We examined three hospitals within the same state: a 247-bed community facility 3 miles from MHRI (Hospital 1); a 719-bed academic Level 1 trauma center 7 miles away (Hospital 2); and a 129-bed community facility 40 miles away (Control).
We compared a two-year pre-closure baseline period (11/30/2015-11/30/2017) with the first 11 months post-closure following a one-month washout period (1/1/2018-11/30/2018). We evaluated three ED operational metrics: monthly ED volume, length of stay (LOS), and left without being seen (LWBS) rates. We also describe patient demographic data and statewide ambulance diversion rates pre- and post-closure.
Results: In the first 11 months following MHRI closure, the average monthly ED volume at Hospital 1 increased by 14%. The ED LOS increased by 24% for discharged patients and 24% for admitted patients. The LWBS rate increased 123%. The average monthly time on diversion increased 374%. The share of patients from MHRI's primary catchment increased 38%. Hospital 2 also experienced increases in LOS: 15% for discharged and 18% for admitted patients. LWBS increased by 76%. There was a -2% decrease in ED volume. The time on diversion increased by 60%. The control hospital experienced smaller changes in metrics: LOS (5% for discharged, 4% for admitted), LWBS (-2%), and volume (1%).
Conclusion: In the first year after a community hospital closure in RI, LOS, LWBS rates, and diversion time increased at two nearby hospitals, but not at a distant control, suggesting resultant operational strain at local EDs. Knowing the extent of operational demand that results from a hospital closure may aid planning in future hospital closure situations. Further resource investment is required to support this increased clinical demand.