JMIR Public Health Surveill. 2026 Aug 11;12:e87597. doi: 10.2196/87597.
ABSTRACT
BACKGROUND: COVID-19 profoundly affected emergency department (ED) services and patient outcomes. However, changes in ED length of stay (LOS) and intensive care unit (ICU) admission risk before, during, and after the pandemic remain poorly understood.
OBJECTIVE: We investigated and compared ED LOS and ICU admission risk among patients in the ED resuscitation area in 2019, 2021, and 2023, respectively before, during, and after the pandemic.
METHODS: This single-center retrospective cohort study enrolled patients in the ED resuscitation area of a regional emergency center in central China. ED LOS was compared across pandemic phases using the Kruskal-Wallis H test with Bonferroni correction. Cox proportional hazards models were applied to evaluate the association between phases and ICU admission risk. Sensitivity analyses included 2021 stratification, exclusion of extreme ED LOS values, and E-value analysis to validate result robustness.
RESULTS: A total of 44,161 patients were included. Stratified analyses demonstrated significant temporal variations in ED LOS across disposition groups. Among discharged patients, median ED LOS increased from a prepandemic 248 (IQR 149-404.8) minutes to a midpandemic 262.0 (IQR 153-423.8) minutes (P=.047) and remained elevated in the postpandemic phase (median 258, IQR 152-451 minutes; P<.001). Among patients admitted to general wards, median ED LOS nearly doubled during the pandemic (from 186, IQR 96-431 to 350, IQR 188-610 minutes) and partially decreased to a postpandemic 293 (IQR 113-840) minutes (all P<.001). Among ICU-admitted patients, median ED LOS increased from a prepandemic 95 (IQR 57-169) minutes to a midpandemic 171 (IQR 77-306) minutes and a postpandemic 105 (IQR 50-215) minutes (all P<.003). The overall ICU admission rate was 15.8% (6982/44,161). Multivariate Cox regression with 2019 as the reference showed that the postpandemic phase was independently associated with an increased ICU admission risk (hazard ratio [HR] 1.079, 95% CI 1.012-1.150; P=.02), whereas the pandemic phase was not (HR 1.032, 95% CI 0.967-1.100; P=.34). In female patients, the postpandemic phase was associated with a higher ICU admission risk (HR 1.143, 95% CI 1.027-1.273; P=.02). All sensitivity analyses verified the robustness of the main findings (all P<.05).
CONCLUSIONS: The COVID-19 pandemic exerted differential effects on ED LOS across disposition groups. The postpandemic period was associated with an increased risk of ICU admission, especially among female patients, whereas no significant change was observed during the pandemic. As a novel gender-stratified analysis covering the full pandemic timeline, this study distinguishes itself from prior research by revealing the sustained impacts of pandemic control policies on emergency care. Our findings deliver key evidence for patient risk stratification and offer practical guidance to optimize health care resource allocation in the postpandemic era. Health care systems should strengthen early identification of high-risk ED patients and rationalize critical care resource deployment accordingly.
PMID:42579810 | DOI:10.2196/87597
