Active patient queue dynamics in a district hospital emergency centre: a cross-sectional analysis
Related Experiment Video
Updated: Nov 2, 2026
Clinical Application of Phase Angle and BIVA Z-Score Analyses in Patients Admitted to an Emergency Department with Acute Heart Failure
Published on: June 30, 2023
Mary Elizabeth Hoffe1, Clint Hendrikse1
1Division of Emergency Medicine, University of Cape Town, Cape Town, South Africa.
Introduction:
Emergency Centre crowding is a persistent challenge, particularly in low- and middle-income countries, where resource constraints exacerbate mismatches between demand and capacity. Clinician productivity is central to EC flow and is typically measured using patients per hour (PPH), but this does not capture concurrent workload. The active queue (AQ), operationally defined as unresolved clinician-attributed patient episodes at any point during a shift, may complement PPH by capturing the concurrent nature of EC clinical work. This study aimed to describe AQ dynamics at a district-level EC and identify the clinician, shift, and flow factors associated with variation in queue size.
Methods:
A retrospective cross-sectional secondary analysis was conducted using data from 1 289 clinician shifts across three four-week periods in 2019 at a South African district-level EC. AQ was derived from timestamped patient flow data. Associations with clinician, shift, and system-level factors were assessed using one-way ANOVA and Spearman's rank correlation.
Results:
Median AQ was 1.50 (IQR 1.00-2.25), with a median peak AQ of 3.00 (IQR 2.00-4.00). AQ varied significantly by clinician category, employment type, and shift type (all p < 0.001), with higher values among senior clinicians and during night shifts. AQ remained stable at a median of 2.00 during the active working period before declining in the final shift quarter, while PPH declined progressively throughout. AQ and PPH were moderately correlated (rho = 0.43, p < 0.001). AQ was positively associated with patients waiting at shift start (rho = 0.19, p < 0.001) but not with patient acuity, despite acuity strongly suppressing PPH (rho = -0.44, p < 0.001).
Conclusions:
Emergency clinicians routinely manage concurrent patient loads, with AQ demonstrating a consistent mid-shift plateau. AQ complements PPH by capturing the concurrent dimension of clinician workload that throughput measures alone cannot reflect, with relevance to staffing and operational planning in district-level EC settings.
