JMIR Publications

ABSTRACT

Background:

Emergency department (ED) crowding compromises patient safety and increases clinical workload. Available metrics often fail to identify real-time physical and cognitive workloads driven by complex patient needs. Clinician perception may provide a complementary measure.

Objective:

To address current limitations by examining the relationship between real-time ED-perceived crowding (EDPC) and established objective metrics.

Methods:

We conducted a prospective study from 25 November to 13 December 2024 in a tertiary ED operating with reduced bed capacity and attending-only staffing due to a national medical workforce shortage. Twenty-eight clinicians (attending physicians and charge nurses) recorded EDPC hourly using a smartwatch-based Ecological Momentary Assessment system. Inter-rater agreement was assessed using Gwet’s AC2. Hourly median EDPC values were compared with objective crowding metrics derived from electronic health records: Emergency Department Occupancy Ratio (EDOR), National Emergency Department Overcrowding Score (NEDOCS), and Emergency Department Workload Index (EDWIN). Statistical associations were evaluated using Pearson correlation and time-lag analyses.

Results:

We analyzed 1,846 ED visits and 447 hourly evaluations. The inter-rater reliability of EDPC was high (Gwet’s AC2=0.742). EDOR showed the strongest correlation with EDPC (r=0.56). NEDOCS showed moderate correlations (r=0.47–0.51), whereas EDWIN showed minimal or no significant correlation. EDPC, EDOR, and NEDOCS followed similar trajectories, whereas EDWIN fluctuated with physician shift changes. EDPC was more strongly associated with maximum values than with mean or median values.

Conclusions:

EDPC aligns most closely with EDOR and NEDOCS, particularly peak values, indicating that clinician-perceived crowding reflects short-term surges. EDWIN shows poor alignment, suggesting that staffing-sensitive metrics may not accurately reflect perceived burden in environments with significant shifts in medical personnel and bed ratios. In conclusion, EDPC captures operational and cognitive burdens not fully represented by existing indices. Smartwatch-based real-time measurement of EDPC is feasible and may offer a complementary signal to monitor acute crowding.