Daily, weekly, and monthly dynamics of crowding in emergency department of university clinical centre Serbia: a one-year analysis of NEDOCS score.

BACKGROUND

Emergency department (ED) crowding is increasingly recognized as a global public health issue, while evidences from South-East Europe remain limited. The aim of this study was to characterize daily, annual, weekly, and monthly patterns of ED crowding in a high-volume tertiary hospital.

METHODS

We conducted a retrospective observational study of ED crowding in the largest ED in Serbia over 360 consecutive days (8,640 one-hour intervals). Crowding was measured hourly using the National Emergency Department Overcrowding Score (NEDOCS), derived from routinely collected data from health information system and expressed on a 0-200 scale, later grouped into six standard categories. Descriptive analyses assessed annual distribution, circadian and weekly variation, seasonal trends, and patient age structure. Between-group differences in NEDOCS scores were assessed using the Kruskal-Wallis test. Effect sizes were estimated using eta-squared (η²).

RESULTS

The majority of patients in ED (43.14%) were older than 60. Almost 90% of patients (89.5%) waited up to 4 h for the first check up by doctor. The median annual NEDOCS score was 43.87 (IQR 51.92 ), corresponding to a "busy" operational state. Level 2 ("busy") occurred in 39.4% of hourly intervals and level 1 ("not busy") in 29.5%, while overcrowding (levels 4-6) was observed in 5.6% of intervals. Crowding was significantly higher during daytime than nighttime hours (p < 0.001), with a pronounced morning peak. Weekday crowding exceeded weekend levels, peaking on Tuesdays. Seasonal variation showed higher crowding during winter and transitional months.

CONCLUSIONS

ED crowding demonstrates predictable temporal patterns that mirror broader population health dynamics. Routine high-frequency monitoring may inform strategic staffing, adaptive capacity planning, and governance mechanisms aimed at strengthening emergency care performance and system resilience.