Answers are grounded in this paper and cite their sources.
Background Emergency department (ED) crowding is a global problem, and boarding is one of its major manifestations. Multicenter studies using longitudinal operational data capable of distinguishing day-to-day fluctuations from persistent between-hospital differences remain scarce, particularly in public health systems. This study characterizes boarding and crowding variability across hospitals and over time, examining their associations with bed dynamics, admission conversion, and sector-specific occupancy. Methods We conducted a multicenter observational study using a hospital-day panel from the pre-intervention baseline of the “Lean nas Emergências” project. Ten Brazilian public EDs were followed for 120 consecutive days (November 2025-February 2026), yielding 1,200 potential hospital-days. Boarding was defined as the interval between admission authorization (AIH) issuance and arrival at the inpatient bed; crowding was measured with the National Emergency Department Overcrowding Score (NEDOCS). Mixed-effects models with Mundlak within-between decomposition were applied to separate daily within-hospital variation from between-hospital differences, allowing for the examination of associations with bed availability and admission conversion. Results A total of 45,340 ED visits and 16,543 hospital admissions were recorded. Median boarding time was 18.6 h (interquartile range [IQR], 3.9–61.5) and varied approximately 40-fold across hospitals, from 1.3 to 54.2 h. Approximately 76% of the estimated variability occurred between hospitals (intraclass correlation coefficient [ICC], 0.76; 95% CI, 0.47–0.87). Three hospitals remained at or above the upper bound of the NEDOCS interpretive scale for nearly the entire study period; among the seven hospitals with sufficient score variability, day-to-day within-hospital variation predominated (ICC, 0.26; 95% CI, 0.05–0.46), whereas analysis of untruncated NEDOCS scores across all ten hospitals showed substantial between-hospital heterogeneity (ICC, 0.91; 95% CI, 0.74–0.95). Previous-day bed release showed no statistically conclusive association with next-day boarding (estimated 0.65% reduction per bed; 95% CI, − 1.55% to + 0.26%) or admission conversion (OR, 1.003 per bed; 95% CI, 0.999–1.007); estimates were consistent in weighted and patient-level analyses. In contrast, each 1-percentage-point increase in ward occupancy was associated with a 0.9% increase in same-day boarding (95% CI, 0.4% to 1.5%). Admission conversion was also associated with greater daily acuity (OR, 1.18 per SD; 95% CI, 1.14–1.23) and a higher surgical/trauma case mix (OR, 1.04 per SD; 95% CI, 1.01–1.07). Conclusions Boarding exhibited persistent between-hospital heterogeneity, while crowding varied predominantly over time among the seven hospitals with sufficient score variability. Additionally, higher ward occupancy was associated with longer same-day boarding. No statistically conclusive association was identified between greater bed release and next-day boarding or admission conversion, reflecting complex throughput dynamics.
No takes yet. Share an insight, caveat, or question.
Filho et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: