Parameters affecting length of stay in a pediatric emergency department: a retrospective observational study

Prolonged emergency department (ED) length of stay (LOS) is used as a proxy for ED overcrowding and is associated with adverse outcomes of patients requiring therapy and reduced patient satisfaction. Our aim was to identify and quantify variables which affect ED-LOS. Patients admitted to the pediatric ED of a large regional Swiss hospital during a 1-year period were analyzed for LOS (in minutes). Predictor variables included patient-associated parameters (time of admission and discharge, ED occupancy, triage score, diagnosis, and demographic data) and external factors (weekday, time, and season). A total of 4885 visits were included in a multivariable logistic regression analysis. Median LOS was 124 min. The most important factors associated with prolonged LOS were physician referral (adjusted odds ratio [OR], 1.97; 95% confidence interval [CI], 1.47-2.62); morning admissions, especially before noon (OR, 1.92; 95% CI, 1.23-3.07); and gastrointestinal infections (OR, 1.38; 95% CI, 1.08-1.76). Upper airway infections (OR, 0.37; 95% CI, 0.27-0.49) and triage level 5 (OR, 0.18; 95% CI, 0.06-0.61) were inversely associated with ED-LOS. Together with ED occupancy, these factors did significantly contribute to log LOS in a stepwise backward multiple regression model (p < 0.001).

Several parameters are associated with prolonged ED-LOS. Notably, morning arrivals represent possible targets for strategies to reduce LOS. What is Known: • Prolonged length of stay (LOS) may affect care delivered to admitted patients in the emergency department (ED) and is well studied in the setting of adult patients with high acuity conditions. • Little is known about parameters which impact LOS in European pediatric EDs. What is New: • Several predictors of prolonged LOS could be identified in a European pediatric setting. • Our results indicate that prolonged LOS is associated with modifiable factors like morning and summer admission, which have the potential to be addressed by modification in staffing, infrastructure, and higher attention to faster processing.