Analyzing Factors Affecting Emergency Department Length of Stay—Using a Competing Risk-accelerated Failure Time Model

Emergency department (ED) length of stay (LOS) is associated with ED crowding and related complications. Previous studies either analyzed single patient disposition groups or combined different endpoints as a whole. The aim of this study is to evaluate different effects of relevant factors affecting ED LOS among different patient disposition groups. This is a retrospective electronic data analysis. The ED LOS and relevant covariates of all patients between January 2013 and December 2013 were collected. A competing risk accelerated failure time model was used to compute endpoint type-specific time ratios (TRs) for ED LOS. A total of 149,472 patients was included for analysis with an overall medium ED LOS of 2.15 [interquartile range (IQR) = 6.51] hours. The medium LOS for discharged, admission, and mortality patients was 1.46 (IQR = 2.07), 11.3 (IQR = 33.2), and 7.53 (IQR = 28.0) hours, respectively. In multivariate analysis, age (TR = 1.012, P < 0.0001], higher acuity (triage level I vs level V, TR = 2.371, P < 0.0001), pediatric nontrauma (compared with adult nontrauma, TR = 3.084, P < 0.0001), transferred patients (TR = 2.712, P < 0.0001), and day shift arrival (compared with night shift, TR = 1.451, P < 0.0001) were associated with prolonged ED LOS in the discharged patient group. However, opposite results were noted for higher acuity (triage level I vs level V, TR = 0.532, P < 0.0001), pediatric nontrauma (TR = 0.375, P < 0.0001), transferred patients (TR = 0.852, P < 0.0001), and day shift arrival (TR = 0.88, P < 0.0001) in the admission patient group. Common influential factors such as age, patient entity, triage acuity level, or arrival time may have varying effects on different disposition groups of patients. These findings and the suggested model could be used for EDs to develop individually tailored approaches to minimize ED LOS and further improve ED crowding status.