Association between adopting emergency department crowding interventions and emergency departments' core performance measures

Abstract

ObjectivesTo estimate the association between adopting emergency department (ED) crowding interventions and emergency departments' core performance measures. MethodsWe analyzed the National Hospital Ambulatory Medical Care Survey (NHAMCS) data from 2007 to 2015. The outcome variables are ED length of stay for discharged and admitted patients, boarding time, wait time and percentage of patients who left ED before being seen (LWBS). The independent variables are whether or not a hospital adopted each of the 20 crowding interventions. Controlling for patient-level, hospital level and temporal confounders we analyze and report results using multivariable logit model. ResultsBetween 2007 and 2015, NHAMCS collected data for 269,721 ED visit encounters, representing a nationwide of about 1.18 billion separate ED visits. Of 20 crowding interventions we tested, using adopting bedside registration (OR = 0.89, 95% CI = 0.75–0.98, P < .05), electronic dashboard (OR = 0.86, 95% CI = 0.76–0.98, P < .05), kiosk check-in technology (OR = 0.56, 95% CI = 0.41–0.83, P < .001), physician based triage (OR = 0.86, 95% CI = 0.73–0.99, P < .05) full capacity protocol (OR = 0.91, 95% CI = 0.79–0.99, P < .05) are associated with decrease in the odds of prolonged wait time. Adopting kiosk check-in (OR = 0.55, 95% CI = 0.35–0.85, P < .05) is associated with a decrease in the odds of prolonged boarding time. Using wireless communication devices (OR = 0.77, 95% CI = 0.57–0.97, P < .05), bedside registration (OR = 0.77, 95% CI = 0.64–0.094, P < .05) and pooled nursing (OR = 0.84, 95% CI = 0.72–0.98, P < .05) are associated with decrease in the odds of a patient LWBS. ConclusionsMajority of interventions did not significantly associated with ED' core performance measures.