Comparison of Rapid-Response Systems Across Multisite Locations of a Pediatric Hospital System

Abstract

OBJECTIVES: Rapid response (RR) systems reduce mortality and cardiopulmonary arrests outside the ICU. Patient characteristics, RR practices, and hospital context and/or mechanism influence post-RR outcomes. We aim to describe and compare RR function and outcomes within our institution’s multiple sites. METHODS: We conducted a 3-year retrospective study to compare RR use, clinical characteristics, and outcomes between our hospital’s central campus (CC) and 2 satellite campuses (SCs). RR training and procedures are uniform across all campuses. RESULTS: Among the 2935 RRs reviewed, 1816 occurred during index admissions at the CC and 405 occurred at SCs. CC, when compared with SCs, had higher age at RR (3.2 years vs 1.4 years), prevalence of complex chronic conditions (62.4% vs 34.4%), surgical complications (20.2% vs 5%), severity of illness, and risk of mortality (P < .001). CC had higher daytime RR activations, longer time from admission to RR, and more activations by nurses (P < .001). Respiratory diagnoses were most prevalent uniformly, but cardiac, neurologic, and hematologic diagnoses were higher at CC (P < .001). Cardiac and/or respiratory arrests during RR and transfers to the ICU were similar. Cardiorespiratory interventions post-RR, hospital length of stay, and mortality were higher and ICU stay was shorter (P < .01) in the CC. Outcomes were mainly affected by patient characteristics and not RR factors on multivariate analysis. CONCLUSIONS: Patient illness severity, RR characteristics, and outcomes are significantly different in our multisite locations. Outcomes are predominantly affected by patient severity and not RR characteristics. Standardized RR training and procedures likely balance the effect of varying RR characteristics on eventual outcomes.