B54-40 Variation in Early Response Outcomes by Level of Care: A Baseline Quality-Improvement Analysis

Abstract Introduction Early Response Teams (ERTs) are designed to intervene before a patient’s clinical deterioration progresses to cardiac arrest. ERTs are activated when staff note abnormal vital signs or other signs of acute decline. Prior studies demonstrate higher Code Blue survival rates among patients on telemetry or progressive care units (PCUs) compared with medical-surgical (M/S) floors, however little is known about early response outcomes. Identifying differences has important implications for early recognition and resource allocation. We aim to characterize ERT activation outcomes by level of care to inform targeted quality-improvement (QI) initiatives. Methods This retrospective study included ERT activations at University Hospital in Newark, New Jersey, between January-July 2025. Events in the emergency department (ED), PACU, psychiatry and obstetrics wards, or radiology scanners were excluded. Activations were categorized by level of care: general wards (medical-surgical; telemetry; ED observation) versus progressive care units. The primary outcome was immediate disposition (transferred to higher level of care; remained on unit; expired). Descriptive statistics were generated for all variables. Group comparisons were performed using χ² tests. Risk ratios with 95% confidence intervals (CIs) estimated relative transfer differences. Logistic regression was used to calculate unadjusted odds ratios (ORs) for transfer by location. Analyses were conducted in Python using two-sided α = 0.05. Results 364 ERT activations were included. Events occurred most frequently on M/S units (n = 165, 45%) and PCUs (n = 158, 43%), with fewer on telemetry (n = 32, 9%) and observation units (n = 9, 2%). After ERT activation, 182 patients (50%) remained on their unit, 172 (47%) were transferred to a higher level of care, and 10 (3%) expired. Patients on general wards were significantly more likely to be transferred than those in the PCU (53% vs 39%; χ² = 5.38, p = 0.02; RR = 1.34). Unadjusted logistic regression confirmed this finding (OR 1.73, 95% CI 1.11-2.69, p = 0.015). Conclusion ERT activations on general wards were significantly more likely to be immediately transferred than those in PCUs, suggesting system-level variation in recognition and escalation practices. A possible interpretation is that PCUs are better equipped to stabilize high-acuity patients without transfer; alternatively, patients on M/S units may experience delayed recognition of deterioration. This data encourages various QI initiatives such as optimizing bed triage and enhancing ward-based stabilization capacity through simulation training and standardized escalation pathways. Future analyses should incorporate patient-level covariates and link ERT outcomes to hospital disposition. This abstract is funded by: None