C56-35 The Impact of Time of Day on an Early Response Team

Abstract Rationale The Early Response Team (ERT) system includes identification, assessment/intervention, review of data and outcomes, and lastly, governance and improvement. At our urban tertiary care hospital, medical staff identify clinical deterioration, and ERT responds on a 24-hour basis. Data suggests that activations are less likely during the early morning hours, and mortality increases in 7am hour while other studies suggest increased mortality between 6am-8am, and 11pm-12am. This study evaluates ERT outcomes by time of day, looks to identify patterns and potential areas for improvement. Methods A retrospective study at University Hospital in Newark, NJ analyzed Electronic Health Record data for all ERTs from January-July 2025. Events were grouped by time — shift change (6-10 AM/PM) vs. mid-shift. Primary outcomes were ERT frequency and transfers to a higher level of care. Descriptive statistics were collected and group comparisons were performed by using Chi-square testing. Risk ratios (RRs) with 95% confidence intervals (CIs) estimated relative transfer differences. Logistic regression models were applied to estimate unadjusted odds ratios (ORs) for likelihood of transfer to a higher level. Analyses were performed in Python with a threshold of α = 0.05. Results 387 ERT activations were analyzed; 143 (37%) occurred during change-of-shift and 244 (63%) during mid-shift hours. If activations were uniformly distributed across a 24-hour period—eight hours of change-of-shift and sixteen hours of mid-shift—129 and 258 events, respectively, would be expected. The observed counts did not differ significantly from expected values (χ² = 2.17, p = 0.14). Immediate transfer to a higher level of care occurred in 44% of change-of-shift activations and 51% of mid-shift activations. The difference in transfer rate between groups was not statistically significant (χ² = 1.58, p = 0.21). In unadjusted logistic regression, change-of-shift events were associated with lower—but nonsignificant—odds of transfer compared with mid-shift (OR 0.75, 95% CI 0.50-1.14, p = 0.17). Conclusion ERT activations during change-of-shift were not significantly different in frequency or disposition compared to mid-shift. While not statistically significant, a modest excess of ERT activations occurred during change-of-shift periods relative to expected frequencies. This may reflect transient vulnerabilities during handoffs or increased vigilance of overlapping staff. The trend toward fewer transfers during handoff periods may reflect either temporary stabilization from increased staff overlap or delayed escalation during transitions. These findings suggest broadly consistent response patterns across shifts but highlight shift transitions as an operational period worth targeting quality improvement review. This abstract is funded by: None