Impact of ICU Admission Thresholds on Outcomes in Rapid Response System Activations: A Multicenter Study in Japan

Abstract

Background The variability in ICU admission rates for patients activated by the Rapid Response System (RRS) is substantial and differs significantly across institutions. This study explores the disparities in ICU admission thresholds and their impact on patient outcomes. Methods A multicenter retrospective observational study was conducted using a Japanese in-hospital emergency registry, focusing on patients for whom the RRS was activated from 2018 to 2022. We calculated the ICU admission rate (ratio of ICU admissions to RRS activations) and the Standardized ICU Admission Ratio (SIAR: ratio of actual to predicted ICU admissions) for each institution (N = 35). The relationship between SIAR and patient outcomes, specifically death or Cerebral Performance Category (CPC) at 30 days, was analyzed using multivariable analysis with the Generalized Estimating Equation (GEE) model. Results The study included 8,794 patients, with 26.9% admitted to the ICU. The median ICU admission rate was 0.33 (1st quantile: 0.21, 3rd quantile: 0.47), and the median SIAR was 0.98 (1st quantile: 0.75, 3rd quantile: 1.17). Univariable analysis indicated that a higher SIAR significantly correlated with a lower incidence of CPC ≥ 3 or death at 30 days (P = 0.037) and showed a trend towards lower mortality at 30 days (P = 0.059). The GEE model revealed that the odds ratio of SIAR for death at 30 days was 0.89 (95% CI = 0.72 to 1.09; P = 0.30), and for CPC ≥ 3 or death at 30 days was 0.78 (95% CI = 0.64 to 0.95; P = 0.015). Conclusions This study demonstrates a significant association between higher SIAR and improved patient outcomes, suggesting that lower ICU admission thresholds during RRS activations may enhance patient prognosis.