Prognostic Value of the Richmond Agitation-Sedation Scale for Mortality and Admission in Older Emergency Department Patients.

Background and Objectives: Altered arousal may provide rapid prognostic information in older emergency department (ED) patients. Originally developed to monitor agitation and sedation in critically ill patients, the Richmond Agitation-Sedation Scale (RASS) provides a rapid measure of arousal. However, its prognostic value relative to the Glasgow Coma Scale (GCS) in older ED patients remains uncertain. We evaluated whether RASS was associated with clinical outcomes and provided information beyond GCS. Materials and Methods: This study was a single-center retrospective cohort study of nontraumatic ED patients aged ≥ 65 years from January 2024 through October 2025. Only the first ED visit per patient was included. The primary outcome was in-hospital mortality; secondary outcomes were hospital and intensive care unit (ICU) admission. RASS was modeled categorically to preserve deviations in both directions from 0. Multivariable logistic regression and DeLong AUC comparisons were performed. Results: Among 10,208 patients, 352 (3.4%) died, 2301 (22.5%) were hospitalized, and 856 (8.4%) were admitted to the ICU. Mortality increased from 2.2% at RASS 0 to 59.4% at -5 and 45.5% at +3. Hospital and ICU admission rates also increased in both directions from RASS 0 (all trend p < 0.001). After adjustment for age, sex, vital signs, and GCS, both reduced arousal (RASS ≤ -4: aOR, 4.93; 95% CI, 2.10-11.61) and agitation (RASS ≥ +2: aOR, 5.16; 95% CI, 2.46-10.85) remained associated with mortality. RASS and GCS had similar discrimination for mortality (AUC, 0.683 vs. 0.679; p = 0.72), whereas their combination improved discrimination (AUC, 0.712; both comparisons p < 0.001). Among patients with GCS 15, mortality was 9.9% with abnormal RASS versus 1.9% with RASS 0. Conclusions: RASS demonstrated a bidirectional association with adverse outcomes and provided prognostic information comparable to and complementary to GCS. Its rapid assessment may help identify high-risk older ED patients, including those with GCS 15.