ACCURACY OF S REMS AS COMPARED TO RAPS IN PREDICTING THE PROGNOSIS IN NON-SURGICAL PATIENTS PRESENTING TO EMERGENCY DEPARTMENT IN A TERTIARY CARE HOSPITAL

Abstract

Early identification of non-surgical emergency patients at high risk of deterioration and death remains a major challenge, particularly in resource-constrained emergency departments. We prospectively evaluated and compared the prognostic performance of the Rapid Emergency Medicine Score (REMS) and Rapid Acute Physiology Score (RAPS) for predicting in-hospital mortality in 700 adults presenting with heterogeneous non-surgical emergencies at a tertiary care hospital in India. Both scores were calculated at initial emergency assessment and evaluated using risk stratification and receiver operating characteristic analysis. Overall, in-hospital mortality was 12.4%. Mortality increased progressively across REMS risk categories, from 0% in low-risk patients to 16.0% in those classified as high risk, while corresponding mortality with RAPS increased from 6.6% to 25.0%. Non-survivors had significantly higher REMS and RAPS scores than survivors (10.86 ± 2.48 vs. 8.08 ± 2.82 and 5.03 ± 2.03 vs. 4.02 ± 1.91, respectively; both P < 0.001). Importantly, REMS demonstrated superior discriminatory performance compared with RAPS (AUC 0.77 vs. 0.64). These findings indicate that incorporating age and oxygenation into physiological risk assessment provides additional prognostic information. REMS may therefore offer a simple, rapid, and clinically applicable approach for early mortality risk stratification and prioritization of monitoring and critical care resources in heterogeneous non-surgical emergency population.