698

Abstract

Introduction: The Modified Early Warning Score (MEWS) is a tool for bedside evaluation based on five physiological parameters: systolic blood pressure, pulse rate, respiratory rate, temperature and mental status. It is a simple physiological scoring system that identifies patients at high risk of deterioration and helps in triaging patients to appropriate level of medical care. Raised MEWS is associated with increased morbidity and mortality [1]. Whether MEWS can be utilized as an adjunct tool along with clinical judgment, to identify patients who will benefit from early intensive care unit (ICU) admission (step down vs. step up) from the emergency room has not been looked at before. Methods: A retrospective, single-center study was conducted at a large tertiary care hospital. Patients admitted to a 12-bed medical step down unit (SDU) in a 1-year period were identified. 102 patients admitted to the SDU, needing transfer to the Medical ICU (MICU), within the first 24 hours were selected as CASES. 93 patients who were admitted to the SDU but did not need transfer to the MICU in the first 24 hours were randomly selected by an independent researcher. These served as CONTROLS. The data regarding demographics and comorbidities were collected based on documentation in the chart on admission. The MEWS at the time of admission to the unit was calculated retrospectively for both groups based on chart documentation. Using the Chi-square test, we compared whether having a higher MEWS (more than or equal to 5) correlated with transfer from SDU to MICU in the first 24 hours of admission. Results: There was no significant difference between the cases and controls with respect to their demographics and comorbidities, prior to the admission to the SDU. 40/102 (39.2%) of patients in the "CASE" group, and 13/93 (14.0%) in the "CONTROL" group, had a MEWS ≥ 5 (χ2 p<0.001). 40/53 (75.4%) of patients with a MEWS ≥ 5 and 62/142 (43.6%) of patients with MEWS < 5 were transferred to the unit. The odds ratio with 95% confidence interval (OR, 95% CI) that MEWS ≥ 5 was associated with going to ICU was 3.90 (1.95-8.0) p<0.001. A ROC curve was constructed with the above data, with the null hypothesis being "using MEWS ≥ 5 is equivalent to guessing," resulted in an area under the curve (AUC) of AUC = 0.659 ± 0.053, p=0.001. Conclusions: 1. Patients with MEWS ≥ 5 demonstrated a 3.9 times increased risk of transfer to the MICU within the first 24 hours of admission (p < 0.001). 2. A prospective randomized controlled trial is required to validate MEWS as a prediction tool for patients who may benefit from early MICU admissions, thus favoring a "STEP-DOWN" model. 3. The impact of MEWS ≥ 5 on length of stay and total cost of hospitalization will be studied in phase 2 of this project.