The Modified Early Warning Score (MEWS) As a Predictor of Unanticipated ICU Admission at the Time of Rapid Response

Document Type


Publication Date


Publication Title

CHEST Journal





First page number:


Last page number:



PURPOSE: The Modified Early Warning Score (MEWS) is a validated predictor of patient deterioration and the need for admission to a higher level of care. No published studies exist regarding MEWS’ ability to prognose the need for intensive care at the time of clinical deterioration. Thus, we aimed to determine a MEWS threshold that obviates the need for medical intensive care unit (MICU) admission and to determine any relationship between MEWS, mortality, and MICU length of stay (LOS) after rapid response. METHODS: 489 of 1156 patient charts met inclusion criteria (i.e., unplanned non-MICU admission for a medical illness, clinical deterioration resulting in a rapid response, and documented MEWS parameters) resulting in two populations: Those transferred to MICU (n=229) and those not transferred (n=260) after rapid response. MEWS was calculated for each patient via documented chart parameters. Correlations were calculated among the MEWS parameters, t-tests were used to compare MEWS and LOS between groups, logistic regression was used to model potential predictors of hospital death, and ROC curve analysis was used to determine potential MEWS thresholds; analyses were completed in SPSS (v.25). RESULTS: ROC curve analysis indicated maximum sensitivity and specificity for MEWS values between 5 and 7, peaking at a MEWS of 7, suggesting a threshold for MICU admission after rapid response. The logistic regression to predict mortality suggested that MICU admission after patient deterioration portended to a 2.7 times higher risk of mortality (OR 2.7, CI 1.6-4.6). The MEWS parameters were highly correlated; MEWS vs. SBP (r=-0.18;p<0.001); MEWS vs. pulse (r=0.57;p<0.001); and MEWS vs. respiratory rate (r=0.33;p<0.001). Comparison of the two groups indicated that MEWS (t=-4.344;CI:-1.271,-0.479) and MICU LOS (t=-2.671;CI:-2.45,-0.366) were significantly greater in those who died after upgrade to the MICU. CONCLUSIONS: The MEWS is a valid predictor of the need for MICU admission at the time of patient deterioration resulting in a rapid response. A MEWS threshold of 7 may be more sensitive and specific when considering MICU admission than the current threshold of 5. We found that an elevated MEWS resulting in unplanned MICU admission carried a nearly 3-fold increased risk of mortality, differing from published values in studies evaluating patients prior to decompensation. Elevated MEWS at the time of upgrade correlates with an increased MICU LOS and risk of mortality, like studies performed in patients after MICU admission. We also determined MEWS components that correlated strongly with the potential need for MICU admission, including decreased systolic blood pressure, elevated heart rate, and increased respiratory rate. CLINICAL IMPLICATIONS: The MEWS is a bedside clinical tool that predicts a patient’s need for unplanned MICU admission, the risk of mortality, and MICU LOS after clinical deterioration.


MEWS; Patient deterioration; Critical care; MICU admission; Length of stay; Rapid response


Critical Care | Medical Specialties | Medicine and Health Sciences



UNLV article access

Search your library