The prognostic value of the Modified Early Warning Score in critically ill patients: a prospective, observational study

Abstract

Context: The Modified Early Warning Score is a validated assessment tool for detecting risk of deterioration in patients at risk on medical and surgical wards.

Objective: To assess the prognostic ability of the Modified Early Warning Score in predicting outcome after critical care.

Design: A prospective observational study.

Setting: A tertiary care general ICU.

Patients: Five hundred and eighteen patients aged at least 16 years admitted to the ICU at Linköping University Hospital were included.

Intervention: The Modified Early Warning Score was documented on arrival at the ICU and every hour for as long as the patient was breathing spontaneously, until discharge from the ICU.

Main outcome measures: The primary endpoint was mortality in the ICU. Secondary endpoints were 30-day mortality, length of stay and readmission to the ICU.

Results: Patients with a Modified Early Warning Score of at least six had significantly higher mortality in the ICU than those with a Modified Early Warning Score <6 (24 vs. 3.4%, P < 0.001). A Modified Early Warning Score of at least six was an independent predictor of mortality in the ICU [odds ratio (OR) 5.5, 95% confidence interval (CI) 2.4-20.6]. The prognostic ability of the Modified Early Warning Score on admission to the ICU [area under the curve (AUC) 0.80, 95% CI 0.72-0.88] approached those of the Simplified Acute Physiology Score III (AUC 0.89, 95% CI 0.83-0.94) and the Sequential Organ Failure Assessment score on admission (AUC 0.91, 95% CI 0.86-0.97). A Modified Early Warning Score of at least six on admission was also an independent predictor of 30-day mortality (OR 4.3, 95% CI 2.3-8.1) and length of stay in the ICU (OR 2.3, 95% CI 1.4-3.8). In contrast, the Modified Early Warning Score on discharge from the ICU did not predict the need for readmission.

Conclusion: This study shows that the Modified Early Warning Score is a useful predictor of mortality in the ICU, 30-day mortality and length of stay in the ICU.