What question did this study set out to answer?

May 14, 2026Open Access

Glasgow Coma Scale (GCS) Score and Full Outline of UnResponsiveness (FOUR) Score in Predicting Outcomes in Patients Admitted to the Medical Intensive Care Unit of Patan Hospital

Key Points

This study aims to compare the predictive performance of GCS and FOUR scores for in-hospital mortality in MICU patients.
Prospective observational study including 340 MICU patients (182 survivors and 158 non-survivors).
Recorded GCS and FOUR scores upon admission and assessed discrimination using AUROC curves.
Conducted multivariable logistic regression to evaluate associations with in-hospital mortality.
FOUR score: AUROC 0.799 (95% CI: 0.753-0.844) vs. GCS: AUROC 0.731 (95% CI: 0.678-0.783), p<0.001.
FOUR score sensitivity: 68%, specificity: 77%, P=0.74; GCS sensitivity: 55%, specificity: 84%.
Both scores remained independently associated with in-hospital mortality post-adjustment.

Abstract

Background Accurate assessment of consciousness is essential for prognostication in critically ill patients. The Glasgow Coma Scale (GCS) is widely used but has limitations in intubated patients and does not assess brainstem reflexes or respiratory pattern. The Full Outline of UnResponsiveness (FOUR) score addresses these limitations. This study compared the predictive performance of GCS and FOUR scores for in-hospital mortality among non-traumatic medical intensive care unit (MICU) patients at Patan Hospital. Methods This prospective observational study included 340 MICU patients (182 survivors and 158 non-survivors). Admission GCS and FOUR scores were recorded. Discrimination was assessed using the area under the receiver operating characteristic (AUROC) curves with 95% confidence intervals (CIs), and optimal cutoffs were determined using Youden's index. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Calibration was assessed using Brier score, calibration slope and intercept, and expected calibration error. Multivariable logistic regression analysis was additionally performed to assess the independent association with in-hospital mortality after adjustment for clinically relevant covariates. Results The FOUR score demonstrated modestly better discrimination for the prediction of in-hospital mortality compared with GCS, with AUROC curves of 0.799 (95% CI: 0.753-0.844) and 0.731 (95% CI: 0.678-0.783), respectively. The difference in AUROC curves was statistically significant (DeLong p<0.001). Optimal cutoffs were 12.5 for FOUR and 10.5 for GCS. Sensitivity, specificity, PPV, and NPV were 68%, 77%, 72%, and 74% for FOUR and 55%, 84%, 75%, and 68% for GCS, respectively. Calibration assessment using Brier score, calibration slope and intercept, and expected calibration error demonstrated more favorable calibration for the FOUR score. In multivariable logistic regression analysis, both scores remained independently associated with in-hospital mortality after adjustment for age, sex, intubation status, comorbidities, and diagnostic category. Conclusions The FOUR score demonstrated modestly better discrimination and more favorable calibration than GCS for the prediction of in-hospital mortality among non-traumatic MICU patients. In addition to improved prognostic performance, the FOUR score offers practical advantages in intubated patients due to the inclusion of brainstem reflexes and respiratory assessment. Further multicenter external validation studies are warranted.

Ask AI

Helpful

Bookmark

View Full Paper