Comparison of the prognostic value of coma scales among health-care professionals: a prospective observational study.

OBJECTIVE: To assess the predictive value of the Full Outline of Unresponsiveness (FOUR) Score and the Glasgow Coma Scale (GCS), investigating whether they are comparable in predicting short- and long-term functional outcome and if their predictive ability remains unaffected by the raters' background and experience. METHODS: Patients treated in the Neurosurgery Department and the Intensive Care Unit in need for consciousness monitoring were assessed between October 1st, 2018, and December 31st, 2020, by four raters (two consultants, a resident and a nurse) using the two scales on admission and at discharge. Outcome was recorded at discharge and at 6 months. Areas under the receiver operating characteristic curve (AUCs) were calculated for the prediction of mortality and poor outcome, and the identification of coma. RESULTS: Eighty-six patients were included. AUCs values were > 0.860 for all outcomes and raters. No significant differences were noted between the two scales. Raters' experience did not affect the scales' predictive value. Both scales showed excellent accuracy in identifying comatose patients (AUCs > 0.950). The difference between admission and discharge values was not a reliable predictor. CONCLUSION: Both the FOUR Score and GCS are reliable predictors of short- and long-term outcome, with no clear superiority among them. The application of the FOUR Score by inexperienced raters is equally reliable, without influencing negatively the predictive value.

Clinimetric study and review of the Reaction Level Scale.

OBJECTIVES: Although the Reaction Level Scale (RLS) is still used for the assessment of the level of consciousness in distinct centers, its clinical characteristics and significance have been incompletely researched. In the current study, the clinimetric properties, the prognostic value, and the impact of the raters' background on the application of the RLS, in comparison with the Glasgow Coma Scale (GCS), are investigated. MATERIALS AND METHODS: A systematic review on the available clinical evidence for the RLS was first carried out. Next, the RLS was translated into Greek, and patients with neurosurgical pathologies in need of consciousness monitoring were independently assessed with both RLS and GCS, by four raters (two consultants, one resident, and one nurse) within one hour. Interrater reliability, construct validity, and predictive value (mortality and poor outcome, at discharge and at 6 months) were evaluated. RESULTS: Literature review retrieved 9 clinimetric studies related to the RLS, most of low quality, indicating that the scale has not been thoroughly studied. Both versions of the RLS (original and modified) showed high interrater reliability (κw >0.80 for all pairs of raters), construct validity (Spearman's p > .90 for all raters), and prognostic value (areas under the curve >0.85 for all raters and outcomes). However, except for broader patients' coverage, it failed to show any advantage over the GCS. CONCLUSIONS: The RLS has not succeeded in showing any advantage over the GCS in terms of reliability and validity. Available evidence cannot justify its use in clinical practice as a substitute to the widely applied GCS.

Cross-Cultural Adaptation and Validation of the Greek Version of the "Full Outline of Unresponsiveness Score": A Prospective Observational Clinimetric Study in Neurosurgical Patients.

BACKGROUND: The Full Outline of Unresponsiveness (FOUR) score is a clinical instrument for the assessment of consciousness which is gradually gaining ground in clinical practice, as it incorporates more complete neurological information than the Glasgow Coma Scale (GCS). The main objectives of the current study were the following: (1) translate and cross-culturally adapt the FOUR score into Greek; (2) evaluate its clinimetric properties, including interrater reliability, internal consistency, and construct validity; and (3) evaluate the reliability of assessments among health care professionals with different levels of experience and training. METHODS: The FOUR score was initially translated into Greek. Next, patients with neurosurgical pathologies in need of consciousness monitoring were independently assessed with the GCS and FOUR score within one hour by four raters who had different levels of experience and training (two board-certified neurosurgeons, a neurosurgery resident, and a registered nurse). Interrater reliability, internal consistency, and construct validity were evaluated for the scales using weighted Cohen's κ (κw) and intraclass correlation coefficients (ICC), Cronbach's α and Spearman's ρ values, respectively. RESULTS: A total of 408 assessments were performed for 99 patients. The interrater reliability was excellent for both the FOUR score (ICC = 0.941) and GCS (ICC = 0.936). The values of κw exceeded 0.90 for all pairs, suggesting that the FOUR score can be reliably applied by raters with varying experience. Among the scales' components, FOUR score's brainstem and respiratory items showed the lowest, yet high enough (κw > 0.60), level of agreement. The interrater reliability remained excellent (κw > 0.85, ICC > 0.90) for all diagnosis and age groups, with a trend toward higher FOUR score values in the most severe cases (ICC = 0.813 vs. 0.723). Both the FOUR score and GCS showed high internal consistency (Cronbach's α > 0.70 for all occasions). The FOUR score correlated strongly with GCS (Spearman's ρ > 0.90 for all raters), suggesting high construct validity. CONCLUSIONS: The Greek version of the FOUR score is a valid and reliable tool for the clinical assessment of patients with disorders of consciousness. It can be applied successfully by nurses, residents, and specialized physicians. Therefore, its use by medical practitioners with different levels of experience and training is strongly encouraged.