Interrater reliability of the Glasgow Coma Scale scoring among nurses in sub-specialties of critical care.

The Glasgow Coma Scale (GCS) is used as an assessment tool to measure the levels of consciousness and coma in patients. This research investigated the reliability of scoring the GCS among registered nurses (RNs) working in five different sub-specialty clinical areas of critical care; general intensive care, neurosurgical intensive care, coronary care, emergency room and post anaesthetic recovery room. Seven video recordings were made of six patients (one patient was recorded twice) having their level of neurological response assessed utilising the GCS. Seventy five RNs (15 from each sub-specialty) viewed each of the GCS assessments on the videotape and rated each patient on the scoring sheet provided. Analysis was performed for all RN subjects as a single group as well as separately for each of the five groups under investigation. The ratings for the first six videos were used to test interrater reliability and the scores from videotape four and seven (same patient) were used to calculate intrarater reliability. Based on comparison with expert scores, of the 75 participants, 38 responded correctly to eye opening responses; only 26 responded correctly to the motor response ratings. However, a better accuracy was achieved in the verbal response category with 67 participants responding correctly. Education qualifications and previous neurosurgical experience were statistically significant with regard to the nurses' accuracy of GCS assessment with p values of 0.004 and 0.043 respectively. The results were consistent with previously published studies demonstrating the motor response rating is most problematic in relation to rate accuracy.