A study of the reliability of the Nociception Coma Scale

Abstract

Objective: In this study, we investigated the reliability of the Nociception Coma Scale which has recently been developed to assess nociception in non-communicative, severely brain-injured patients. Design: Prospective cross-sequential study. Setting: Semi-intensive care unit and long-term brain injury care. Subjects: Forty-four patients diagnosed as being in a vegetative state (n=26) or in a minimally conscious state (n=18). Interventions: Patients were assessed by two experts (rater A and rater B) on two consecutive weeks to measure inter-rater agreement and test-retest reliability. Main measures: Total scores and subscores of the Nociception Coma Scale. Results: We performed a total of 176 assessments. The inter-rater agreement was moderate for the total scores (k = 0.57) and fair to substantial for the subscores (0.33 ≤ k ≤ 0.62) on week 2. The test-retest reliability was substantial for the total scores (k = 0.66) and moderate to almost perfect for the subscores (0.53 ≤ k ≤ 0.96) for rater A. The inter-rater agreement was weaker on week 1, whereas the test-retest reliability was lower for the least experienced rater (rater B). Conclusions: This study provides further evidence of the psychometric qualities of the Nociception Coma Scale. Future studies should assess the impact of practical experience and background on administration and scoring of the scale.