Validation to Brazilian Portuguese of the coma recovery scale-revised.

Abstract

Background:  To improve the diagnostic accuracy of the state of consciousness of patients with severe brain injury, Giacino et al. introduced the Coma Recovery Scale (CRS) in 1991, which underwent revision in 2004, resulting in the revised CRS scale (CRS-R).

Objective:  To determine the concurrent validity, as well as inter- and intrarater agreement of the CRS-R's adaptation to Brazilian Portuguese.

Methods:  This study involved a sample of 30 patients with severe brain injury. Concurrent evaluations were also performed with the Glasgow Coma Scale (GCS) and the Full Outline of UnResponsiveness (FOUR) scale. A total of seven rehabilitation experts were recruited to assess the inter- and intrarater reliability agreement.

Results:  Interrater reliability was moderate to high for auditory, visual, motor, verbal, communication, and arousal subscales (Cohen weighted kappa = 0.765 to 0.892; p < 0.001). Significant inter and intrarater intraclass correlation coefficients were observed for the total CRS-R scores, all of which were statistically significant (p < 0.001). Also, total CRS-R scores exhibited a high correlation with the total GCS and FOUR scores, indicating acceptable concurrent validity (p < 0.001).

Conclusion:  The Brazilian Portuguese version of CRS-R can be reliably administered by trained examiners. This study demonstrated substantial to almost perfect interrater agreement for the total score and subscales, as well as high concurrent validity between the Brazilian Portuguese version of CRS-R and the other two standardized behavioral scales.