Interrater reliability of 2 sedation scales in a medical intensive care unit: a preliminary report

BACKGROUND: Critical care nurses must assess the effectiveness of sedatives and analgesic agents in order to titrate doses. OBJECTIVES: To measure the interrater reliability of 2 sedation scales used to assess patients in medical intensive care units. METHODS: The interrater reliabilities of the Motor Activity Assessment Scale and the Luer sedation scale were compared prospectively in 31 patients receiving mechanical ventilation in an 18-bed medical intensive care unit of a tertiary care institution. Three registered nurses, 1 clinical pharmacist, and 1 physician simultaneously and independently followed a standardized procedure to rate each patient by using the 2 scales. Scales were randomly ordered to counteract ordering effect. Analysis of variance with post hoc Duncan multiple range tests was used to detect bias; a correlation coefficient matrix was used to examine degree of association among raters; and the intraclass correlation coefficient was measured to control for multiple raters. RESULTS: No significant bias was detected with either scale. The Motor Activity Assessment Scale had less variation (Pearson r = 0.75-0.92) than did the Luer scale (Pearson r = 0.37-0.94) and had a stronger intraclass correlation coefficient (0.81 vs 0.79). CONCLUSIONS: The Motor Activity Assessment Scale showed the highest consistency among raters.