Abstract
Background: Many sedation scales and tools have been developed and compared for validity in critically ill patients. However, selection and use of sedation scales vary among intensive care units.
Objective: The aim of this study is to compare the reliability of 2 sedation scales-Ramsay Sedation Scale and Richmond Agitation-Sedation Scale (RASS)-in the adult intensive care unit.
Method: Four hundred twenty-five patients were recruited in the study. Informed consent had been obtained from each patient guardian/relative. However, only 290 patients (68.24%) completed the study and were independently assessed for sedation effect by investigator and bedside nurses simultaneously using Ramsay scale and RASS.
Results: Agreement between the nurse and investigator scores on Ramsay scale (weighted [kappa] = 0.449, P < .001) indicated weak level of agreement. Agreement between the nurse and investigator on RASS (weighted [kappa] = 0.879, P < .001) indicated a strong level of agreement. Cronbach [alpha] analysis showed that 10 items of RASS had an excellent level of internal consistency ([alpha] = .989) compared with good level of internal consistency of Ramsay scale ([alpha] = .828).
Discussion: Richmond Agitation-Sedation Scale showed excellent interrater agreement compared with weak interrater agreement of Ramsay scale. The results also support that RASS has consistent agreement with clinical observation and practice among different observers. The results suggest that use of RASS is linked to a more reliable assessment of sedation levels in the intensive care unit.