Pain assessment in critically ill patients who are intubated, sedated, and unable to verbalize their needs remains a challenge. No universally accepted pain assessment tool is used in all intensive care units.
To examine concurrent validation of scores on the Critical-Care Pain Observation Tool for a painful and a non-painful procedure and to examine interrater reliability of the scores between 2 nurse raters.
A prospective, repeated-measures within-subject design was used. A convenience sample of 35 patients was recruited to achieve enrollment of 30 patients during a 5-month period. Observational data were collected on patients intubated after cardiac surgery during routine turning and during dressing changes for central catheters.
Raters’ mean scores did not increase significantly during dressing changes (increase, +0.25; 95% CI, −0.07 to 0.57; P = .12) but did increase significantly during turning (increase, +3.04; 95% CI 2.11–3.98; P < .001). The degree to which mean scores increased was significantly greater during turning than during dressing changes (increase, +2.80; 95% CI, 1.84–3.75; P < .001). The Fleiss-Cohen weighted κ for the inter-rater reliability of the ratings of research nurses was 0.87 (95% CI, 0.79–0.94).
The results support previous research investigations on validity and reliability of the Critical-Care Pain Observation Tool for evaluating pain in intubated, critically ill adults.