Validating a pain assessment tool in heterogeneous ICU patients: Is it possible?

Non-communicative adult ICU patients are vulnerable to inadequate pain management with potentially severe consequences. In German-speaking countries, there is limited availability of a validated pain assessment tool for this population. Aim The aim of this observational study was to test the German version of the Critical-Care Pain Observation Tool (CPOT) in a heterogeneous adult ICU population. Methods The CPOT's feasibility for clinical use was evaluated via a questionnaire. For validity and reliability testing, the CPOT was compared with the Behavioural Pain Scale (BPS) and patient's self-report in 60 patients during 480 observations simultaneously performed by two raters. Results The feasibility evaluation demonstrated high satisfaction with clinical usability (85% of responses 4 or 5 on a 5-point Likert scale). The CPOT revealed excellent criterion validity [agreement between CPOT and BPS 94.0%, correlation of CPOT and BPS sum scores r = 0.91 (P < .05), agreement of CPOT with patient self-report 81.4%], good discriminant validity [mean difference of CPOT scores between at rest and non-painful stimulus 0.33 (P < .029), mean difference of CPOT scores between at rest, and painful stimulus 2.19 (P < .001)], for a CPOT cut-off score of >2 a high sensitivity and specificity (93% and 84%), high positive predictive value (85%), and a high negative predictive value (93%). The CPOT showed acceptable internal consistency (Cronbach's alpha 0.79) and high inter-rater reliability [90% agreement, no differences in CPOT sum scores in 64.2% of observations, and correlation for CPOT sum scores r = 0.72 (P < .05)]. Self-report obtained in patients with delirium did not correlate with the CPOT rating in 62% of patients. Conclusion This is the first validation study of the CPOT evaluating all of the described validity dimensions, including feasibility, at once. The results are congruent with previous validations of the CPOT with homogeneous samples and show that it is possible to validate a tool with a heterogeneous sample. Further research should be done to improve pain assessment and treatment in ICU patients with delirium.