Cornell Assessment of Pediatric Delirium: Turkish translation and validation

Abstract

Background: Hypoactive delirium may go unrecognized unless routinely screened. At present, there is no valid screening tool for delirium in the Turkish language. This study was conducted to translate the Cornell Assessment of Pediatric Delirium (CAPD) into Turkish and to evaluate its validity and reliability. Methods: In this is validation study, CAPD assessments were conducted by pediatric intensive care unit nurses and compared with assessments by a child psychiatrist. Results: A total of 76 patients were included, 37 participants (48.6%) were younger than 24 months, and 22 participants (28.9%) had developmental disabilities. Prevalence of delirium was 25.0% (n=19). Inter-rater agreement for the identification of delirium by psychiatrists was strong and reliable, with a Cohen's kappa value of 0.86 (95% confidence interval [CI]: 0.72-0.99). Inter-rater reliability for nurses was also significant, with a Cohen's kappa of 0.74 (95% CI, 0.57-0.91). Inter-rater reliability ranged from 0.64 to 0.84 for each CAPD item except item 6, indicating reliable scoring. Sensitivity and specificity improved when the CAPD cut-off score was increased from 9 (100% and 95%, respectively) to 11 (100% and 98.02%, respectively). Subgroup analyses showed high sensitivity and specificity in patients with developmental delay (96%) and in patients under 2 years of age (96%) when the CAPD cut-off score was 9. However, specificity decreased slightly to 93% in patients under 6 months of age. Conclusion: The Turkish CAPD, the first delirium screening scale translated into Turkish, has demonstrated validity and reliability in screening for delirium in children of all ages, including those with developmental disabilities.