Confusion Assessment Protocol: Italian Cross-Cultural Adaptation and Validation

Background: This study validated the Italian version of the Confusion Assessment Protocol (CAP), a tool designed to assess Post-Traumatic Confusional State (PTCS) in patients with severe acquired brain injury (sABI) who are not evaluable with standard neuropsychological evaluations. Objectives: The primary aim of this study was to promote the CAP as a tool for assessing patients who are not still eligible for standard neuropsychological evaluation and to adapt it to Italian-speaking sABI patients by translating it into Italian and conducting a cross-cultural adaptation and evaluating its psychometric properties. The secondary objective was to correlate the CAP scores with broader functional scales, such as the Levels of Cognitive Functioning Assessment Scale (LCF) and Disability Rating Scale (DRS). Methods: A total of 42 sABI patients were enrolled at IRCCS Fondazione Santa Lucia. The CAP was translated and culturally adapted using international back-translation guidelines. Cross-cultural validity was assessed in 20 patients. The final version was administered by three trained raters over two days to evaluate inter- and intra-rater reliability. Results: The Italian version of the CAP demonstrated high internal consistency and substantial inter-rater reliability for key symptoms, including night-time sleep disturbances, decreased daytime arousal, and psychotic-type symptoms. Cognitive impairment showed moderate inter-rater agreement, likely due to symptom fluctuations typical of this recovery phase. The convergent validity of the CAP was confirmed through its correlations with the Levels of Cognitive Functioning (LCF) and the Disability Rating Scale (DRS), demonstrating its clinical utility in integrating cognitive and behavioral symptom assessments. Conclusions: The Italian version of the CAP is a reliable and valid tool for assessing PTCS in sABI. Future developments should address limitations related to symptom intensity, behavioral domains, and differential symptom weighting.