Psychometric properties of the Child and Adolescent Trauma Screen (CATS) in a sample of Swedish children

Abstract

Objective The aim of this study was to investigate the psychometrics of the Swedish version of the Child and Adolescent Trauma Screen (CATS). This was to obtain access to an international instrument to identify symptoms of post-traumatic stress in children and adolescents according to the new Diagnostic and Statistical Manual for Mental Disorders, 5th edition (DSM-5). Method A total of 591 young Swedish schoolchildren aged 13–17 years old were given the CATS together with the Trauma Symptom Checklist for Children (TSCC), and the Linköping Youth Life Experience Scale (LYLES-Y). A clinical group of 42 children who had experienced a potential trauma was also given the same questionnaires. Statistical analyses were carried out for the purpose of reliability, different kinds of validity, and confirmatory factor analysis (CFA) concerning the construct of the CATS. Results The CATS showed good internal consistency on all four subscales from α = 0.73–0.89 and moderate stability (intra class correlation [ICC] = 0.57–68). The four-factor model for PTSD indicated good fit, reliability, and convergent validity. The CATS correlated strongly with the PTSD subscale on the TSCC (r = 0.82) and there were significant differences between the nonclinical and clinical groups. Conclusions The study shows that the Swedish translation of CATS has satisfactory psychometric properties, including acceptable sensitivity and specificity. The CATS could therefore be used as a screening tool both outside and within a clinical setting.