RESUMO
The use of patient-reported measures in assessing mental health symptoms is common in both the research and clinical fields. With regard to assessing posttraumatic stress symptoms, there are specific versions of measures designed for child and adolescent populations in accordance with the fourth and fifth editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV and DSM-5, respectively). Different clinical thresholds, numbers of items, and score ranges may present obstacles for clinicians and researchers attempting to compare self-report ratings across different versions of a measure. The current study aimed to produce a score conversion crosswalk between two child/adolescent self-report measures of posttraumatic stress disorder (PTSD): the UCLA PTSD Reaction Index for DSM-IV (RI-IV) and DSM-5 (RI-5). Using item response theory (IRT), we calibrated both measures separately to derive scaled scores. The discrimination parameters ranged from 0.57 to 2.08 (SE = 0.09-0.17) for RI-IV and from 0.73 to 2.11 for RI-5 (SE = 0.07-0.13). The scaled scores were connected with equipercentile linking. Total scores based on common items between the two measures were used as anchors to enhance the linking results. A total of 1,486 children and adolescents completed the measure: 571 respondents filled out the RI-IV and 915 respondents filled out the RI-5. The results allow linked scores to be compared to establish recommended clinical cutoffs and help elucidate the implications of changes in the diagnostic criteria for the measurement of self-reported PTSD symptoms in children and adolescents.
