Abstract
The factor structure of DSM‐5 posttraumatic stress disorder (PTSD) has been extensively debated, with evidence supporting the recently proposed seven‐factor hybrid model. However, few studies examining PTSD symptom structure have assessed the implications of these proposed models on diagnostic criteria and PTSD prevalence. In the present study, we examined seven alternative DSM‐5 PTSD models within a confirmatory factor analysis (CFA), using the Child PTSD Symptom Scale–Self‐Report for DSM‐5 (CPSS‐5). Additionally, we generated prevalence rates for each of the seven models by using a symptom‐based diagnostic algorithm and assessed whether substance abuse, depression, anxiety symptoms, and daily functioning were differentially associated with PTSD depending on the model used to derive the diagnosis. Participants were 317 adolescents aged 13–17 years (M = 15.93, SD = 1.23) who had experienced a DSM‐5 Criterion A trauma and/or childhood adversity. The CFA results showed good fit indices for all models, with the seven‐factor hybrid model presenting the best fit. The rates of PTSD diagnosis varied according to each model. The four‐factor DSM‐5 model presented the highest rate (30.6%), and the seven‐factor hybrid model presented the lowest rate (17.4%). Similar to the CFA analysis, the inclusion criteria for the diagnosis based on the hybrid model also presented the strongest associations with daily functional impairment, odds ratio (OR) = 1.48, 95% CI [1.25, 1.75]; and adverse childhood experiences, OR = 1.46, 95% CI [1.16, 1.82]. Research and clinical implications of these results are discussed, and suggestions for future investigation are presented.