A seven-factor model of the Posttraumatic Stress Disorder (PTSD) Checklist for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (PCL-5) has been shown to be the best fitting model in community, military, and undergraduate samples. This study replicates and adds to these findings by comparing four-, six-, seven-, and higher order factor models via confirmatory factor analysis in a partial hospitalization program. Convergent and discriminant validity were assessed using measures of behavioral activation of ruminative worry. Participants (n = 228) were predominantly White (87%), heterosexual (80%), and female (86%), with a mean age of 36 (SD = 12.5). The mean total score of the PCL-5 in our sample was 55.4 (a = .90, SD = 1.4). Confirmatory factor analysis was conducted in R Version 3.5.0 using the lavaan and semTools packages. Factor analysis supports the superiority of the seven-factor model (comparative fit index = .95, Tucker-Lewis Index = .93, Gamma hat = .96, root mean square error of approximation = .05, standardized root mean square residual = .04) to the four-, six-, and higher order factor models. Convergent validity was demonstrated through latent correlations between the seven-factor model, ruminative worry, and avoidance. Discriminant validity was demonstrated through latent correlations between the seven-factor model, behavioral activation, and absence of worry. This study provides additional evidence in support of the seven-factor model of the PCL-5 in a partial hospitalization program for PTSD, including strong relationships between negative alterations in cognitions and mood, anhedonia, rumination, and decreased behavioral activation. Future research should assess the utility of using the seven-factor model with longitudinal data, ideally in conjunction with measures representing clinical targets (e.g., quality of life).
- Confirmatory factor analysis
- Posttraumatic stress disorder