Discriminant validity and gender differences in DSM-5 posttraumatic stress disorder symptoms

Background The posttraumatic stress disorder (PTSD) literature is replete with investigations of factor structure, however, few empirical studies have examined discriminant validity and the moderating role of gender on factor structure and symptom expression. This study aimed to address these gaps. Methods An online, population-based study of 3175 Australian adults was conducted. This study analyzed data from 642 participants who reported a traumatic event. Overall, 10.2% (13.4% females, 7.6% males) met diagnostic criteria for current PTSD. Results Confirmatory factor analyses indicated that eight factor models provided excellent fit to the data. The DSM-5 model, anhedonia and hybrid models provided strong fit to the data, based on statistical fit indices and parsimony. The models' factors were significantly associated with a number of external correlates. Factor structure was gender invariant for the three models, albeit significant latent mean-level differences were apparent in relation to the intrusion/re-experiencing and alterations in arousal and reactivity factors. Bonferroni-adjusted Wald chi-square tests indicated significant gender differences in four DSM-5 PTSD symptoms: females reported significantly higher rates of negative beliefs, diminished interest, restricted affect and sleep disturbance symptoms compared to men. Limitations Response rate to the survey was low. However, the number of respondents who completed the survey was high and population weights were employed to account for self-selection biases and aid generalizability. Conclusions The findings provide support for the DSM-5, anhedonia and hybrid models compared to alternative models based on DSM-5 symptoms. Discriminant validity analyses indicated similar patterns of significant associations with the transdiagnostic factors, potentially suggesting that all the PTSD factors are related to non-specific distress. Further research investigating how gender influences PTSD symptom expression is warranted, including possible gender differences in symptom item interpretation.