(PS12-C69) A Latent Profile Analysis of Post-trauma Outcomes in Help-seeking Intimate Partner Violence Survivors: Discrepancies in Self-report and Clinician-assessed Pathology

Heterogeneity in trauma-related mental health outcomes, including PTSD, depression, and anxiety, makes understanding patterns of post-trauma symptom response imperative for both prevention and intervention. A collection of studies, using latent profile analysis (LPA) has explored patterns of symptom response, although most studies rely on self-report measures. Owing to discrepancies between self-report and clinician-assessed psychopathology, it is unclear whether derived profiles can be used to inform applied efforts. In the current report, latent profiles, derived from self-report measures, were compared with clinician-rated severity of PTSD, depression, and anxiety, in a sample of women who had experienced intimate partner violence. This sample may be particularly relevant for this investigation, given high levels of overall distress.  

Participants included 387 help-seeking women (M_age = 37.25, SD_age = 12.81) who endorsed experiences of IPV. Self-report measures included the Impact of Event Scale-Revised, the Beck Depression Inventory-II, and the Beck Anxiety Inventory. Clinician-administered interviews included the Clinician-Administered PTSD Scale for DSM-IV and the Anxiety Disorders Interview Schedule for DSM-IV.

Latent Profile Analyses revealed a 3-profile model as the optimal solution, characterized by Low (22.48%), moderate (37.98%), and high (39.53%) symptomology. Multinomial logistic regression comparing membership on clinician-rated symptom severity scores noted the high symptomology profile was characterized by the highest clinician-rated symptom severity scores. Higher clinician ratings of outcome variables increased likelihood of membership in the high versus moderate profiles. The moderate profile was characterized by moderate levels of clinician-rated symptom severity scores. Higher clinician ratings for outcome variables increased likelihood of membership in the moderate versus low profile. The same pattern emerged for likelihood of membership in the high versus low profiles such that higher clinician ratings for outcome variables increased likelihood of membership in the high versus low profile. All between-group comparisons were significant at p < .001, apart from the moderate versus high profiles regarding PTSD (p = .015).

Results align with prior findings denoting 3 distinct profiles of low, moderate, and high symptomology, reflecting the heterogeneity in post-trauma responses. Additionally, findings demonstrate that while some correspondence between self-report profiles and clinician assessed measures was observed, differences in relative severity of PTSD, depression, and anxiety emerged within each profile. These results suggest that self-report and clinician-administered measures may capture different depictions of which mental health issues are most salient following interpersonal trauma exposure.