(PS10-33) PTSD, Depression, and Alcohol Use in Military Sexual Trauma Survivors: A Latent Profile Analysis

Abstract Text: Alcohol use and depressive disorders consistently present alongside posttraumatic stress disorder (PTSD), and their cooccurrence has been implicated in increased risk for functional impairment and poor outcomes, including suicide risk. These outcomes are heightened in survivors of military sexual trauma (MST; sexual assault or harassment experienced during military service), as interpersonal trauma is associated with increased symptom severity and comorbidity. As a heterogeneous condition, distinct symptom patterns of PTSD and its comorbidity might suggest different underlying etiology and different long-term health consequences. Latent profile analysis is a data-driven method to identify profiles of individuals based on reporting patterns on continuous measures. As such, it has been used to identify distinct typologies in PTSD-substance use comorbidity. However, profiles of PTSD, alcohol use, and depression have yet to be developed, despite the high prevalence of their cooccurrence, and no previous research has identified typologies of these outcomes in survivors of MST. Further, previous research has shown sex differences in PTSD, depression, and alcohol use in MST survivors, indicating that profile membership may differ by biological sex.

The proposed study seeks to address these gaps by using latent profile analysis to elucidate the distinct PTSD, depression, and alcohol use symptom patterns among MST survivors. Data will be drawn from an anonymous, online survey of 600 U.S. military service members and veterans (50% male) with history of MST. Participants completed measures of posttraumatic stress (PTSD Checklist for DSM-V, PCL-5), depression (Patient Health Questionnaire-9, PHQ-9), and alcohol use (Alcohol Use Disorder Identification Test-C, AUDIT-C), as well as a demographic questionnaire.

Latent profile analysis will be conducted in R using the tidyLPA package. To gain a deeper understanding of how these symptoms may be related to each other, PTSD will be included at cluster level. Further, given previous research supporting a two-factor structure of the PHQ-9 (with somatic and cognitive-affective factors), a factor analysis will be done on the PHQ-9 in this data. Then, depression will be included in the latent profile analysis on a factor level. As we will be using the three-item AUDIT-C, alcohol use will be included as a total score reflecting alcohol use frequency. Since profiles of PTSD, depression, and alcohol use together have yet to be considered, we will fit 1 to 5 profile models and select the optimal model based on several fit indices (e.g., entropy, BIC, saBIC, LMR test) and interpretability. Then, if more than 2 profiles are found, we will explore biological sex as a predictor of profile membership.

Results of the proposed study may help shed light on what, specifically, drives different PTSD, depression, and alcohol use presentations in MST survivors, and if this differs by sex. Such different profiles have the potential to inform the development and refinement of treatment based on different symptom presentations and underlying etiology. Overall, the proposed study has the potential to contribute greatly to the literature on the treatment of posttraumatic comorbidity in MST survivors.