Posttraumatic stress disorder and measurement invariance in a sample of sexual assault survivors: Are symptom clusters stable over time?

Abstract

Many sexual assault victims develop posttraumatic stress disorder (PTSD) after their assault, the symptoms of which can last for months or years. Although severity of PTSD symptoms generally decreases over time, it is unclear whether symptom clusters decrease at a consistent pace or whether certain symptom clusters resolve more quickly than do others. Put in terms of measurement invariance, do all symptom clusters maintain stable contributions to the underlying construct of PTSD over time? This article responds to that question by analyzing longitudinal data collected from a diverse sample of sexual assault survivors. We conducted a longitudinal confirmatory factor analysis to assess measurement invariance of the Posttraumatic Stress Diagnostic Scale. We also tested the impact of time since the assault and revictimization between measurement occasions on posttraumatic stress symptoms. Results indicated that although 3 out of 4 symptom clusters remained stable, the contribution of the dysphoria symptom cluster to the latent factor of PTSD varied over time. Findings also suggested that whereas time since the assault was not a significant predictor of posttraumatic stress symptoms, revictimization between Wave 1 and Wave 2 surveys did predict posttraumatic stress at the 2nd measurement occasion. Trauma researchers and practitioners are advised to monitor item-level movement on dysphoria scales rather than focus on the average cluster score, because items in this symptom cluster may change in a less predictable way than do items in other symptom clusters. Lack of attention to this issue may lead to inaccurate or incomplete conclusions about how PTSD symptom clusters evolve over time. (PsycInfo Database Record (c) 2020 APA, all rights reserved).