(PS4-B40) Disentangling How Our Treatments Work: Examining Time-lagged Change in PTSD Symptom Clusters in PE and CPT

Background: Prolonged Exposure (PE) and Cognitive Processing Therapy (CPT) are gold standard treatments for Post-Traumatic Stress Disorder (PTSD). However, the mechanisms of change in these therapies remain poorly understood. Better understanding of the temporal sequence of change in PTSD symptom clusters may inform our understanding of these mechanisms, including the possibility that reductions in one cluster contribute to reductions in another. The primary aim of the present study was to examine the lagged associations among symptom clusters during the first 12 weeks of treatment in individuals receiving PE or CPT.

Method: We examined weekly Post-Traumatic Stress Checklist - 5 (PCL-5) scores from adult patients with PTSD receiving either PE or CPT (n = 36) at an anxiety and traumatic stress outpatient psychiatry clinic. Analyses assessed PCL-5 scores at baseline and treatment weeks 1-12. We ran four time-lagged mixed-effect models to predict each of the four symptom clusters (i.e., (B) intrusions, (C) avoidance, (D) negative alterations in mood or cognitions, (E) hyperarousal). In each model, predictors were mean-centered time-lagged symptom clusters B-E, including the lagged outcome variable to account for autoregressive effects, treatment week, and treatment type.

Results: Intrusions were the only symptom cluster that prospectively predicted other symptom clusters the following week after controlling for autoregressive effects and the effects of other symptom clusters. Lower intrusions predicted lower avoidance (b = .09, p = .038), negative alternations in mood and cognitions (b = .29, p =  .013) and hyperarousal (b = 0.28, p = .002) the following week. Conversely, no other symptom cluster uniquely and prospectively predicted intrusions. 

Discussion: We found that lower intrusions were a robust predictor of subsequent reduction in all other PTSD symptom clusters. This finding suggests the possibility that one notable mechanism of change in PE and CPT may be an indirect effect on avoidance, negative alternations in mood and cognitions, and hyperarousal that occurs via the impact of these treatments on the frequency and intensity of intrusion symptoms. Notably, these results replicated the findings of Maples-Keller et al. (2017) and Li et al. (2020), whose studies found cluster B symptoms to have the greatest impact on changes in other symptom clusters. Future research should examine these effects separately for PE and CPT. Further, future studies should investigate which components of these treatments impact symptom clusters to understand if and how they affect symptom clusters independently.