There is growing evidence that change in distress is an indicator of change during Prolonged Exposure (PE) for posttraumatic stress disorder (PTSD). However, temporal sequencing studies investigating whether change in distress precedes PTSD symptom decline are lacking. These studies are essential since the timeline between indicators of change and treatment outcome is a key assumption for mediation. The aim of the present study was to assess the temporal relationship between within- and between-session change in subjective distress and PTSD symptom decrease. We analyzed session data from 86 patients with PTSD. Data were analyzed using dynamic panel models. We distinguished temporal effects (within-persons) from averaged effects (between-persons). Results regarding the temporal effect showed that within-session change in subjective distress preceded PTSD symptom improvement while the reversed effect was absent. Averaged within-session change in subjective distress was also related to PTSD symptom improvement. Results regarding the temporal effect of between-session change in subjective distress showed that it did not precede PTSD symptom improvement. Averaged between-session change in subjective distress was related to PTSD symptom improvement. This study provides evidence for within- but not between-session change in subjective distress as indicator of change during PE. We also found that the way of modeling potential indicators of change affects results and implications. We recommend future studies to analyze mediators during treatment using temporal rather than averaged effects.
Keywords: PTSD; change in distress; dynamic panel model; prolonged exposure; temporal sequencing; working mechanism.
Copyright © 2021. Published by Elsevier Ltd.