ObjectiveThe current study sought to examine the relationship between changes in distress for items on in-vivo exposure hierarchies and posttraumatic stress disorder (PTSD) symptom change over the course of exposure therapy. MethodsActive duty army soldiers (N = 108) were recruited from a military base in the U.S. and were enrolled in a randomized clinical trial comparing Prolonged Exposure (PE), Virtual Reality Exposure (VRE), and a wait-list control for the treatment of PTSD stemming from deployments to Iraq or Afghanistan. PTSD diagnosis followed DSM-IV-TR criteria. Outcome measures were assessed via self-report and clinician interview. The relationships between in-vivo exposure distress, imaginal exposure distress, and PTSD symptoms, were examined in a factor of curves model for participants in the treatment conditions. ResultsAnalyses revealed that, when controlling for one another, changes in in-vivo exposure distress were significantly associated with changes in PTSD symptoms (β = 0.75, 95% CI [0.60, 0.90]), while changes in imaginal exposure distress were not (β = 0.03, 95% CI [-0.27, 0.33]). The model also revealed that after accounting for the shared variation in trajectories of change, symptom clusters did not have unique variation, meaning that symptom clusters did not change independently. ConclusionResults suggest the possibility that in-vivo exposures are more closely tied to changes in overall PTSD symptoms than imaginal exposures during exposure therapy. Furture research should incorporate more frequent measurement of in-vivo exposure distress to better elucidate these relations over the course of treatment.