Predictors of treatment noninitiation, dropout, and response for cognitive behavioral therapy for trauma nightmares.

Objective: This report aims to extend previous findings on predictors of treatment dropout and response within nightmare treatment. Method: Factors predicting treatment noninitiation, dropout, and response were examined in 70 trauma-exposed individuals treated in a randomized controlled trial of 2 conditions (exposure [EX] and nonexposure) of a therapy for trauma nightmares. Results: Together, treatment noninitiation and dropout was predicted by younger age for the EX condition only and by fewer nights with nightmares at baseline for both conditions. Treatment noninitiation was predicted by fewer nights with nightmares. Treatment dropout was not predicted by any variables. For the EX condition, more baseline nightmares predicted more posttreatment nightmares, whereas more reported nights with nightmares at baseline predicted fewer nightmares at posttreatment. Baseline sleep quality and total posttraumatic stress disorder (PTSD) severity predicted posttreatment sleep quality and PTSD severity, respectively. In the nonexposure condition, more traumatic events predicted fewer nightmares at posttreatment, whereas more nightmares predicted more nightmares at posttreatment. Baseline daytime sleepiness predicted worse posttreatment sleep quality, and total PTSD symptom severity predicted posttreatment PTSD severity. Conclusions: Baseline symptom severity on constructs targeted in treatment are the best indicators of response. Treatment initiation appears to be more significant than dropout. Identifying reasons for treatment noninitiation is needed to maximize engagement.