Predictors of cognitive behavioral therapy outcomes for insomnia in veterans with post-traumatic stress disorder

Background Insomnia is a well-recognized co-morbid condition in veterans with post-traumatic stress disorder (PTSD) with negative personal and social consequences. Cognitive behavioral therapy (CBT) is considered an efficacious treatment, yet little attention has been devoted to treatment response in this population. The aim of this study was to identify factors that may predict clinical response to CBT for insomnia (CBT-I) in veterans with PTSD. Methods A retrospective chart review of 136 veterans with PTSD-related insomnia was conducted. Epworth Sleepiness Score (ESS), PTSD Checklist (PCL), and Insomnia Severity Index (ISI) were assessed at baseline. We converted prescribed antidepressant and hypnotic dosages before and after CBT-I to dose equivalent of fluoxetine diazepam, respectively. A 6-point reduction or greater in ISI scores at 6-month follow-up visit was defined as CBT-I responsiveness. Results CBT-I responsiveness was observed in 47% of veterans with PTSD. Seventy-seven percent completed treatment. Lack of perceived benefit was the most given reason for failure to return for follow-up. In contrast to hypnotics, antidepressants usage decreased in those who had experienced benefit from CBT-I (p=0.001). Younger age, non-white race, and use of hypnotics prior to behavioral therapy were independently associated with lack of response to CBT-I. Conclusions While CBT-I ameliorates insomnia in veterans with PTSD, the use of hypnotics prior to instituting behavioral therapy may negatively affect the response rate to CBT-I. Future studies should examine whether racial and cultural influences on the generation of insomnia in veterans with PTSD affects the response to CBT-I.