Cognitive-Behavioral Treatment for Posttraumatic Nightmares: An Investigation of Predictors of Dropout and Outcome

This study examined factors predicting treatment dropout and outcome in 124 male Vietnam War veterans with chronic, severe posttraumatic stress disorder treated in a randomized controlled trial of two cognitive-behavioral group therapies for combat-related nightmares. Though significant bivariate predictors of dropout in the imagery rehearsal condition included non-African American race, use of selective serotonin reuptake inhibitors, more traumas, and lower perceived treatment credibility, none of these variables uniquely predicted dropout in multivariate analyses. In the sleep and nightmare management condition, only low avoidance symptoms predicted dropout. Use of benzodiazepines and higher reexperiencing symptoms predicted posttreatment nightmare frequency in imagery rehearsal; although baseline sleep quality and higher avoidance symptoms predicted posttreatment sleep quality. In sleep and nightmare management, only poorer sleep quality predicted posttreatment nightmares, although poorer baseline sleep quality and higher avoidance symptoms predicted posttreatment sleep quality. Practical and clinical implications, including the use of "socialization" strategies (e.g., patient testimonials, in-depth explanation of treatment rationale), are discussed.