Actuarial prediction of psychotherapy retention among Iraq-Afghanistan veterans with posttraumatic stress disorder.
The present study aimed to identify predictors of treatment retention in a sample of Operation Enduring Freedom/Operation Iraqi Freedom/Operation New Dawn (OEF/OIF/OND) veterans with posttraumatic stress disorder (PTSD) who were referred for PTSD-focused treatment through completion of a Veterans Affairs (VA) specialty clinic introductory information session. A total of 124 returning veterans (89% male, 53% Caucasian, 40% African American, 2% Latino; average age = 37 years) participated in an introductory session intended to facilitate informed decision making about treatment selection for PTSD. To evaluate patient, therapist, and system characteristics that were associated with risk of prematurely dropping out of psychotherapy for PTSD, we used recursive partitioning or "classification tree" methods commonly used to derive actuarial models of risk for high or low scores on a particular outcome when the set of independent or predictor variables is large. Findings revealed interactions among predictors involving access to care, readiness for change, histories of traumatic brain injury, and previous PTSD treatment. Results from the exploratory recursive model indicated that participation in therapy was highest when veterans entered psychotherapy within 68 days of the information session, believed that they needed help, and had a history of traumatic brain injury, while participation was lowest when entry into treatment exceeded 68 days and belief in needing help was low. Effects associated with partitions in the recursive model were substantial, with Cohen's d statistics ranging from .60 to 1.75. Results of the present effectiveness study implicate the importance of access to care as well as motivation for treatment in the returning cohort of OEF/OIF/OND veterans seeking help for PTSD. (PsycINFO Database Record
Fleming, CE; Kholodkov, T; Dillon, KH; Belvet, B; Crawford, EF
Volume / Issue
Start / End Page
Pubmed Central ID
Electronic International Standard Serial Number (EISSN)
Digital Object Identifier (DOI)