Response to psychotherapy for posttraumatic stress disorder: the role of pretreatment verbal memory performance

Objective: Neuropsychological studies have consistently demonstrated impaired verbal memory in posttraumatic stress disorder (PTSD). Trauma-focused treatment for PTSD is thought to rely on memory, but it is largely unknown whether treatment outcome is influenced by memory performance. The aim of the study, therefore, was to examine the relationship between verbal memory performance and treatment response to trauma-focused psychotherapy.
Method: Participants were referred to our outpatient clinic and recruited between December 2003 and January 2009 upon diagnosis of PTSD according to DSM-IV. Secondary analyses of a randomized controlled trial comparing eye movement desensitization and reprocessing therapy (n = 70) and brief eclectic psychotherapy (n = 70), a cognitive-behavioral intervention, are reported. Response to treatment was measured by self-reported PTSD symptom severity (Impact of Event Scale—Revised) over 17 weeks. Pretreatment verbal memory measures (California Verbal Learning Test, Rivermead Behavioral Memory Test) were included in the mixed linear model analyses in order to investigate the influence of memory on treatment outcome.
Results: Pretreatment encoding, short-term retrieval, long-term retrieval, and recognition performance were significantly associated with treatment response in terms of self-reported PTSD symptom severity for both treatments (P ≤ .013). Receiver operating characteristic curves predicting treatment response with pretreatment memory indices showed that 75.6% of the patients could be correctly classified as responder.
Conclusions: Poor verbal memory performance represents a risk factor for worse treatment response to trauma-focused psychotherapy. Memory measures can be helpful in determining which patients are unable to benefit from trauma-focused psychotherapy. Future research should explore how treatment perspectives of patients with poor verbal memory can be improved.