INTRODUCTION: Dysfunctional appraisals about traumatic events and their sequelae are a key mechanism in posttraumatic stress disorder (PTSD). Experimental studies have shown that a computerized cognitive training, cognitive bias modification for appraisals (CBM-APP), can modify dysfunctional appraisals and reduce analogue trauma symptoms amongst healthy and subclinical volunteers.

OBJECTIVE: We aimed to test whether CBM-APP could reduce dysfunctional appraisals related to trauma reactions in PTSD patients, and whether this would lead to improvements in PTSD symptoms.

METHODS: We compared CBM-APP to sham training in a parallel-arm proof-of-principle double-blind randomized controlled trial amongst 80 PTSD patients admitted to an inpatient clinic. Both arms comprised a training schedule of 8 sessions over a 2-week period and were completed as an adjunct to the standard treatment programme.

RESULTS: In intention-to-treat analyses, participants receiving CBM-APP showed a greater reduction in dysfunctional appraisals on a scenario task from pre- to posttraining (primary outcome) assessments, compared to those receiving sham training (d = 1.30, 95% CI 0.82-1.80), with between-group differences also found on the Posttraumatic Cognitions Inventory (PTCI; d = 0.85, 95% CI 0.39-1.32) and the PTSD Checklist for DSM-5 (PCL-5; d = 0.68, 95% CI 0.23-1.14), but not for long-term cortisol concentrations (d = 0.25, 95% CI -0.28 to 0.78). Reductions in dysfunctional appraisals assessed via the scenario task correlated with reductions on the PTCI, PCL-5, and hair cortisol concentrations from pre- to posttraining time points.

CONCLUSIONS: Results support dysfunctional appraisals as a modifiable cognitive mechanism, and that their proximal modification transfers to downstream PTSD symptoms. These findings could open new avenues for improving present therapeutic approaches.