Prospective associations between DSM-5 PTSD symptom clusters and suicidal ideation in treatment-seeking veterans.

Posttraumatic stress disorder (PTSD) rates are higher in military veterans than in the civilian population. Meta-analyses have found strong and consistent associations between PTSD and suicide risk. Several studies have demonstrated a concurrent reduction in suicidal ideation (SI) with reduction of PTSD symptoms during trauma-focused treatment. However, it is unclear whether changes in specific PTSD symptom clusters are most strongly associated with these changes in SI. This study prospectively examined associations between PTSD symptom clusters and SI to better specify mechanisms of change during treatment. Participants were 160 veterans (87% male, 63% Caucasian, 64% combat trauma) who completed a course of evidence-based trauma-focused therapy at a VA hospital. The Patient Health Questionnaire-9 and Posttraumatic Stress Disorder Checklist-5 were used to assess depression, SI frequency, and PTSD symptoms. Binary logistic regression analyses found that the cognitive/mood alteration cluster was the only significant independent predictor of SI at termination. Post hoc analysis of variance Bonferroni tests indicated those who decreased SI frequency had a greater reduction in intrusive, cognitive/mood alteration, and hyperarousal symptoms relative to those who increased or had no change in SI. A within-cluster item analysis revealed that baseline symptom D3 (blame self/others) was the only significant independent predictor for baseline SI, whereas baseline symptom D6 (detachment) was the only significant independent predictor for SI at termination. This discrepancy may be explained by reductions in guilt during treatment, as 79% of the sample elected to receive cognitive processing therapy. Given these associations, PTSD patients with SI may benefit from a treatment emphasis on reducing cognitive/mood alteration symptoms. (PsycINFO Database Record (c) 2018 APA, all rights reserved).