Identification of and At-risk Interventions for Pre-deployment Psychophysiologic Predictors of Post-deployment Mental Health Outcomes

Abstract

Abstract : Objectives and Rationale: The primary objectives of this study (Warriors Achieving Resilience or WAR) were to develop objective pre-deployment predictors of PTSD and test two pre-deployment resiliency interventions. Objective predictors include: 1) physiologic reactivity to combat-related virtual reality environments and white noise startle and 2) cognitive bias assessment. We also will test two pre-deployment resiliency interventions: 1) video game-based heart rate variability biofeedback training and 2) computerized cognitive bias training. Objective assessment and training measures are more reliably measured and could be early indicators of resilience/vulnerability. Study Design: Longitudinal prediction study and randomized controlled trial to prevent post-deployment PTSD symptoms. 600 soldiers were consented with the goal of completing 500 pre-deployment assessments. Pre-deployment data collection for was completed for 427 soldiers, 411 soldiers deployed. Subjects were randomized to one of three groups: heart rate variability (HRV) biofeedback training, cognitive bias feedback training, or no additional training. iPods with respective apps were provided to the soldiers in the resilience training groups. Follow-up data was collected at 3- and 12- months post-deployment, 308/411 (75%) and 236/411 (57%), respectively. PTSD symptom severity was measured using the PTSD Checklist. Major Findings: Pre-deployment HRV (measured by the standard deviation of beat to beat intervals or SDNN) was a significant predictor of 3-month post-deployment PTSD symptoms. Low HRV is associated with increased physical and emotional stress. The time between pre-deployment SDNN and post-deployment PTSD symptom measurement was approximately one year. There was an expected inverse relationship between pre-deployment SDNN and 3-month post-deployment PTSD symptoms (beta=-0.002, p=0.006) in a multivariate model with a continuous measure of PTSD symptoms.