The specific aims of the proposed study are to apply innovative longitudinal design and new quantitative methods for frequently assessed or intensively measured data to examine previously unstudied processes of change in posttraumatic stress disorder (PTSD) symptoms among military veterans during the weeks and months following their return from war-zone duty in Iraq and Afghanistan. In Ecological Proximal Assessment (EPA) participants make frequent self-reports that are proximal in time to the phenomena under study using hand-held computers (PDAs) so that behavior can be assessed in its natural setting, thus minimizing retrospective recall bias. The proposed study aims to combine intensive data measurement and multilevel modeling to examine change in PTSD symptoms among military veterans during the weeks and months following return to their communities that may reveal important mechanisms of PTSD emergence, chronicity and recovery.
The specific aims of the project include examining intra-individual temporal associations among the PTSD symptom clusters and time-varying predictors of panic attack symptoms and daily stressors. Additionally, inter-individual non-time-varying baseline indicators of depression and anxiety and participant sex/gender will be examined as moderators of change in PTSD symptoms over time. Participants in this study will be 40 male and female veterans seeking treatment at the National Center for PTSD, VA Boston Healthcare System following return from war-zone duty in Iraq or Afghanistan. Following baseline interview, using PDAs participants will complete one daily assessment of PTSD symptoms, panic attack symptoms, and daily stressors each day for 30 days. Following completion of the daily PDA assessments, four additional telephone assessments will be conducted of the repeated measures (PTSD, panic, stressors) each at one-month intervals. Change in symptoms indicating processes of PTSD emergence, chronicity, and recovery will be examined using multilevel random coefficients regression for intensively measured data. These analyses will include multilevel cross-lagged autoregression, functional multilevel modeling that integrates local linear regression estimation, as well as individual growth modeling. The proposed research intends to use traditional and contemporary multilevel modeling methods to examine relationships among the variables in atypical ways that can illuminate processes of PTSD symptom recovery, emergence, and chronicity among military veterans during the weeks and months following return from war-zone duty. ? ? ?
