PTSD: National Center for PTSD
Development and Psychometric Evaluation of the Deployment Risk and Resilience Inventory-2 (DRRI-2)
Development of the Deployment Risk and Resilience Inventory-2 (DRRI-2)
The DRRI-2 is the result of a multiyear psychometric endeavor that involved the application of both classical test theory (CTT) and item response theory (IRT) analytical strategies. This project included three key phases. Phase I began with an initial assessment of the content validity of DRRI scales based on a comprehensive literature review and focus groups with OEF/OIF Veterans. These groups included both men and women, Veterans deployed from Active duty and National Guard/Reserves, and representation from all branches of service. Based on this information, several new measures were then constructed to address newly identified content domains, and new items were developed to more comprehensively represent content domains addressed within existing measures. In addition, items that were deemed to be less pertinent to contemporary Veterans (e.g., items highly specific to the first Gulf War) were eliminated or rewritten to be more broadly relevant across Veteran cohorts.
In Phase II, both original and revised DRRI items were administered to a national sample of 469 male and female OEF/OIF Veterans using a mail survey procedure. Overall, results supported the psychometric quality of the proposed DRRI-2 scales. Specifically, evidence was provided for the internal consistency reliability of updated DRRI scales and expected associations were observed between these scales and a measure of PTSD symptom severity, providing support for criterion-related validity. Importantly, estimates of internal consistency reliability and criterion-related validity were slightly higher than that observed for the original DRRI scales despite the overall reduction in scale length. However, IRT analyses pointed to the need for further revision to better capture the full construct continuum for several constructs.
Phase III involved administering a revised set of scales to a second national sample of 1,046 male and female OEF/OIF Veterans using the same mail survey procedure. Both CTT and IRT analyses were used to identify final items sets. A systematic evaluation of the psychometric quality of the final DRRI-2 scales (14 revised scales and 3 new scales) suggested that the revisions improved upon the inventory. Specifically, CTT analyses confirmed that the scales have high internal consistency reliability and the finalized scales showed moderate to strong bivariate associations with PTSD symptom severity. In addition, an examination of incremental validity revealed that the new DRRI-2 scales add unique variance in the prediction of PTSD symptom severity above and beyond existing DRRI-2 scales, indicating that the inclusion of these additional measures provide for a more comprehensive assessment of deployment-related risk and resilience. Finally, as compared to the original DRRI scales, the DRRI-2 scales are 15% shorter, on average.
In summary, the DRRI-2 represents a psychometrically sound, yet efficient, suite of scales that can be used to capture predeployment, deployment, and postdeployment risk and resilience factors with implications for the postdeployment mental health and functioning of Service members and Veterans. It is our hope that these measures will be used to further knowledge of the role that these factors play in postdeployment health and inform interventions aimed at reducing risk and enhancing resilience within these populations.