The Prevalence and Latent Structure of Proposed DSM-5 Posttraumatic Stress Disorder Symptoms in US National and Veteran Samples

The Diagnostic and Statistical Manual, Fourth Edition (DSM-IV) is currently undergoing revisions in advance of the next edition, DSM-5. The DSM-5 posttraumatic stress disorder workgroup has proposed numerous changes to the PTSD diagnosis. These include the addition of new symptoms, revision of existing ones, and a new four-cluster organization (Friedman, Resick, Bryant, & Brewin, 2011). We conducted two Internet-based surveys to provide preliminary information about how proposed changes might impact PTSD prevalence and clarify the latent structure of the new symptom set. We used a newly developed instrument to assess event exposure and lifetime and current DSM-5 PTSD symptoms among a nationally representative sample of American adults (N = 2,953) and a clinical convenience sample of U.S. military veterans (N = 345). Results from both samples indicated that the originally proposed DSM-5 symptom criteria (i.e., requiring 1 B, 1 C, 3 D, and 3 E symptoms) yielded considerably lower PTSD prevalence estimates compared with DSM-IV estimates. These estimates were more comparable when the DSM-V D and E criteria were relaxed to 2 symptoms each (i.e., the revised proposal). Confirmatory factor analyses (CFA) indicated that the factor structure implied by the four-symptom criteria provided adequate fit to the data in both samples, and a DSM-5 version of a dysphoria model (Simms, Watson, & Doebbeling, 2002) yielded modest improvement in fit. Item-response theory and CFA analyses indicated that the psychogenic amnesia and new reckless/self-destructive behavior symptom deviated from the others in their respective symptom clusters. Implications for final formulations of DSM-5 PTSD criteria are discussed.