Objective: This study used data from a recent randomized clinical trial (RCT) that found differences between women veterans and civilians in posttraumatic stress disorder (PTSD) treatment response, with civilians demonstrating greater improvement than did veterans. Despite having similar PTSD severity scores at baseline, veterans scored roughly 18 points higher than civilians did on the Clinician-Administered PTSD Scale (CAPS) at posttreatment (p < .01). This study sought to identify the clinical and treatment variables that were associated with the differential response to treatment demonstrated by the women in the RCT. Method: Veteran (n = 21) and civilian (n = 105) women with PTSD received cognitive processing therapy (CPT) for PTSD. These secondary data analyses used structural equation modeling to investigate the role of 7 clinical and treatment variables to explain the reduced treatment response to CPT in veterans compared to civilians. Results: Using structural equation modeling, we found that differences in CAPS scores at posttreatment were largely mediated by negative posttraumatic cognitions, as measured by the Posttraumatic Cognition Inventory (PTCI). Although veterans and civilians had similar PTCI scores at baseline, civilians had significantly lower PTCI scores at posttreatment, which predicted lower CAPS scores at posttreatment. This mediation appeared to be at least in part explained by lower treatment expectancies by veterans compared to civilians. Conclusions: Future research should be focused on further understanding and addressing these 2 treatment outcome predictors in an effort to reduce the gap in PTSD treatment outcomes between veterans and civilians.