Posttraumatic Stress Disorder and Spinal Cord Injury

Objectives: To identify the prevalence of posttraumatic stress disorder (PTSD) after spinal cord injury (SCI) in a sample averaging over 2 decades postinjury at assessment. Related objectives are to confirm the factor structure, compare subscales with those reported in a nonclinical sample, and identify the relationship of PTSD with depression. Design: Survey. Setting: A medical university in the Southeastern United States. Participants: Participants were initially identified through specialty hospitals in the Midwest and Southeastern United States. A cohort of adults (N=927) with traumatic SCI of at least 1 year duration at enrollment in 2002 to 2003 and a minimum of 7 years at the time of assessment completed the study materials. Interventions: Not applicable. Main Outcomes Measures: PTSD was measured by the Purdue Posttraumatic Stress Disorder Scale-Revised, and depression was measured by the Patient Health Questionnaire 9-item. Results: PTSD was reported by less than 10% of the participants. Item endorsement decreased as a function of years postinjury, primarily because of low rates of endorsement among those 21 or more years postinjury. Confirmatory factor analysis did not result in an acceptable fit for subscales, item sets, or factors previously reported in the literature. Participants scored higher than a nonclinical sample (reported in the literature) on the arousal and avoidance subscales but lower on the re-experiencing subscale. Item endorsements were lower for the first set of items that relate directly to the SCI itself, with the highest item endorsement for "have difficulty remembering important aspects of event." PTSD rarely occurred in the absence of a depressive disorder. Conclusions: PTSD does not appear to be highly prevalent in long-term SCI survivors, and endorsement of items related to re-experiencing and even recalling the injury are rare. Because SCI often is accompanied by mild traumatic brain injury, difficulty recalling the event may have an organic rather than psychologic component.