Diagnostic utility and factor structure of the PTSD Checklist in older adults

ABSTRACT Background: Little research has examined the diagnostic utility and factor structure of commonly used posttraumatic stress disorder (PTSD) assessment instruments in older persons. Methods: A total of 206 adults aged 60 or older (mean age = 69 years; range = 60-92), who resided in the Galveston Bay area when Hurricane Ike struck in September 2008, completed a computer-assisted telephone interview two-five months after this disaster. Using the PTSD Checklist (PCL), PTSD symptoms were assessed related both to this disaster and to participants' worst lifetime traumatic event. Total PCL scores were compared to PCL-based, Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV)-derived probable diagnoses of PTSD to determine optimal cut scores. Confirmatory factor analyses (CFAs) were conducted to evaluate PTSD symptom structure. Results: Receiver operating characteristic analyses indicated that a PCL score of 39 achieved optimal sensitivity and specificity in assessing a PCL-based, algorithm-derived DSM-IV diagnosis of worst event-related PTSD; and that a score of 37 optimally assessed probable Ike-related PTSD. CFAs revealed that a recently proposed five-factor model - comprised of re-experiencing, avoidance, numbing, dysphoric arousal, and anxious arousal factors - provided a better fitting representation of both worst event- and disaster-related PTSD symptoms than alternative models. Current Ike-related anxious arousal symptoms demonstrated a significantly stronger association with current generalized anxiety than depressive symptoms, thereby supporting the construct validity of this five-factor model of PTSD symptomatology. Conclusions: A PCL score of 37 to 39 may help identify probable PTSD in older persons. The expression of PTSD symptoms in older adults may be best characterized by a recently proposed five-factor model with distinct dysphoric arousal and anxious arousal clusters.