Operating characteristics of the PTSD Checklist in a military primary care setting.

Abstract

The Department of Defense (DoD) is implementing universal behavioral health screening for all DoD health-care beneficiaries presenting to military primary care clinics. The PTSD Checklist-Civilian Version (PCL-C; Weathers, Litz, Herman, Huska, & Keane, 1993) is used for the identification of posttraumatic stress disorder (PTSD); however, the operating characteristics of the PCL-C remain unstudied in this population. This study examined the operating characteristics of the PCL-C in a sample of 213 patients from 3 Washington, D.C., area military primary care clinics. Blinded raters independently assessed PTSD using the PTSD Symptom Scale Interview (Foa, Riggs, Dancu, & Rothbaum, 1993) as the diagnostic criterion standard. The receiver operating characteristic curve revealed that PCL-C scores accounted for 92% of the area under the curve. A PCL-C score of 31 optimized sensitivity (0.93) and specificity (0.90), and the multilevel likelihood ratio was 5.50 (95% confidence interval [2.26, 13.37]). Internal consistency (0.97) and test-retest reliability (0.87 after a median 13 days) were strong. Results suggest that a PCL-C score of 31 is the optimal cutoff score for use in a military primary care setting serving active duty service members, dependents, and retirees. These findings offer military primary care providers preliminary data to interpret PCL-C scores and to inform treatment decisions as part of routine clinical practice.