2 - (IOP 19) Diagnostic Utility of the PTSD Checklist for DSM-5 (PCL-5) in Treatment-seeking Veterans: A Comparison Across Demographic and Clinical Characteristics

The PTSD Checklist for DSM-5 (PCL-5) is the most widely used measure of PTSD in both clinical and research settings. Although numerous studies have demonstrated support for the diagnostic utility of PCL-5 scores, little work has been done to compare diagnostic utility of the PCL-5 across diverse populations. Based on observed group differences in rates and severity of PTSD in prior research, it is possible that the diagnostic accuracy of PCL-5 scores may also differ between groups. The goal of this investigation was to test for group differences in diagnostic utility of PCL-5 scores across multiple demographic and clinical characteristics. The sample included 1,237 treatment-seeking veterans (Mean Age = 48.7 [range: 20-94]; 90% Male; 55% White) who completed a battery of self-report measures including the PCL-5, Patient Health Questionnaire-9 (PHQ-9), and Alcohol Use Disorders Identification Test (AUDIT), as well as the Clinician Administered PTSD Scale for DSM-5 (CAPS-5), as part of routine clinical care. PCL-5 scores were significantly higher among veterans who were younger (vs. older than 65; d = 0.41, 95% CI = [0.29, 0.54], p < .001), female (vs. male; d = 0.26, 95% CI = [0.07, 0.45], p = .005), identified as a race other than white (vs. white; d = 0.28, 95% CI = [0.17, 0.39], p < .001), had at least moderately severe depression (PHQ-9 scores ≥15 vs. < 15; d = 1.51, 95% CI = [1.38, 1.63], p < .001) and hazardous alcohol use (AUDIT scores ≥ 8 vs. < 8; d = 0.18, 95% CI = [0.06, 0.30], p = .003). PTSD diagnosis was defined using the CAPS-5 DSM-5 algorithm. Compared to veterans with less severe depression scores, those with at least moderately severe depression scores had significantly higher rates of CAPS-5 PTSD (85% vs. 66%, p < .001) and PCL-5 scores with significantly lower diagnostic accuracy (AUCs: 0.61, 95% CI = [0.55, 0.67] vs. 0.71, 95% CI = [0.66, 0.76], p = .011). Moreover, a higher cutoff score on the PCL-5 was needed to achieve the optimal efficiency for predicting CAPS-5 PTSD in the presence of moderately severe depression (49 [sensitivity (Se); 0.88, specificity (Sp): 0.25; efficiency (Eff): 0.79; κ(.5): 0.14] vs. 36 [Se: 0.76; Sp: 0.56; Eff: 0.69; κ(.5) = 0.31]). On the other hand, rates of CAPS-5 PTSD were lower in older compared to younger veterans (67% vs. 81%, p < .001), but accuracy of the PCL-5 in predicting CAPS-5 PTSD was similar across age groups (AUCs: 0.72, 95% CI = [0.66, 0.78] vs. 0.70, 95% CI = [0.65, 0.74], respectively, p = .545). CAPS-5 PTSD and PCL-5 diagnostic accuracy were also comparable across male and female veterans, racial categories, and alcohol use groups (ps ≥ .154). Overall, results supported consistent diagnostic utility of the PCL-5 across groups except in the context of depression. Moderately severe depression was linked to lower diagnostic accuracy of PCL-5 scores and a need to use higher PCL-5 cutoffs to achieve optimal efficiency in identifying PTSD. Given the high comorbidity of PTSD and depression, these findings have important implications for screening and assessment practices in clinical care.