Psychometric evaluation of the weekly version of the PTSD checklist for DSM-5

Abstract

The posttraumatic stress disorder (PTSD) Checklist for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5; PCL-5) was designed and validated to track symptoms over the past month (PCL-5-M), yet an untested ad hoc weekly version (PCL-5-W) is commonly used to track changes during treatment. We used archival data of clinical trials for the treatment of PTSD in veterans to assess the construct validity of PCL-5-W. Both PCL-5-M and PCL-5-W were found to have configural measurement invariance across four consecutive administrations. The results also indicated at least partial metric and scalar invariance for each version. The reliability estimates of PCL-5-M and PCL-5-W at each time point were equivalent. However, we found a discrepancy with regard to concurrent validity; correlations with the nine-item Patient Health Questionnaire may be meaningfully different between PCL-5-M and PCL-5-W. Nevertheless, overall, the results suggest that PCL-5-W can be validly used to assess PTSD symptoms over time, but factor scores may need to be tracked alongside total scores to address validity concerns.