(PS4-B35) Construct Validity of the Weekly Version of the PTSD Checklist for DSM-5

True evidence-based psychotherapy is more than following a treatment approach that is supported by empirical evidence; it also entails assessing and tracking change, or the lack thereof, over the course of treatment. In theory, repeated (e.g., weekly) outcome monitoring provides evidence to guide clinical and shared decision-making about whether expected change is occurring, and if not, a new course of action needs to be charted (i.e., measurement-based care). In PTSD treatment, the PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2013) is the most common method used to track change in both clinical and research contexts. The PCL-5 was designed and validated to assess PTSD symptoms over the last month. Ad hoc weekly versions of the PCL-5 (sometimes indexed to “since the last visit”) have been created to track change (e.g., Bragesjö et al., 2021), but no study to date has assessed the reliability and validity of these less than monthly versions of the test. In this study, we plan to determine if the PCL-5 – weekly version (PCL-5-W) measures the same latent construct as the monthly version (PCL-5-M), specifically examining the lateral invariance between the two forms. Secondary data analyses will be conducted using PCL-5 data from four time points (i.e., baseline, session 1 – session 3), culled from the datasets for 6 South Texas Research Organizational Network Guiding Studies on Trauma and Resilience (STRONG STAR) Consortium and the Consortium to Alleviate PTSD (CAP) clinical trials (~N = 940). Preliminary analyses were conducted using data from one of these trials; an examination of intensive treatment for combat-related PTSD (N = 234). Confirmatory factor analyses (N = 234) replicated a 7-factor structure (e.g., Wortmann et al., 2016) for the PCL-5-M at baseline (N = 234), χ²(210) = 1750.03, p < .001, CFI = .94, TLI = .92, SRMR = .05, RMSEA = .05, and a 7-factor structure for the PCL-5-W at 3-weeks post-baseline (n = 175), χ²(149) = 257.861, p < .001, CFI = .96, TLI = .95, SRMR = .03, RMSEA = .07. The internal consistency reliability of the PCL-5-W at 3-weeks post-baseline was excellent (PCL-5-W, α = .97). The stability in the number of PCL-5 factors across monthly and weekly versions, and the excellent internal consistency for the PCL-5-W, provides initial evidence that the PCL-5-M and PCL-5-W measures the same latent construct. Using all CAP trials, we will be examining the structural invariance between versions of the PCL-5, and the temporal invariance of the PCL-5-W over 3 time points, using nested χ² tests to examine equality of factor structure, factor loadings, and indicator error variances between PCL-5 versions and for the PCL-5-W across time.  We will also examine the convergent validity of the PCL-5-W by examining its correlation with a repeated measure of depression symptoms. We predict strong, positive correlations between the PCL-5-Ws, PHQ-9s, and the PCL-5-M. The results of these analyses will represent the first test of the construct validity of a weekly symptom tracking version of the PCL-5.