(PS12-A10) Identifying Strengths in CBT Skills for Treatment Personalization: Comparing the Utility of Three Approaches

Background: Improving treatment efficiency is essential to increasing access to mental health services. One promising approach is to tailor the delivery of existing interventions to clients’ pre-treatment competencies in CBT skills (i.e., personal strengths; Flückiger et al., 2023). However, it is unclear how best to identify these personal strengths. Across various samples, researchers have used clinical interviews (Cheavens et al., 2012), pre-treatment data (Fisher et al., 2019), and standardized responses to conceptually relevant self-report measures (Sauer-Zavala et al., 2019). We examined the degree to which these three approaches identified personal strengths in a single sample and conducted preliminary tests of symptom change during the identified strengths to validate these identifications.

Methods: Participants (N = 50; Mage = 32.83, 68% female, 82% White) were treatment-seeking adults with an anxiety or related disorder receiving core modules from the Unified Protocol for Transdiagnostic Treatment of Emotional Disorders (UP; Barlow et al., 2018) as part of an ongoing parent sequential multiple assignment randomized trial that will conclude data collection in July 2023. We compared three methods of assessing strengths at baseline: (1) a structured interview with a study assessor; (2) a self-report measure of skill use based on the Styles of Emotion Regulation Questionnaire (Murphy et al., 2021) adapted to the UP; and (3) a set of five validated self-report measures (Beliefs About Emotions Scale; Southampton Mindfulness Questionnaire; UP Cognitive Skills Questionnaire; MEAQ – Behavioral Activation Scale; Anxiety Sensitivity Index) that conceptually map onto the mechanism targeted by each UP module. Participants then reported on anxiety symptoms (Overall Anxiety Symptoms and Impairment Scale) before each session. We calculated Fleiss’ kappa to assess agreement among the three assessment approaches and conducted paired t-tests of individual slopes representing within-person change in anxiety symptoms between each approach to test if any approach identified a module that led to consistently larger reductions in anxiety symptoms.

Results: There was significant disagreement among the strengths assessments, k = -.09, p = .03. The strengths identified via the five self-report measures were associated with the greatest reduction in anxiety symptoms, M = -1.37, SD = 1.50, compared to the interview, M = 0, SD = 1.62, t(6) = 2.47, p = .048, d = 1.23. The strengths identified by the skills measure, M = -.88, SD = 2.96, did not significantly differ from either the five measures, t(4) = -.56, p = .60, d = .30, or the interview, t(6) = 1.65, p = .15.

Discussion: These results suggest relatively little agreement among three common ways of assessing baseline strengths in CBT. Of the methods examined, our preliminary results indicate that standardized scores on validated self-report measures of constructs being targeted in treatment are more accurate at detecting baseline CBT strengths than a structured interview or self-report measure of baseline skill frequency.