(PS5-D76) Multilevel Predictors of Adherence to Modular Cognitive Behavior Therapy for Children

Background

The extent to which therapists deliver evidence-based treatments (EBTs) as intended depends on session, client, therapist, organizational, and sociopolitical factors. For example, treatment interference (e.g., emergencies) is associated with less EBT adherence. In the absence of structured guidance, therapists often respond to interference without using EBT strategies, which slows the rate of clinical improvement. MATCH, a modular CBT for youth, provides structured guidance on managing treatment interference related to comorbidity. This study examined session, client, therapist, and organizational predictors of adherence to MATCH delivered in community settings.

Methods

This study included 97 therapists serving 863 clients across 9408 unique sessions. Therapist-reported adherence to MATCH was measured using three criteria: 1) Did the session cover content consistent with the client’s target problem (e.g., exposure for anxiety)? 2) Did the session cover content in the expected sequence for the client’s target problem (e.g., psychoeducation followed by problem solving for depression)? 3) Did the session cover content in the expected sequence and with the expected participant(s) for the client’s target problem (e.g., praise followed by active ignoring with a caregiver for disruptive behaviors)?

Results

Content

Multilevel logistic regression results showed that more session interference (B = .34, p < .001), greater baseline internalizing symptoms (B = .02, p = .05), and more days in treatment (B = .001, p < .001) were related to greater adherence to MATCH content for the target problem. Low levels of general CBT skills was related to greater adherence to MATCH content for the target problem (B = -.17, p = .05).

Content and Sequence

Therapist adherence to MATCH content and sequence for the target problem was greater when there was more session interference (B = .31, p < .001), more days between sessions (B = .01, p < .001), more days in treatment (B = .0005, p = .02), and positive perception of organizational culture (e.g., sense of belonging, shared values; B = .29, p < .05).  Fewer crises (B = -.17, p = .02) and low levels of general CBT skills (B = -.16, p = 0.03) were related to higher adherence to MATCH content and sequencing.

Content, Sequence, and Participant

Therapist adherence to MATCH content, sequence, and expected participants for the target problem was greater when there was more session interference (B = .34, p < .001), more days between sessions (B = .01, p < .001), and positive perception of organizational culture (B = .35, p = .03). 

Discussion

Results suggest that session interference is a positive predictor of adherence regardless of the adherence criteria used. This finding indicates that the design of MATCH, which offers flexibility and structured guidance to address comorbid mental health problems, may aid in managing interference while maintaining fidelity in community settings. The negative relationship between CBT skills and adherence also suggests that a traditional skill set may challenge therapists’ ability to implement more flexible, modular treatments. Future studies should explore predictors of adherence to modular EBTs that provide guidance on addressing other types of treatment interference.