Can a Single Measure Estimate Protocol Adherence for Two Psychosocial Treatments for Youth Anxiety Delivered in Community Mental Health Settings

Abstract Treatment adherence measurement can be time and resource-intensive in clinical trials, so the ability to measure protocol adherence for two distinct treatment programs with a single measure may benefit the field. The present study sought to determine if the Therapy Process Observational Coding System – Revised Strategies Scale (TPOCS-RS) could assess protocol adherence to two youth treatment programs. Treatment sessions (N = 796) from 55 youth (M age = 9.89 years, SD = 1.71; range 7-15 years; 55.0% White; 46.0% female) with primary anxiety problems treatment by 39 clinicians (M age = 40.54 years, SD = 9.56; 50.0% White; 80.0% female) were independently scored by coders using observational treatment adherence and alliance measures. The youth received one of three treatments: (a) Standard (i.e., cognitive-behavioral treatment program), (b) Modular (i.e., a program with cognitive-behavioral and parent training components), or (c) Usual Care. Consultants filled out a self-report measure of protocol adherence within the Standard and Modular conditions. Interrater reliability, ICC(2,2) for the various items for the full sample ranged from .17 to .92 (M ICC = .67; SD = .17). Scores from a TPOCS-RS subscale that mapped onto the specific content of the treatment protocols used in the Standard and Modular conditions evidenced convergent validity with the consultant-report adherence measure and discriminant validity with the alliance measure. The model-specific TPOCS-RS subscales also discriminated between the Standard and Modular treatments and Usual Care. This study provides initial evidence that (a) the TPOCS-RS has utility in estimating protocol adherence in different treatment programs and (b) support the score validity of the self-report consultation records.