(PS5-32) Factor Structure of an Observer-rated Caregiver-clinician Alliance Scale in Treatment for Youth Anxiety

Caregivers play a crucial role in youth psychosocial treatment. The caregiver-clinician alliance (herein called the caregiver alliance) is often conceptualized as containing three components: an affective bond, agreement on therapeutic activities, and agreement on treatment goals. Previous research has demonstrated that a positive caregiver alliance predicts treatment attendance and improved youth clinical outcomes. However, few studies have examined the factor structure of existing measures used to assess the caregiver alliance. As such, empirical work has not established if the three alliance components are relevant to the caregiver alliance.

To address this knowledge gap, the current study investigates the factor structure of a measure widely used to assess the caregiver alliance called the Therapy Process Observational Coding System for the Child Psychotherapy-Alliance (TPOCS-A) Caregiver scale. The scale includes 16 items that assess the bond, task, and goal components of the alliance. To accomplish this goal, 482 treatment sessions attended by caregivers were coded with the TPOCS-A Caregiver scale by graduate students. Sessions were pulled from two effectiveness trials: (a) the Youth Anxiety Study (YAS) that included 38 youth with primary anxiety disorders (M age = 10.83, SD = 2.11; 57.9% female) and 29 clinicians (M age = 33.86 years, SD = 10.30; 82.8% female), who were randomized to Coping Cat or usual clinical care and (b) Child STEPS Multisite Trial that included 55 youth with primary anxiety problems (M age = 9.89 years, SD = 1.71; 54.5% female) and 39 clinicians (M age = 40.24 years, SD = 9.39; 79.5% female) who were randomized to Coping Cat, MATCH, or usual clinical care.

Interrater reliability based on the intraclass correlation coefficient (ICC[2,2]) for the TPOCS-A Caregiver items ranged from .04 to .60. An Exploratory Factor Analysis conducted with the YAS sample yielded a three-factor solution. A Confirmatory Factor Analysis conducted with the Child STEPS Multisite Trial data confirmed the three-factor solution.

The three factors focus on affective and goal-oriented aspects of the alliance but do not map onto the classic bond, task, and goal components. Our findings suggest that the structure of the caregiver alliance may differ from the child-clinician alliance. These findings have clinical implications for clinicians to consider the agreement of goals and problems, positive affect, and negative interaction when working with caregivers to treat youth anxiety.