(PS8-C57) Youth-caregiver Problem Severity Ratings: How Does Agreement Differ by Psychopathology Subtype and Relate to Functional Impairment in a Large, Multiethnic Public Mental Health Sample?

Multi-informant reporting is a commonly utilized method for understanding youth psychopathology. Gathering information from multiple perspectives yields benefits, not only by broadening available data, but by allowing insight into cross-informant reporting patterns that might nuance clinical decision-making (e.g., assignment of mental health diagnoses, ongoing treatment planning, assessing and responding to emergency situations). Previous studies have demonstrated that in both research and clinical settings, informants often provide diverging perspectives when reporting about youth emotional and behavioral concerns. When these informant discrepancies emerge at treatment intake, they have been linked to poor youth outcomes and response to intervention. The current study builds upon previous research by leveraging the Ohio Scales, Short Form, Problem Severity domain (an empirically supported outcome measurement tool with identical item content across youth- and caregiver-report versions; Ogles et al., 1996) to expand investigations of informant discrepancies to a large, multiethnic public mental health sample. This study includes two aims. First, we assessed the degree of youth-caregiver rater agreement on the Problem Severity domain’s total and subscale scores at intake into a state public mental health system. Second, we examined the extent to which youth-caregiver rater agreement at intake predicted youth treatment outcomes as indicated by their discharge scores on a measure of functional impairment, the Child and Adolescent Functional Assessment Scale (CAFAS; Hodges, 1989). 

Participants included 266 dyads of caregivers and their children receiving public mental health services through our state’s Department of Health division for child and adolescent mental health. Youth were ethnically diverse, 59% male, an average of 12.96 years old (SD = 3.4), and predominantly treated for mood or disruptive behavior difficulties.

Cronbach’s alpha coefficients for youth- and caregiver-report Problem Severity domain scores were run prior to our major analyses, and fell in the acceptable to excellent range for youth-report (Total scale = .90, Externalizing = .86, Delinquency = .78, Anxiety = .80, Depression = .87), and questionable to excellent range for caregiver-report (Total scale = .90, Externalizing = .90, Delinquency = .68, Anxiety = .78, Depression = .81). Concerning our first aim, youth and caregiver ratings showed medium and positive correlations at both the total and subscale levels (Total scale r = .30, p < .01; Externalizing r = .40, p < .01; Delinquency r = .41, p < .01; Anxiety r = .30, p < .01; Depression r = .35, p < .01). Regarding our second aim, forthcoming polynomial regression analyses examining the relationship between intake youth-caregiver Problem Severity domain rater agreement and discharge CAFAS scores will be presented in full at the conference. The current study’s results will contribute to research exploring youth-caregiver rater agreement as a meaningful tool to inform treatment delivery efforts, especially as it pertains to evidence-based practices, large public mental health systems, and behavioral health emergencies. Study limitations and future directions will also be discussed.