UC Berkeley

The aim of this dissertation was to examine the extent of association, if any, between strengths (including those of the client, the client's family, and the client's environment) and psychiatric symptom severity for "youth" (children and adolescents aged 5-22) in contact with a mental health and/or substance abuse agency in a public System of Care. This dissertation examined a public systems dataset that included administrative assessment data for every client aged 5-22 (n=2049) who entered into a public System of Care from June 2010-August 2013. The assessment data was collected by clinicians at entry to the system (initial assessment), six months after initial assessment, and twelve months after initial assessment. Behavioral/social learning theory and related research in applied and experimental settings (e.g. Goldiamond, 1974; Madden, 2013; Staats, 2012) would suggest that growth in strengths would be associated with reduction in symptoms. However, time order cannot be determined with the current data and thus only associational relationships are discussed.

This dissertation is a "three paper" dissertation including a dissertation introduction, three inter-related papers, and a dissertation conclusion. The dissertation introduction provides a background to the entire dissertation and is intended to orient to the reader to terminological and conceptual issues relevant to all three papers (each paper might also cover some of the literature included in the dissertation introduction, as relevant to the specific paper).

Paper one reviews the literature on prevalence of emotional and behavioral problems among youth in the United States, orients the reader to the history and current policies of Systems of Care (SOC), and reviews the literature on strength-based assessment. Then, paper one reports the prevalence of clinician-reported strengths in the public System of Care, explores the psychometric properties of the assessment's Strengths section, and describes three alternative modeling options to explore group differences (by gender, age, and ethnicity) of youth strengths. Youth assessed at entry to the System of Care were reported to enter with considerable strengths, similar to other studies reporting on prevalence of strengths in a System of Care (e.g. CMHI, 2011; Dunleavey et al., 2011). The confirmatory factor analysis found that the hypothesized two factor model was a better fit for the data than a single factor model, though the fit of the two factor model was only adequate and the interpretation of the factors were questionable. Individual strength items were assessed as the most informative for future analyses, compared to the factors or the composite score.

Paper two examines whether, at initial assessment, client strengths are associated with psychiatric symptoms. Higher levels of family relationships, peer relationships, and educational strengths were significantly associated with lower levels of psychiatric symptoms, after accounting for variation due to clinician, program, reported trauma exposure and youth demographics.

Paper three reviews some studies that report on the relationship between strengths and mental health outcomes in Systems of Care samples. This paper then examines 1) the average change in psychiatric symptom severity twelve months after entry (initial assessment) to the System of Care, 2) whether initial strengths are associated with rate of change in psychiatric symptoms, 3) whether rate of change in strengths at six month and twelve month assessment are associated with rate of change in psychiatric symptoms, and 4) explores the contribution of variations related to clinician, youth, and time. Results suggest that self-reported growth in peer relationships, relationship permanence, and education strengths twelve months after entry to the System of Care was associated with larger reported reductions in symptoms at twelve months, adjusting for clustering, time, and youth demographics. Only associational relationships are reported; causal direction cannot be established in the current data. The percentage of youth referred for services, where they were referred from and to, the percentage who went to services, type, quantity, and quality of interventions offered and used (if any), and drop-out rates and reasons are not reliably tracked in the System of Care.

This dissertation focuses on one aspect of service delivery in public mental health systems, namely the role of strengths, if any, in Systems of Care for youth with emotional and behavioral problems. The dissertation conclusion discusses overall themes that emerged, notes limitations relevant to the entire dissertation, and discusses implications for social workers and others.