Amy L. Stevens

Examining Change in Therapeutic Alliance to Predict Youth Mental Health Outcomes

The objective of the study was to examine the link between therapeutic alliance and youth outcomes. The study was conducted at a group home with 112 youth with a disruptive-behavior diagnosis. Therapeutic alliance was collected routinely via youth and staff report. Outcome data were collected using youth and staff reports of externalizing behavior as well as behavioral incidents occurring during care. Outcome data were collected following intake into services and at 6 and 12 months of care. Data were analyzed to examine (a) whether youth behavior problems at intake were predictive of therapeutic alliance and (b) whether changes in alliance were predictive of subsequent youth outcomes. These were conducted with a 6-month service-delivery model and replicated with a 12-month model. There was some support for the first hypothesis, that initial levels of youth externalizing behavior would be related to alliance ratings; however, most of the effects were marginally significant. The second hypothesis, that changes in therapeutic alliance would be related to subsequent youth outcomes, was supported for the 6-month model, but not the 12-month model. Changes in therapeutic alliance may be predictive of youth outcomes during care. Additional research into examining therapeutic alliance trajectories is warranted to improve mental health services for youth.

Convergent Validity of the Strength-Based Behavioral and Emotional Rating Scale with Youth in a Residential Setting

Strength-based assessment has been identified as an appropriate approach to use in planning treatment and evaluating outcomes of youth in residential settings. In previous research, the Behavioral and Emotional Rating Scale-2, a standardized and norm-referenced strength-based measure, has demonstrated adequate reliability and validity with youth served in community and educational settings. The purpose of the present study was to examine the internal reliability and convergent validity of the BERS-2 by comparing the test to the Child Behavior Checklist and the Symptoms and Functioning Severity Scale. The results indicate that the scores from the BERS-2 are internally consistent and converge with other behavioral and emotional measures which, taken together, suggest that the BERS-2 could be acceptable for assessing the emotional and behavioral strengths of youth in residential settings. Study limitations and future research directions are identified.

Psychometrics of the Symptoms and Functioning Severity Scale for High-Risk Youth:

Youth in residential care have significant mental health needs, which require regular progress monitoring; however, very few emotional or behavioral assessments have been examined with this unique, high-risk population. This study examined the psychometrics of the Symptoms and Functioning Severity Scale (SFSS), a brief 24-item measure designed to assess the emotional and behavioral status of youth. This study examined the SFSS ratings from 143 youth with a disruptive behavior diagnosis living in a group-home facility in the Midwest and 52 of their service providers. Overall, the findings suggest that the psychometrics of the SFSS, when rated by staff or youth, were similar to the original outpatient clinical samples. More specifically, the Rasch analyses indicate that the SFSS items and the overall scale are performing adequately, and the confirmatory factor analyses replicated the two-factor structure for staff. However, the fit of the two-factor model was less compelling for youth ratings. In all, the brief SFSS seems a promising measure for assessing problem severity for youth in residential care.

Measurement Properties of the Motivation for Youth Treatment Scale with a Residential Group Home Population

BackgroundA client’s motivation to receive services is significantly related to seeking services, remaining in services, and improved outcomes. The Motivation for Youth Treatment Scale (MYTS) is one of the few brief measures used to assess motivation for mental health treatment.ObjectiveTo investigate if the psychometric properties of the MYTS previously found for youth receiving outpatient treatment would be replicated in a sample of youth entering residential care.MethodsThe study was conducted at a large residential group care agency in the Midwest with 145 youth who completed the 8-item MYTS upon intake into the group homes. To investigate the psychometrics of the MYTS, we used classical test theory, Rasch modeling, and confirmatory factor analysis approaches.ResultsMost of the psychometric properties were strong and replicated previous research in outpatient settings. The primary divergent finding was that the two subscales, problem recognition and treatment readiness were not significantly correlated in the residential sample, in contrast to the moderate correlations found in outpatient samples.ConclusionsThe MYTS can be recommended for use in residential settings, however, researchers and practitioners should carefully examine the sub-scale scores. Future research needs to replicate these findings in residential settings and further explore why the two subscales are minimally correlated.

Validation of the Symptoms and Functioning Severity Scale in Residential Group Care

Tests that measure the emotional and behavioral problems of children and youth are typically not normed and standardized on youth diagnosed with disruptive behavior, particularly those youth in residential care. Yet professional standards mandate that before instruments are used with a specific population the psychometric properties need to be studied and re-established: specifically, psychometric properties, including validity, need to be evaluated (AERA, APA, and NCME, The standards for educational and psychological testing. AERA, Washington, DC, 1999). The purpose of the present study was to assess the validity characteristics of the Symptoms and Functioning Severity Scale (SFSS; Bickman et al., Manual of the Peabody Treatment Progress Battery, Vanderbilt University, Nashville, TN, 2010), a widely used test developed for use in outpatient clinics, with youth in a residential care program. The convergent validity of the SFSS was established with the large correlations (0.78–0.86) with the CBCL. Several binary classification analyses including specificity, area under the receiver operating characteristic curve, positive and negative likelihood ratios, and the Youden Index supported the validity of the SFSS. However, the sensitivity index was somewhat low indicating the test may produce a high level of false negatives. Limitations, future research and implications are discussed.

Overweight and Obesity Among Youth Entering Residential Care: Prevalence and Correlates

Little is known about the prevalence of clinical weight problems for youth living in residential care. Therefore, this study examined the prevalence and correlates of overweight and obesity in a large sample of youth (N = 1709) entering a residential care program. Results indicated that 48% of youth were overweight or obese at the time of intake, which is much higher than national pediatric rates. Females had higher rates of overweight/obesity, as did youth referred from moderately restrictive placements such as foster care. Youth who were overweight/obese had greater internalizing symptoms at intake. Clinical and research implications are discussed.

Positive and Negative Peer Influence in Residential Care

The potential for negative peer influence has been well established in research, and there is a growing interest in how positive peer influence also impacts youth. No research, however, has concurrently examined positive and negative peer influence in the context of residential care. Clinical records for 886 residential care youth were used in a Hierarchical Linear Model analysis to examine the impact of negative and positive peer influence on naturally occurring patterns of serious problem behavior over time. Negative peer influence, where the majority of youth in a home manifested above the average number of serious behavior problems, occurred 13.7% of the time. Positive peer influence, where the majority of youth manifested no serious problem behaviors for the month, occurred 47.7% of the time. Overall, youth problem behavior improved over time. There were significantly lower rates of serious problem behavior in target youth during positive peer influence months. Conversely, there were significantly higher rates of serious problem behaviors in target youth during negative peer influence months. Negative peer influence had a relatively greater impact on target peers’ serious behavior problems than did positive peer influence. Caregiver experience significantly reduced the impact of negative peer influence, but did not significantly augment positive peer influence. Months where negative peer influence was combined with inexperienced caregivers produced the highest rates of serious problem behavior. Our results support the view that residential programs for troubled youth need to create circumstances that promote positive and control for negative peer influence.

Developmental effects of childhood household adversity, transitions, and relationship quality on adult outcomes of socioeconomic status: Effects of substantiated child maltreatment

The degree to which child maltreatment interacts with other household adversities to exacerbate risk for poor adult socioeconomic outcomes is uncertain. Moreover, the effects of residential, school, and caregiver transitions during childhood on adult outcomes are not well understood. This study examined the relation between household adversity and transitions in childhood with adult income problems, education, and unemployment in individuals with or without a childhood maltreatment history. The potential protective role of positive relationship quality in buffering these risk relationships was also tested. Data were from the Lehigh Longitudinal Study (n = 457), where subjects were assessed at preschool, elementary, adolescent, and adult ages. Multiple group path analysis tested the relationships between childhood household adversity; residential, school, and caregiver transitions; and adult socioeconomic outcomes for each group. Caregiver relationship quality was included as a moderator, and gender as a covariate. Household adversity was negatively associated with education level and positively associated with income problems for non-maltreated children only. For both groups, residential transitions was negatively associated with education level and caregiver transitions was positively associated with unemployment problems. Relationship quality was positively associated with education level only for non-maltreated children. For children who did not experience maltreatment, reducing exposure to household adversity is an important goal for prevention. Reducing exposure to child maltreatment for all children remains an important public health priority. Results underscore the need for programs and policies that promote stable relationships and environments.