Insuring the integrity and validity of social work interventions: the case of the subsidized guardianship waiver experiments

Longitudinal Evaluation of the Neurosequential Model of Therapeutics with Adoptive Families in Tennessee

PURPOSE

This study evaluated effectiveness of the Neurosequential Model of Therapeutics (NMT) with adoptive families who received post adoption services in Tennessee.

METHODS

Researchers obtained a sample of 552 families who received post adoption services in the U.S. state of Tennessee. Most families (77%) had adopted children through public child welfare services. A quasi-experimental design examined wellbeing outcomes for an NMT group (n = 319) versus a services-as-usual (SAU) group (n = 233) in intent-to-treat analyses. Then, the SAU group was contrasted to an NMT subgroup that had high adherence to the NMT model (n = 109) in an analysis of treatment-on-the-treated.

RESULTS

Intent-to-treat models indicated no differences on outcomes between the NMT and SAU groups. However, the results of treatment-on-the-treated analyses showed slightly greater reduction on the Behavior Problems Index over time for the NMT with high adherence group as compared to SAU.

DISCUSSION AND CONCLUSION

Adoptive families may face challenges that could be addressed through developmentally sensitive, trauma-informed services, such as NMT. The results of this study suggest that the NMT might benefit adoptive families if greater attention is paid to implementation adherence, or fidelity.

Age and other risk factors related to reentry to care from kin guardian homes

Although kinship guardianship is an increasingly important foster care exit pathway for children in the United States, research on the factors leading to kinship guardianship breakdown is lacking. This study examines the factors associated with guardianship breakdown for children who exited foster care to kinship guardianship in California between 2003 and 2010 (N = 18,831). Specifying time-dependent Cox relative risk models, children's age trajectories are directly accounted for in the analysis. This allows differentiation between duration dependence (i.e., time spent in guardianship) and children's development (expressed as age). Overall, 17.3% of children reentered care by 2017. Early adolescents, age 13-15 years (HR = 1.63, p < .001), and late adolescents, age 16-17 years (HR = 1.93, p < .001), had an increased hazard of reentry compared with children under the age of six. Children with a history of mental health concerns had more than twice the hazard of reentering than children without such a history (HR = 2.18, p < .001). Our findings indicate that transition to adolescence was associated with increased risk of reentry into care, highlighting the need for guardianship support services leading up to, and during, this child developmental stage.

Predictors of Consent in a Randomized Field Study in Child Welfare

PURPOSE

Randomized controlled trials (RCTs) are often viewed as the "gold standard" for proving the efficacy and effectiveness of new interventions. However, some are skeptical of the generalizability of the findings that RCTs produce. The characteristics of those willing to participate in research studies have the potential to affect the generalizability of its findings. This study examined factors that could influence consent among families recruited to participate in a randomized field trial in a real-world child welfare setting.

METHODS

This study tested the Parent Management Training Oregon Model for children in foster care with serious emotional disturbance. It employed a post-randomization consent design, whereby the entire sample of eligible participants, not just those who are willing to consent to randomization, are included in the sample. Initial eligibility assessment data and data from the federally mandated reporting system for public child welfare agencies provided the pool of potential predictors of consent. Bivariate and multivariate analyses were conducted to identify statistically significant predictors of consent.

RESULTS

Being a dual reunification family was the most significant factor in predicting consent. Unmarried individuals, younger, female parents, cases where parental incarceration was the reason for removal and cases where the removal reason was not due to their children's behavioral problem(s) were also more likely to participate.

DISCUSSION

As one of the first research studies to examine predictors of consent to a randomized field study in child welfare settings, results presented here can act as a preliminary guide for conducting RCTs in child welfare settings.

Formative Evaluation: A Case Study of the Illinois Birth-to-Three IV-E Waiver

Few interventions addressing early childhood trauma have adequate evidence. The Illinois Birth-to-Three Waiver is a 5-year federal project to reduce trauma and increase permanency for young children. In this article the authors describe the explorative and formative phases, including: (a) developing a well-built research question; (b) working with purveyors of evidence-supported interventions (ESIs); (c) adapting ESIs to a large public child welfare system; (d) developing fidelity measures, assessment data and outcomes; and (e) managing the challenges of implementing a randomized controlled trial. Public and private agency staff work in conjunction with university partners to merge implementation science, program evaluation, and intervention research. The project's theory of change stipulates that working with the triad of birth parent(s), foster parent(s), and focal child is critical to achieving project outcomes. The authors articulate the importance of working closely with an interdisciplinary team to achieve these outcomes, and the challenges associated with the early stages of implementation.