Download PDF

In 2015, an estimated 683,000 children in the United States were victims of abuse and neglect, and 428,000 children were living in foster care. For children in foster care, the harmful effects of maltreatment are often intensified by exposure to additional traumatic events, such as separation from family members and multiple placement disruptions. For example, while in care, children experience an average of 3.2 placement changes. In turn, they may exhibit more severe problems with attachment, behavior problems, and mental health, including higher rates of posttraumatic stress disorder compared with their peers who remain at home or are not maltreated. Child welfare systems that are sensitive and responsive to trauma are better positioned to support the safety, permanency, and well-being of children in foster care.8 Nevertheless, the vast majority of child welfare agencies do not provide resource parents (foster parents and kinship caregivers) adequate preparation to manage the daily challenges of caring for children who are exposed to trauma and to support their recovery. To help address this gap, The Duke Endowment worked with Child Trends to evaluate a pilot implementation of a new training for resource parents and child welfare staff—ARC Reflections. The training is based on the principles of Attachment, Self-Regulation, and Competency (ARC), a widely disseminated, evidence-informed framework and clinical trauma treatment model.

This report presents findings from a mixed methods evaluation of the implementation of ARC Reflections in five counties in North Carolina. The evaluation focused on identifying essential elements of ARC Reflections, assessing fidelity to the model, and highlighting adaptations that may increase the model’s effectiveness in child welfare settings. An additional component of the evaluation was an exploratory analysis of associations between ARC Reflections and placement stability and resource home retention. Prior to presenting findings, we provide an overview of ARC Reflections and the ARC model on which it is based.

Key Findings

Overall, resource parents, child welfare leaders, and caseworkers held positive perceptions of ARC Reflections. The results of the evaluation revealed the following strengths:

• ARC Reflections was implemented with high fidelity. Trainers followed the format of the training 90 to 100 percent of the time (except in one county, where trainers elected to reduce the use of icebreakers after the first few sessions).

• Resource parents found the training interesting, useful, and practical. A high percentage of resource parents (73 to 99 percent) agreed the training was interesting and balanced, presenters were clear, and activities were helpful. In focus groups, they reported learning useful and practical tools and approaches to caring for children exposed to trauma.

• Resource parents gained knowledge and skills related to child trauma. Resource parents’ scores on the Resource Parent Knowledge and Beliefs Scale11 increased significantly from pre- to post-training, showing improvements in trauma-informed parenting, tolerance of children’s misbehavior, and parenting efficacy. These gains were maintained at follow-up three months after the implementation period ended.

• Family protective factors increased. Resource parents completed the Protective Factors Survey12 prior to the training and again at follow-up and reported a significant increase in praising children when they behaved well and in feelings of closeness between resource parent and child.

• Fewer trained resource homes closed compared with untrained homes. The percentage of resource homes with resource parents trained in ARC Reflections that closed for negative reasons (i.e., reasons other than adoption or guardianship) as of three months after the implementation period was significantly smaller than the percentage of homes that closed for negative reasons during the year prior to implementing ARC Reflections (2 percent vs. 16 percent).

• Fewer children exited trained homes compared with children in untrained homes. A significantly smaller percentage of children in trained resource homes exited for negative reasons (i.e., move to another placement setting, transferring to another agency, running away) as of three months after the implementation period compared with children in untrained homes during the year prior to implementation (7 percent vs. 43 percent).

• Several child welfare agencies plan to continue offering ARC Reflections. Three of the five counties plan to offer ARC Reflections training for resource parents beyond the pilot period. They will also offer the ARC Reflections training for new child welfare staff. One county plans to provide booster trainings for existing staff, and another county will offer trainings for staff who work with kinship caregivers. Adequate resources (e.g., financial, number of trained staff), leadership support, and staff “buy-in” appear to be factors that support sustainability of the training.

In addition to lessons learned and suggestions for improvement, findings also elucidated specific challenges to implementing ARC Reflections in child welfare:

• A more integrated approach to training would further support a trauma-informed system. Child welfare staff and leaders suggested that training all staff and resource parents would facilitate shared knowledge of how to understand and address child trauma.

• Additional support for implementing ARC Reflections may enhance outcomes. Trainers and child welfare leaders
indicated that they would benefit from additional instruction, supervision, and coaching, such as more guidance on training caseworkers, booster sessions, and follow-up with resource parents during and after training. Caseworker training was the least developed component of the ARC Reflections curriculum, and all counties expressed the desire for more guidance and direction on how and when to train caseworkers. These types of support may allow for more consistent and higher quality integration of ARC Reflections and trauma-informed care into child welfare practice.

• Careful selection of trainers is important. Child welfare staff reported that, although they were generally satisfied with trainers, successful implementation required careful selection of trainers, prioritizing high-quality trainers with prior training experience. One county also suggested including a resource parent as a co-trainer, a practice utilized in several other child welfare trauma training initiatives.

• Limited time and resources in child welfare agencies impeded sustainability. Child welfare staff and several resource parents reported that the length and duration of the training was a barrier to sustaining ARC Reflections given other commitments and concurrent initiatives. In addition, some agencies did not have the financial resources to cover costs such as child care and food for resource parents during training sessions, which were important to resource parent attendance and engagement. Given many competing priorities, successful implementation of ARC Reflections requires child welfare agenciesto give high priority to trauma training.

Implications and Conclusions

The results of this evaluation indicate that ARC Reflections was successfully implemented in the five North Carolina counties. Resource parents, child welfare leaders, and caseworkers found the training to be useful and practical, and resource parents made significant gains in knowledge and skills related to caring for children who have experienced trauma. A preliminary analysis also suggested the ARC Reflections training was associated with improvements in resource home retention and children’s placement stability. However, more rigorous study is needed to confirm these findings and to attribute positive outcomes directly to the training. Lessons learned include the importance of gaining support from leadership and staff early in the implementation process, selecting experienced trainers, providing system-wide training, and ensuring that training is accessible and convenient for resource parents.