ABSTRACT
There is much to be learned about “what it takes” to scale-up evidence-based practices in real-world community settings. Although increasing attention has focused on factors that influence the adoption, implementation, and sustainability of evidence-based practices (EBPs) in child and family service sectors, controlled studies that directly examine these factors empirically in real-world settings are limited. This chapter describes the need for scale-up of EBPs and provides a description of a large-scale randomized implementation trial of an evidencebased mental health intervention delivered through county social service systems. In this trial, 51 counties in California and Ohio were randomly assigned to one of two implementation conditions for one of the few evidence-based training programs for foster parents, called Treatment Foster Care Oregon (TFCO; formerly Multidimensional Treatment Foster Care; MTFC). TFCO is a community-based alternative to placement in residential or group care settings for children and adolescents with severe behavioral and/or emotional problems (www.mtfc.com). The two study conditions were: 1) the experimental condition called the Community Development Teams (CDT), an implementation strategy that brings together cohorts of 5-8 counties that use peer-to-peer learning to facilitate successful implementation and outcomes, or 2) the no-treatment control condition called the “business as usual” model of individualized single site/agency implementation (IND). As will be described, TFCO is a top-tier evidence-based mental health intervention for high-needs youth in the juvenile justice and child welfare systems. In this chapter we describe the need for the scale-up of practices such as TFCO, the background of the TFCO model, the implementation trial design, a comparison of implementation outcomes for
sites randomly assigned to the CDT or IND conditions, and an innovative method for measuring implementation milestone progress. We also discuss considerations when conducting EBP scale-ups, including monitoring and supporting model fidelity, infrastructure building for both developers and sites/ agencies, and sustainability considerations.
