全球智库信息集成服务系统

A growing body of research shows that well-designed and well-implemented home visiting program can help prevent child maltreatment. Home visiting programs can also improve the health of mothers and their children, strengthen parenting skills, and improve school readiness.

In 2008, the Children’s Bureau in the Administration for Children & Families in the U.S. Department of Health and Human Services entered into cooperative agreements with 17 organizations in 15 states to support the implementation of home visiting programs that have potential to prevent child maltreatment. The agency identified three goals:

1. Support implementation with fidelity to the home visiting program models
2. Support scale-up of the home visiting models—replicating the program model in a new service area, adapting the model for a new target population, or increasing the enrollment capacity in an existing service area
3. Support sustainability of the home visiting model beyond the end of the funding period

Mathematica Policy Research and Chapin Hall at the University of Chicago conducted a national cross-site evaluation of the Supporting Evidence-Based Home Visiting to Prevent Child Maltreatment (EBHV) initiative. Using a mixed-methods approach, we designed the national cross-site evaluation to (1) examine the degree to which system change occurred, (2) document the fidelity with which the program models were implemented, and (3) identify implementation strategies and challenges.

Key Findings

The study team made several key findings: 

• Implementing agencies consistently met many fidelity standards, but struggled to maintain caseloads and deliver services at the intended intensity. Sustaining full caseloads for home visitors, retaining participants, delivering services at the intended dosage, and achieving consensus on goal-setting proved challenging.
• Fidelity of implementation varied more by implementing agency than by home visiting model, demonstrating the significance of the characteristics of agencies and local service networks.
• Higher-risk families were more likely than those at lower risk to leave the program early. Younger, more economically disadvantaged and potentially more socially isolated participants (as suggested by their single-parent status) left multiyear home visiting programs early or, if enrolled in short-term programs, did not successfully complete them.
• Changes in the national and local contexts influenced the order in which infrastructure-building activities were carried out. Due to the roll-out of a new national home visiting initiative (known as the Maternal, Infant, and Early Childhood Home Visiting Program) and the economic downturn, the 17 grantees engaged in a new round of planning activities midway through the initiative (2011), and focused on stabilizing their funding for implementation. 
• During the late implementation phase of the initiative (2012), grantees focused efforts on building sustaining infrastructure. Partners of grantees that built sustaining infrastructure were more likely to perceive that the initiative’s goals had been achieved.
• Partners’ reports of the quality of their collaboration with one another were associated with achieving the initiative’s goals of implementation with fidelity, scale-up, and sustainability.

Mathematica and Chapin Hall also conducted a study of the costs of evidence-based home visiting programs, in partnership with Casey Family Programs. The research team noted the following:

• On average, implementing agencies spent $6,583 serving a family from program entry through exit. Costs per family ranged widely among agencies, from a minimum of $2,122 to a maximum of $13,962.
• Costs varied by program model and other agency characteristics, including model implementation status (new or continuing), agency location (rural, suburban, or urban), and agency type (government, medical center, or private nonprofit). Across all agencies in the cost study sample, the average estimated annual cost to operate a home visiting program was $580,972. Agencies served an average of 110 families during the year.