Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Purpose: Child welfare system reform efforts seek to reduce the need for foster care by investing in community and family-based preventive services. New York City's child welfare reinvestment plan aims to do this by identifying neighborhoods at high risk for child welfare involvement based on evidence that neighborhood social and economic characteristics are linked to child maltreatment. If the demand for foster care varies widely across neighborhoods, then directing resources to the areas with the greatest need is fiscally efficient as well. This study, based on work in New York City, addresses the following questions: 1) how does foster care utilization vary across neighborhoods and over time? 2) How much of these differences can be explained by social structural factors?

Method: Administrative data from the Administration for Children's Services in New York City and census child population data from the U.S. Census are used to derive community district placement rates of first-time foster care entry cohorts in 2000 (n=5,768) and 2004 (n=3477). Multivariate ecological models estimate the relationship between placement rates (and the change in rates) and covariates including census-derived measures of social and economic need, as well as historical placement rates.

Findings: The percent unemployed residents and percent black residents in districts and borough are significantly related to higher placement rates in 2000. The 2000 placement rate is the best predictor of the 2004 rate. The change in the district placement rate between years is best explained by the percent of female-headed households and the percent of black residents in 2000, as well as the median household income and the percent of owner-occupied housing units in 2005. Residuals maps from these models suggest areas where the rate of placement and the change in the placement rate is estimated to be higher or lower than predicted.

Implications: Districts where the placement rate did not change as much as expected present rich opportunities for preventive services and reducing the need for foster care. Districts where the placement rate changed more than expected are areas may already be benefiting from local social networks that are bringing placement rates down, even in the face of challenging social environments. This information is being used to redistribute funds for foster care and preventive services according to neighborhood need in New York City.