Using Administrative Data to Assess Child Safety in Out-of-Home Care
Garnier, Philip C., Poertner, John, Child Welfare
This article describes efforts to produce useful safety measures from administrative data. A measure similar to that proposed by the U.S. Department of Health and Human Services is presented and compared to a measure that takes into account the length of time children are in placement. These measures are also reported for out-of-home care placement types. The challenges posed in constructing such measures from extant data are discussed.
Social workers remove children from parents and place them in out-of-home care because it is their judgment that those children are at risk for (continued) abuse or neglect. What, however, is the risk of abuse or neglect to those children in outof-home care? The popular media abound with stories of neglect, abuse, and even death among children who are in the care of foster parents, group homes, and institutions. These stories do little more than inflame public opinion. They furnish little background regarding the prevalence of such outcomes for children placed in out-of-home care, nor do they indicate the relative risk of abuse and neglect within a child's home of origin or the rate of abuse and neglect in the general population.
The professional literature provides little help. Despite the fact that child welfare professionals have for many years placed children in family foster care and other out-of-home care arrangements, a significant and reliable body of research on the safety of children in these living arrangements has not developed. Prior to 1984, reports of abuse or neglect of children in out-of-home placements were often counted as violations of placement licensing standards rather than as incidents of abuse or neglect. In 1984, the National Center on Child Abuse and Neglect began requiring states participating in their basic grant program to treat reports of abuse and neglect of children in out-of-home care like all other reports of abuse and neglect [Rosenthal et al. 1991].
The relative lack of data on the safety of children in the care of public welfare agencies may be understandable, given the difficulty of acquiring such data. First, most child welfare jurisdictions do not have information systems that allow them to produce these statistics. Even in states that have the capacity to report safety data, there are considerable challenges in producing useful measures from administrative databases. Second, direct collection of this information is very expensive.
Two recent initiatives should contribute to the understanding and assessment of the safety of children in-out-of home care. The first, Section 203 of the Adoption and Safe Families Act of 1997 [PL. 105-89], mandates that the U.S. Department of Health and Human Services develop a common set of child welfare outcome indicators to be the basis of annual reports to Congress. Among the indicators to be developed is a measure of abuse and neglect in out-of-home care. Specifically, Child Welfare Outcome 2 asks, "Of all children who were in foster care during the reporting period, what percentage was the subject of substantiated or indicated maltreatment by a foster parent or facility staff?"
The second initiative is the Statewide Automated Child Welfare Information System (SACWIS) authorized by the Social Security Act of 1996. SACWIS provides financial incentives for states to develop, as its name implies, automated information systems that can track and report systemwide child welfare outcomes, including safety.
This study reports on the development of indicators for the safety of children in out-of-home care using administrative data from the Illinois Department of Children and Family Services. First, rates of abuse and neglect are reported for a four-year period covering state fiscal years 1996 through 1999. Overall rates of abuse and neglect are reported for the four years as well as the relative rates across different types of out-of-home care living arrangements. Second, the measurement of abuse or neglect as a simple percentage is examined, and an alternative is presented that accounts for the length of time a child is exposed to risk of abuse or neglect. …