Over the course of their childhood, many US children will experience maltreatment that is confirmed by child protective services (CPS) or will be placed in foster care. According to recent estimates, roughly 1 in 8 US children will ever experience CPS-confirmed child maltreatment, with roughly 1 in 8 ever being placed in foster care (with higher risks of both events for African American and Native American children). Knowledge about how these cumulative risks vary geographically remains quite limited; however, which is unfortunate since these child welfare events are risk factors for poor child health and wellbeing, as well as for later adult health and wellbeing. The goal of this project is o advance the understanding of state- and county-level variations in the cumulative prevalence of three child welfare events-having CPS investigate an allegation of maltreatment, having CPS confirm that maltreatment did occur, and being placed in foster care. We plan to accomplish these goals in two stages. In the first, we will use life tables and data from the National Child Abuse and Neglect Data System (NCANDS) and Adoption and Foster Care Analysis and Reporting System (AFCARS) to generate the first state-level estimates of the cumulative proportion of US children who will ever experience (a) a CPS investigation, (b) a confirmed maltreatment case, and (c) foster care placement by age 18 (ages 8 to 12 for a CPS investigation). We will also estimate each risk by race/ethnicity and sex. In the second, we will use the same data and techniques to generate the first county-level estimates of the cumulative proportion of US children who will ever experience (a) a CPS investigation, (b) a confirmed maltreatment case, and (c) foster care placement at any point before age 18 (ages 8 to 12 for having a CPS investigation). We will again estimate each risk by race/ethnicity and sex. Because the NCANDS and AFCARS data do not provide county- level identifiers for counties with fewer than 1000 annual investigations (for NCANDS) or foster care placements (for AFCARS), the second stage of our project will involve a unique user agreement between the PI of this project and Michael Dineen, the chief data analyst for both the NCANDS and AFCARS data at the National Data Archive on Child Abuse and Neglect (NDACAN) at Cornell University, that will generate county- level estimates for all US counties. By providing the first state- and county-level estimates of the cumulative proportion of children who ever experience these three child welfare events, we greatly extend research on how the risk factors children face for poor health and wellbeing vary geographically and also inform policies and programs targeted at reducing the risks of maltreatment and foster placement.
Using demographic techniques and administrative data on all children who experience a Child Protective Services (CPS) investigation, confirmed maltreatment, or foster care placement, this project will produce six scientific papers that generate state- and county-level estimates of children's cumulative exposure to these three child welfare events by race/ethnicity and sex. In so doing, this project will dramatically enhance scientific knowledge regarding the spatial concentration of these child welfare events, generating important knowledge regarding the unequal distribution of these risk factors for poor child health and wellbeing.
