Child Welfare Outcomes for Young Children of Color with Substance Removals: An Intersectional Analysis of Longitudinal National Data

Background/Purpose:  The opioid epidemic has resulted in increasing attention to the effect of parental substance use disorders on child welfare system involvement, including foster care utilization. One earlier study using national data reported substantial recent increases in the number of infants removed due to parental drug use and observed that the rate of infants entering foster care depended on both state policy and prevalence of neonatal abstinence syndrome. Additionally, national data suggests that young children (under age 4) entered foster care at double the rate of older children in 2017. Unknown is whether young children of color with substance removals are also entering care at increasing rates and whether disparities exist in child welfare outcomes. This study aimed to determine rates of young children of color (ages 0-3) with substance removals in foster care across every state and territory between 2007-2017 and evaluate the impact of race, age, and substance removal status on child permanence.

Method: Adoption and Foster Care Analysis and Reporting System (AFCARS) data identified all children whose most recent foster care placement started between 2007-2016.  Cases were tracked through September 30, 2017.  Longitudinal descriptive characteristics were explored across states and nationally. Cluster robust Cox proportional hazard models were employed to test the primary research questions while accounting for nesting within states. Two randomly selected 25% sub-samples were analyzed (n=607,172). Child length-in-care was calculated based on foster care exit date for children who exited prior to the last date of observation.  Time-in-care care for children who remained out of home (censored cases) was calculated by subtracting the last date of observation from their most recent removal date.  Likelihood of reunification and permanence were estimated across an 8-level categorical variable accounting for all combinations of three dichotomized variables: child age (0-3/4+), race (white/non-white), and removal due to substance use (yes/ no). Model covariates included sex, number of prior foster care episodes, diagnosed disability status, number of federal benefits received, and rural/urban continuum code.

Results: Between 2007-2016, the number and proportion of white children with substance removals (ages 0-3 and 4+) in foster care consistently increased.  However, children of color, specifically ages 0-3, faced disadvantages respecting foster care outcomes.  Results of the Cox models revealed that permanency was more likely for all other groups of young children, and older white children with substance removals, compared to young children of color with substance removals. The only group at significantly higher risk of failure to achieve permanency was children of color ages 4+ with substance removals.  Regarding reunification, every group was significantly and substantially more likely to reunify than young children of color with substance removals.

Implications: Federal and state responses to the opioid crisis have substantially expanded resources to families with substance use issues in child welfare.  Results of this study suggest that white families may be benefiting from these resources to a greater extent than families of color.  Given geographic and cultural differences in drug use patterns, expansion of opioid-specific services may be contributing to this phenomenon.