A Difference-in-Difference Analysis: Does State-Level Medicaid Expansion Reduce Racial Disproportionality in the U.S. Child Welfare System?

Background and Purpose:

In the United States, racial disproportionality (RD) in healthcare and the child welfare system (CWS) is a well-known issue. Recently, Medicaid expansion (ME) has been associated with decreasing racial disparities in some healthcare outcomes, such as mental healthcare service utilization, and maternal and infant health outcomes. ME was also associated with significant decreases in the number of neglect reports and foster care admissions due to neglect (e.g. Brown et al., 2019; Beland et al., 2021). To date, however, no study has investigated the impact of Medicaid expansion on RD in the CWS. This study begins to fill this gap.

Methods:

Our analysis used data from three sources: For all 50 states, National Child Abuse Neglect System Child Files were combined with demographic data from the American Community Survey from 2008-2019, and the state’s ME status from the Kaiser Family Foundation. First, we calculated the disparity indices (Shaw et al., 2008) for overall child maltreatment, physical abuse-only, and neglect-only reports for five different racial/ethnic groups: Asian American, Black/African American, Hispanic/Latino/e/x, multiracial, and Native American/Alaskan. Second, we conducted a synthetic difference-in-difference analysis (Arkhangelsky et al., 2021; Porreca, 2022) with three treatment groups: those who expanded ME in 2014, 2015, and 2016. The control group consisted of the states that had not expanded Medicaid by 2018.

Results:

Our results indicate that disparities across different forms of maltreatment reports did not change due to Medicaid expansion as only 8.9% (n=4) of our results were statistically significant. The following significant results should be interpreted with caution. The 2014 expansion group saw an increase of 0.32 in the disparity of physical abuse reports between Black/African American and White children (95% CI: 0.02, 0.62). The 2015 expansion group saw a decrease of 0.66 in the disparity of overall maltreatment reports between Native American/Alaska Native and White children (95% CI: -1.30, -0.02), and a decrease of 0.55 in the disparity of neglect reports between Black/African American and White children (95% CI: -1.05, -0.05). The 2016 expansion group saw a decrease of 0.18 in the disparity of maltreatment reports between Hispanic/Latino/e/x and White children (95% CI: -0.30, -0.05).

Conclusions and Implications:

To our knowledge, this is the first study to investigate the impact of ME on RD in the CWS, and while our results need to be interpreted with caution, the fact that we did not find as significant a reduction in RD as expected could indicate several things. More inter-system collaboration and research is warranted to explore why the decrease in RD in health outcomes and decreases in neglect reports/foster care entries due to neglect do not translate to lower RD in the CWS. Further, parents may be eligible but either do not have Medicaid or have Medicaid but are not accessing services for a myriad of reasons. Future research should investigate persisting barriers to healthcare access for parents involved in the CWS to inform policy and practice adjustments to successfully combat RD in the CWS, so that all families may thrive.