Evaluation of Racial Disparities in Suspected Child Abuse among Insured Children with Head Injury

Abstract

Previous studies demonstrate racial disparities in child abuse evaluations even after controlling for health insurance coverage. We conducted a cross-sectional study using the Military Health System Data Repository (MDR) of Military Health System (MHS) beneficiaries born between fiscal years 2016 and 2018 to evaluate racial disparities and other factors in the suspicion of child abuse. We observed beneficiaries for 2.5 years after birth and assessed the incidence of head injury using diagnostic codes. Among children with head injury codes, we performed a multivariable logistic regression analysis to measure the association between race and the concurrent use of the diagnostic code for suspected child abuse (SCA) or for the performance of a skeletal survey. There were 195,893 infants included and 45,269 (23.1%) underwent evaluations for head injury. Less than one percent ( n = 424) concurrently had the diagnostic code for SCA or a skeletal survey performed within 2 weeks of the head injury evaluation. When controlling for other factors, race was not associated with SCA. Higher military rank was independently associated with decreased odds of SCA. Racial disparities in SCA may be mitigated in the MHS, and further evaluation is needed. Military rank could be a factor in SCA disparities and warrants further study.

Our Commitment to Promoting Diversity, Equity, Inclusion, and Justice in Child Maltreatment

“…In an attempt to disentangle these issues through research, there have been calls from scholars to “disaggregate” samples into same-race groups to empirically evaluate the benefits of disaggregation within a child welfare sample. Jones and Banaag et al, (2023) used National Child Abuse and Neglect Data System and U.S. Census data to employ sample and variable-creation disaggregation. The authors found that both kinds of disaggregation were demonstrably beneficial and should be considered for future research using national geographically structured child maltreatment data.…”