Standardized Prenatal Substance Exposure Protocols and Racial Disparities in Child Welfare Reporting: Evidence from Connecticut

Background/Purpose: Black infants with prenatal substance exposure (PSE) are disproportionately reported to Child Protective Services (CPS), reflecting broader racial disparities in child welfare reporting. Standardizing hospital practices for identifying and referring infants with PSE to CPS has been proposed as a strategy to mitigate these disparities. In 2019, Connecticut implemented a standardized system for tracking and responding to PSE cases, incorporating risk assessment questions, family care plans, and de-identified notifications to CPS. This presented an opportunity to test whether a standardized, universal risk assessment approach is associated with a reduction in racial disparities in maltreatment reports involving PSE.

Research Questions:

• How did CPS reporting rates for PSE differ between Black and non-Black infants in Connecticut before and after the 2019 implementation of a standardized system?
• Was this implementation associated with a reduction in racial disparities in CPS reporting rates?

Methods: Using a difference-in-differences (DiD) approach, this study conducted an aggregate-level analysis of administrative child welfare and birth certificate data from January 2018 to July 2022. The sample included 9,383 CPS reports and 151,690 birth records. Month/year and racial group combinations (Black vs. non-Black) were used as observational units. Unadjusted and adjusted DiD models estimated maltreatment report rates for PSE for both racial groups before and after practice implementation (Research Question 1). Adjusted OLS regressions including an interaction term assessed the association of implementation with racial disparities (Research Question 2). Models controlled for maternal alcohol consumption, absence of prenatal care, and neonatal abstinence syndrome diagnosis. Primary analyses used the full population, while secondary analyses were restricted to a more socioeconomically homogenous subgroup of publicly insured mothers.

Results: Before implementation, Black infants were reported to CPS at significantly higher rates than non-Black infants (28.76 vs. 11.65 per 1,000 births; p<0.01). After implementation, reporting rates for Black infants decreased to 23.50 per 1,000 births while non-Black rates remained stable at 11.33 per 1,000 births. The racial disparity declined by 5.39 reports per 1,000 births in the full sample (not significant) and by 10.92 reports per 1,000 births in the public insurance subsample (p<0.05). This reduction in disparity occurred over a year and a half after implementation.

Conclusions and Implications: Connecticut’s standardized system for identifying and referring infants with PSE was associated with a reduction in racial disparities in CPS reporting among mothers with public insurance. However, the delayed reduction in disparity and the lack of a significant reduction in the full sample suggest a need for cautious interpretation. These findings highlight the intersection of racism, socioeconomic stress, and child maltreatment. Standardized protocols may help reduce racial disparities within socioeconomically vulnerable populations who are most likely to interact with child welfare systems.