Identifying newborn discharge to child protective services: Comparing discharge codes from birth hospitalization records and child protection case files.

PURPOSE: Newborn removal by North America's child protective services (CPS) disproportionately impacts Indigenous and Black families, yet its implications for population health inequities are not well understood. To guide this as a domain for future research, we measured validity of birth hospitalization discharge codes categorizing newborns discharged to CPS. METHODS: Using data from 309,260 births in Manitoba, Canada, we compared data on newborns discharged to CPS from hospital discharge codes with the presumed gold standard of custody status from CPS case reports in overall population and separately by First Nations status (categorization used in Canada for Indigenous peoples who are members of a First Nation). RESULTS: Of 309,260 newborns, 4562 (1.48%) were in CPS custody at hospital discharge according to CPS case reports and 2678 (0.87%) were coded by hospitals as discharged to CPS. Sensitivity of discharge codes was low (47.8%), however codes were highly specific (99.8%) with a positive predictive value (PPV) of 81.4%, and a negative predictive value (NPV) of 99.2%. Sensitivity, PPV and specificity were equal for all newborns but NPV was lower for First Nations newborns. CONCLUSIONS: Canadian hospital discharge records underestimate newborn discharge to CPS, with no difference in misclassication based on First Nations status.

Infant rates of child protective services contact and termination of parental rights by first nations status from 1998 to 2019: An example of intergenerational transmission of colonial harm.

BACKGROUND: Rates of child removal by child protective services (CPS) in Manitoba are the highest in Canada with a profoundly disproportionate impact on First Nations families. Despite infants constituting the highest proportion of children affected, no research has examined population-level rates of infant contact with CPS. OBJECTIVE: We examined the incidence of infant contact with different levels of CPS, including termination of parental rights (TPR), according to First Nations status. PARTICIPANTS: We identified 217,261 infants (47,416 First Nations; 169,845 non-First Nations) born between 1998 and 2014 in Manitoba, Canada and residing in the province until at least age 5. METHODS: We used linked administrative data to calculate population-level rates of contact with different levels of CPS by First Nations status, including an open file before age 1, out-of-home placement before age 1, and TPR before age 5. RESULTS: Overall 35.8 % of First Nations infants had an open file, 8.5 % experienced out-of-home placement, and 5.4 % experienced TPR. Among other infants, 8.5 % had an open file, 1.3 % experienced out-of-home placement and 0.7 % experienced TPR. The rate of early-stage contact increased the fastest among First Nations infants, with a rise of 22.4 % in our study period, compared to a rise of 1.7 % among all other infants. CONCLUSIONS: CPS contact was exceptionally high among First Nations infants compared to other infants, with early-stage contact accelerating most dramatically over time. Findings support calls to greatly reduce the disruption of system contact in the lives of First Nations families.