What Can We Learn About the Incidence of Foster Care Placement from Birth Records?

Finding new and effective ways to monitor for child maltreatment can help to shape prevention and intervention efforts and inform the policies and programs designed to keep children safe. The Cross Jurisdiction Model Replication (CJMR) project examined a new way to predict foster care placement through a risk prediction model built from population-level and anonymized birth records. After developing and validating the model in California, the CJMR project team explored if, and to what extent, the model could then be used to predict foster care placement in additional jurisdictions (Alaska and Kentucky). This brief summarizes the development, validation, and replication of the risk prediction model and shares key findings and their implications for the field.