Child Maltreatment Incidence Data Linkages (CMI Data Linkages)
Prepared for:
U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research, and Evaluation
Using Hospital Data to Predict Child Maltreatment Risk
OPRE report number: 2021-254
Associated Project
Clients
Key Findings
- The project found a strong association between the risk level (generated by the CPS risk model) and maltreatment-related injuries in hospital records.
- No relationship was found between the risk level (generated by the CPS risk model) and medical encounters that were not maltreatment-related.
- The findings not only validate a model trained to predict future CPS involvement, but also indicate that maltreatment injuries are a credible indicator of child harm when viewed from the lens of the child protection system. Relationship between risk for placement and hospital encounters.
This report is an overview of the Children’s Data Network/Rady project, conducted as part of the Child Maltreatment Incidence Data Linkages (CMI Data Linkages) project. The project incorporates hospital records from San Diego County into existing linked data, including child protective services (CPS) records and vital birth and death records, to determine whether adding hospital records increased the predictive power of an existing predictive risk model (PRM) in identifying children and youth who experienced child maltreatment. The data were used to validate statewide predictive risk models built to predict future child protection involvement and explore the use of hospital records as predictors in predictive risk models (PRMs) focused on child maltreatment.
More like this from Mathematica
Efficiency Meets Impact.
That's Progress Together.
To solve their most pressing challenges, organizations turn to Mathematica for deeply integrated expertise. We bring together subject matter and policy experts, data scientists, methodologists, and technologists who work across topics and sectors to help our partners design, improve, and scale evidence-based solutions.
Work With Us