Leveraging predictive analytics in child protection case closures: informed decisions for child welfare practice - the leveraging predictive analytics in child protection case closures: informed…

The Department of Health and Human Services Administration for Children and Families awarded the Executive Office of the State of New Jersey $599,970 in a cooperative agreement on September 30, 2026, under the Child Abuse and Neglect Discretionary Activities program (CFDA 93.670) to design, validate, pilot, and evaluate a predictive analytics decision-support tool for child protective services casework.

The New Jersey Department of Children and Families will develop the tool to support investigation caseworkers and supervisors in deciding whether to close a family's case following a child protective services investigation or determine that additional support is warranted. The tool integrates predictive analytics with existing child welfare administrative data, safety and risk assessments, policy, frontline staff professional judgment, and existing workflows. The project employs a phased approach including assessment and preparation, model development and validation, pilot testing, and evaluation, guided by stakeholder input throughout. The department will establish a framework for data quality, monitoring and evaluation, transparency, and workforce support to ensure appropriate tool use, performance, and sustainability.

Over half of CPS reports involve families with prior involvement with the Division of Child Protection and Permanency, with more than one in four involving families whose cases closed within the past year. The tool aims to strengthen earlier identification of families who may need additional support before concerns escalate to subsequent maltreatment findings or out-of-home placement.

The period of performance runs through September 29, 2029. Work is performed in New Jersey.

Leveraging predictive analytics in child protection case closures: informed decisions for child welfare practice - the leveraging predictive analytics in child protection case closures: informed decisions for child welfare practice project is administered by the new jersey department of children and families (DCF) to enhance decision-making among frontline child protective services (CPS) staff. The purpose of the project is to design, validate, pilot, and evaluate a predictive analytics decision-support tool to support investigation caseworkers and supervisors in deciding whether to close a family's case following a CPS investigation or determine that additional support may be warranted. Many families who come to DCF's attention have current or prior involvement with DCF's division of child protection and permanency (CP & P), new jersey's child protection and welfare agency. Over half of CPS reports involve families with a previously closed CP & P case. More than one in four reports involve families whose cases closed within the past year. This suggests there is an opportunity to strengthen earlier identification of families who may need additional support before concerns escalate to subsequent maltreatment findings or out-of-home placement. Using a phased approach- to include assessment and preparation, model development and validation, pilot testing, and evaluation- and guided by stakeholder input at each phase, DCF will create a transparent, practical, and useful predictive analytics decision-support tool, using DCF's child welfare administrative data, that is seamlessly integrated with information derived from existing safety and risk assessments, policy, frontline staff's professional judgement, and existing workflows to improve decision-making. Through the intentional creation and implementation of a framework for data quality, monitoring and evaluation, transparency, and workforce support, DCF will ensure the tool's appropriate use, performance, and sustainability overtime. The agency will also set the groundwork to maintain and evaluate the tool's capability, replicability, and translatability to other areas of child welfare practice beyond the grant period. This project represents a strategic investment in data-driven decision-making, promising to deliver long-term benefits for the families and children served by DCF.