Key Points
- Most child welfare agencies in the United States depend on manual tools to assess children’s risk of
abuse and neglect. These tools are time consuming to complete, prone to human error and influence,
and rely on pre-determined questions and weights that have minimal predictive accuracy. - Automated predictive risk models represent an important advancement in risk assessment, relying on
information already captured in case management systems. They generate real-time risk classifications
that are more accurate, more consistent, and far less resource-intensive than traditional manual tools. - In the last decade, predictive risk models and predictive analytics, more generally, have been
successfully implemented to support child welfare staff across the decision-making continuum. Use
cases include or could include:
o Hotline screening of maltreatment allegations;
o Supervisory oversight of active child abuse and neglect investigations;
o Monitoring in-home cases to promote safety and well-being;
o Targeting real-time quality assurance and accountability reviews;
o Informing risk-based staffing assignments;
o Increasing caregiver recruitment and retention rates for both relatives and non-relatives; and
o Improving caregiver and child matching. - Findings from randomized controlled trials and quasi-experimental studies indicate that these tools
have the potential to improve child safety,1,2
create time savings for workers,3
reduce disparities,1,4 and
produce greater visibility and accountability for decisions. - By integrating predictive risk modeling into child welfare screening and assessment processes, states
can better fulfill their responsibility to provide consistent and data-driven services that improve
outcomes for children and families.