This brief describes the innovative uses of IDS data, how it shapes decision-making on the ground and how it improves outcomes for vulnerable children, families and adults. The human services departments in Allegheny County, Pennsylvania and Washington state are leaders in using an integrated data system to develop predictive modeling tools used by frontline case managers in the child welfare system and in the Medicare-Medicaid systems. These departments also share IDS data with their service provider networks to empower caseworkers and improve service coordination.
Integrated Data Systems are the perfect tool for developing Predictive Risk Models
Findings & Stats
Early evidence suggests predictive modeling supports improved screening and targeting of cases.
Allegheny County’s predictive model found in historical data that 27% of the cases identified as highest risk with the predictive modeling tool had been screened out by call screeners.
Almost half (48%) of the children identified as the lowest risk by the Allegheny County predictive model had been screened in by previous call screeners and investigated with over 33% being re-referred and placed in foster care.
Early results from a pilot program in Washington state providing intensive care coordination and using the predictive modeling tool saved $21 million in Medicare expenditures.
Statements & Quotations
An IDS has become part of our culture here. These new practices are intended to get data where it needs to be— in the hands of caseworkers.
– Erin Dalton, Allegheny County Department of Human Services, Office of Data Analysis, Research, and Evaluation
In Washington state our experience has been that the return from IDS has been well worth the investment.
– David Mancuso, Research and Data Analysis Division, Department of Social and Health Services
Subscribe to our newsletter to get our data, reports and news in your inbox.