Vol. 14, No. 1
April 2009

Assessing Risk Assessment in North Carolina

One of the seven fundamental strategies of MRS is the redesign of in-home services and case management. Under this strategy, the families with the greatest needs receive the most intensive services, while those with fewer needs receive less intense services. Ideally, families at lower risk can still receive voluntary, preventive support services, while greater resources are available for families where children are at greater risk for maltreatment.

Of course, for this strategy to be realized, social workers have to be able to accurately assess the risks, strengths, and needs of each family that comes into contact with child protective services. In 2002, North Carolina adopted the Structured Decision Making (SDM®) tools statewide in the hopes of achieving accurate, consistent assessments.

Currently, the Division of Social Services is working with experts from the Children’s Resource Center, a nonprofit research organization that developed the Structured Decision Making Tools, to evaluate how well SDM tools measure risk factors for families in North Carolina, and how our assessments can be improved.

A Validation Study
To do this, the CRC conducted a validation study of one of the SDM tools, the Risk Assessment (the DSS-5230). What does this mean? Researchers attempted to determine whether the risk level families were assigned using the Risk Assessment (high, moderate, or low risk) actually correlated with how likely those families’ children were to experience another episode of maltreatment.

In other words, the CRC and the Division asked: are families rated as “high risk” actually at the highest risk for experiencing repeat maltreatment? Likewise, are those families rated “low risk” during their assessment actually those least likely to experience repeat maltreatment?

To find out, CRC selected a random sample of approximately 2,000 family case files from 23 counties across the state. They reviewed all of the SDM tools completed for the selected families, then looked to see whether the families had any repeat episodes of maltreatment.

The SDM tools were originally developed in Michigan by looking at risk factors from a sample of families substantiated for maltreatment in the early 1990s. North Carolina’s new validation study will help ensure that we base our tools on current data from a wider range of families in our own state, where demographics and child welfare practices are different than they are in Michigan.

This validation study demonstrates the Division’s continuing commitment to family-centered MRS strategies and to supporting high quality, evidence-based child welfare practice. It will also provide some important information for continued improvement. For example, the validation study should help us identify which specific risk factors put children in our state at the greatest risk for maltreatment. At the same time, it can help us identify which strengths and needs either increase or mitigate children’s risk for maltreatment. This can help us more effectively target prevention and intervention efforts.

Upcoming Changes
The CRC has completed its validation study and, based on what it learned, recommended changes to North Carolina’s Risk Assessment (DSS-5230) and Risk Reassessment (DSS-5226) tools. These recommendations will soon be reviewed by the project’s advisory group, the NC Association of County Directors of Social Services’ Children’s Services Committee, and the Division. If these changes are approved, revised Risk Assessment and Risk Reassessment tools should be available later in 2009.

As agencies seek to refine their family-centered child welfare practices, workers need practical, reliable tools to help make critical decisions about the service needs of families. This project’s evidence-based information should serve as an important supplement to the family-centered assessment skills of workers in the field.