Program Profile: Structured Decision Making (SDM)

This is a decision-support system designed to reduce subsequent abuse and neglect of children and expedite permanency for children who have been removed from home. The program is rated Promising. At the 12-month follow up, treatment group children experienced statistically significant higher rates of permanency status, compared with control group children. However, there were no differences between groups on rates of reentry to the foster care system after reunification.

A Promising rating implies that implementing the program may result in the intended outcome(s).

Program Goals/Program Theory

Structured Decision Making (SDM) is a case management system that provides objective and structured assessment tools and decision-making guidelines for child welfare agencies that handle cases of child abuse and neglect. The purpose of the model is to increase the level of consistency and validity of decisions through the use of evidence-based assessments, improve the decision-making process by introducing structure at critical decision points for a case and ensure that the limited resources available in the system are targeted to families with the highest likelihood of subsequent abuse or neglect of a child.

SDM operates under the theory that the systematic use of structured case assessment and decision-making procedures should improve the services delivered to children and families and ultimately produce better case outcomes. A fundamental principle of the SDM model is that an important first step in developing decision-making procedures is to break down large, complex decisions into their basic parts. The model provides caseworkers with different tools to use for each decision point, because there are different issues that are addressed at each stage of the case.

Program Components

The SDM decision support system includes several components. One component is a series of tools that child welfare workers can use to assess families and organize a response at specific decision points in the case, from intake to reunification. When a child first enters the foster care system because of allegations of abuse or neglect, child welfare workers conduct a family risk assessment to determine parents’ strengths and needs in 12 areas of parent or family functioning (such as substance abuse and parenting skills) that are important to ensuring a child’s safety and well-being. The assessment helps caseworkers develop a treatment plan that builds on parents’ strengths while also addressing potential barriers to reunification in the parents’ functional areas. Caseworkers develop a similar treatment plan for children as well.

In addition to the initial assessment, caseworkers complete a three-part reunification assessment that evaluates the progress of the treatment plan at subsequent case reviews. The assessment involves an evaluation of the family’s current functioning and any progress made to reduce barriers to reunification, an evaluation of parental visitation based on the frequency and quality of contact with children, and a safety assessment of the family’s home. The safety assessment allows caseworkers to determine if there is any imminent threat to a child’s safety and well-being if they are returned to the parents’ home.

After completing the reunification assessment, workers consult a decision tree that provides structured guidelines about updating the treatment plan and case goals, based on the assessment findings and the length of time children have spent in placement. The decision tree incorporates a welfare agency’s policies regarding the use of the reunification assessment findings to make decisions about returning a child home or seeking another form of permanency.

SDM includes two management-related components as well. One component is a workload measurement and accounting system, which ensures an equitable workload among case workers based on their needs for workload demand. The other component is management information the uses various types of data (such as aggregate family assessment data and workload data) to assist managers in planning, monitoring, budgeting, and evaluation.

Study 1

Johnson and Wagner (2005) used a quasi-experimental design to look at the effectiveness of a Structured Decision Making (SDM) support system for foster care cases. The researchers examined whether children entering the foster care system in agencies that were piloting the SDM case management model achieved permanency more quickly than they would have if the new case management procedures had not been implemented. The 3-year evaluation compared the outcome results among 18 counties in Michigan: nine counties that implemented a pilot version of SDM and nine counties that did not.

The nine pilot counties were matched to nine non-pilot counties based on demographic characteristics, including population size, percentage of families in rural areas, percentage of families by race, percentage of families receiving public assistance, and percentage of families below the federal poverty level. Counties were also matched on administrative characteristics of the county’s child welfare agency, such as foster care caseload size, the ratio of direct service cases per foster care worker, and the percentage of county caseload that was managed by private agencies. The counties were comparable on most characteristics. Pilot counties did, however, have a significantly higher number of children initially placed in a shelter home (15.7 percent versus 4.7 percent) instead of a foster home, which could have affected the outcome results since children that are initially placed in foster homes are more likely to attain permanency sooner.

The nine pilot counties had a total of 841 cases. In those cases, the ethnicity of the child was 76 percent white, 18 percent African American, and 5 percent “other”/unknown. In almost half the cases (49.8 percent), the child was between 5 and 14 years old. The nine non-pilot counties had 871 cases. In those cases, the ethnicity of the child was 49 percent white, 44 percent African American, and 7 percent “other”/unknown. In almost 48 percent of the cases, the child was between 5 and 14 years old.

The study looked foster care outcomes in at the counties for an 18-month period prior to implementation of SDM. Preimplementation findings showed that children entering foster care in the pilot counties were no more likely to enter a permanent placement within 15 months of entering foster care then were children in comparison counties. Children in pilot counties were somewhat less likely to return home or be freed for adoption than children in the comparison counties.

The postimplementation period looked at outcomes for children who entered foster care during an 18-month period (from April 1998 through September 1999) with a goal of returning home. The primary outcome measure was a child’s permanency status 15 months after placement (including placement with a relative, foster home, shelter home, or other). Each child was followed for a period of 15 months, which started the date the child entered care. Permanency was considered achieved for a case if the child was reunited with the parents, had entered a permanent placement with another family member, was adopted or freed for adoption by termination of parental rights, or had another permanent arrangement (such as a permanent foster placement agreement).

The study also examined whether children in the pilot counties were more likely to reenter foster care after reunification than those children in the comparison counties. For those children that had returned home from 15 months after placement, data was collected for a 12-month follow-up period, beginning at the time of reunification, to determine if the children reentered foster care.

The study relied on bivariate and multivariate analysis, including logistic regression, which controlled for case characteristics that may have influenced case outcomes. No subgroup analysis was conducted.

For additional information about Structured Decision Making, please see: Children’s Research Center. 2008. The Structured Decision Making Model: An Evidence-Based Approach to Human Services. Madison, Wisc.: Children’s Research Center.

These sources were used in the development of the program profile:

These sources were used in the development of the program profile: