September 2012Vol. 13, No. 8Performance-Based Contracting in Illinois

Performance-based contracting in child welfare allows public agencies to contract with private agencies to provide services in an environment that focuses on child and family outcomes (e.g., residing in a permanent home) rather than on outputs (e.g., number of children served). The Illinois Department of Children and Family Services (DCFS) began using performance-based contracting in 1997 for its foster care case management. This initiative proved to be successful, but no formal evaluation was ever completed. In 2007, the National Quality Improvement Center on the Privatization of Child Welfare Services awarded an evaluation grant to DCFS, the Child Care Association of Illinois (a private child welfare agency), and the Children and Family Research Center of the University of Illinois (the project evaluator) to develop, implement, and evaluate a performance-based contracting initiative. The initiative would evaluate performance-based contracting for DCFS's residential treatment program and independent living opportunities/transitional living program (ILO/TLP). In 2012, 39 agencies participated in the project. This article provides information about the residential treatment component of the project.

A key factor to any performance-based contracting initiative is the development of the performance measures. Working with the existing Illinois Child Welfare Advisory Committee, as well as an array of experts in social work, statistics, and other fields from local universities, DCFS developed the performance measures, fiscal structure, and other components. Residential treatment providers are assessed based on the percentage of children in residential treatment who remain in a favorable postdischarge placement (i.e., less restrictive or neutral discharge) for 90 days or more and on the percentage of time in residential treatment that a child is not on the run, in detention, or in a psychiatric hospital. A "no decline" policy was established to prevent providers from only accepting children who are more likely to boost their performance ratings. Additionally, providers can outline the characteristics of the children they can best serve, and the Centralized Matching Team refers children to the providers that are best prepared to care for them.

The project instituted a risk adjustment model to account for the differences in the characteristics of the population served by each agency (e.g., child history, gender, age). Brice Bloom-Ellis, the Statewide Residential Quality Assurance Manager at DCFS, highlighted the importance of "striking a balance between simplicity and complexity." The initiative needs to be straightforward enough that providers can fully understand their role and that neither the providers nor DCFS are overly burdened by its administration, but the initiative also needs a degree of intricacy that allows it to capture the realities and desired outcomes of the child welfare system.

The first year of the project included financial incentives and penalties for providers based on how they fared on their performance measures. Staff from DCFS and Lawrence Hall Youth Services, one of the service providers, noted that the financial incentives and penalties were critical to grabbing providers' attention at the beginning of the project. Upon hearing about the move to performance-based contracting for residential treatment and ILO/TLP, providers' initial response was that "this too shall pass," but the financial incentives and penalties forced them to evaluate their current business and service practices.

The incentives and penalties were discontinued in the second year, however, due to a fiscal crisis in Illinois. Staff hoped the project would be cost-neutral, but it ended up paying out more in incentives than it took back with penalties. Even without the financial incentives and penalties, though, DCFS still regularly collects and assesses data on provider performance. DCFS uses the data to make decisions about contract renewals, with three agencies (five contracts) having been terminated since the beginning of the initiative. Additionally, 18 agencies have implemented quality improvement plans based on their performance data. Providers also want to improve their performance to serve children and families better and because of a sense of organizational pride. DCFS distributes annual provider rankings on the measures, and the providers want to compare well against similar contracts. Mary Hollie, CEO of Lawrence Hall Youth Services, said that many of the providers have built the performance measures into their organizational quality improvement processes and use the data to assess and adapt their services.

The evaluation of this project uses mixed methods, including onsite facility visits, an implementation survey of provider staff, focus groups, and a document review. The evaluation is ongoing, but the following are preliminary findings:

In-placement stability has remained relatively constant over the course of the project.
The percentage of all youth served who were negatively discharged has decreased from 16.5 percent in fiscal year 2009 to 14.1 percent in fiscal year 2011. A negative discharge is when a youth disrupts from care due to running away, a detention placement, psychiatric hospitalization, or a shelter placement and subsequently does not return to the same residential agency contract. Additionally, a negative discharge includes being transferred to another contract agency at the same or a more restrictive placement level. Approximately half of all negative discharges throughout the project are due to youth running away.
The percentage of all youth served who were favorably discharged has increased from 22.1 percent in fiscal year 2009 to 29.4 percent in fiscal year 2011.
These improvements may be caused by several factors, including the initiation of the project, implementation of quality improvement plans with struggling contracts, and the termination of poorly performing contracts.
Agencies that did not perform well did not tend to have a well-defined treatment model or had staff that could not articulate the treatment model.
Staff in agencies that performed lower than average tended to blame the youth for their poor performance.
Urban group homes did not perform as well as other provider groups. The project established a workgroup to analyze these findings and make recommendations for improvement.

Many thanks to Brice Bloom-Ellis, Statewide Residential Quality Assurance Manager, Illinois DCFS and Mary Hollie, CEO, Lawrence Hall Youth Services for providing information for this article.