| [Previous] | [Table of Contents] | [Next] |
In 1993, the New York State Department of Social Services (DSS) and the New York City Child Welfare Administration (CWA) began testing a new approach to the financing of services to foster children and their birth families based on concepts from managed care. Before this initiative, private child welfare agencies providing foster care services were reimbursed by the state for each day a child was in foster care. Under the new approach, agencies were promised a flat amount for a 3-year period which was intended to cover the costs of providing services for an identified group of children. It was hypothesized that the change in the payment system would achieve earlier permanency for children through service continuity, intensified discharge planning, and aftercare services. The initiative involved six agencies. An experimental design was implemented involving comparison groups in five of the agencies, with random assignment in three. The new financing mechanism was to be used in the experimental groups while services to the comparison group would be paid for under the old system. Funds could be used for foster care costs and any other services the agencies believed would achieve earlier permanency.
The original intent in the initiative was to collect follow-up data from workers and perhaps families and to analyze administrative data on these families to compare the foster care experiences of children and costs in the experimental and comparison groups. However, resource limitations prevented the collection of follow-up data from workers and families and also limited the analyses of administrative data. The Evaluation of Family Preservation and Reunification Services has supplemented the evaluation efforts of New York.
The purpose of the evaluation was to document the implementation of the program and its effects on percentage of case closings, average days in foster care, family functioning, and child behavior. The evaluation included follow-up interviews with a sample of workers and client families, analysis of administrative data, and interviews with administrative, supervisory, and front line staff about the implementation of the demonstration. Administrative data on foster care careers of children were examined.
Six New York City child welfare agencies participated in the HomeRebuilders program: Harlem Dowling, Little Flower, Miracle Makers, New York Foundling, St. Christopher-Ottilie, and St. Christopher-Jennie Clarkson. The sample for the demonstration was initially selected by New York State from its records of foster care cases in the selected agencies. Criteria for selection were agency specific. The criteria were developed to fit the particular composition of each agency's foster care population.
The evaluation began about 2 years after the demonstration began and as it was facing a precipitous ending 6 months early. Involvement in the HomeRebuilders' experiment in midstream has resulted in some design limitations. While agency caseworkers had completed structured baseline assessments of the families at the beginning of the experiment, these were incomplete and were inconsistently available across agencies. For these reasons, it was decided not to use these assessments and rely solely on follow-up data. As a result it has been possible to assess differences between experimental and comparison group cases at Time 2, but change with respect to family functioning and child behavior between Time 1 and Time 2 could not be measured. Furthermore, entering the experiment at a time that a fair number of caseworkers had already left the agencies and the experiment faced termination inevitably resulted in gaps of information about the startup of the demonstration.
The experiment was originally designed with procedures for random assignment of children to either a comparison group that received the regular services offered at the agency or an experimental group that used the HomeRebuilders model. Three agencies, Harlem Dowling, Little Flower, and Miracle Makers, used random assignment procedures. Two agencies, New York Foundling and St. Christopher-Ottilie, compared children in two different offices. At New York Foundling the Bronx office was considered the experimental office and followed the HomeRebuilders approach while the Queens office continued with practice as usual. St. Christopher-Ottilie also divided its caseload geographically with the experimental group at the Queens office and the comparison group served by the Brooklyn office. At St. Christopher-Jennie Clarkson, the entire foster boarding home population was selected to participate in the HomeRebuilders project.
Based on the selection criteria, the state selected children for participation in both the experimental and comparison groups. There were about 3000 cases in the demonstration. Analysis of the administrative data used the entire sample. For the interviews, a sample of caretakers was drawn, with the goal of conducting 500 face-to-face interviews with caretakers and 500 telephone interviews with their caseworkers. To conserve resources, it was determined that interviews would only be conducted with the experimental and control cases in agencies that randomly assigned cases to these two groups. Therefore, no interviews were conducted with New York Foundling comparison cases. St. Christopher-Jennie Clarkson did not have a comparison group. Caretakers served by St. Christopher-Ottilie were interviewed because it was initially thought that the agency had applied random assignment procedures to selecting experimental and control cases. Caseworkers from St. Christopher-Ottilie were not interviewed.
A total of 433 caretaker interviews were completed and caseworkers of 407 cases were interviewed. This represents 73 percent of the caretaker and 70 percent of the caseworker cases that were released for interview.
This report documents the demonstration's implementation process and reports the evaluation findings. Chapter 2 describes the implementation of the demonstration. Chapter 3 presents findings from the analysis of administrative data and interviews with caretakers and caseworkers, conducted approximately 3 years after the demonstration began. Appendix A provides a detailed description of the design and implementation of the follow-up evaluation, Appendix B presents additional tables, and the study questionnaires are in Appendix C.
| [Previous] | [Table of Contents] | [Next] |