Home > Programs & Funding > Profiles of the Title IV-E Child Welfare Waiver Demonstration Projects > Arizona - Expedited Reunification Services

Arizona

Demonstration Type:	Expedited Reunification Services1
Approval Date:	June 30, 2005
Implementation Date:	April 17, 2006
Expected Completion Date:	March 31, 2011
Interim Evaluation Report Expected:	November 30, 2008
Final Evaluation Report Expected:	September 30, 2011

Target Population

The target population for Arizona's demonstration includes title IV-E-eligible and non-IV-E-eligible children (1) in out-of-home placement for no more than nine months in a congregate or licensed foster care setting (e.g., shelter facilities, group homes, residential placements, and licensed foster homes); (2) for whom reunification is the case plan goal; (3) whose caregivers agree to participate in the waiver demonstration; and (4) for whom a juvenile court concurs with a plan of expedited reunification.

Jurisdiction

Arizona's demonstration involves two project phases. Phase I was implemented for a 15-month period in randomly selected Child Protective Services (CPS) units in the Mesa, Thunderbird, and Tempe Child Welfare Offices in Maricopa County. Phase II began in January 2008 and expanded the demonstration to three additional offices in Maricopa County: Avondale, Glendale, and Talavi.

Intervention

Arizona's demonstration is testing innovative child welfare services that focus on expediting reunification for children in congregate and licensed foster care settings. Participants in the waiver demonstration have access to a variety of services:

1. Intensive home-based strategies and interventions, which include counseling (individual, family, and marital therapy), family assessments, case planning, and intensive case management in accordance with the child's safety plan and family assessment. Services also include counseling and skill development related to conflict resolution, anger management, communication and negotiation, parenting education, stress management, home management, job readiness, and linkages to community-based resources.
2. Child and Family Teams (CFTs), which provide a framework for facilitating the reunification of children in out-of-home placement with their caregivers. CFT participants include the Family Reunification Specialist, extended family, family friends, the child welfare case manager, and other significant persons in the family's life. These teams work together to support the family in the assessment, planning, intervention, and aftercare phases of the intervention.
3. Flexible funds, which are available for use when the CFT and/or Family Reunification Specialist identify basic or immediate family needs that cannot be met through existing resources. The use of flexible funds is specific to the individual needs and circumstances of each family. Examples of the use of flexible funds include provision for basic physical needs (e.g., food, clothing, shelter, or furniture); home repairs; financial support for a parent mentor; and counseling, therapeutic, or similar services that would otherwise be unavailable to the family.

Evaluation Design

Arizona's evaluation is testing the hypothesis that intensive home-based early reunification services will (1) reduce children's length of stay in congregate and licensed foster care settings; (2) decrease the likelihood of re-entry into out-of-home care; (3) prevent the recurrence of child abuse and neglect; and (4) improve family well-being and functioning. The State's evaluation approach in Phase I involved a modified comparison group design in which CPS units in the three Maricopa County CPS offices—Mesa, Thunderbird, and Tempe—were randomly selected to serve as experimental and control groups. Within each of the three participating CPS offices, the State chose two case management units to comprise the experimental group and one case management unit to serve as the control group.

At the onset of the project's implementation, existing cases from CPS units in the experimental group that met the demonstration's eligibility criteria were offered enhanced demonstration services. A matching group of comparison cases receiving "traditional services" were selected based on case and demographic characteristics that most closely matched those of the existing experimental group cases. New child protection cases were then randomly assigned to CPS units in either the experimental or control group; cases assigned to the experimental group received enhanced services, while cases assigned to the control group received a standard set of traditional child welfare services. The original purpose behind this approach was to minimize contamination of the research design that might occur if CPS workers carried mixed caseloads of experimental and control cases.

During the implementation in Phase I, contamination became a less serious issue because most enhanced waiver services are provided to families by contracted service providers rather than by the CPS workers themselves. Based on this observation and the preliminary evaluation findings from Phase I, the State eliminated the distinction between experimental and control CPS units for Phase II and now uses a standard experimental design in which new cases are randomly assigned to an experimental condition (eligible for enhanced waiver services) or to a control condition (ineligible for enhanced services). In addition, the sample for Phase II will not include children already in out-of-home placement (the existing case cohort) but will be limited to new CPS cases.

Sample Size

For Phase I, the State estimated that approximately 250 existing cases would be assigned to the experimental group at the start of the demonstration. However, at the project's onset, out of 357 potential cases, only 64 existing cases were found eligible based on screening criteria developed during the initial planning stage. The primary reason for the difference between estimated and actual cases assigned to the experimental condition was the initiation of a district-wide effort to reduce the number of children in congregate care settings through placements with unlicensed relatives. The success of this initiative significantly reduced the pool of children eligible to participate in Arizona's waiver demonstration.

For Phase II, the State estimates that approximately 20 cases will be randomly assigned each month across the six demonstration sites. Of these 20 cases, ten will be assigned to the experimental group and ten will be assigned to the comparison group. Overall, the State estimates that 800 cases will be served over the remaining years of the waiver.

Process Evaluation

Arizona's evaluation includes interim and final process evaluations that describe how the demonstration was implemented and how enhanced services differed from traditional services received by families in the control group. Questions addressed by the process evaluation include the following:

• What was the logic model for the demonstration and did it change over time?
• What were the processes for planning, organizing, implementing, and monitoring the project?
• What were the characteristics of staff involved with the project and what was their level of involvement?
• How were services delivered to families? What types of services were received and what was the duration of services?
• What was the role of the juvenile courts in the project? What was the nature and intensity of collaboration between the courts, the State, and local child welfare agencies?
• What contextual factors may have affected the implementation and outcomes of the project?
• What were the demographic and other important characteristics of participating families?
• What barriers were encountered during the implementation of the project and what steps were taken to address these barriers?

As part of the process evaluation, Arizona's evaluation contractors are conducting site visits to complete formal interviews with social workers and supervisors in participating CPS offices, as well as with staff from contracted service providers involved in the delivery of intensive reunification services. In addition, the evaluation team is reviewing a sample of case files to obtain information regarding the case planning process, services needed and provided, and the involvement of the family and child in permanency decision making. Annual focus groups with caregivers are also conducted.

Outcome Evaluation

Arizona's outcome evaluation compares the experimental and control groups for significant differences in the following outcome measures:

• The proportion of children reunified with their parents/caregivers;
• Length of stay in congregate care placements or in other out-of-home placement settings;
• The proportion of children with a subsequent alleged or substantiated maltreatment report;
• The proportion of children who re-enter out-of-home care; and
• The proportion of children and parents/caregivers who experience improvement in well-being and functioning as measured by the North Carolina Family Assessment Scale (NCFAS).

In conducting its analysis, Arizona's evaluation contractor is examining differences in outcomes by various subgroups (e.g., child age and presenting problems).

Cost Analysis

Arizona's cost analysis compares the costs of key services received by children in the experimental group with the costs of traditional services received by children in the control group. To the extent feasible, the State is conducting a cost-effectiveness analysis to identify costs per successful outcome for the experimental group versus the control group.

Evaluation Findings

Process Evaluation

As of March 31, 2008, a total of 271 cases had been enrolled in the demonstration, of which 21 parents/caregivers declined to participate. Of the remaining 250 cases, 102 cases were from the existing case cohort (with 58 assigned to the experimental group and 44 to the matched comparison group) and 148 were new randomly assigned cases (with 61 assigned to the experimental group and 87 to the control group).

Outcome Evaluation

Safety and permanency outcomes available as of March 2008 are highlighted below:

• Forty cases (16 percent) had one or more new maltreatment reports after entering the demonstration. In the matched case cohort, this number includes 12 of 58 experimental group cases (21 percent) and 9 of 44 comparison cases (20 percent), while in the randomized cohort it includes 8 of 61 experimental group cases (13 percent) and 11 of 87 control cases (13 percent).
• Targeted children from 69 of 250 cases (28 percent) were reunified with their parent/caregiver, of which 53 (77 percent) remained with their parent/caregiver as of March 31, 2008. Reunification rates appear to be slightly higher for experimental group children than for matched comparison or control group children. In the matched case cohort, 21 of 58 experimental cases (36 percent) were reunified compared with 11 of 44 matched comparison cases (25 percent). In the randomized cohort, 10 of 61 experimental cases were reunified (16 percent) compared with 11 of 87 control cases (13 percent). Data on statistical significance are pending.
• Of the 69 target children initially reunified, 16 children (23 percent) were returned to out-of-home placement. In the matched case cohort, these numbers include 7 of 28 reunified experimental cases (25 percent) and 5 of 16 reunified comparison cases (31 percent). In the randomized cohort it includes 2 of 12 reunified experimental group cases (17 percent) and 2 of 13 reunified control cases (15 percent).
• Experimental group children appear to spend slightly less time in out-of-home placement than matched comparison or control group children. In the randomized cohort, experimental group children spent an average of 210 days in placement compared with 221 days for children in the control group. In the matched case cohort, experimental group children spent an average of 266 days in placement compared with 271 days for matched comparison cases. Data on statistical significance are pending.

Additional outcome findings will become available as implementation continues.

Web Links

Annual evaluation reports for the State's expedited reunification demonstration project for 2004–2007 are available at the following Web site: http://www.cabhp.asu.edu/projects/.

¹ Based on information submitted by the State as of March 2008. Back

Back to Table of Contents