Latest Findings from Assisted Guardianship Waivers: Tennessee and Wisconsin Presentation to the 12th Annual Title IV-E Child Welfare Demonstration Projects.

Latest Findings from Assisted Guardianship Waivers:

Tennessee and Wisconsin

Presentation to the 12th Annual Title IV-E Child Welfare Demonstration

Projects Meeting June 17, 2008

Tennessee Subsidized Guardianship Waiver

Demonstration:Preliminary Outcomes

Leslie Cohen Children and Family Research Center

Viola Miller, Commissioner TN DCSMark Testa, Director CFRCJune 17, 2008

Enter page title here!Fostering ResultsNational Perspective

Currently, 35 states and DC have subsidized Guardianship Programs (funding sources vary)

Since 1996, 12 states have been given waiver authority to test a federally supported guardianship program: Tennessee, Delaware, Illinois, Iowa, Maryland, Minnesota, Montana, New Mexico, North Carolina, Oregon, Wisconsin, and Virginia.

Enter page title here!Fostering ResultsNational Perspective

Legislation to provide a federally subsidized guardianship program has been proposed in the 110th Congress: H.R.2188 Kinship Caregiver Support Act S.661 Kinship Caregiver Support Act S.3038 Adoption Incentive and Relative

Guardianship Support Act of 2008

Availability of SG boosted legal permanence

At wave I of the evaluation (1998) there was a 8% permanency advantage for children in the experimental group who were offered the choice of subsidized guardianship compared to children in the control group.

Experimental Control0%

20%

40%

60%

80%

100%

AlreadyReunified

Already SG Already AA

}8.3%

Illinois

At wave II of the evaluation (2000) there was still a 6% permanency advantage.

Are These Results Generalizable?

Experimental Control0%

20%

40%

60%

80%

100%

AlreadyReunified

Already SG Already AA

}

6.0%

Illinois

Availability of SG boosted legal permanence

In October of 2005, the Tennessee Department of Children’s Services (DCS) received HHS approval to operate a five-year demonstration of permanent guardianship as a federally subsidized permanency option under title IV-E.

Tennessee Subsidized Guardianship Waiver

Demonstration Project

Waiver demonstrations should demonstrate that a policy change would be in the best interest of children:

The goal of Tennessee’s Subsidized Permanent Guardianship Demonstration is to improve permanency and safety outcomes for children and families in approved relative and kin settings. The State is using the waiver authority to test whether the introduction of a subsidized permanent guardianship benefit will result in an increase of permanence and safety for children and an improvement in a range of child outcomes such as reduced length of stay in foster care and improved stability of substitute care.

Tennessee Subsidized Guardianship Waiver

Target Population

The target population consists of all title IV-E eligible and non-IV-E eligible children in Tennessee, aged 0 to 17.75 years old, who meet the following criteria: have been in foster care for at least nine (9) months (of the latest 12 months); live in a approved relative setting or resided with the same kin caregiver continuously for at least six (6) months (allowing for temporary absences from the home); and for whom reunification and adoption are not viable permanency options.

Target Population

Implementation of the demonstration began in December of 2006 and will initially occur in 16 counties (see Map 1). In addition, the State made eligible for assignment all children statewide who have a goal of planned permanent living arrangement (PPLA) and meet other program requirements.

Guardianship Program will initially occur in 16 counties throughout the State.

Approximately 1,500 children will be a part of the five-year waiver demonstration.

Enter page title here!Ten Objectives

Evaluate the impact of the subsidized permanent guardianship waiver on:

children’s safety. the number of children in PPLA. permanence for children with relatives and kin. the length of stay in foster care. the rates of reunification and adoption. the rate of re-entry into foster care.

Enter page title here!Ten Objectives (cont’d)

In addition, the evaluation will: Track the number and proportion of guardianships

that are disrupted or dissolved and the reasons for the displacement.

Assess the implementation of the subsidized guardianship waiver.

Estimate the overall savings accrued from a greater level of permanence achieved by the treatment group.

Experimental Design

The classic experimental design used in this study is the best way to determine causal connections between interventions and outcomes.

The control group will receive the “regular services” and full range of permanency options in effect in Tennessee prior to October 2006 for which they are eligible—including reunification, subsidized adoption, and PPLA. Those in the treatment group will be offered the additional option of subsidized permanent guardianship. This is a value added program because nothing is taken away from the control group.

Thus, we will study the effects of the treatment services relative to services that would have been provided in the absence of the subsidized guardianship option.

Group Assignment Criteria

Group Assignment occurs when: The child has been in the custody of the state

for 9 months or more immediately prior to establishing subsidized permanent guardianship and is likely to remain in care.

The child has lived with a relative or kin caregiver for at least six months immediately prior to establishing subsidized permanent guardianship.

The child’s committed county must be Shelby, Davidson, or an Upper Cumberland County.

Group Assignment Criteria

Relative vs. KinRelative:

The child and the caregiver must be related by blood, marriage, or adoption.

Kin: A person with whom the child has a significant relationship that pre-exists placement such as godparent, friend, neighbor, church member, minister, teacher

or

A relationship that develops over time after placement has been made.

Group Assignment

A child between the ages of 0 and 17.75 is eligible for group assignment when he or she:

is in the foster care system for 9 months is living with an approved relative caregiver or kin for 6 months has a committed county of Shelby, Upper Cumberland, or Davidson

When eligibility is met the child

will be randomly assigned to

the experimental group

or control group at a 1:1 ratio

Experimental Group Eligible for

subsidized perm. guardianshipand eligible

for all traditional permanency alternatives

Control GroupNot eligible for

subsidized perm. guardianship.Eligible

for all traditional permanency alternatives, including permanent

guardianship

Randomization

Randomization was successful in balancing (within the bounds of chance variation) the characteristics of children and their caregivers who were assigned to the demonstration and cost neutrality groups.

Because the two groups look statistically similar at the start of the waiver, any differences that later emerge with respect to permanency outcomes and days spent in foster care can reasonably be attributed to the offer of subsidized guardianship to the demonstration group.

Randomization

  Demonstration Cost Neutrality Difference

Child age at assignment 11.0 yrs. 11.1 yrs. -0.1

Age at removal 7.2 yrs. 7.4 yrs. -0.2

Female 48.1% 48.4% -0.2%

White 32.4% 30.0% 2.4%

Black 65.9% 65.6% 0.3%

Caregiver age at interview 47.9 yrs. 48.5 yrs. -0.6

White 32.8% 31.1% 1.6%

Black 63.9% 66.7% -2.8%

Grandparent-grandchild 26.7% 23.8% 2.9%

Aunt/Uncle-niece/nephew 25.7% 28.2% -2.5%

Other Relatives 16.2% 15.0% 1.2%

Non-biological kin 31.4% 33.0% -1.6%

Sample N 293 273  

Randomization

 Site at Assignment Demonstration Cost-neutrality Difference

Davidson 35.8% 37.7% -1.9%

Shelby 41.0% 40.3% 0.7%

Upper Cumberland 23.2% 22.0% 1.2%

Sample N 293 273  

The demonstration and cost neutrality groups are well balanced with respect to geographical distribution: nearly equivalent proportions of children were assigned in the principal demonstration sites of Davidson, Shelby, and Upper Cumberland.

Outcomes

The December 2007 data extract for Tennessee shows a 12.9 percentage point higher rate of discharge to permanent homes from foster care in the demonstration group compared to the cost neutrality group. This is an important and statistically significant difference that replicates similar results previously reported for Illinois and currently for Milwaukee, Wisconsin-- the two jurisdictions that also operate a IV-E waiver program similar to Tennessee’s program. Experimental Control

0%

20%

40%

60%

80%

100%

GuardianshipReunificatioon

Custody to KinAdopted

}12.9%

Tennessee(Dec., 2007)

Enter page title here!Fostering ResultsOutcomes

78 Guardianships were completed between December 2006 and November 2007

Of the 78 guardianships finalized in the demonstration group, 28 were completed in the first quarter of 2007, 22 in the second quarter of 2007, and 20 in the third quarter of 2007. As of November 30, 8 guardianships were completed during the fourth quarter of 2007.

There are substantial differences among demonstration sites with respect to how the offer of subsidized guardianship has impacted permanency outcomes:

Davidson- 28.6% of assigned children were discharged to permanent guardianship, which boosted the net permanency gain over the cost neutrality group to 11.6%.

Shelby- 33.3% of assigned children were discharged to permanent guardianship, which boosted the net permanency gain over the cost neutrality group to 19.5%.

Upper Cumberland- 11.8% of assigned children were discharged to permanent guardianship, which resulted in a statistically insignificant difference of 2.5% between the two groups

Outcomes

Determine whether the comparatively lower utilization of subsidized guardianship in Upper Cumberland reflects implementation shortfalls or other contextual differences that may limit the generalizability of the results for Davidson and Shelby to other geographical areas in Tennessee.

Next Steps

Expansion Sites

On July 1, 2008, the State will roll-out the demonstration to an additional 24 counties. An additional 12 counties will be brought in on October 1, 2008, and the remainder of the State will be included in 2009.

Approximately 1,500 children will be a part of the five-year waiver demonstration.

WISCONSIN SUBSIDIZED GUARDIANSHIP ASSESSMENT

AND EVALUATIONInterim Evaluation Report

Presentation to the 12th Annual Title IV-E Child Welfare Demonstration Projects

Meeting June 17, 2008

Presenters

Denise Revels Robinson Director, Bureau of Milwaukee Child Welfare

David Sorenson Performance Planner, Office of Program Evaluation and

Planning, Wisconsin Department of Health and Family Services

George Gabel Evaluation Project Director, Westat

Overview

Federal Waiver became effective on September 23, 2004 and was approved for a 5-year period.

State enabling legislation was approved as part of the 2005 state budget bill, Wisconsin Act 25.

Program implementation began in October 2005 in Milwaukee County.

DHFS completed a RFP for evaluation of the subsidized guardianship program.

Westat, based in Rockville, Maryland, was awarded the Evaluation Contract in November 2005.

Eligibility Requirements

Target population: Children currently in a licensed relative foster care placement who have been in out-of-home care for 9 months.

It has been determined that children will not be reunified and adoption as a permanency option has been discussed.

In rare instances, children placed with a licensed person who is recognized by the child as part of the extended family (distant relative, god-parent, etc.) will also be included in the program (like-kin).

In rare instances, children who have been in care for less than 9 months may be eligible for the subsidized guardianship program. These cases will be determined on a case-by-case basis.

Designed to increase the permanency options available to licensed relatives by creating the ability to provide payment in the amount of the foster care payment.

Waiver Specifics

Initial focus of subsidized guardianship is on Milwaukee County only.

Expansion to other counties may be considered at a later date on a voluntary basis.

Services are provided by the State of Wisconsin, Bureau of Milwaukee Child Welfare.

Implementation of subsidized guardianship is divided into 2 separate phases.

Phase 1: Special Experimental Group The “Grandfathered” Group

Children with a guardianship order previously transferred (including all children through December 31, 2005) who remained in foster care and met eligibility requirements.

209 children 128 converted to subsidized guardianship 0 closed to adoption 20 closed for other reasons 61 remain open in foster care

Close to aging out Working towards reunification AWOL Placed with relative guardians who do not want to end services

Children were not subject to random assignment

Phase 2: Random Assignment

Children, who as of January 1, 2006, met eligibility criteria and did NOT have a guardianship order in place.

Once all eligibility criteria are met, children are randomly assigned to the intervention (experimental) or comparison (control) group.

Siblings/children in the same case are assigned together. Caregivers are notified of their assignment by letters.

320 children received a program assignment

Intervention (Experimental) = 157

Comparison (Control) = 163

Phase 2: Requirements

Children living in licensed family foster homes• Adoption rule-out not required• 9 months in foster care• Currently with licensed relative

Children living with like-kin• Adoption has been ruled out• 9 months in foster care• Approval of “like-kin” circumstance by supervisor and BMCW

Program Manager

Fast-track children • Reasonable efforts of reunification not required or sibling of a child

already in subsidized guardianship• Approval by supervisor and BMCW Program Manager

Phase 2: Where We Are TodayWhere We Are Today

Children to be determined eligible for the remainder of the waiver period (through October 2010).

It is anticipated that approximately 80 children will be assigned each year of the waiver for a total of 400 children over the five year period.

Community Support

BMCW established communication with partners in the community early in the process.

BMCW worked closely with Milwaukee Children’s Court and legal parties to discuss legal procedural issues related to implementation.

BMCW has actively involved the Governor Appointed Partnership Council and Executive Committee in discussions and reports regarding the implementation of subsidized guardianship.

Evaluation Team

Westat

Children and Family Research CenterSchool of Social WorkUniversity of Illinois at Urbana-Champaign

Institute for Research on PovertyUniversity of Wisconsin-Madison

Data Collection

315 caregiver phone surveys (Phase II) 95% response rate

Case manager survey (Phase II) 80% response rate on Part 1 67% response rate on Part 2

eWiSACWIS data on 529 children (Phase I&II) Focus groups with case managers, caregivers,

youth, judges and court staff June-October 2006 (Phase I) June-August 2007 (Phase II)

Implementation of the Waiver

The waiver has garnered support among the staff, courts, and families.

Adoption rule-out is not required, but in practice, many case managers and court staff believe rule-out is still necessary.

Families and case managers are concerned about potential loss of services.

Many intervention group families did not hear about the new SG option. rule-out practice lack of case manager training

Concern from court and workers about difficulty of converting SG to adoption. DA loses grounds for TPR after CHIPS order is dismissed caregivers may have to pay out of pocket for a private TPR

Who are the caregivers?

Majority are African American

Nearly ¾ work full- or part-time

Nearly ¾ attended a religious service in the last month

Over half had college education

Almost all felt they had enough money to take care of

their families

Almost all had support to help care for their children if

the caregiver became ill

Who are the children?

Majority are African American

Overall, half are between the ages of 6-13, ¼ are under

6 years old, and ¼ are 14 or older

Overall, half are male and half are female

Nearly half have IEPs

Over half have some type of reported disability

A positive outlook for the children

Nearly all children are in good or excellent health

Nearly half of the children between 6-13 years old display high levels of pro-social behavior

Over half of the children’s caregivers believe that their child will go to college

What are we trying to learn?

Does the waiver... increase rates of permanence for children in foster

care? increase the use of licensed relatives as placement

resources? reduce the number of disrupted placements for

children in foster care? not change the rate of reunification and adoption for

children in foster care? not increase the costs of providing foster care?

The answers so far...Has the waiver increased rates of permanence for children in foster care?

The answers so far...

Has the waiver increased the use of licensed relatives as placement resources?

The rate of licensed relative homes appears to have gone downCase managers report concerns about burden

of licensing on families Family might not qualify Family might end up in comparison group

The answers so far...

Has the waiver reduced the number of disrupted placements for children in foster care?

No statistical difference in placement disruption between the intervention and control groups

To date, only one child in subsidized guardianship from Phase I has returned to foster care; none have returned from Phase II

The answers so far...

Has the waiver not changed the rate of reunification and adoption for children in foster care?

Rate of adoption remains the same BMCW decided not to notify families with children

with existing TPR about SGMainly assignments from January 2006

Needs to be watched closely for new assignments

What to watch for....

Will the culture around adoption rule-out change?

Will all families be informed about the waiver?

Will the increase in permanence continue?

Will cost neutrality be maintained?

Permanence to guardianship

Phase I Exempt-ions

Phase IIExempt-ions

Phase II assignments  

20061st Q

20062nd Q

20063rd Q

20064th Q

20071st Q

20072nd Q

20073rd Q Total

Guardianships 117 11 26 3 1 1 1 0 0 160

Guardianship Permanency

Ratio 117:1 1.83:10.34:

1 .75:1 .06:1 .09:1 .01:1 0:1 0:11.19:

1

The rate of subsidized guardianship in comparison with reunification and adoption has declined through Phase II

Next steps for the evaluation

Continue with random assignment through September 2010

Continue caregiver interviews through April 2010 assignments

Continue obtaining cost data Medicaid expenditures Revenues from child support orders

Additional analysis Reentry Incidents of maltreatment Social capital

If funding is available, in-person survey of all caregivers and children

Questions?

For further information

DCFS:David Sorenson (608) 261-8896 [email protected]

BMCW:Denise Revels-Robinson [email protected]

(414) 220-7029 Westat:

George Gabel (301) 251-4223 [email protected] Hill (301) 738-3534 [email protected]

CFRC:Mark Testa (312) 641-2505 [email protected]

IRP:Kristen Shook Slack (608)263-3671 [email protected]

Discussion Topic

Availability of SG boosted legal permanence

At wave I of the evaluation (1998) there was an 8% permanency advantage for children in the experimental group who were offered the choice of subsidized guardianship compared to children in the control group.

Increase in both adoptions and guardianships

Experimental Control0%

20%

40%

60%

80%

100%

AlreadyReunified

Already SG Already AA

}8.3%

Illinois

External Validity: SG boosted legal permanence

At wave I of the evaluation (2007) there was a 20% permanency advantage for children in the experimental group who were offered the choice of subsidized guardianship compared to children in the control group.

Increase in both adoptions and guardianshipsExperimental Control

0%

20%

40%

60%

80%

100%

AlreadyReunified

Already SG Already AA

}20.1%

Milwaukee, WI

External Validity: SG boosted legal permanence

After one year of implementation (2007) there was a 13% permanency advantage for children in the experimental group who were offered the choice of subsidized guardianship compared to children in the control group.

But net gain came at the expense of adoptionsExperimental Control

0%

20%

40%

60%

80%

100%

AlreadyReunified

Already SG Already AA

}12.9%

Tennessee

At wave II of the evaluation (2000) there was still a 6% permanency advantage, but perhaps two-thirds of the completed guardianships might have eventually converted into adoptions in the absence of the option.

Is the net gain in permanence

worth the loss in adoptions?

Future net gains also came at expense of adoptions in Illinois

Experimental Control0%

20%

40%

60%

80%

100%

AlreadyReunified

Already SG Already AA

}

6.0%

Illinois

Wider availability of guardianship is prompting a rethinking of the meaning of permanence

Original meaning of permanence as lastingRooted in the psychology of bonding and attachment that defines permanence as a lifelong relationship that arises out of feelings of belongingness among persons.

Newer meaning of permanence as binding Rooted in law that defines permanence as a lifelong commitment that is legally enforceable.

Redefinition demotes guardianship as a permanency goal

Newer thinking establishes a hierarchy of permanency goals. Requires ruling-out of reunification and adoption prior to pursuing guardianship. Guardianship is less binding because it is more easily vacated by the caregiver & more vulnerable to legal challenge by birth parents than termination of parental rights and adoption.

Binding vs. Lasting

Permanence as Binding Commitment needs to be legally binding to qualify truly as permanence. Hierarchy of permanency goals: reunification, adoption, followed by

guardianship. Strict interpretation of “rule-out” that adoption needs to be ruled out

independently of the desires of the family. Supported by federal waivers, IL statute, ABA, NCJFCJ

Permanence as Lasting Relationship not certain to last forever but intended to last indefinitely. Least restrictive (most family like) principle. Full disclosure of permanency options that allows kin to choose option that

best fits cultural norms of belongingness. Consistent with ASFA.

Issue

Are the biological bonds and social attachments of kinship sufficiently lasting to ensure a family's intention to raise a child to adulthood?

OR

Must the commitment be made legally binding through formal adoption for the intention to endure?

Questions

Do the intentions of adoptive parents to raise a child to adulthood differ from the intentions of guardians?

Are guardian homes any more likely to disrupt than adoptive or foster homes?

Do private wards express any lesser sense of belonging to the family than children who are adopted?