GAO

United States General Accounting Office
Report to Congressional Requesters

FOSTER CARE

Recent Legislation Helps States Focus on Finding Permanent Homes for Children, but Long-Standing Barriers Remain

June 2002

GAO-02-585

Contents

June 28, 2002

The Honorable Wally Herger  
Chairman, Subcommittee on Human Resources
Committee on Ways and Means
House of Representatives

The Honorable Max Baucus  
United States Senate

The Honorable Lincoln D. Chafee
United States Senate

The Honorable Larry E. Craig  
United States Senate

The Honorable Mike DeWine  
United States Senate

The Honorable Charles E. Grassley
United States Senate

The Honorable Mary L. Landrieu
United States Senate

The Honorable John D. Rockefeller IV
United States Senate

In the mid-1990s, some children languished in temporary foster care while prolonged attempts were made to reunify them with their families. In response to concerns about the length of time children were spending in foster care, the Congress enacted the Adoption and Safe Families Act of 1997 (ASFA). ASFA contained two key provisions that were intended to help states more quickly move the more than 800,000 children estimated to spend some time in foster care each year to safe and permanent homes. One of these provisions, referred to as "fast track," allows states to bypass efforts to reunify families in certain egregious situations. The other provision, informally called "15 of 22," requires states to file a petition to terminate parental rights (TPR) when a child has been in foster care for 15 of the most recent 22 months.¹ In addition, ASFA emphasized the importance of adoption when foster children cannot safely and quickly return to the care of their families. Toward that end, ASFA established incentive payments for states that increase their adoptions. In addition, the law provided a new source of funds for states to use to promote and support adoptions through the Promoting Safe and Stable Families (PSSF) program.

Because of your interest in how states have implemented ASFA and what its impact has been on permanency outcomes for children in foster care, you asked us to review key provisions in the legislation. Specifically, you asked us to determine (1) how the outcomes and characteristics of children in foster care have changed since ASFA was enacted, (2) how states have used ASFA's fast track and 15 of 22 provisions, (3) how states are spending new adoption-related funds provided by ASFA, and (4) what states are doing to address barriers to achieving permanency.

To conduct our work, we analyzed national child welfare data sets and statistical reports. To supplement these statistics, we surveyed all 50 states and the District of Columbia regarding foster care outcomes,² the use of ASFA's fast track and 15 of 22 provisions, expenditures of ASFA adoption- related funds, and efforts to address barriers to finding permanent homes for children. We received responses from 46 states, although they did not all respond to every question.³We also reviewed a variety of Department of Health and Human Services (HHS) Administration for Children and Families' (ACF) documents and interviewed federal officials and child welfare experts on ASFA's impact. In addition, we visited six states— Illinois, Maryland, Massachusetts, North Carolina, Oregon, and Texas—to obtain first-hand information on ASFA's effect on state child welfare agencies. We selected these states to represent a range of locations and levels of performance under the adoption incentive program. We also chose to visit some states that had implemented innovative practices in their child welfare systems. We conducted our work between June 2001 and April 2002 in accordance with generally accepted government auditing standards. A more detailed discussion of our scope and methodology appears in appendix I.

¹ASFA allows for some exceptions to this provision. For example, states are not required to file a TPR for a child who has been in foster care for 15 months if the child is in a relative's care.

²Throughout this report, references to state survey responses include the District of Columbia.

³In addition, we requested survey data by federal fiscal year (FY) for 1999 and 2000. However, of the 46 states responding to our survey, 22 provided data for time periods other than FY 1999 and FY 2000, such as calendar year (CY) or state fiscal year (SFY).

Results in Brief

While the annual number of adoptions increased by 57 percent since ASFA was enacted, changes in other foster care outcomes and the characteristics of children in foster care cannot be identified due to the lack of comparable pre- and post-ASFA data. Foster care adoptions increased from approximately 31,000 in 1997 to over 48,600 in 2000, continuing a trend of increasing finalized adoptions that started before ASFA. However, data limitations make it difficult to determine either ASFA's role in this increase or changes in other foster care outcomes. For example, reliable pre-ASFA child welfare data for several states are available from the University of Chicago, but they cannot be matched against the post-ASFA data available from HHS due to differences in measurement techniques. Nevertheless, examination of HHS data between 1998 and 2000 provides insight into the characteristics and outcomes of children in foster care after ASFA was enacted. For example, children who left foster care between 1998 and 2000 spent a median of nearly 1 year in care. Of these children, those who were adopted spent more time in foster care—a median of approximately 3 1/2 years. Children most frequently returned home after a stay in care, but approximately 33 percent of children who reunified with their families in 1998 re-entered foster care within 3 years. Although these data illustrate the experiences of foster children since ASFA, recent improvements in HHS data make it difficult to determine if changes observed after 1998 are the result of changes in data quality or actual changes in the outcomes and characteristics of foster children.

Although data on states' use of ASFA's two key permanency provisions— fast track and 15 of 22—are limited, some states described circumstances that hindered the broader use of these provisions. In response to our survey, very few states were able to provide specific information on the number of children affected by these provisions and HHS does not require states to collect this data. Survey data from a few states suggest that they use the fast track provision infrequently. Most of the states we visited reported using fast track for a small number of children, primarily those who have experienced serious abuse or whose parents previously had parental rights to other children terminated. However, state officials described several court-related issues, such as reluctance on the part of some judges to allow the state to bypass reunification efforts, which prevented these states from using it in more cases. In addition, survey responses from nine states indicated that they exempted a number of children—ranging from 31 to 2,919 in 2000—from the 15 of 22 provision. The states we visited similarly reported exempting many children, such as adolescents and children with serious medical needs for whom locating adoptive parents would take a long time. Other children exempted included those placed with relatives and those who were expected to reunify shortly with their families. The states we visited reported that the timeframe of the 15 of 22 provision was helpful in facilitating more timely decisions on permanent placements for children.

In general, states are most frequently using the new adoption-related funds provided by ASFA to recruit adoptive parents and provide post adoption services. For example, Connecticut used its adoption incentive funds to buy advertisements on Spanish language television to recruit adoptive families for older Hispanic children and sibling groups to address a shortage of families for these children. Likewise, North Carolina used its PSSF funds to hire private agencies to recruit adoptive families for certain groups of children. Oregon, on the other hand, used its PSSF funds to create a new, statewide resource center to provide information and referral services, support groups, and educational workshops to adoptive families. In addition to recruitment and post adoption programs, we found that states have spent these ASFA funds on a variety of other child welfare activities, including hiring and training social workers.

The states we visited have implemented a variety of practices to address long-standing barriers to achieving permanency for foster children. These barriers include court delays and insufficient court resources, difficulties in recruiting adoptive families for children with special needs, obstacles and delays in placing children in permanent homes in other jurisdictions, and poor access to some services families need to reunify with their children. States are testing different approaches to address these barriers, but limited data are available on the effectiveness of these practices. For example, Massachusetts has developed a mediation program to help families and potential adoptive parents agree on the permanent plan for a child, thereby avoiding the time associated with a court trial and an appeal of the court's decision. However, only preliminary evaluation results are available on this program. Under HHS's child welfare demonstration waiver program, four states are experimenting with approaches to improve access to treatment services for parents with substance abuse problems. Final evaluation results for the first waiver project were published in April 2002, but results for most of the existing waiver projects will not be available for a few years.

To obtain a clearer understanding of how ASFA';s two key permanency provisions are working, we are recommending that the Secretary of HHS consider ways to obtain information on states' use of ASFA's fast track and 15 of 22 provisions. In commenting on this report, ACF generally agreed with our findings and recommendation. ACF noted that it has established a team to review data issues and that the team plans to evaluate the feasibility of including data on ASFA's fast track and 15 of 22 provisions in its existing adoption and foster care data system.

Background

Two titles under the Social Security Act provide federal funding targeted specifically to foster care and related child welfare services.⁵ Title IV-E provides an open-ended individual entitlement for foster care maintenance payments to cover a portion of the food, housing, and incidental expenses for all foster children whose parents meet certain federal eligibility criteria.⁶ Title IV-E also provides payments to adoptive parents of eligible foster children with special needs.⁷ Special needs are characteristics that can make it more difficult for a child to be adopted and may include emotional, physical or mental disabilities, emotional disturbance, age, being a member of a sibling group, or being a member of a minority race. Title IV-B provides limited funding for child welfare services to foster children, as well as children remaining in their homes. In federal fiscal year 2001, total Title IV-E spending was $5.6 billion and total Title IV-B spending was $576 million.⁸

Two key provisions introduced by ASFA were intended to help states move into safe, permanent placements those foster children who are unable to safely return home in a reasonable amount of time. Under the fast track provision, states are not required to pursue efforts to prevent removal from home or to return a child home if a parent has (1) lost parental rights to that child's sibling; (2) committed specific types of felonies, including murder or voluntary manslaughter of the child's sibling; or (3) subjected the child to aggravated circumstances, such as abandonment, torture, chronic abuse, or sexual abuse. In these egregious situations, the courts may determine that services to preserve or reunite the family—that is, the "reasonable efforts" requirement established in the Adoption Assistance and Child Welfare Act of 1980 (Public Law 96-272)— are not required. Once the court makes such a determination, the state must begin within 30 days to find the child an alternative permanent family or other permanent arrangement. In addition, the Abandoned Infants Assistance Act of 1988, as amended in 1996, requires states to expedite the termination of parental rights for abandoned infants in order to receive priority to obtain certain federal funds.

The second provision requires states to file a TPR with the courts if (1) an infant has been abandoned; (2) the parent committed any of the felonies listed in the fast track provision; or (3) the child has been in foster care for 15 of the most recent 22 months.⁹ States may exempt children from this requirement if the child is placed with a relative; the state has not provided services needed to make the home safe for the child's return; or the state documents a compelling reason that filing a TPR is not in the child's best interest.

ASFA also contained other provisions to help states focus on the length of time children were remaining in care. For example, ASFA requires states to hold a permanency planning hearing for each child in foster care at least every 12 months, during which the court determines the future plans for a child—for example, whether the state should continue to pursue reunification with the child's family or begin to pursue some other permanency goal. Prior to ASFA, these permanency hearings had been required to occur by the 18th month of a child's stay in care. For those children who will not be reunified with their families, ASFA also requires that the permanency plan document the steps taken to place the child and finalize the adoption or legal guardianship. At a minimum, the permanency plan must document any child-specific recruitment efforts taken to find an adoptive family or legal guardian for a child.

In addition, ASFA includes three provisions that are specific to interjurisdictional adoption issues. These provisions (1) require assurances in state plans¹⁰ that a state will not delay or deny the placement of a child for adoption when an approved family is available in a different state or locality, (2) require assurances that the state will develop plans for the effective use of cross-jurisdictional resources to facilitate permanent placements of waiting children, and (3) make ineligible for certain federal funds any state that is found to deny or delay the placement of a child for adoption when an approved family is available in another jurisdiction.

⁴Examples of HHS standards include the percent of foster children who are the subject of a substantiated or indicated report of abuse or neglect by a foster parent and the percent of children who re-enter foster care within 12 months of a prior foster care episode.

⁵In addition, Title XX provides funds under the social services block grant that may be used for many purposes, including child welfare.

⁶In addition, certain judicial findings must be present for the child in order for the child to be eligible for Title IV-E foster care maintenance payments.

⁷To qualify for an adoption subsidy under Title IV-E, a state must determine that the child cannot or should not return home; a state must make a reasonable, but unsuccessful effort to place the child without the subsidy; and a specific factor or condition must exist that makes it difficult to place the child without a subsidy. States have the discretion to define the specifics of the special needs category.

⁸Throughout this report, fiscal year refers to federal fiscal year, unless noted otherwise.

⁹Under ASFA a child is considered to have entered care on the earlier of (1) the date of the first judicial finding that the child has been subjected to child abuse or neglect or (2) 60 calendar days after the date on which the child is removed from the home.

¹⁰State plans are required under Title IV-E of the Social Security Act in order for states to receive federal foster care funds for maintenance of foster children, specific administrative costs associated with foster care programs, and adoption assistance.

¹¹ASFA also makes ineligible for certain federal funds any state that fails to promptly grant a fair hearing to an individual who alleges such a violation.

New Funds Introduced by ASFA

ASFA also authorized a new funding source dedicated to adoption-related activities. Prior to ASFA, the Congress established the family preservation and family support program under subpart 2 of Title IV-B of the Social Security Act, authorizing funds to provide two categories of services: family preservation and community-based family support services. ASFA reauthorized the program, renaming it Promoting Safe and Stable Families (PSSF) and adding two new funding categories: adoption promotion and support services and time-limited family reunification services. HHS program instructions specify that states must have a strong rationale for spending less than 20 percent of their PSSF funds on each of the four defined categories. The Congress authorized $305 million for the PSSF program in fiscal year 2001. A research firm found that state expenditures of federal PSSF funds on adoption promotion and support activities increased from $50 million in fiscal year 1999 to $79 million in fiscal year 2001, representing a 58 percent increase.¹² In January 2002, the PSSF program was reauthorized, authorizing $305 million for each of fiscal years 2002 through 2006, along with an additional $200 million in discretionary grant funds for each of those years.

ASFA created the adoption incentive payment program, which financially rewards states for increasing numbers of finalized adoptions. The states have the flexibility to use the incentive payment funds for any child welfare related initiative. To benefit from the incentive payment, states must exceed an adoption baseline established for their particular state. The baseline for the initial award year— fiscal year 1998—was each state's average number of finalized adoptions in fiscal years 1995, 1996, and 1997. After fiscal year 1998, a state's baseline is based on any previous fiscal year that has the largest number of finalized adoptions, beginning with fiscal year 1997. States receive a fixed payment of $4,000 for each foster child who is adopted over the baseline and an extra $2,000 for each adopted child characterized as having a special need. States have earned a total of more than $127 million in incentive payments for adoptions finalized in fiscal years 1998, 1999 and 2000.

ASFA expanded the use of federal child welfare demonstration waivers that allow states to test innovative foster care and adoption practices. In 1994, the Congress gave HHS the authority to establish up to 10 child welfare demonstrations that waive certain restrictions in Titles IV-B and IV-E of the Social Security Act and allow broader use of federal funds. ASFA authorized 10 additional waivers in each year between fiscal years 1998 and 2002 to ensure that more states had the opportunity to test innovations. States with an approved waiver must conduct a formal evaluation of the project's effectiveness and must demonstrate the waiver's cost neutrality—that is, a state cannot spend more in Title IV-B and IV-E funds than it would have without the waiver. Projects generally are to last no more than 5 years. Although funding for this program is scheduled to end in fiscal year 2002, the Congress expects to consider its reauthorization later this year.

Child Welfare Data Systems

HHS compiles data on children in foster care and children who have been adopted from state child welfare agencies in the Adoption and Foster Care Analysis and Reporting System (AFCARS).¹³ HHS is responsible for collecting and reporting data and verifying their quality. States began submitting AFCARS data to HHS in 1995. Twice a year, states are required to submit data on the characteristics of children in foster care, foster parents, adopted children and adoptive parents. Prior to AFCARS, child welfare data was collected in the Voluntary Cooperative Information System (VCIS), operated by what was then called the American Public Welfare Association.¹⁴ Since reporting to VCIS was not mandatory, the data in the system were incomplete. In addition, the data submitted were inconsistent because states used reporting periods and definitions for various data elements that differed from each other.

¹²James Bell Associates, Analysis of States' Annual Progress and Services Reports and Child and Family Services Plans (1999-2001): The Family Preservation and Family Support Services (FP/FS) Implementation Study (Arlington, Va., 2002).

¹³The Children's Bureau, under ACF, is the HHS component responsible for compiling these data.

¹⁴In 1998, the American Public Welfare Association became the American Public Human Services Association (APHSA).

Limited Data are Available to Measure Changes in the Outcomes and Characteristics of Children Since ASFA

The number of annual adoptions has increased since the implementation of ASFA; however, data limitations restrict comparative analysis of other outcomes and characteristics of children in foster care.¹⁵ Foster care adoptions grew from 31,004 in fiscal year 1997 to 48,680 in fiscal year 2000, representing an increase of 57 percent since ASFA was enacted.¹⁶ However, current data constraints make it difficult to determine what role ASFA played in this increase. The lack of reliable and comparable pre- and post-ASFA data at this time limits our ability to analyze how other foster care outcomes or children's characteristics have changed. For example, reliable pre-ASFA child welfare data are available from the University of Chicago for a limited number of states,¹⁷ but they cannot be matched against the post-ASFA data available from HHS. Current data do, however, provide some information about the characteristics and experiences of foster children after ASFA. For example, children leaving care between 1998 and 2000 spent a median of approximately 1 year in care. Of these children, those who were adopted spent more time in foster care—a median of approximately 3 1/2 years. Children most frequently returned home after a stay in care, but about 33 percent of those children re-entered foster care within 3 years.

Adoptions Have Increased Since ASFA, but Changes in Other Outcomes Unclear

Adoptions from state child welfare foster care programs have increased nationwide by 57 percent since ASFA was implemented, while changes in other outcomes are less clearly discernible. Adoptions began to increase prior to the enactment of federal child welfare reforms (see fig. 1). For example, between 1995 and 2000, annual adoptions of children in foster care increased by 89 percent from approximately 26,000 to nearly 49,000.¹⁸Adoptions generally increased between 8 percent and 12 percent year to year between 1995 and 2000, except in 1999 when they increased by 29 percent over 1998 adoptions. This increase in overall adoptions of children in foster care is accompanied by a parallel increase in the adoptions of children with special needs.

¹⁵HHS officials believe that the adoption data available as early as 1995 are more reliable than early data on other outcomes because of the incentive payments that states can earn for increasing adoptions. They noted that states made great efforts to improve the accuracy of their adoption data when the incentive payment baselines were established. For example, HHS initially estimated about 20,000 adoptions for FY 1997, but after states reviewed and submitted their adoption data, as required for participation in the adoption incentive program, they reported about 31,000 adoptions.

¹⁶All HHS data are presented in terms of federal fiscal year.

¹⁷The University of Chicago has data that tracks foster children in 12 states. One state has provided data since 1975, 7 states have provided data since the 1980s, and 4 states have provided data since the 1990s.

¹⁸The data represented in this analysis were compiled by HHS and the states in response to the adoption incentive payment program created by ASFA. Initial eligibility for participation in the incentive program was based on states reporting adoption data from FY 1995 through FY 1997. An average of these years was taken to establish a baseline number of adoptions. Adoption data prior to FY 1995 are considered unreliable.

Note: The percentages in parentheses represent the increase in finalized adoptions over the previous year.

Source: Data from HHS's Children's Bureau used to determine adoption incentive payments.

In at least one of the 3 years following the implementation of ASFA, all states increased their adoptions over the average number of adoptions finalized between 1995 and 1997. This average represents the baseline established for each state for participation in the adoption incentive program. A comparison of the states' baselines with their average number of adoptions for the 3 years following ASFA shows that 10 states at least doubled the annual average number of adoptions between 1998 and 2000 (see table 1).

Source: Data from the HHS's Children's Bureau used to determine adoption incentive payments.

The role ASFA played in the increase in adoptions after 1997, however, is unclear. Similarly, whether the number of foster children being adopted will continue to rise in the future is unknown. While ASFA may have contributed to the adoptions of these children, other factors may have also played a role. For example, HHS officials have noted that earlier state child welfare reform efforts may be linked to the observed increase in adoptions. To better understand why adoptions have increased and to evaluate ASFA's impact, HHS has asked the University of Chicago to use its data from several states to track groups of foster children over time to determine if the percentage of children adopted from foster care has changed and if adoptions occur more quickly now. Since it can take several years for foster children to be adopted, and ASFA has only been in existence for a few years, evidence of ASFA's effect may not be available for some time.

ASFA's effect on other foster care outcomes, such as family reunifications, is also difficult to determine. Lack of comparable and reliable data on foster children, before and after ASFA, make it difficult to know how ASFA has affected the child welfare system. While HHS officials report that some data are reliable and can provide a picture of children in foster care post-ASFA, they state that the child welfare data covering pre-ASFA periods are not reliable. According to HHS data specialists, early data available suffered from problems such as low response rates and data inconsistencies.

Since 1998, however, HHS data specialists have observed marked improvements in the data submitted to HHS by states and attribute the changes to several factors. These factors include the provision of federal technical assistance to the states on data processing issues and the use of federal financial penalties and incentives.¹⁹ HHS data specialists also note that states are improving their data in response to the use of outcome measures in the Child and Family Services Reviews and the annual publication of child welfare outcomes for each state. According to HHS, these data improvements make it impossible to determine whether observed changes in outcomes from one year to the next are the result of changes in data quality or changes in state performance. HHS expects that the data will stabilize over time and can eventually be used as a reliable measure of state performance.

Although HHS cannot provide reliable pre-ASFA data, research conducted at the University of Chicago provides reliable pre-ASFA information on some foster care outcomes for 12 states. However, the University of Chicago's pre-ASFA data cannot be compared with HHS's post-ASFA data. Unlike other child welfare data sources that collect periodic data on children in care,²⁰ the University of Chicago's system follows all of the children entering foster care in an individual year and collects data on them until they leave to determine their final outcomes. This approach provides accurate information on the experiences of all foster children over time and does not over represent the experiences of certain children, such as those who stay in care for extended periods of time.²¹ However, the use of this different measurement technique prevents comparisons of the University of Chicago's pre-ASFA data with HHS's post-ASFA data.

¹⁹In January 2002, the Assistant Secretary of the Administration for Children and Families at HHS repealed the penalties issued for non-compliance with AFCARS submissions. The decision came after 12 states jointly appealed the penalties and the presiding officer of the hearing, appointed by the Assistant Secretary, recommended their withdrawal. HHS data specialists expressed concern about the consequences the repeal might have on AFCARS data quality if noncomplying states are not penalized. However, HHS officials report that they will continue to monitor data quality and state compliance in the absence of penalties.

²⁰Other child welfare data sources include AFCARS and VCIS.

²¹AFCARS and VCIS are point-in-time data systems and therefore only capture children who are in foster care at the time the data are reported. As a result, these systems can report either the experiences of children currently in care or the experiences of children who have exited care, but cannot track the experiences of all children served in foster care over a certain timeframe. When collecting data on children currently in care, children with longer stays are over represented in the data set, biasing the measured outcomes. Similarly, when collecting data on children who exit care, the overall lengths of stay for children exiting care tend to appear shorter due to the over representation of children with shorter lengths of stay, biasing the outcomes in the opposite direction. Consequently, the AFCARS and VCIS outcome measures vary depending on the group observed.

Current Data Describe Characteristics and Experiences of Children Who Exited Care between 1998 and 2000

Although pre-ASFA data are limited and more time is needed to determine how ASFA has affected the child welfare system, current data do shed some light on the characteristics and experiences of the more than 741,000 children who exited foster care between 1998 and 2000.²²According to HHS data for this time period, children left foster care after a median length of stay of approximately 1 year. Prior to leaving foster care, children typically lived in 1 or 2 foster care placements and a very small portion of them were abused or neglected by their foster care providers. Most foster children reunified with their families; however, approximately 33 percent of the children who went home to their families in 1998 subsequently returned to foster care within 3 years, for reasons such as additional abuse and neglect at home. A smaller percentage of children left foster care through adoption. The majority of children adopted from foster care were under age 12 and classified as having special needs. Limited evidence suggests that few adopted children returned to the child welfare system.

Foster Care Trends

About half of the children leaving foster care exit within one year; however, the data show slight changes in the length of children's stays during 1998-2000 (see table 2). In 1998, the median length of stay for children exiting care was 11 months—by 2000, it had risen to 12 months. In contrast, the median length of stay for adopted children dropped from 43 months in 1998 to 39 months in 2000 (see table 3). Determining whether these shifts represent real changes in the amount of time children spend in foster care or whether they simply reflect the recent improvements in HHS data is difficult. Twenty-three states reported in our survey that in fiscal year 2000 adopted children spent an average of 18 months living with the family that eventually adopted them prior to their adoption being finalized.²³

^aBased on data from 44 states (including the District of Columbia and Puerto Rico).

^bBased on data from 51 states (including the District of Columbia and Puerto Rico).

^cBased on data from 51 states (including the District of Columbia).

^d"Data missing" indicates that a child exited foster care, but the length of stay is not available.

Source: AFCARS data from HHS's Children's Bureau.

²²According to HHS's AFCARS data, 223,255 children exited foster care in FY 1998 from 44 states (including the District of Columbia and Puerto Rico); 250,950 exited foster care in FY 1999 from 51 states (including the District of Columbia and Puerto Rico); and 267,344 exited foster care in FY 2000 from 51 states (including the District of Columbia).

²³One of the 23 states reporting on the length of time children lived with the family that eventually adopted them provided CY 2000 data.

Note: Percentages do not always total to 100 due to rounding.

^aBased on data from 49 states (including the District of Columbia and Puerto Rico).

^bBased on data from 52 states (including the District of Columbia and Puerto Rico).

^cBased on data from 52 states (including the District of Columbia and Puerto Rico).

^dSome of the children adopted after less than a year in foster care are infants. During our site visit to Texas, child welfare workers reported that infants abandoned by their families are generally adopted by the time they are 6 or 7 months old.

Source: AFCARS data from HHS's Children's Bureau.

The amount of time children spend in foster care varies from state to state. For example, the median length of stay for children exiting care in Delaware was about 5 months in 2000, while in Illinois it was close to 4 years in 2000. Differences in state foster care stays may be linked to child welfare practices. For example, higher adoption rates can play a role in increasing the median length of stay figures, since adopted children stay in foster care for longer periods of time. Conversely, higher reunification rates can play a role in decreasing the median length of stay, since reunified children spend less time in foster care. In Delaware, most children who left care reunified with their families and only a small percentage were adopted. In contrast, Illinois had lower reunification rates and one of the highest yearly adoption rate averages in the country. Illinois officials explained that they work extensively with families to prevent the need for foster care and only bring children into care when these efforts have failed. Consequently, although data are not available, Illinois officials believe that the children in their care are less likely to reunify with their families than foster children in other states that they believe may not work as extensively with families before children are removed from their homes.

Before they leave foster care, most children live in one or two different placements (see table 4). Many children have only one placement during their foster care stay, but a few experience five or more placements.

Adopted children tend to experience a greater number of foster care placements (see table 5). Adopted children may have more foster care placements than other children, in part, because of their longer foster care stays. According to some researchers, children experience more placements the longer they are in foster care.²⁴

^aBased on data from 44 states (including the District of Columbia and Puerto Rico).

^bBased on data from 51 states (including the District of Columbia and Puerto Rico).

^cBased on data from 51 states (including the District of Columbia).

^d"Data missing" indicates that a child exited foster care, but the number of placements is not available.

Source: AFCARS data from HHS's Children's Bureau.

²⁴Kathy Barbell and Madelyn Freundlich, Foster Care Today (Washington, D.C.: Casey Family Programs, 2001).

Note: Percentages do not always total to 100 due to rounding.

^aBased on data from 49 states (including the District of Columbia and Puerto Rico).

^bBased on data from 52 states (including the District of Columbia and Puerto Rico).

^cBased on data from 52 states (including the District of Columbia and Puerto Rico).

^d"Data missing" indicates that a child was adopted from foster care, but the number of foster care placements is not available.

Source: AFCARS data from HHS's Children's Bureau.

During their foster care stays, a small percentage of children are abused or neglected by their caregivers. Our survey results²⁵ indicate that the median percentage of children abused or neglected while in foster care during 1999 and 2000 was 0.60 percent and 0.49 percent, respectively. Maltreatment rates in foster care in 2000 ranged from a high of 2.74 percent in the District of Columbia to a low of 0.02 percent in Nebraska.²⁶

On average, less than one-third of the children in foster care exit each year. Children exit foster care in a number of ways, including reunifying with their families,²⁷ being adopted, emancipation,²⁸ or entering a guardianship arrangement²⁹ (see table 6). Upon leaving foster care, most children returned home to the families they had been living with prior to entering foster care. However, a number of these children re-entered foster care for a number of reasons, such as additional abuse and neglect by their families (see table 7).

²⁵Of the 46 states that responded to our survey, 16 provided data on the percentage of foster children who had a substantiated report of abuse or neglect in FY 1999, 2 provided data for CY 1999, 1 provided data for SFY 1999, 18 provided data for FY 2000, 1 provided data for CY 2000, and 2 provided data for SFY 2000.

²⁶According to HHS's Child Welfare Outcomes 1999: Annual Report, the median percentage of foster children abused and neglected while in care was 0.5 percent for the 20 states reporting. Maltreatment rates in foster care range from a high of 2.3 percent in Rhode Island to a low of 0.1 percent in Arizona, Delaware and Wyoming.

^aBased on data from 44 states (including the District of Columbia and Puerto Rico).

^bBased on data from 51 states (including the District of Columbia and Puerto Rico).

^cBased on data from 51 states (including the District of Columbia).

^d"Data missing" indicates that a child exited foster care, but the exit destination was not reported.

Source: AFCARS data from HHS's Children's Bureau.

²⁷In AFCARS, reunification is defined as the child returning to the family with whom the child had been living prior to entering foster care.

²⁸A child is emancipated from foster care when the child reaches majority age according to state law.

²⁹Guardianship arrangements occur when permanent legal custody is awarded to an individual, such as a relative.

^aBased on data from 44 states (including the District of Columbia and Puerto Rico).

^bBased on data from 51 states (including the District of Columbia and Puerto Rico).

^cBased on data from 51 states (including the District of Columbia).

^dNot applicable.

Source: AFCARS data from HHS's Children's Bureau.

Although most children reunify with their families, the second most common way of exiting foster care is through adoption. The children adopted from foster care have a wide variety of characteristics, yet the data indicate some general themes. Most children adopted from foster care have at least one special need that may make placing a child with an adoptive family challenging.³⁰ On average, 85 percent of the children adopted in 1998, 1999, and 2000 were classified as having at least one special need that would qualify them for adoption subsidies under Title IVE.³¹Eighteen states reported in our survey that, on average, 32 percent of the children adopted from foster care in 2000 had three or more special needs.³² In addition, according to HHS data, children adopted from foster care are equally likely to be male or female, slightly more likely to be black (see table 8), and much more likely to be under age 12 (see table 9). The gender and race/ethnicity distributions of children in foster care are similar to those for children who are adopted. However, the age distribution differs between the two groups of children. For example, in 1999, approximately 46 percent of the children in foster care were 11 years or older.

Note: Percentages do not always total to 100 due to rounding.

^aBased on data from 49 states (including the District of Columbia and Puerto Rico).

^bBased on data from 52 states (including the District of Columbia and Puerto Rico).

^cBased on data from 52 states (including the District of Columbia and Puerto Rico).

^d"Data missing" indicates that a child was adopted from foster care, but the race or ethnicity is not available.

Source: AFCARS data from HHS's Children's Bureau.

³⁰Our discussion of special needs in this report focuses on the determination of special needs of adopted children for eligibility for adoption assistance under Title IV-E of the Social Security Act. Therefore, an equivalent measure is not available for children in foster care.

³¹While current AFCARS data indicate that 85 percent of children adopted in FY 1998 through FY 2000 have at least one special need, the adoption incentive data presented earlier in figure 1 indicate a lower proportion of children with special needs adopted from foster care. An HHS official explained that this discrepancy is due to the timing of the measurement of the data. The data in figure 1 were measured at an earlier point in time and the numbers are not updated since adoption incentive payments are awarded to states based on the number of finalized adoptions reported at the end of each fiscal year. The data presented here reflect changes to AFCARS based on states resubmitting previous data after the initial reporting periods and are as recent as April 2002. In addition, the HHS official explained that different groups of children are included in the AFCARS and adoption incentive program data regarding children with special needs. Only children who are eligible for adoption assistance under Title IV-E are included in the special needs category for the adoption incentive program. AFCARS data on children with special needs, however, include children who are eligible for adoption assistance either through the Title IV-E program or through an individual state's adoption assistance program.

³²One of the 18 states reporting on the number of children adopted with three or more special needs provided data based on CY 2000.

^aBased on data from 49 states (including the District of Columbia and Puerto Rico).

^bBased on data from 52 states (including the District of Columbia and Puerto Rico).

^cBased on data from 52 states (including the District of Columbia and Puerto Rico).

Source: AFCARS data from the HHS's Children's Bureau.

As noted for other outcomes, the lack of reliable and national pre-ASFA data make it difficult to determine whether the rate at which adoptions encountered problems has changed since ASFA was enacted.³³> However, limited data suggest that problems occur in a small percentage of foster care adoptions. According to our survey,³⁴ about 5 percent of adoptions planned in fiscal years 1999 and 2000 disrupted prior to being finalized.³⁵ States also reported that approximately 1 percent of adoptions finalized in these years legally dissolved at a later date and that about 1 percent of the children who were adopted in these years subsequently re-entered foster care.³⁶ However, little time has elapsed since these adoptions were finalized and some of these adoptions may legally dissolve at a later date. HHS data similarly indicate that about 1 percent of the children entering foster care each year have previously been adopted. States reported in our survey that adopted children return to foster care for different reasons, including abuse or neglect by their adopted families, behavior problems which are too difficult for their adoptive families to handle, or a child's need for residential care.

³³Studies on adoption problems prior to ASFA's enactment have several limitations, such as small samples sizes, coverage of a few locations, and a focus on narrowly defined groups of children.

³⁴Of the 46 states that responded to our survey, 17 provided data on adoption disruptions for FY 1999, 2 provided data for CY 1999, 18 provided data on adoption disruptions for FY 2000, and 2 provided data for CY 2000.

³⁵For example, Illinois reported a 12.4 percent disruption rate for FY 1999 on our survey. In comparison, one study (Robert Goerge and others, Adoption, Disruption, and Displacement in the Child Welfare System, 1976-1995 (Chicago: Chapin Hall Center for Children at the University of Chicago, 1995)) reviewed all planned and finalized adoptions in Illinois between 1981 and 1987. During that time, an average of 9.9 percent of adoption plans for Illinois foster children disrupted.

³⁶Of the 46 states that responded to our survey, 18 provided data on dissolutions for FY 1999, 19 provided data on dissolutions for FY 2000 and 1 provided data for CY 2000. In addition, 21 states provided survey data on foster care re-entries by children adopted in FY 1999, 22 provided data for children adopted in FY 2000, and 1 provided data for children adopted in CY 2000. Our survey results indicate that of all the children with finalized adoptions in FY 1999, 0.55 percent returned to foster care in FY 1999 or 2000. Of all the children with finalized adoptions in FY 2000, 1.43 percent returned to foster care in FY 2000.

Although Little Data are Available on Key ASFA Permanency Provisions, Some States Describe Circumstances That Limit Broader Use of These Provisions

While few states were able to provide data on the numbers of children affected by ASFA's fast track and 15 of 22 provisions, some reported on circumstances that make it difficult to use these provisions for more children. In addition, HHS collects very little data on the use of these provisions. Data from four states that provided fast track data in response to our survey indicate that they do not use this provision frequently. Officials at our site visits told us that they use the fast track provision for a small number of children, primarily those who have experienced serious abuse or whose parents had involuntarily lost parental rights for other siblings. However, they described several court-related issues that make it difficult to fast track more children, including court delays and a reluctance on the part of some judges to relieve the state from reunification efforts. Survey responses from the few states that provided data on the 15 of 22 provision indicate that these states do not file TPRs for many children who are in care for 15 months. Officials at the six states we visited believe that ASFA's 15 of 22 time standard has helped them make more timely permanency decisions, but reported that they exempt many children from this requirement for a number of reasons, including difficulties in finding adoptive parents.

While Data on Fast Track Are Limited, Some States Report Court-Related Issues That Hinder the Use of the Fast Track Provision for More Children

Few states were able to provide data on their use of the fast track provision in response to our survey and HHS does not collect this data from the states. As a result, we do not have sufficient information to discuss the extent to which states are using this provision. As shown in table 10, the data from a handful of states suggest the infrequent use of fast track. In fiscal year 2000, for example, about 4,000 children entered the child welfare system in Maryland, but only 36 were fast tracked. Child welfare officials in the six states we visited told us that they used ASFA's fast track provision for a relatively small number of cases. Three states indicated that they fast tracked abandoned infants, while four states reported using fast track for cases involving serious abuse, such as when a parent has murdered a sibling; however, some state officials also noted that few child welfare cases involve these circumstances. In addition, five states reported that they would fast track certain children whose parents had involuntarily lost parental rights to previous children if no indication exists that the parents have addressed the problem that led to the removal of the children.

^aIn its survey response, Massachusetts reported the total number of children who were fast tracked, but did not provide data on the different fast track categories.

Source: GAO survey and AFCARS data.

Officials in five of the states we visited described several court-related issues that hindered the greater use of the fast track provision. However, because of the lack of data on states' use of fast track, we were unable to determine the extent of these problems. Officials in these states told us that some judges or other legal officials are at times reluctant to approve a state's fast track request. According to officials in Massachusetts, North Carolina, and Maryland, some judges believe that parents should always be given the opportunity to reunify with their children. According to child welfare staff for a county in North Carolina, the courts had recently denied the county's request to fast track several cases and ordered the county to provide services to the families involved. In one case, a judge approved a fast track request involving a child who had suffered from shaken baby syndrome, but refused a similar request on a sibling who was born a few months after the shaking episode. County staff stated that the judge's decision was based on the fact that the parents had not hurt the newborn and should be given an opportunity to demonstrate their ability to care for this child.

Three states we visited described other court problems related to the fast track provision. For example, state officials in North Carolina told us that delays in scheduling TPR trials in the state undermine the intent of fast tracking. They noted that the agency may save time by not providing services to a family, but the child may not be adopted more quickly if it takes 12 months to schedule the TPR trial. Officials in Massachusetts expressed similar concerns about court delays experienced in the state when parents appeal a court decision to terminate their parental rights. Finally, a Massachusetts official explained that the state is cautious about using the fast track provision due to concerns that not providing services to parents could undermine their TPR case. According to a Massachusetts official, the state obtained court approval to fast track a case, but subsequently lost the TPR trial in part because the judge found that the parents did not receive services to help them reunify with their child.

Other difficulties in using fast track to move children out of foster care more quickly are related to the specific categories of cases that are eligible to be fast tracked. Officials in five states told us that they look at several factors when considering the use of fast track for a parent who has lost parental rights for other children. In some of these cases, a different birth father may be involved. Child welfare officials told us that they are obligated to work with the father to determine if he is willing and able to care for the child. According to Maryland officials, if the agency is providing services to the father to facilitate reunification, pursuing a fast track case for the mother will not help the child leave foster care more quickly. In addition, child welfare officials in Massachusetts and Illinois emphasized that a parent who has addressed the problems that led to a previous TPR should have an opportunity to demonstrate the ability to care for a subsequent child. For example, they would not necessarily fast track a substance-abusing mother who lost custody of a previous child if she has engaged in treatment and addressed her parenting issues.

Regarding the fast track category involving parents who have been convicted of certain felonies, child welfare officials in Massachusetts and Texas described this provision as impractical due to the time it takes to obtain a conviction. Massachusetts officials told us that, in most cases, the children are removed at the time the crime is committed and judges will not approve the fast track in these cases until the parent is actually convicted, which is usually at least a year after the actual crime. As a result, the state must provide services to reunify the family until further evidence of the parent's unfitness is documented. Finally, in Massachusetts, Texas, and Maryland, officials reported that it can be difficult to prove that a parent subjected a child to aggravated circumstances, such as torture or sexual abuse. According to these officials, the time and effort to go through additional court hearings to demonstrate the aggravated circumstances is not worthwhile; instead, the child welfare agency chooses to provide services to the family.

In response to our survey, three states provided information about why they did not fast track cases that fell into one of the fast track categories, citing reasons that were similar to those reported by our site visit states. For example, Minnesota estimated that in 25 percent of the cases, the state was working to reunify the children with the noncustodial parent. In an additional 25 percent of the cases, the court did not approve the state's request to fast track the case. In the remaining cases, Minnesota did not consider fast track to be in the child's best interests. A Minnesota official explained that in certain circumstances, the agency would try to reunify a family, even if the parents had subjected the child to aggravated circumstances or lost custody of a previous child. For example, if a parent assaulted a child resulting in a broken bone—which would be considered aggravated circumstances under Minnesota law—the agency might not consider a TPR to be in the child's best interests if the assault was a single incident for which the parent accepted responsibility and the child has otherwise had a positive relationship with his or her parent. In addition, the state might not fast track a child born to a mother who had lost custody of a previous child, if the TPR occurred years before and the mother's circumstances had since improved.

Although Little Data Exist on 15 of 22, Some States Report That Many Children Are Exempted

Most states do not collect data on their use of ASFA's 15 of 22 provision. In response to our survey, only nine states were able to provide information on the number of children for whom the state filed a TPR due to the 15 of 22 provision or the number of children who were exempted from this provision. In addition, HHS does not systematically track this data. As part of its Child and Family Services Reviews (CFSR), HHS collects some limited information on the 15 of 22 provision. Specifically, HHS asks each state to discuss its compliance with the 15 of 22 provision and directly assesses compliance during its on-site review of a limited number of case records, if the case under review involves a child who has been in care for 15 months.³⁷

For most of the states that provided data on their use of the 15 of 22 provision in response to our survey, the number of children exempted from the provision greatly exceeded the number of children to whom it was applied (see table 11). For example, while Oklahoma filed over 1,000 TPRs primarily because the child had been in foster care for 15 of the most recent 22 months, it did not file a TPR for an additional 2,900 children. Similarly, in 1999, we reported on states' efforts to review all children who were already in foster care for 15 months when ASFA was enacted to determine if a TPR should be filed or to document an exemption if a TPR was not appropriate, as required by ASFA. The 12 states that had data reported exempting 60 percent of the children they reviewed.³⁸

^aNot available.

^bWest Virginia reported that it filed 375 TPRs for children who had been in foster care for 15 of the most recent 22 months. However, being in care for 15 of the most recent 22 months was not necessarily the primary reason for filing the TPR.

Source: GAO survey.

³⁷In addition, prior to the CFSR site visit to a state, ACF provides the state with an approximation of the number of children potentially impacted by the 15 of 22 provision, based on a state's AFCARS data.

³⁸U.S. General Accounting Office, Foster Care: States' Early Experiences Implementing the Adoption and Safe Families Act, GAO/HEHS-00-1 (Washington, D.C.: Dec. 22, 1999).

Officials in all six site visit states told us that establishing specific timeframes for making permanency decisions about children in foster care has helped their child welfare agencies focus their priorities on finding permanent homes for children more quickly. Two of the states we visited—Texas and Massachusetts—created procedures prior to ASFA to review children who had been in care for a certain length of time and decide whether continued efforts to reunify a family were warranted. Other states had not established such timeframes for making permanency decisions before the 15 of 22 provision was enacted. The director of one state child welfare agency told us that, prior to ASFA, the agency would work with families for years before it would pursue adoption for a child in foster care. In response to ASFA's requirement to hold permanency hearings every 12 months for children in foster care, five of the states we visited emphasized that they now try to make decisions about a child's permanent placement by the time the child has been in care for 12 months. The director of one state child welfare agency noted that the 15 of 22 provision does not fit well with other child welfare timeframes—he stated that having more frequent permanency hearings would force states to make more timely decisions and would be less administratively awkward to implement.

Officials in Oregon, Maryland, and North Carolina stated that that the pressure of these new timeframes has helped child welfare staff work more effectively with parents, informing parents up front about what actions they have to take in the next 12 to 15 months in order to reunify with their children. Conversely, private agency staff in three states expressed concern that pressure from these timeframes could push the child welfare agency and the courts to make decisions too quickly for some children. In one state, for example, staff with a private agency that recruits adoptive families for the state were worried that making decisions so quickly may lead to more children re-entering foster care after being adopted or reunified with their families.

Child welfare officials in the six states we visited described several circumstances under which they would not file a TPR on a child who was in care for 15 of the most recent 22 months. In five of the six states, these officials told us that the provision is difficult to apply to children with special needs for whom adoption may not be a realistic option, such as adolescents and children with serious emotional or behavioral problems. Officials from Maryland and North Carolina reported that, in many cases, the child welfare agency exempts these children from the provision because either the agency or the courts do not consider it to be in theirbest interest to be legal orphans—that is, to have their relationship to their parents legally terminated, but have no identified family ready to adopt them.³⁹ State officials in Oregon told us that state law requires that parental rights be terminated solely for the purpose of adoption, so as to avoid creating legal orphans. Officials in other states said that while the child welfare agency would like to pursue a TPR, some courts are not willing to do so unless a potential adoptive family has been identified for the child.

Officials in four states noted that many adolescents remain in long-term foster care. In some cases, they have strong ties to their families, even if they cannot live with them, and will not consent to an adoption.⁴⁰ In other cases, the teenager is functioning well in a stable situation with a relative or foster family that is committed to the child but unwilling to adopt.⁴¹ For example, officials in a child welfare agency for a county in North Carolina told us about a potentially violent 16-year old foster child who had been in a therapeutic foster home for 10 years. The family was committed to fostering the child, but did not want to adopt him because they did not have the financial resources to provide for his medical needs and because they did not want to be responsible for the results of his actions.

Similarly, four states reported difficulties in recruiting adoptive families for children with severe behavioral or medical problems who will require long-term treatment in a residential facility. State officials in Massachusetts told us that some of these children have such severe problems that they are not ready to live in a family setting. Staff in a county child welfare office in North Carolina told us that mentally ill children whose parents voluntarily place them in state custody because they cannot afford the residential services their children require are generally exempted from the 15 of 22 provision.

In Illinois, child welfare staff told us that some parents need a little more than 15 months to address the problems that led to the removal of their children. If the child welfare agency is reasonably confident that the parents will be able to reunify with their children in a few months, the agency will not file a TPR for a child who has been in foster care for 15 months. Similarly, staff in a county child welfare office in North Carolina told us about two cases involving adolescent children in long-term foster care who became pregnant and had a child while in foster care. In these cases, the adolescent mothers remained in foster care with their children. Staff explained that these young mothers needed more than 15 months to be able to parent their children independently, given their own troubled pasts. As long as the mothers were making reasonable progress in parenting their children, the state would not file a TPR on these infants even though they were in foster care for more than 15 months. One of the mothers had recently reunified with her child, now 2 years old, and was expected to regain legal custody of the child shortly.

³⁹In North Carolina, child welfare agency staff may recommend to the court that a child be exempted from the 15 of 22 provision; according to state officials, however, North Carolina requires that a judge determine that one of the exceptions exists.

⁴⁰All the states we visited require that children over a certain age consent to their adoption. For example, Maryland requires that children 10 years or older consent to their adoption.

⁴¹ASFA specifically allows states to exempt children placed with relatives from the 15 of 22 provision.

Child welfare officials in four states observed that parents must have access to needed services, particularly substance abuse treatment, soon after a child enters care in order for the child welfare system to determine if reunification is a realistic goal by the time a child has been in care for 15 months. Officials in Texas, Oregon, and Maryland reported that the lack of appropriate substance abuse treatment programs that address the needs of parents makes it difficult to get parents in treatment and stable by the 15th month. Juvenile court judges in Massachusetts and Oregon told us that they would not necessarily pursue a TPR when a child has been in care for 15 of the most recent 22 months if parents are engaged in substance abuse treatment and showing progress toward reunification.

State officials in Massachusetts, North Carolina, and Maryland noted that delays in scheduling TPR trials and delays in hearing appeals of TPR decisions can undermine the use of the 15 of 22 provision to achieve permanency for children in a timely manner. For example, Massachusetts officials noted that appeals of TPR decisions face significant delays— simply scheduling the appeal trial can take a year.

In response to our survey, a few states provided explanations regarding why they did not file a TPR on children who had been in care for 15 of the most recent 22 months. The reasons reported by seven states were similar to those reported during our site visits, although they varied significantly among the seven states (see table 12). For example, the District of Columbia estimated that it did not file a TPR for about 60 percent of the children who were in care for 15 months because the state expected that these children would soon be reunified with their parents. In contrast, Rhode Island reported that 600 children were in care for 15 months without having a TPR filed and estimated that 67 percent of them were adolescents with permanent plans of either independent living or long- term foster care.

Note: Percentages do not total to 100 percent due to rounding.

^aMinnesota reported data for CY 2000.

Source: GAO survey.

New ASFA Adoption-Related Funds Most Commonly Used to Recruit Adoptive Families and Provide Post Adoption Services

States reported in our survey that they most commonly used their adoption incentive payments and PSSF adoption promotion and support services funds to recruit adoptive parents and to provide post adoption services. For example, Arizona has used its incentive payments to fund performance-based contracts that reward agencies for finding adoptive families for groups of siblings, children aged 10 or older and those from minority groups.⁴² Utah, on the other hand, has used its PSSF funds to sponsor a post adoption Web site for adoptive families. In addition to recruitment and post adoption services, we found that states have spent these ASFA funds on a variety of other child welfare activities, including hiring and training social workers.⁴³

Our survey results on states' use of new adoption-related funds mirror findings from a recent study,⁴⁴ which found that the top two uses of incentive payments were for the recruitment of adoptive families and the provision of post adoption services (see table 13). For example, states are using ASFA's adoption-related funds to pursue a variety of activities to recruit adoptive parents. Child welfare officials in all of the states we visited reported that they are struggling to recruit adoptive families for older children and those with severe behavioral or medical problems. To meet this challenge, states are investing in activities designed to match specific foster children with adoptive families, as well as general campaigns to recruit adoptive families. Child specific recruitment efforts include: featuring children available for adoption on television, hosting matching parties for prospective adoptive parents to meet children available for adoption, and taking pictures and videos of foster children to show to prospective families. Massachusetts used its incentive payments to fund recruitment videos to feature the 20 children who had been waiting the longest for adoptive families, while Nebraska used its incentive funds to improve the profiles of waiting children on its state Web site.

⁴²State specific examples regarding the use of these funds come from either our survey, state Annual Progress and Service Reports to HHS, or our site visits.

⁴³Our review of state Annual Progress and Services Reports to HHS found that states have most frequently used their adoption incentive payments for recruitment, post adoption services and training.

⁴⁴Cornerstone Consulting Group, Inc., A Carrot Among the Sticks: The Adoption Incentive Bonus (Houston, 2001).

Note: Of the 46 states that responded to our survey, 34 provided data on the use of FY 1999 and FY 2000 incentive payments and 26 provided information on the use of FY 2000 PSSF funds.

^aOne state provided data for CY 1999.

^bOne state provided data for CY 2000.

^cPost adoption services include: counseling, respite care (short-term specialized child care to provide families with temporary relief from the challenges of caring for adoptive children), support groups, adoption subsidies, and adoption preservation services.

^dThis activity was not included as a response for this funding source.

^eOne state provided data for SFY 1999 and SFY 2000.

Source: GAO survey.

General recruitment efforts being funded by states include: promoting adoption through National Adoption Month events, hiring additional recruiters, and partnering with religious groups. For example, Maryland has used its PSSF funds to partner with faith-based organizations to recruit adoptive families primarily for black children, while Colorado used its incentive payments to hire a public relations firm to develop a campaign to recruit minority parents. According to our survey results, 18 states are using PSSF funds to create or expand both child specific recruitment efforts and general recruitment programs.⁴⁵

States are also investing adoption incentive payments and PSSF funds in services to help adoptive parents meet the challenges of caring for children who have experienced abuse and neglect. Adoptive parents sometimes have difficulties managing the emotional and behavioral problems of children from foster care.⁴⁶ Some researchers believe that post adoption services may help stabilize these adoptive families.⁴⁷ However, available research on post adoption services is largely descriptive, with little information on the effectiveness of such services.⁴⁸During our site visits, officials in Massachusetts and Illinois pointed out that the population of adopted children had increased significantly in recent years and that the availability of post adoption services was essential to ensure that these placements remain stable.

Approximately 60 percent of the states responding to our survey used their adoption incentive payments or their PSSF funds or both for post adoption services. Our survey results show that 21 states used PSSF dollars to initiate or expand post adoption counseling and support groups.⁴⁹ In addition, 20 states reported using PSSF dollars to create or expand services to preserve adoptions and help adoptive families maintain their new relationships. Thirteen states also reported that they are providing respite services with PSSF adoption promotion and support services dollars. In addition to these core post adoption services, some states noted both in our survey and in other reports that they are providing a range of other services to adoptive families, including information and referral networks, mentoring, and recreational opportunities. For example, California has used some of its adoption incentive funds to pay for therapeutic camps and tutoring sessions for adopted children. In addition, Minnesota has used PSSF funds to teach adoptive parents how to care for children with fetal alcohol syndrome and children who find it difficult to become emotionally attached to caregivers.

⁴⁵Of the 46 states that responded to our survey, 34 provided data on new and expanded services funded by PSSF adoption promotion and support services dollars. The 18 states creating or expanding child specific recruitment efforts are not the same 18 states that are creating or expanding general recruitment programs.

⁴⁶Elizabeth Oppenheim, Shari Gruber, and Doyle Evans, Report on Post-Adoption Services in the States (Washington, D.C.: The Association of Administrators of the Interstate Compact on Adoption and Medical Assistance, Inc., 2000).

⁴⁷Richard P. Barth and others, "Contributors to Disruption and Dissolution of Older-Child Adoption,"Child Welfare, vol. LXV, no. 4 (1986).

⁴⁸Richard P. Barth, Deborah A. Gibbs, and Kristin Siebenaler, Assessing the Field of Post- Adoption Services: Family Needs, Program Models, and Evaluation Issues (Washington, D.C.: April 10, 2001), http://aspe.hhs.gov/hsp/PASS/lit-rev-01.htm (downloaded Feb. 11, 2002).

⁴⁹The 21 states using PSSF dollars to initiate or expand post adoption counseling are not the same 21 states using PSSF dollars to initiate or expand post adoption support groups.

Although the 46 states responding to our survey reported that they are most frequently using the money for the activities described above, over two-thirds of them also reported that they are investing some of these funds in other services.⁵⁰ Many states are using PSSF funds to provide preadoptive counseling to help children and parents prepare for the emotional challenges of forming a new family. Similarly, some states are using incentive payments and PSSF funds to train foster families, adoptive families, and service providers. For example, Arkansas used incentive money to help families attend an adoptive parent conference and Nevada used PSSF dollars to fund an adoption-training curriculum in Spanish. Likewise, Montana used incentive payments to provide adoption training to therapists who agree to provide services to children in foster care. Kentucky, on the other hand, has used incentive funds to train judges and attorneys on adoption matters.

In addition, we found that some states are taking advantage of the flexibility allowed in the use of adoption incentive payments to increase the number of people working on child welfare cases. During our site visit to Oregon, child welfare officials told us that the lack of legal resources has inhibited the state's ability to quickly pursue court cases against birth parents to terminate their parental rights and thereby free a child for adoption. To address this issue, Oregon has used its adoption incentive payments to contract for additional lawyers to litigate these cases. According to our survey results, 6 states have used the incentive payments to hire or contract additional legal staff and 13 states have used these funds to hire or contract additional social workers.⁵¹Noting that state adoption numbers may level off in the future, a recent report questioned the sustainability of investments made with adoption incentive payments.⁵²Similarly, three states we visited told us that they did not believe they would continue to increase adoption levels and would therefore not earn future incentive payments, and one of these states had limited its use of incentive funds to one time, nonrecurring expenses.

⁵⁰Of the 46 states that responded to our survey, 33 reported using their adoption incentive payments or PSSF adoption promotion and support services funds for services other than recruitment or post adoption services. One of these states provided CY data for its use of adoption incentive payments and PSSF adoption promotion and support services funds and one state provided SFY data.

⁵¹Of the 46 states that responded to our survey, 34 provided data on the use of FY 1999 and FY 2000 adoption incentive payments. One state that reported using incentive payments to hire or contract additional legal staff and social workers provided SFY data.

⁵²Cornerstone Consulting Group, Inc., A Carrot Among the Sticks.

States Develop Practices in Response to Long-Standing Barriers That Continue to Hamper Efforts to Promote Permanency for Foster Children

States have been developing a range of practices to address long-standing barriers to achieving permanency for children in a timely manner—many of which have been the subject of our previous reports. Both independently and through demonstration waivers approved by HHS,⁵³states are using a variety of practices to address barriers relating to the courts, recruiting adoptive families for children with special needs, placing children in permanent homes in other jurisdictions, and the availability of needed services. For example, with a demonstration waiver, Maryland is testing whether the provision of comprehensive and coordinated drug treatment services to parents will improve their access to services and reduce the length of time their children spend in foster care. Because few of these practices have been rigorously evaluated, however, limited information is available on their effectiveness.

Systemic Court Problems Continue to Delay Child Welfare Cases

Our previous work, all the states we visited, and over half of our survey respondents identified problems with the court system as a barrier to moving children from foster care into safe and permanent homes. For example, 29 states reported in our survey that the child welfare system did not have enough judges or court staff, 28 reported that not enough training was available for judges or other court personnel, and 23 reported the existence of judges who were not supportive of ASFA's goals.⁵⁴ In 1999, we reported on systemic problems that hinder the ability of courts to produce decisions on child welfare cases in a timely manner that meet the needs of children.⁵⁵ The barriers included inadequate numbers of judges and attorneys to handle large caseloads, the lack of cooperation between the courts and child welfare agencies, and insufficient training of judges and attorneys involved in child welfare cases.

⁵³See appendix II for a description of current waiver projects and the status of their evaluations.

⁵⁴Of the 46 states that responded to our survey, 40 reported on the insufficient number of judges and court staff, 41 reported on the lack of training for judges and court staff, and 39 reported on judges not being supportive of ASFA's goals. Of the 29 states that reported an insufficient number of judges or court staff, 18 reported that the problem represented a moderate, great, or very great hindrance to finding permanent homes for children. Of the 28 states that reported insufficient training for court staff, 20 reported that the problem represented a moderate or great hindrance. Of the 23 states that reported that judges were not supportive of ASFA's goals, 10 reported that the problem represented a moderate, great, or very great hindrance.

⁵⁵U.S. General Accounting Office, Juvenile Courts: Reforms Aim to Better Serve Maltreated Children, GAO/HEHS-99-13 (Washington, D.C.: Jan. 11, 1999).

During our visit to Massachusetts, state officials told us that the courts experienced significant delays in court hearings and appeals due to a lack of court resources. As an alternative to court proceedings, Massachusetts implemented a permanency mediation program—a formal dispute resolution process in which an independent third party facilitates permanency planning between family members and potential adoptive parents in a nonadversarial setting. Three other states we visited—Texas, Oregon, and Maryland—have implemented similar mediation programs. By avoiding trials to terminate parental rights, Massachusetts officials reported that permanency mediation helps reduce court workloads and more effectively uses limited court resources. In addition, they told us that the mediation process eliminates appeals because a joint permanency decision is made between the birth parents and the adoptive parents that both parties can accept. For example, an open adoption between the birth and adoptive parents is a common outcome of permanency mediation, allowing the birth parents to continue some type of relationship with their child after adoption. A preliminary evaluation of the Massachusetts program suggested that cases involved in the mediation program needed less time and fewer court resources to reach an agreement than cases that go to trial. However, the evaluation did not directly compare outcomes, such as the length of time a child spent in foster care, for mediation and nonmediation cases.

To improve collaboration between child welfare and court staff, two states we visited developed ongoing committees to address barriers to achieving permanency for children in foster care. For example, Massachusetts created a committee comprised of staff from the courts, the Attorney General's office, and the child welfare agency to identify and address court delays affecting child welfare cases. This committee has studied delays in the process for appealing child welfare decisions and has implemented several changes to streamline the process. Illinois has several ongoing committees composed of court and child welfare agency staff to address a variety of legal barriers that delay the placement of a child in a safe and permanent home.

Texas officials identified court barriers in rural areas that negatively affect both the timeliness and quality of child welfare proceedings—specifically, the lack of court time for child welfare cases and the lack of judges with training and experience in child welfare issues. In response to these barriers, Texas developed the visiting judge cluster court system, an approach in which a judge trained in child welfare issues is assigned to a cluster of rural counties. The judge travels from county to county presiding over all child welfare cases. This approach can create more court time in rural areas and allows knowledgeable and experienced judges to make the best possible decisions for children in foster care. While Texas officials believe this approach has been helpful in moving children to permanency, no formal evaluation of the approach has been conducted.

Permanency for Children with Special Needs Hindered by Lack of Adoptive Families

Officials in five states we visited, along with the majority of the respondents to our state survey, reported that difficulties in recruiting families to adopt children with special needs is a major barrier to achieving permanent placements for these children.⁵⁶ The National Center for Resource Family Support notes that the lack of foster and adoptive families to meet the needs of children in care is one of the biggest challenges facing child welfare agencies across the nation.⁵⁷ In Texas and Illinois, social work staff and state officials noted that the children currently in foster care are older and have more severe problems, making it increasingly difficult in find adoptive homes for the children in care.

Our survey revealed that states relied on three main activities to recruit adoptive families for children who are waiting to be adopted: listing a child's profile on state and local Web sites, exploring adoption by adults significantly involved in the child's life, and featuring the child on local television news shows.⁵⁸ Other recruitment efforts cited by the states we visited included profiling children in need of adoptive families in local newspapers, holding regular meetings during which social workers across the state exchange information on children in their communities who need an adoptive family and local families available to adopt, and holding adoption parties during which children available for adoption are introduced to families who are waiting to adopt a child. In Massachusetts, the child welfare agency established a successful collaboration with a local company that sponsored adoption fairs for children with special needs, donated space for meetings, and provided advice on effective marketing techniques. In Illinois and Maryland, staff use databases to match children with a goal of adoption with families waiting to adopt a child.

Several states we visited are also using recruitment campaigns targeted to particular individuals who may be more likely to adopt children with special needs. However, a report on recruitment efforts in Illinois noted that little information exists on what kinds of families are likely to adopt children with specific characteristics.⁵⁹The child welfare agencies in Maryland, North Carolina, Texas, and Illinois are collaborating with local churches to recruit adoptive families specifically for minority children. In addition, Illinois conducted a recruitment campaign at local hospitals to identify adoptive families for children with complex medical needs; however, of the 14 children adopted as a result of the campaign, only one had a complex medical need. While the states we visited used a variety of recruitment efforts to find families for special needs children, they generally did not collect data on the effectiveness of their recruiting efforts.

During our site visit, Illinois social workers discussed the importance of consulting with people involved in a child's life, such as coaches and teachers, to identify those who might be interested in adopting a child. However, the Illinois recruitment report found that many adoption workers did not have the experience or skills to carry out such child- specific recruitment activities effectively. To address this, the state has established a training program for social workers on specialized recruitment activities.

ASFA requires states to document the individualized recruitment efforts undertaken for a child waiting for an adoptive family. The states we visited used several documentation methods, such as making notes in a child's case record, using tracking forms, and using computerized databases that document all actions taken on a child's case. For example, state officials in Oregon recently created a new document that social workers must use to record efforts made to recruit adoptive families for foster children. In Massachusetts, if a child has a goal of adoption and no identified adoptive family, the social worker is required to submit an electronic referral form within a specified timeframe to the regional recruitment office.

⁵⁶Of the 46 states responding to our survey, 43 states reported on the sufficiency of adoptive homes for children with special needs. All 43 states reported that they did not have enough adoptive homes for children with special needs, with 39 reporting that the problem represented a moderate, great, or very great hindrance to finding permanent homes for children.

⁵⁷National Center For Resource Family Support, Casey Family Programs, http://www.casey.org/cnc/recruitment/index.htm (downloaded Apr. 9, 2002).

⁵⁸Of the 46 states responding to our survey, 44 provided information on recruitment activities used in their states.

⁵⁹Jane Elmore and Diane DeLeonardo, Report on the Status Of Foster and Adoptive Parent Recruitment In the Illinois Child Welfare System (Springfield, Ill.: n.p., 2002).

In addition to the activities described above, some demonstration waivers are testing different approaches to finding permanent homes for children in foster care. Seven states are using demonstration waivers to pay subsidies to relatives and foster parents who become legal guardians to foster children in their care. These states hope to reduce the number of children in long-term foster care by formalizing existing relationships in which relatives or foster parents are committ