This is the accessible text file for GAO report number GAO-07-816 entitled 'African American Children in Foster care: Additional HHS Assistance Needed to Help States Reduce the Proportion in Care' which was released on July 30, 2007. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to the Chairman, Committee on Ways and Means, House of Representatives: United States Government Accountability Office: GAO: July 2007: African American Children In Foster Care: Additional HHS Assistance Needed to Help States Reduce the Proportion in Care: GAO-07-816: GAO Highlights: Highlights of GAO-07-816, a report to the Chairman, Committee on Ways and Means, House of Representatives Why GAO Did This Study: A significantly greater proportion of African American children are in foster care than children of other races and ethnicities, according to HHS and other research. Given this situation, GAO was asked to analyze the (1) major factors influencing the proportion of African American children in foster care, (2) extent that states and localities have implemented promising strategies, and (3) ways in which federal policies may have influenced African American representation in foster care. GAO’s methodologies included a nationwide survey; a review of research and federal policies; state site visits; analyses of child welfare data; and interviews with researchers, HHS officials, and other experts. What GAO Found: A higher rate of poverty is among several factors contributing to the higher proportion of African American children entering and remaining in foster care. Families living in poverty have greater difficulty accessing housing, mental health, and other services needed to keep families stable and children safely at home. Bias or cultural misunderstandings and distrust between child welfare decision makers and the families they serve are also viewed as contributing to children’s removal from their homes into foster care. African American children also stay in foster care longer because of difficulties in recruiting adoptive parents and a greater reliance on relatives to provide foster care who may be unwilling to terminate the parental rights of the child’s parent—as required in adoption—or who need the financial subsidy they receive while the child is in foster care. Most states we surveyed reported using strategies intended to address these issues, such as involving families in decisions, building community supports, and broadening the search for relatives to care for children. HHS provides information and technical assistance, but states reported that they had limited capacity to analyze data and formulate strategies, and states we visited told us they relied on assistance from universities or others. States reported that the ability to use federal funding for family support services was helpful in keeping African American children safely at home and that federal subsidies for adoptive parents helped move children out of foster care. However, they also expressed concerns about the inability to use federal child welfare funds to provide subsidies to legal guardians. As an alternative to adoption, subsidized guardianship is considered particularly promising for helping African American children exit from foster care. States were also concerned about the lack of flexibility to use federal foster care funds to provide services for families, although states can use other federal funds for this purpose if they consider it a priority. Figure: Proportion of Children in Foster Care Settings, End of Fiscal Year 2004: [See PDF for Image] Source: GAO analysis of AFCARS and Census data. [End of figure] What GAO Recommends: GAO suggests that Congress consider amending current law to allow subsidies for legal guardianships. HHS believes its proposal for restructuring child welfare funding would give states the option to do this, but the viability of this proposal is uncertain. GAO also recommends that HHS further assist states in addressing disproportionality. In its comments, HHS noted that GAO’s recommendation was consistent with its efforts to provide technical assistance to states, but did not address the specific actions GAO recommended. GAO continues to believe that further assistance is important for helping states address disproportionality. [Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-07-816]. To view the full product, including the scope and methodology, click on the link above. For more information, contact Denise M. Fantone at (202) 512-7215 or fantoned@gao.gov. [End of section] Contents: Letter: Results in Brief: Background: States Report Poverty and Difficulty in Finding Permanent Homes Are among Major Factors Influencing African Americans' Entry and Length of Stay: States Implemented a Range of Strategies Considering Promising for Addressing Disproportionality, but Fewer States Specifically Focus Attention on Issue: States Reported That Some Current Federal Policies May Reduce the Disproportion of African American Children in Care, While Other Policies May Increase It: Conclusions: Matter for Congressional Consideration: Recommendation for Executive Action: Agency Comments and Our Response: Appendix I: Objectives, Scope, and Methodology: Appendix II: Disproportionality Indexes of Children in Foster Care by Race and State: Appendix III: Comments from the Department of Health and Human Services: Appendix IV: GAO Contact and Staff Acknowledgments: Tables: Table 1: Types of Temporary and Permanent Homes for Children in Child Welfare: Table 2: Key Federal Legislation Affecting Foster Care Programs: Table 3: Strategies Used by States That Address Factors Contributing to Disproportionality: Table 4: Disproportionality Index by State of Children Ages 17 Years of Age or Under in Foster Care as of Last Day of Fiscal Year 2004: Figures: Figure 1: Proportion of Children by Race in Foster Care Settings, End of Fiscal Year 2004: Figure 2: Federal Child Welfare Funding by Funding Source, FY 2004: Figure 3: State Views on Poverty-Related Factors Affecting Higher Entry of African American Children to Foster Care: Figure 4: State Views on Support and Preventive Services Affecting African American Children's Higher Entry into Foster Care: Figure 5: State Views on Issues of Distrust and Bias or Cultural Misunderstanding Affecting African American Children's Greater Entry to Foster Care: Figure 6: State Views on Factors Affecting Longer Time in Foster Care for African American Children --Difficulty in Finding Permanent Homes: Figure 7: State Views on Factors Affecting Greater Time in Foster Care for African American Children --Difficulties in Achieving Reunification: Figure 8: Strategies Intended to Improve Decisions by Reducing Bias: Figure 9: Strategies to Improve Access to Support Services: Figure 10: Strategies to Reduce Length of Stay in Foster Care: Figure 11: Children Entering Foster Care in California in 2005: Figure 12: Children in Foster Care in California in 2005: Figure 13: States' Views on Impact of Funding Policies on Disproportionality: Figure 14: States' Views on Impact of Adoption and Guardianship Policies on Disproportionality: Figure 15: Adoption Rates for All Children, 2001 through 2005: Figure 16: States' Views on Impact of Federal Policies on Licensing and Time Frames for Making Permanency Decisions on Disproportionality: Abbreviations: ACF: Administration for Children and Families: ASFA: Adoption and Safe Families Act: CFSR: Child and Family Services Reviews: CWLA: Child Welfare League of America: HHS: Health and Human Services: IEP: Interethnic Placement Act: MEPA: Multiethnic Placement Act: NRC: national resource center: OJJDP: Office of Juvenile Justice and Delinquency Prevention: SSBG: Social Security Block Grant: TANF: Temporary Assistance for Needy Families: United States Government Accountability Office: Washington, DC 20548: July 11, 2007: The Honorable Charles B. Rangel: Chairman: Committee on Ways and Means: House of Representatives: Children of all races are equally as likely to suffer from abuse and neglect, according to the Department of Health and Human Services' (HHS) National Incidence Study of Child Abuse and Neglect (NIS); however, HHS data show that a significantly greater proportion of African American children enter and remain in foster care than children of other races and ethnicities. African-American children across the nation were more than twice as likely to enter foster care compared with White children in 2004, and African American children remained in foster care about 9 months longer. On the last day of fiscal year 2004, African American children totaled 162,911--or 34 percent--of the 482,541 children in foster care, according to HHS data--about twice their proportions in the general child population. Although there is great variability among and within the states, data from nearly all states show some disproportionate representation of African American children in foster care. State data also show patterns of disproportionate representation in foster care for Native American children and, in certain localities, Hispanics and Asian subgroups are also disproportionately represented to some extent.[Footnote 1] About 60 percent of children who enter foster care do so through reports of child abuse or neglect provided to a state's child welfare system.[Footnote 2] Child welfare staff screen and investigate reports of child maltreatment, and make decisions about whether a child can remain safely at home, with or without family support services, or must be immediately removed and placed in foster care. The decision to place a child in foster care is subsequently presented before a judge who evaluates the evidence for removal from home and either corroborates or overturns the decision. After entering a child in foster care, child welfare staff develop case plans that are approved by the courts outlining steps parents must take before a child can return home or, for children the courts decide cannot safely be returned home, establishing other permanency goals for them, such as adoption or legal guardianship. States have the primary responsibility for establishing the legal and administrative structures and programs of their child welfare services; however, federal legislation and regulations establish a framework within which states make their programmatic and fiscal decisions. The Department of Health and Human Services (HHS) is the principal federal agency that provides federal oversight of states' child welfare systems. HHS administers about $8 billion in funds each year that are dedicated to support states' child welfare systems. HHS also administers social services block grant programs, such as Temporary Assistance for Needy Families (TANF), and states generally spend about $12 billion of these funds to provide direct social services. States use these block grant funds to benefit various populations, including child welfare families. Concerned about why African American children are overrepresented in foster care, you asked us to analyze: (1) The major factors that have been identified as influencing the proportion of African American children entering and remaining in foster care compared to children of other races and ethnicities; (2) The extent that states and localities have implemented strategies that appear promising in addressing African American representation in foster care; and: (3) The ways in which key federal child welfare policies[Footnote 3] may have influenced African American representation in foster care. To address these three objectives, we used multiple methodologies, including administering a state survey; conducting site visits; interviewing researchers and federal agency officials; conducting a literature review; and analyzing federal legislation and policies. Although we focused on African American children in this report, we also noted points of similarity or difference with children of other races and ethnicities as appropriate.[Footnote 4] Specifically, we conducted a nationwide Web-based survey of state child welfare administrators in 50 states and the District of Columbia between November 2006 and January 2007 and received responses from 48 states.[Footnote 5] In developing the survey, we relied upon a literature review to identify issues, such as factors that contribute to disproportionality, as well as interviews with child welfare researchers and others. To obtain a more in-depth understanding of issues, we conducted site visits to California, Illinois, Minnesota, New York, and North Carolina, where we interviewed state and local child welfare officials, juvenile court judges, and others involved in the child welfare systems. In addition we conducted telephone interviews with Texas state and local child welfare officials, service providers, and a judge. When viewed as a group, the states we visited reflected diversity in their rates of African American representation in foster care, strategies and initiatives used to address this disproportionality, program administration (state administered and county administered), and geographic location. In addition, the states we selected collectively covered nearly one-third of children in foster care across the nation. To extend our understanding, we interviewed child welfare researchers identified through our literature review and through recommendations from child welfare officials and stakeholders for their knowledge on issues of racial disproportionality in foster care. We also interviewed HHS officials responsible for foster care programs and related data, as well as federal officials at the Justice Department, Office of Juvenile Justice and Delinquency Prevention (OJJDP), which is required by law to address racial disproportionality in the juvenile justice systems. In addition, we conducted an extensive literature review of research on racial disproportionality in foster care and strategies used by states and others to address this issue. In reporting our findings, we drew upon research publications our methodologists considered generally reliable and methodologically sound. We analyzed federal child welfare legislation and policies relevant to foster care that our literature review and interviews had indicated might have an impact on racial disproportionality. Finally, we also analyzed HHS data on foster care and adoptive children that state child welfare agencies submit biannually to the agency under its foster care and adoption reporting system. We also confirmed the reliability of these data for our purposes. We conducted our work between June 2006 and June 2007 in accordance with generally accepted government auditing standards. Results in Brief: A higher rate of poverty and challenges in accessing support services, as well as racial bias and difficulties in finding appropriate permanent homes, were identified in our review as the main factors influencing the proportion of African American children in foster care. Thirty-three states in our survey cited high rates of poverty among African Americans as a factor influencing children's entry into foster care. Nationally, African Americans are nearly four times more likely than others to live in poverty. Studies have shown that families living in poverty have difficulty accessing needed services that can help support families and keep children who may be vulnerable to abuse and neglect safely at home. However, research suggests that poverty does not fully account for differing rates of entry into foster care. State child welfare directors we surveyed also responded that bias or cultural misunderstanding and distrust between child welfare decision makers and the families they serve also contribute to the removal of children from their homes. Once African American children are removed from their homes, their lengths of stay in foster care average 9 months longer than those of White children. The challenges in accessing services, such as substance abuse treatment and subsidized housing, also contributed to longer lengths of stay for children whose goal is to reunify with their families. For children who cannot be reunified with their families, state officials reported difficulties in finding them appropriate permanent homes, in part because of the challenges in recruiting adoptive parents, especially for youth who are older or have special needs. An additional factor is that African Americans are more likely to rely on relatives to provide foster care. Although this type of foster care placement, known as kinship care, can be less traumatic for children and reduce the number of placements and chance of their re- entry into foster care, it is also associated with longer lengths of stay. Most states in our survey reported implementing some strategies that experts have identified as promising for African American children and noted several factors they considered fundamental to any attempt to address racial disproportionality. Researchers and officials stressed that no single strategy would fully address the issue, but that strategies to increase access to support services, reduce bias, and increase the availability of permanent homes all hold some promise for reducing disproportionality. For example, 38 states reported collaborating with neighborhood-based organizations to expand the availability of support services. Most states sought to reduce bias by including the family in making key decisions and by recruiting and training staff with the skills to work with people of all ethnicities. To move children more quickly from foster care to permanent homes, more than half of states performed a diligent search for relatives of children in foster care who might be willing to provide permanent homes, recruited African American adoptive families, and offered subsidies to guardians who were not willing to adopt, as is currently allowed for adoptive families. However, fewer states reported focusing attention on disproportionality itself by, for example, enacting state legislation or establishing councils on racial disproportionality. Although research on the effectiveness of strategies has been limited, public and private officials in the forefront of research and implementation said that the ability to analyze data, work across social service agencies, and sustain leadership was fundamental to any attempt to address racial disproportionality. HHS has taken steps to help states in their efforts to address disproportionality through outreach and technical assistance. However, state child welfare directors generally reported in our survey that additional support in analyzing data on disproportionality and disseminating strategies is needed. Child welfare officials in states considered to be at the forefront of addressing disproportionality told us they relied on technical assistance in analyzing data from universities and funds from a private foundation to help them devise strategies to address disproportionality. According to our survey results, federal policies that provide for family support services and promote adoption were generally considered helpful in reducing the proportion of African Americans in foster care, but policies that limit the use of foster care funding for family support services and legal guardianship were reported to have a negative effect. Half of the state child welfare directors we surveyed reported that federal block grants used to provide services to families, such as substance abuse treatment, contribute to reducing the proportion of African American children in foster care. However, even more child welfare directors in our survey reported that policies governing the use of funds specifically intended for children in foster care increase the proportion of African American children in foster care. More specifically, many state child welfare directors expressed concerns about the cap on funds for preventive services and the lack of flexibility to use funds meant for foster care and adoption for other purposes, including services to families at-risk of having their children removed, such as parenting classes. Among policies that affect states' ability to find permanent homes for children, states generally reported that adoption policies have been helpful, such as the requirement to recruit minority adoptive parents and providing subsidies to families adopting children that states have identified as having special needs. However, states still face challenges in recruiting sufficient numbers of willing and qualified adoptive families for African American children. In addition, state and local officials also reported wanting federal support for legal guardianship. States responding to our survey considered the federal policy recognizing legal guardianship as helpful in enabling children to exit foster care, but policies limiting the use of federal funds to pay subsidies to guardians, similar to those provided to adoptive parents, as a barrier. States were less definitive about the impact of federal policies that impose time frames on permanency decisions. These time frames may shorten the time children remain in care but may also impede states' ability to reunify children with their parents. Our draft report recommended that HHS pursue specific measures to allow adoption assistance payments to be used for subsidizing legal guardianship. In commenting on the draft report, HHS disagreed, stating that the administration had already proposed an alternative funding approach, known as the Child Welfare Program Option. Under this proposal, states could choose to remain under the current foster care funding structure or instead receive a flexible capped grant that they could use for a wide range of child welfare services and supports, including subsidizing guardianships. The current adoption assistance program would remain the same under this proposed option. However, although HHS has presented this broad restructuring of child welfare funding in its budget proposals each year since 2004, no legislation has been offered to date to authorize it. Moreover, if enacted, it is unknown how many states would choose a capped grant that would allow greater program flexibility instead of the current title IV-E foster care entitlement funding. Therefore, in light of these factors, we have deleted our recommendation to HHS and are instead suggesting that Congress consider amending current law to allow subsidies for legal guardianships, as is currently allowed for adoption. Current evidence indicates that allowing such subsidies could help states increase the number of permanent homes available for African American and other children in foster care. We are also making a recommendation that the Secretary of the HHS provide states with additional technical assistance and tools to develop strategies to address disproportionality. In its comments, HHS noted that our recommendation was consistent with its efforts to provide technical assistance to states for addressing disproportionality, but the department did not address the specific actions we recommended. We continue to believe that it is important for HHS to take these actions to help states address this complex issue. Background: The HHS National Incidence Study has shown since the early 1980s that children of all races and ethnicities are equally likely to be abused or neglected; however, African American children, and to some extent other minority children, have been significantly more likely to be represented in foster care, according to HHS data and other research.[Footnote 6] Nationally, African American children made up less than 15 percent of the overall child population in the 2000 Census, but they represented 27 percent of the children who entered foster care during fiscal year 2004, and they represented 34 percent of the children remaining in foster care at the end of that year, as shown in figure 1.[Footnote 7] Figure 1: Proportion of Children by Race in Foster Care Settings, End of Fiscal Year 2004: [See PDF for image] Source: GAO analysis of AFCARS and Census data. [End of figure] African American children were more likely to be placed in foster care than White or Hispanic children, and at each decision point in the child welfare process the disproportionality of African American children grows. Although racial disproportionality is most severe and pervasive for African American children, Native American children also experience higher rates of representation in foster care than children of other races or ethnicities. Just over 2 percent of children in foster care at the end of fiscal year 2004 were Native Americans, while they represented less than 1 percent of children in the United States.[Footnote 8] However, there can be significant variation by state and county. It is especially important to understand local variations for Hispanic and Asian children, since they are underrepresented in foster care nationally and in most states, but are overrepresented in some counties and states. For example, in a single county in California, Hispanic children represented 30 percent of the population but 52 percent of the county's child welfare cases. There are various options for placing children in temporary and permanent homes through the child welfare system. Temporary options include foster care with relatives or non-relatives--whether licensed or unlicensed--and group residential settings. According to HHS, approximately one-fourth of the children in out-of-home care are living with relatives, and this proportion is higher for Hispanic and African American families. For permanent placements, adoption and guardianship are options under federal law in addition to the child's reunification with their parents. One important difference is that adoption entails terminating parental rights, while guardianship does not. (See table 1.) Table 1: Types of Temporary and Permanent Homes for Children in Child Welfare: Type of temporary placement: Foster parents; Definition: Non-related adults who have been trained and licensed/certified to provide shelter and care to a child. Type of temporary placement: Kinship Care[A] - Licensed Foster; Definition: Relative or close family friend who provides shelter and care. Licensed kinship care may involve a training and licensure process for the caregivers and support services. Type of temporary placement: Kinship Care - Unlicensed Foster; Definition: Relative or close family friend who provides shelter and care. Unlicensed kinship care may involve only an assessment process to ensure the safety and suitability of the home along with supportive services for the child and caregivers. Type of temporary placement: Congregate Care; Definition: These settings include community-based group homes, campus- style residential facilities, and secure facilities. Residential programs, and the staff who work in them, are generally focused on working with children who have certain special physical or behavioral needs. Type of permanent placement[B]: Reunification; Definition: Parents reassume their role as the principle caretaker for their children if the courts determine that parents have successfully completed the action required in their case plan. Type of permanent placement[B]: Adoption; Definition: Caretakers who assume legal guardianship of the child through the termination of parental rights. Type of permanent placement[B]: Guardianship; Definition: Caretakers who assume legal guardianship without the termination of a child's parent's rights. Legal guardianship is more durable than a simple transfer of custody to caretakers. Source: HHS, Child Welfare Information Gateway. [A] Kinship care exists both as formal arrangements made through the child welfare system and as informal arrangements made by families outside of child welfare. In this study, we are only referring to formal kinship care through child welfare, either as a temporary placement for a child or a permanent placement that allows the child to exit foster care. [B] Children can also exit foster care through emancipation when they turn 18 years old. This is sometimes referred to as "aging out" of the foster care system. [End of table] Federal Foster Care and Adoption Legislation: In the last decade, several federal laws have been enacted to help states reduce the number of children who enter and remain in foster care. These laws include the Multi-Ethnic Placement Act of 1994,[Footnote 9] as amended in 1996 by the Interethnic Adoption Provisions included in the Small Business Job Protection Act[Footnote 10] (MEPA-IEP). MEPA-IEP is intended to eliminate race-related barriers to adoption by prohibiting foster care and adoption agencies that receive federal funds from delaying or denying placement decisions on the basis of race, color, or national origin of either the adoptive or foster parent or child. MEPA-IEP also required states to diligently recruit potential foster and adoptive families that reflect the ethnic and racial diversity of children in the state who need foster care and adoptive homes. MEPA-IEP was followed by the Adoption and Safe Families Act of 1997 (ASFA),[Footnote 11] which established expedited time frames to place children in permanent homes through reunification, adoption, or guardianship and for terminating parental rights. ASFA recognized that guardianship may be an appropriate permanency option for some children in foster care and it encouraged adoption by establishing adoption incentive payments for states. The Promoting Safe and Stable Families (PSSF) program also created under ASFA supported the need to strengthen and reunify families. This program expanded dedicated funding for services that could help prevent the removal of children from their homes or could expedite children's return home from foster care. In addition to family preservation and community-based support services, PSSF services include time-limited reunification services and adoption promotion and support services. States are required to allocate "significant portions" of their funding for each of these four service categories. (See table 2.) Table 2: Key Federal Legislation Affecting Foster Care Programs: Multi-Ethnic Placement Act of 1994 (MEPA-IEP). Legislation: Prohibition against adoption placements on basis of race; Description: * Prohibited states and other entities that receive federal funding assistance from delaying or denying a child's foster care or adoption placement on the basis of the child or prospective parent's race, color, or national origin. Legislation: Recruiting foster and adoptive parents; Description: * Required that states diligently recruit foster and adoptive parents who reflect the racial and ethnic diversity of the foster care population for a state to remain eligible for federal assistance for child welfare programs. Adoption and Safe Families Act of 1997 (ASFA). Legislation: Expediting decisions about permanent homes for children; Description: * Required that states hold a permanency hearing no later than 12 months after the date a child enters foster care. This requirement was shortened from 18 months in prior law; * Required that states file a petition to terminate parental rights for children who have been in foster care for 15 of the most recent 22 months. States may exempt children from this requirement for multiple reasons, including if the child is placed with a relative. Legislation: Providing incentives to states for increasing adoptions; Description: * Provided financial rewards to states for increasing numbers of finalized adoptions through the adoption incentive payment program. States have the flexibility to use the incentive payment funds for any child welfare related initiative; * Required that in order to receive incentive payments, states must exceed adoption baselines established for their state. States receive a fixed payment of $4,000 for each foster child who is adopted over the baseline, an extra $2,000 for the adoption of each special needs child younger than age 9, and $4,000 for the adoption of each child aged 9 or older. Source: GAO analysis of federal legislation. [End of table] One of the prerequisites to finding children temporary and permanent homes is for states to ensure that criminal background checks have been conducted for prospective foster care and adoptive parents. Congress had prohibited states from receiving federal foster care or adoption assistance support on behalf of eligible children who are placed in the home of a foster or adoptive parent who had certain types of convictions.[Footnote 12] States had been allowed to opt out of certain federal criminal background requirements by providing alternative plans to ensure children's safety, which were assessed as part of an HHS review process. According to HHS, eight states had been approved to use alternative plans.[Footnote 13] However, requirements for conducting federal background checks have recently changed: a provision of the Adam Walsh Safety and Protection Act of 2006,[Footnote 14] which was developed in response to concerns about child predators, establishes additional federal requirements for criminal background checks of prospective foster or adoptive parents[Footnote 15] and eliminates states' ability to opt out of the federal requirements, effective October 2008. Foster Care Financing: Federal funds account for approximately half of states' total reported spending for child welfare services, with the rest of funding coming from states and localities. In fiscal year 2004, total federal spending on child welfare was estimated to be $11.7 billion based on analysis of data from over 40 states.[Footnote 16] These federal funds come from sources that are dedicated to child welfare as well as those that are provided to states under the federal block grant structure for broader purposes. Titles IV-E and IV-B of the Social Security Act are the principal sources of federal funds dedicated for child welfare activities. Title IV-E provides the majority of dedicated federal funds for support payments to foster families, adoption assistance, and related administrative costs on behalf of children who meet certain federal eligibility criteria.[Footnote 17] Title IV-E foster care maintenance and adoption assistance payments are authorized as open-ended entitlements. States may claim federal reimbursement for a specified amount of the costs for every eligible child who is placed in a licensed foster home. In addition, Title IV-E established subsidies paid to families who provide adoptive homes to children who states identify as having special needs that make placement difficult.[Footnote 18] In 2003, 2004, and 2005, states designated more than 80 percent of adoptions as special needs adoptions enabling families to receive federal financial subsidies, according to HHS data. Total federal expenditures, including administrative costs, for Title IV-E programs were about $6.8 billion in fiscal year 2006. Title IV-B authorizes funds to states for broad child welfare purposes, including child protection, family preservation, and adoption services. In contrast to Title IV-E funds, Title IV-B funds are appropriated annually and totaled about $700 million in 2006.[Footnote 19] Federal block grants such as the Temporary Assistance for Needy Families (TANF) and the Social Services Block Grant (SSBG) provide additional sources of funds that states can use for child welfare purposes. Block grants and other sources of non-dedicated funds made up about 44 percent of total federal funds spent on child welfare in 2004, or about $5.2 billion, according to the most recent research available.[Footnote 20] Under these block grants, states have discretion to provide direct social services for various populations, including child welfare families, the elderly, and people with disabilities. (See fig. 2.) Figure 2: Federal Child Welfare Funding by Funding Source, FY 2004: [See PDF for image] Source: Urban Institute 2005 Child Welfare Survey reported in May 2006. Note: This funding analysis is the most recent available that shows federal funding used specifically for child welfare. Total federal spending on child welfare was estimated to be $11.7 billion in state fiscal year 2004. Percentages may not total to 100 due to rounding. [End of figure] In 1994, the Congress authorized the use of flexible funding demonstration waivers to encourage innovative and effective child welfare practices. These waivers, typically authorized for 5 years, allowed states to use Title IV-E funds to provide services and supports other than foster care maintenance payments. For example, states could use waivers to provide subsidies to legal guardians or services to caregivers with substance abuse problems. Waiver demonstrations must remain cost-neutral to the federal government and they must undergo rigorous program evaluation to determine their effectiveness. As of May 2007, according to HHS, 14 states have one or more approved Title IV-E child welfare waiver demonstration projects, involving one or more programmatic components, such as subsidized guardianship. With regard to guardianship specifically, four states have completed demonstrations that involved subsidized guardianships as of May 2007, seven states have active guardianship demonstrations, and one state has not yet implemented its guardianship demonstration. HHS's ability to approve new Title IV-E waivers expired in 2006, however, and Congress has not reauthorized this program. The Administration's fiscal year 2008 budget proposes, for the fifth consecutive year, to implement a "Child Welfare Program Option," which would restructure the Title IV-E foster care program. Under this proposal, states could forego open-ended entitlement foster care funding in exchange for a pre-determined grant. Unlike the open-ended funds, the grant could be spent on the entire range of child welfare purposes and for any child (regardless of the child's federal foster care eligibility status). States taking this option would need to continue to ensure child safety protections, maintain existing state funding for child welfare, and participate in federal assessments of state child welfare programs, known as Child and Family Services Reviews. Under this proposal, the Title IV-E Adoption Assistance program would continue as an entitlement program, according to an HHS official. In 2006, HHS approved two states to pilot the program option over the next 5 years under its Title IV-E waiver demonstration authority, which expired in that year.[Footnote 21] Final evaluation results for these pilots will not be available for at least 5 years. HHS Assessments of State Programs and Technical Assistance to States: States are required to enact policies and meet certain federal standards related to child welfare programs, and HHS evaluates how well state child welfare systems achieve these federal standards through its Child and Family Services Reviews (CFSR). Implemented in 2001, these reviews focus on states' performance in ensuring children's safety, permanency, and well-being over a range of child welfare services, using various outcome measures.[Footnote 22] To address any areas identified as not in substantial conformity with these outcome measures, the state develops a program improvement plan.[Footnote 23] To evaluate states' performance on these measures, HHS also relies, in part, on its Adoption and Foster Care Analysis and Reporting System (AFCARS) to capture, report, and analyze information collected by the states.[Footnote 24] In addition, AFCARS is used to generate annual reports on foster care and adoption programs nationwide. HHS provides states with training and technical assistance to help them develop and implement their CFSR performance improvement plans, build state agency capacity, and improve the state child welfare system. Technical assistance providers in this network include HHS's Children's Bureau, 10 regional offices, and various National Resource Centers. An additional resource is the department's Child Welfare Information Gateway, a Web site that provides access to information and resources to help protect children and strengthen families.[Footnote 25] States Report Poverty and Difficulty in Finding Permanent Homes Are among Major Factors Influencing African Americans' Entry and Length of Stay: A complex set of interrelated factors influence the disproportionate number of African American children who enter foster care as well as their longer lengths of stay, and our review found that poverty and the lack of appropriate homes are particularly influential. Major factors affecting children's entry into foster care included African American families' higher rates of poverty, families' difficulties in accessing support services so that they can provide a safe home for vulnerable children and prevent their removal, and racial bias and cultural misunderstanding among child welfare decision makers. Factors often cited as affecting African American children's length of stay in foster care included the lack of appropriate adoptive homes for children, greater use of kinship care among African Americans, and parents' lack of access to supportive services needed for reunification with their children. Higher Rates of Poverty, Lack of Support Services, and Racial Bias Viewed as Increasing African American Children's Entry into Foster Care: In responding to our survey, states considered three main groups of factors as contributing to African American children's entry into foster care: These groups included high rates of poverty and other poverty-related issues, challenges in accessing supports and services in impoverished communities, and racial bias or cultural misunderstanding among decision makers. Higher Rates of Poverty: Of the many factors that have been found to influence African American children's disproportionate entry into foster care, the most often cited factors that emerged in our survey were African American families' higher rates of poverty and issues related to living in poverty. Poverty-related factors included the large number of single parents among African American households, a high rate of substance abuse, and greater contact with public officials who have mandatory responsibilities to report incidents of abuse and neglect.[Footnote 26] (See fig. 3.) Figure 3: State Views on Poverty-Related Factors Affecting Higher Entry of African American Children to Foster Care: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] In our survey, poverty was cited as a key factor for entry: 33 states reported that high rates of poverty in African American communities may increase the disproportionate number of African American children entering foster care compared to children of other races and ethnicities. Researchers and child welfare officials in states we visited also noted the importance of poverty as a contributing factor. Across the nation, an estimated 23 percent of all African American families lived below the poverty level compared to only 6 percent of Whites, making African Americans nearly four times more likely to live in poverty, according to U.S. Census data.[Footnote 27] Since foster care programs primarily serve children from low-income families, this could account for some of the disproportionate number of African American children in the foster care system. However, our review of our survey results, interviews, and studies indicate that factors unrelated to poverty are also at play in foster care placements. In addition, child welfare directors in 25 states reported that the greater number of African American single-parent households was also a factor contributing to African American children's entry into foster care to at least a moderate extent.[Footnote 28] According to the most recent National Incidence Study, children of single parents had a 77 to 87 percent greater risk of harm than children from two-parent families.[Footnote 29] Across the nation, 35 percent of African American family households were headed by single females with children under 18 years of age compared to 9 percent for Whites and 19 percent for Hispanics, according to U.S. Census data.[Footnote 30] In addition, nearly half of the child welfare directors responding to our survey considered higher rates of substance abuse in African American households as contributing to the proportion of African American children in foster care. (See fig. 3.) Despite this perception, national data show that African Americans have nearly the same rate of substance abuse as Whites.[Footnote 31] However, 65 percent of African American children were removed from their homes because of parental substance abuse and placed in foster care, compared to 58 percent of White children, according to our analysis of AFCARS fiscal year 2004 data.[Footnote 32] Finally, child welfare directors in 14 states responded that African American children's greater contact with officials mandated to report child abuse and neglect played a role in the children's entry to foster care. Several researchers we interviewed noted that low-income families come into contact with a greater number of mandated reporters because they have more interaction with some public services. In fact, as noted in an HHS report, the top three sources of reports to child protective services hotlines in 2003 were educational staff, law enforcement officials, and social services personnel, of which the latter two disproportionately interact with low- income individuals. Challenges in Accessing Supports and Services: In our survey, African American families' challenges in accessing supports and social services was also viewed as influencing African American children's entry to foster care. African American and other families living in impoverished neighborhoods often do not have access to the kinds of supports and services that can prevent problems in the home from leading to abuse or neglect. Such supports and services include affordable and adequate housing, substance abuse treatment, and family services such as parenting skills and counseling. Access to legal representation in courts responsible for making decisions about children reported to have been abused or neglected was also as a factor as influencing African American children's entry into foster care. (See fig. 4.) Figure 4: State Views on Support and Preventive Services Affecting African American Children's Higher Entry into Foster Care: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] According to our survey, child welfare directors in 25 states reported that the lack of affordable housing options was a factor that may increase the proportion of African American children entering foster care to at least a moderate extent. For low-income families, affordable public housing is a critical support that can help families stay together and allow for in-home services, thereby decreasing the chances of children being removed from their families, but in some areas, there is a shortage. For example, child welfare officials in a California county told us they have a waiting list estimated at around 20,000 applicants for public housing.[Footnote 33] Families involved in the child welfare system often live in communities that lack resources and services, including drug treatment services and job training, which they either do not receive or must travel long distances to obtain, according to an HHS study.[Footnote 34] In our survey, 25 states reported that the lack of substance abuse treatment and 24 states reported that the limited access to preventive services were factors that may increase the proportion of African American children entering foster care to at least a moderate extent. The state survey responses may reflect the fact that a higher percentage of African American families live in impoverished neighborhoods that lack such resources. Child welfare officials in all of the states we visited as well as researchers noted that lack of adequate supportive services contributed to disproportionality. For example, during a site visit, a Minnesota child welfare official noted that wealthier families may be able to draw upon support services, like family and substance abuse counseling, that can help keep the children with their families. However, poorer families, without access to supportive services, may have a more difficult time weathering a problem such as substance abuse or emotional issues.[Footnote 35] Even after they are reported to child welfare, families can have difficulty in gaining access to the types of services that would allow a child to remain with the family and risk being removed to foster care. According to HHS's 2005 Child Maltreatment report, about 40 percent of children identified as victims of maltreatment do not receive services such as counseling and family support services.[Footnote 36] With regard to substance abuse treatment services, one study found significant gaps in services for families involved with the child welfare system, with only 31 percent of at-risk children and families with substance abuse problems receiving treatment.[Footnote 37] There is also some evidence that African American families, in particular, are not offered the same amount of support services when they are brought to the attention of the child welfare system. As one study found, race was a significant factor in whether families received mental health related services, even after controlling for age, type of maltreatment, behavior of the child, and gender.[Footnote 38] A Texas state child welfare official reiterated this point, telling us that in her experience, African American children are less likely than children of other races or ethnicities to receive in-home services. According to 20 states responding to our survey, once African American families come into contact with the child welfare system, they often have difficulties obtaining adequate legal representation in court, and this contributed to their disproportionate numbers in foster care. Local court officials and others we interviewed observed that higher income families can afford private legal representation, which can help prevent their child's removal to foster care, but lower-income parents usually do not have this option. In one state we visited, we were told that public attorneys assigned to child welfare families often do not meet parents before they appear in court and have little time to review case files, putting parents at a disadvantage in unfamiliar legal settings. In addition, the Pew Commission on Children in Foster Care found that parents in dependency hearings were often inadequately represented because of a lack of time, preparation, and resources, including attorney compensation.[Footnote 39] Distrust and Racial Bias or Cultural Misunderstanding: Coupled with African American parents' greater distrust of the child welfare system, racial bias or cultural misunderstanding among decision makers also emerged in our survey as major factors contributing to the disproportionate number of African American children entering foster care. These decision makers include mandated reporters, child welfare caseworkers, and those involved in judicial rulings about these children. (See fig. 5.) Figure 5: State Views on Issues of Distrust and Bias or Cultural Misunderstanding Affecting African American Children's Greater Entry to Foster Care: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] Families' distrust of the child welfare system was cited by child welfare directors in 28 states as a factor contributing to the entry of African American children into foster care to at least a moderate extent. According to state child welfare officials and some researchers we interviewed, African American families' distrust of the child welfare system stems from their perception that the system is unresponsive to their needs and racially biased against them. Child welfare officials and researchers said that many African Americans in poor communities perceive child welfare caseworkers as more intent on separating African American parents from their children than on working within their communities to address child safety issues. As an example of how this dynamic might occur, a neighborhood-based service provider we interviewed in California described a situation in which a mother fleeing domestic violence at home did not seek public services for herself and her children--despite the evident need for clothing and therapy--because child welfare had once before removed her children and she did not trust the system to be helpful. These experiences in turn can shape the families' dynamics in their initial contacts with mandated reporters, caseworkers, and judges. Casey Family Programs[Footnote 40] staff in an interview noted that African American families in such circumstances may not seek services because of such distrust, which in turn increases the risk of a child's removal. In our survey, 23 state child welfare directors reported that they considered racial bias or cultural misunderstanding on the part of those reporting abuse or neglect, such as teachers, medical professionals, or police officers, as a factor in the disproportionate representation of African American children entering foster care to at least a moderate extent. In support of this view, some studies have found that medical professionals are more likely to report low-income or minority children to child protective services, even controlling for other factors, such as type of abuse.[Footnote 41] In addition, bias or cultural misunderstanding on the part of child welfare caseworkers and juvenile and family court judges are viewed as playing a role in the proportion of African American children entering foster care. In our survey, child welfare directors in 21 states reported that caseworker bias, cultural misunderstanding, or inadequate training was a factor that contributes to entry. To a lesser extent, bias or cultural misunderstanding was considered a factor in judicial rulings as well. HHS and a few state child welfare officials we interviewed also noted that class and educational differences between caseworkers and families also contributed to cultural misunderstandings. As one researcher noted, even well-meaning decision makers at any stage of the child welfare process may have faulty assumptions about racial, ethnic, or socio-economic groups. Studies that have tried to control for other factors to determine if race or racial bias was a predictor for entry into foster care have produced varied results. One study using California data found that, after controlling for poverty and maltreatment, African American children were more likely to be removed from their homes and placed in foster care compared to White children, when income was accounted for.[Footnote 42] However, another study using Baltimore, Maryland, data found African American children did not have an increased likelihood of being removed from their homes and placed in foster care.[Footnote 43] Although research on racial bias or race as a predictor for entry into foster care is not always consistent, a recent review of the current literature by the Casey-Center for the Study of Social Policy Alliance for Racial Equity in the Child Welfare System concluded that race is an important factor that affects the decision to place children into foster care.[Footnote 44] Difficulties in Finding Permanent Homes and Achieving Reunification May Increase Time in Foster Care for African American Children: In responding to our survey, states considered certain groups of factors as contributing to African American children's length of stay in foster care, thereby increasing their disproportionality: These included challenges in finding appropriate adoptive homes for those unable to be reunified with their families, the impact of kinship foster care on length of stay, and other challenges affecting children's ability to exit foster care to be reunified with their families. Challenges in Finding Appropriate Adoptive Homes: Certain factors made finding permanent homes for African American children more challenging, according to states responding to our survey, thereby contributing to longer lengths of stay for African American children. These factors included an insufficient number of appropriate adoptive homes, difficulties in finding families that will adopt older African American children, and the belief that African American children are more likely to be diagnosed as having special needs. (See fig. 6.) Figure 6: State Views on Factors Affecting Longer Time in Foster Care for African American Children --Difficulty in Finding Permanent Homes: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] State officials from 29 states cited the insufficient number of appropriate adoptive homes as a factor that may increase African American children's length of stay in foster care to at least a moderate extent. For African American children, lengths of stays in foster care averaged 9 months longer compared to White children in 2004, according to our analysis of AFCARS data. This is partly due to the fact that African American children constituted nearly half of the children legally available for adoption in 2004 and waited significantly longer than other children for an adoptive placement. State officials we interviewed described challenges in recruiting appropriate adoptive families for African American children. These challenges include the difficulty many states have in recruiting adoptive families of the same race and ethnicity of the children waiting for adoption and the unwillingness of some families to adopt a child of another race. An additional challenge was finding adoptive African American families who are able to meet state licensing requirements, including housing and background checks, for an appropriate adoptive home. In New York, for example, local officials explained that state requirements for a certain number of bedrooms can prevent poor African American families from being able to meet licensing requirements needed for adoption--this can be especially an issue in high-cost urban areas in which there is limited affordable housing. In three states we visited, child welfare officials also told us that African American families who were interested in adopting were sometimes prevented from doing so because a member of the household had a prior criminal record, even though child welfare officials had determined that the person would not be a risk to the child.[Footnote 45] The age of foster children awaiting adoption also contributes to the challenges in finding appropriate adoptive families, with greater difficulties in placing older children. According to research, prospective adoptive parents are more inclined to adopt younger children, and older children may also have less interest in being adopted. In our survey, 21 states reported that a factor accounting for the longer lengths of stay for African American children waiting to be adopted was that many of them were also older.[Footnote 46] According to a 2003 study comparing a cohort of children whose parents' rights were terminated at the same time, children who were both older and African American had longer wait times between the termination of parental rights and adoption.[Footnote 47] State officials we visited echoed this finding, noting that child welfare agencies have a difficult time trying to find adoptive homes, particularly for older African American children. In our survey, 16 states also considered the greater likelihood of African American children being diagnosed as having medical and other special needs as a factor affecting African American's length of stay to at least a moderate extent. According to HHS data, African American children in foster care in 2004 were only slightly more likely to have been diagnosed as having medical conditions or other disabilities (28 percent) than White children in foster care (26 percent). Children with special needs may require additional support services, and some African American families may have less access to support services that would enable them to take on this extra responsibility. The impact on African American children is supported by HHS adoption data that shows that 23 percent of African American children who were adopted out of foster care had a medical condition or disability, compared to 31 percent of White children in the same category. Kinship Care: African American children are more likely than White and Asian children to enter into the care of relatives.[Footnote 48] Although kinship care is associated with longer lengths of stay, child welfare researchers and officials we interviewed consider these placements to be positive options for African American children because they are less stressful to the child and maintain familial ties. In addition, some researchers associate the use of kinship care with fewer foster care placements and lower rates of re-entry.[Footnote 49] Kinship care also has some drawbacks. For example, a 1999 GAO study found that kinship care might increase a child's risk of harm because caregivers may be unwilling to enforce court-ordered restrictions on parental visits.[Footnote 50] In responding to our survey, 18 states reported that the use of kinship care was a factor contributing to longer lengths of stay in foster care for African American children to at least a moderate extent. This view is supported by research findings indicating that children living with relatives generally spend more time in foster care than children living with non-relatives. For example, a study cited in a HHS report showed that 42 percent of children in foster kinship care remain there for more than 2 years compared to 36 percent of children in non-kin foster care.[Footnote 51] Moving a child from kinship foster care to adoption can be difficult for caregivers who need financial assistance or wish to retain family ties. Several child welfare officials said that there is a financial disincentive to adopt children who are only eligible for financial subsidies and services while they are in foster care, especially for grandparents and others living on a fixed income. Even when states offer financial subsidies to help families adopt these children, relatives may be reluctant to terminate their relatives' rights. An alternative is for these caregivers to provide a permanent home for their relative's children through legal guardianship in which caregivers are afforded legal decision-making authority over the child without terminating the birth parent's rights.[Footnote 52] Difficulty in Achieving Reunification: Some of the same factors that states view as contributing to African American children's entry also contribute to their difficulties in exiting foster care and being reunified with their families. In our survey, nearly half of the states considered the lack of affordable housing, distrust of the child welfare system, and lack of substance abuse treatment as factors contributing to African American children's longer lengths of stay. The lack of such supports and other services in many poor African American neighborhoods contributes to children's longer stays in foster care because services can influence a parent's ability to reunify with their child in a timely manner, according to our survey, interviews, and research. (See fig. 7.) Figure 7: State Views on Factors Affecting Greater Time in Foster Care for African American Children --Difficulties in Achieving Reunification: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] In our survey, nearly half of the states reported a lack of affordable housing options for African American parents, and state and county child welfare officials said that housing issues often delay family reunification, resulting in longer lengths of stay in foster care. According to child welfare officials and researchers we interviewed, poor families can lose their housing once their children have been removed because the TANF program requires children to be living with caregivers for them to qualify for TANF child-only benefits. For example, a county official in California noted that about 70 percent of families in that county experience a housing crisis when their children are removed. If families cannot afford to remain in their homes without TANF benefits, then they must seek other alternatives to create homes suitable for reunification with their children. Furthermore, if families do maintain their housing or find other housing they can afford, the standards that parents must meet before their children can be returned home from foster care are often higher than when the children were removed. According to a private foundation that assists 13 state and local child welfare agencies around the country in addressing disproportionality, a parent living in poverty might be unable to meet housing requirements needed for reunification, such as having a bedroom for each child, even though the appropriateness of the parent's housing had not been the original basis for a child's removal. According to 25 states in our survey, parents' distrust of the child welfare system was also a factor contributing to African American children's longer length of stay in foster care to at least a moderate extent. We were told that African American families in some communities do not trust child welfare agencies because families in their communities have had adversarial relationships with various public organizations, including schools, public health, and criminal justice systems. The lack of substance abuse treatment, mental health services, and other support services in African American communities are additional factors that can slow African American children's reunification with their parents, thereby contributing to longer stays in foster care. According to our survey, 23 state child welfare administrators reported the lack of substance abuse treatment services as factors contributing to African American children's longer stays in foster care to at least a moderate extent. An HHS study found that state officials lack the resources to provide substance abuse and other types of treatment services sufficient to help African American families and those of other racial and ethnic minorities move toward reunification and adoption.[Footnote 53] Court officials in California said that initiatives to refer drug offenders to treatment programs instead of incarceration have increased competition for accessing publicly funded substance abuse programs, adding to the difficulties families may face in making changes needed for reunification. In addition, when services are available, it may take 2 years for a parent to complete a substance abuse treatment program, and entry into such programs may be delayed if there are waiting lists for services. States Implemented a Range of Strategies Considering Promising for Addressing Disproportionality, but Fewer States Specifically Focus Attention on Issue: Most states we surveyed reported implementing a range of child welfare strategies-often good practices generally-that researchers and experts believe may also be promising for reducing the number of African American children in foster care. These strategies are intended to reduce bias in decision making and increase access to supportive services for families and the availability of permanent homes for foster children. Fewer states reported focusing attention on disproportionality through such actions as convening task forces or passing state legislation to study the issue. States that did more directly address disproportionality agreed that certain key elements were central to their efforts: these elements were data analysis to identify problems and strategies to address them, leadership to sustain change across time, and collaboration with different social services agencies to access programs and resources needed outside the child welfare system. HHS has provided some support to states for reducing the proportion of African American children in foster care through conferences, workshops and various Web sites, but states reported that they would benefit from having additional guidance in analyzing information and from the dissemination of strategies that other states have found promising. States Implement a Range of Strategies Expected to Have an Impact on Disproportionality: Researchers and child welfare administrators stressed that no single strategy was sufficient to fully address disproportionality. Some strategies states reported on have the potential to reduce bias or improve decision making. Other strategies are intended to improve access to support services for parents, and still others could increase the availability of permanent homes for children waiting in foster care (see table 3). Table 3: Strategies Used by States That Address Factors Contributing to Disproportionality: Factor: Bias in decision making; Type of Strategy States Are Using: Involving family in planning decisions about children. Factor: Bias in decision making; Type of Strategy States Are Using: Training caseworkers to strengthen their ability to work across cultures. Factor: Bias in decision making; Type of Strategy States Are Using: Conducting outreach or education to mandated reporters on criteria or standards for reporting abuse and neglect. Factor: Bias in decision making; Type of Strategy States Are Using: Recruiting, retaining, and promoting culturally competent staff. Factor: Bias in decision making; Type of Strategy States Are Using: Using risk assessment tools that are considered culturally competent or validated. Factor: Lack of access to support services; Type of Strategy States Are Using: Collaborating with neighborhood- based organizations to provide services. Factor: Lack of access to support services; Type of Strategy States Are Using: Using interagency agreements with other social service agencies to improve families' access to services across programs. Factor: Lack of access to support services; Type of Strategy States Are Using: Providing supports for families judged to be a at lower risk of abusing or neglecting their children instead of removing them from their families through an approach known as Alternative, Dual or Differential Response. Factor: Challenges in finding permanent homes; Type of Strategy States Are Using: Searching for fathers or paternal kin of foster children. Factor: Challenges in finding permanent homes; Type of Strategy States Are Using: Recruiting more African American adoptive families. Factor: Challenges in finding permanent homes; Type of Strategy States Are Using: Providing financial subsidies to guardians willing to permanently parent foster children. Factor: Lack of focused attention; Type of Strategy States Are Using: Establishing councils or advisory committees on disproportionality. Factor: Lack of focused attention; Type of Strategy States Are Using: Providing preventive services targeted at African American families. Factor: Lack of focused attention; Type of Strategy States Are Using: Establishing requirements in contracts to address disproportionality. Source: GAO survey of state child welfare administrators. [End of table] States' Strategies to Address Racial Bias in Decision Making: To help mitigate the influence of racial and other forms of bias in child welfare decision making processes, states implemented a range of strategies such as including family members in discussions of placement options, providing training for case workers to strengthen their cultural competency, implementing tools to help caseworkers make more systematic decisions, and reaching out to educate mandated reporters about reporting requirements. (See fig. 8.) Among these strategies, states expected that including families in the decision making process and training culturally competent staff would most reduce disproportionality. Figure 8: Strategies Intended to Improve Decisions by Reducing Bias: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] All states we surveyed had implemented strategies to include families in the decision making process to some degree, and every state we visited told us they were using this method to help address disproportionality. There may be differences in the extent to which states involved families, ranging from occasional discussions with family members to more formal approaches of "family group conferencing," which follows a specific model of engaging family members in decisions about the child's placement through three phases including monitoring and follow up. This approach can help address caseworker bias, as one researcher explained, because it increases caseworkers' exposure to the lifestyles of the community they are serving and helps bridge misunderstandings. Some studies of this strategy show that it holds promise for African American families. According to an evaluation in Texas, family group decision making led to a reduction in foster care placements and an increase in placements with relatives for all children; these findings were especially pronounced for African American and Hispanic children. Specifically, 32 percent of African American children whose families attended such a conference returned home compared to 14 percent whose families received traditional services. Almost all states (45) reported conducting training for caseworkers to strengthen their understanding of different cultures, known as cultural competency training. Such training could include workshops on cultural differences to enable caseworkers to better interpret behaviors and interactions with their clients. Somewhat fewer states (36) reported using initiatives to recruit and retain culturally competent staff. To address bias among caseworkers, some of the states we visited required that their child welfare workers take an intensive program called "Undoing Racism."[Footnote 54] This program has participants analyze the ways in which structural racism may affect their decisions through dialogue, reflection, role-playing, and presentations.[Footnote 55] In addition, officials pointed out that, beyond cultural understanding, caseworkers need to understand the challenges of living with economic disadvantages so that they can work effectively with their clients. For example, one county agency in Iowa required its child welfare workers to spend 1 day using public transportation to get to social service appointments their clients must attend to better understand the time and transportation constraints some people face. Although most survey respondents expected cultural competency training to have an impact on disproportionality, there is little research linking cultural competency training programs to improved outcomes for African American children. However, one 3-year evaluation of a comprehensive cultural competency program in Washington state, which was initiated specifically to address the causes of disproportionality, found that families served by staff trained in this approach had a higher rate of children returning home than African American children in other areas. Because some state data have shown that disproportionality in foster care starts with the differing rates of reporting among races, most states (37) are also conducting educational outreach for those who work with children, such as teachers, health care providers, and social workers, who are required to report suspected abuse and neglect. These efforts may help ensure that mandated reporters are not inappropriately referring families to child welfare. Illinois conducted a widespread public awareness campaign for mandated reporters about how to identify abuse and neglect. Although some child welfare officials expect this outreach strategy to reduce disproportionality, none of the studies we examined assessed its effectiveness in that regard. Over half of states (29) reported using risk assessment tools, which can help caseworkers make more systematic decisions about a child's safety and the need to remove a child from the home. Caseworkers use these tools when investigating an allegation of child maltreatment to systematically collect information about a family and, based on this information, more objectively assess the level of risk in keeping the child at home. Without such tools, workers may err on the side of unnecessarily removing a child from its family, according to some child welfare officials. Two studies found particular risk assessment tools to be both race-neutral and more accurate predictors of future harm than caseworker judgment alone. A 2004 study of five counties in California found that minority groups often showed a lower risk than Whites when the assessment tool was applied, which means it could help to reduce the representation of some groups in the child welfare system, according to the author.[Footnote 56] However, some researchers express concern that other risk assessments that rely too heavily on information related to social conditions and poverty might actually contribute to racial bias. Despite the promising research about the value of specific risk assessment tools, only about one-third of the child welfare administrators surveyed who were using this strategy expected it to reduce disproportionality. States' Strategies to Improve Families' Access to Services: Child welfare agencies are taking action to improve access to services, such as providing or arranging for mental health treatment, medical care, and housing assistance for low-income people. The strategies states are using for this purpose include collaborating with neighborhood-based services, establishing interagency agreements to improve access to these services, and implementing an alternative approach to the removal of children--known as alternative, dual, or differential response. Just over half of the states who used each strategy reported in the survey that they expected it to reduce disproportionality. (See fig. 9.) Figure 9: Strategies to Improve Access to Support Services: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] Thirty-eight states reported using neighborhood-based support organizations to improve access to and use of support services. Neighborhood-based services can improve access to supports for parents because they are often more conveniently located to parents' homes and more likely to be staffed by people familiar with issues particular to their ethnic community. For example, child welfare officials in Los Angeles County told us that they went door to door in minority neighborhoods to find service providers beyond those with whom they have historically contracted. They subsequently heard that this collaboration helped to increase trust between the community and the child welfare system and increased use of these services. Similarly, one county in North Carolina convened a task force of schools, police, and community groups to examine and identify what support services were available to families. An HHS report synthesizing the views of child welfare workers in eight states noted that working with community-based services holds promise for reducing disproportionality because they are more accessible and provide services in a culturally appropriate context.[Footnote 57] Interagency agreements, used by 34 states, may improve families' ability to obtain services and supports they need from agencies outside of child welfare, which are primarily provided and funded through other state agencies. To address gaps in the provision of services like substance abuse treatment and financial supports,[Footnote 58] agencies can work with one another in any of the following ways: training staff jointly, sharing information and tracking systems, using common intake and assessment forms, coordinating case management, and placing staff from multiple agencies in the same office. Some child welfare officials told us they were reluctant to share information about overlapping clients because of federal privacy laws,[Footnote 59] while other local officials described methods they use to share information across systems. For example, the child welfare agency in San Francisco uses court agreements with other agencies, such as juvenile justice and mental health, to share information about families who are involved in multiple systems, and county officials report that these agreements enable them to better serve these families. Although we found no studies on the effectiveness of inter-agency agreements in reducing disproportionality, many child welfare officials expected that this strategy could have an impact. Another approach used by slightly more than half of states is differential or alternative response, which is a way for states to provide services to families when the risk of abuse and neglect is judged to be lower in lieu of removing children from their homes. Differential response can be used when maltreatment is not related to physical and sexual abuse, but instead to conditions of chronic poverty, chemical abuse, or domestic violence. For example, some California counties have three tracks for assessing families, depending on a family's situation. In the first track, if the case involves abuse and neglect and the risk is considered moderate to high for continued abuse, the caseworker may take action to remove the child with or without the family's consent, court orders may be involved, and criminal charges may be filed. In the second track, if the risk of continued abuse and neglect is lower, families work with representatives of county child welfare agencies to identify services for improving child and family well-being. In the third track, if an allegation is not considered abuse, the family is linked to services in the community through expanded partnerships with local organizations. Evaluations of alternative response in some states have shown this strategy to have promise for addressing the factors that may lead to the disproportionate number of African American children in care. Evaluations in Missouri and Minnesota[Footnote 60] found that use of alternative response increased cooperation between families and the child welfare agency. The Minnesota study also found that families who participated in the alternative response system received significantly greater access to support services and also a lower rate of new maltreatment reports than families in a control group. States' Strategies for Increasing the Availability of Permanent Homes: States are also devising strategies to increase the number of permanent homes for African American children who cannot be reunified with their parents so as to reduce the length of time they remain in foster care and increase the likelihood that they will be adopted. To do this, many states are increasing the search for fathers and paternal kin, making efforts to recruit more African American adoptive parents, and providing financial subsidies for caregivers (often relatives) who are willing to act as permanent guardians for foster children. (See fig. 10.) Figure 10: Strategies to Reduce Length of Stay in Foster Care: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] Almost all states surveyed reported that they take action to search for paternal kin when making decisions on where to place a child. Until recently, caseworkers did not routinely gather information on fathers, according to child welfare workers we spoke with. As foster care agencies have placed greater reliance on placing children with relatives, however, fathers and paternal relatives are increasingly being viewed as potential caregivers. Greater efforts to locate fathers and paternal kin are particularly relevant for African American families who are less likely to have a father living with the family at the time of their involvement with the child welfare agency. Officials we visited in Illinois, North Carolina, and New York told us that they had instituted changes so that searching for paternal kin was routine. One county in North Carolina requires social workers to use a structured protocol in contacting and gathering information from the father about family members as potential resources. This approach can allow fathers and other paternal relatives to take a much more active role in their child's life to prevent out-of-home placements. About two- thirds of states using this strategy expected it to reduce disproportionality, but there is relatively little research on the role of fathers in child welfare cases in general. Although 38 states reported that, to some extent, they are recruiting African American adoptive families, states still face challenges. States are required by law to diligently recruit foster and adoptive parents who reflect the racial and ethnic backgrounds of children. States have adopted various strategies, such as contracting with faith- based organizations and convening adoption support teams, to recruit greater numbers of African American adoptive parents. However, despite these efforts the overall number of African American children adopted by African American parents has not substantially increased in the past 8 years. In addition, HHS's 2001-2004 review found that only 21 of 52 states were sufficiently recruiting minority families, and a recent report found that the recruitment of minority families was one of the greatest challenges for nearly all states. Using subsidized guardianship as an alternative to adoption may hold particular promise for reducing disproportionality, and more than half of the states surveyed (30) reported using this strategy.[Footnote 61] African Americans are more likely than White children to be placed with relatives for foster care, and relative foster care is generally longer term.[Footnote 62] These relative caregivers are also more likely than non-relative foster parents to be low-income.[Footnote 63] They may be unwilling to adopt because they may find it difficult financially to forego foster care payments or because adoption entails terminating the parental rights of their kin. However, states can provide a way for foster children living with relatives to convert this to a more permanent arrangement by creating subsidized guardianship programs. These programs provide financial subsidies for foster parents (often relatives) who agree to become legally responsible for children, but are unable or willing to adopt. When Illinois and California implemented two of the largest of such programs, they subsequently saw an increase in permanent placements for all children.[Footnote 64] After instituting their subsidized guardianship programs, over 40 percent of children who were in long-term relative foster care in both states found permanency. In Illinois, this decrease also coincided with a reduction in disproportionate numbers of African American children in foster care.[Footnote 65] HHS officials also pointed out that these programs can be cost-neutral because the administrative costs associated with maintaining a child in foster care are no longer incurred with permanent legal guardianships. All seven states that used federal waivers to subsidize their guardianship programs with Title IV- E funds did so in a cost-neutral manner, as required by the waivers. Fewer States Implemented Strategies That Focused Attention Specifically on Disproportionality: Although many states we surveyed are employing the types of strategies that hold promise for reducing the proportion of African American children in foster care, fewer states were focusing attention specifically on disproportionality as a policy issue.[Footnote 66] Such strategies included establishing councils on disproportionality, requiring child welfare contractors to address disproportionality, and targeting preventive services to African American families.[Footnote 67] While these strategies may not necessarily be more effective than other strategies, they do represent a public acknowledgment of the issue and may be considered a starting point for further activity. States were much less likely to use these strategies compared with other strategies in our survey. A total of 15 states had established disproportionality councils or commissions that can provide leadership in addressing the issues. According to a report by the National Council of State Legislatures, the Illinois African-American Family Commission has the broad mandate to monitor legislation and programs, and assist in designing new programs on behalf of African American families, as well as facilitate the participation of African Americans in establishing community-based services. In addition, 13 states reported in our survey that they were targeting preventative services to African American families, and 11 states had some requirements for contracted agencies to address disproportionality. For example, child welfare officials in Kentucky reported that they were making a concerted effort to contract with service providers that can demonstrate their knowledge or understanding of the issue of disproportionality. Data, Leadership, and Working across Social Service Systems Are Key Elements to Address Disproportionality: Collecting, analyzing, and disseminating data were considered fundamental aspects of states' efforts to address disproportionality. These data can include not only disproportionality rates (as described in appendix II), but also information that identifies the extent to which disproportionality occurs, among ages, along the child welfare process, and geographically. For example, a researcher at the University of California at Berkeley has used state data to show that African American infants enter at a much higher rate than other children (see fig. 11). However, this disproportionality grows as children get older, because African American children are also less likely to exit foster care; and the foster care population for all ages then becomes disproportionally African American (see fig. 12). Child welfare officials in most of the states we visited used their data to show that as a child moved through the child welfare process from having a case reported, then investigated, then being removed from the home, disproportionality increased at each decision point. Lastly, researchers in Illinois learned that disproportionality rates were actually higher in the rest of the state than in Cook County, the main urban county containing Chicago and over half of the state's foster care population. Figure 11: Children Entering Foster Care in California in 2005: [See PDF for image] Source: Center for Social Services Research, University of California at Berkeley. [End of figure] Figure 12: Children in Foster Care in California in 2005: [See PDF for image] Source: Center for Social Services Research, University of California at Berkeley. [End of figure] Using data is considered crucial in identifying where disproportionality occurs in the child welfare process in order to devise strategies to most effectively address the issue. For example, when they analyzed state-level data, Texas officials realized that it was difficult to find foster care placements in close proximity to the birth family, making it difficult for African American children to be reunified with their families. To address this problem, Texas provided automated support for tracking the vacancies of foster homes and facilities. Data can also be useful for building consensus among community leaders, practitioners and policy-makers. Researchers in Illinois shared data on disproportionality with child welfare supervisors and caseworkers to increase their awareness that once an African American child is removed from the home, they are more likely to spend longer time in foster care. In Guilford County, North Carolina, child welfare officials shared data to show teachers, who are also mandated reporters, how disproportionality increases as a child moves from being referred, to investigated, to placed into foster care. Despite the importance of data collection, 18 states we surveyed reported that they were not regularly using data in their efforts to address disproportionality. In states we visited, child welfare officials also agreed that sustained leadership was fundamental to the process of identifying and addressing disproportionality. Members of the Child Welfare League of America's Cultural Competence and Racial Disproportionality and Disparity of Outcomes Committee told us that initiatives generally take root through the efforts of a person or organizations that champion the issue. All of the states we visited had some support from the Casey Foundation, and four states were involved with Casey's Breakthrough Series Collaborative, which focuses specifically on having child welfare professionals test new ideas and strategies to address racial disproportionality.[Footnote 68] Without such leadership, officials who have many competing priorities may be reluctant to tackle a politically sensitive issue. For example, child welfare workers in one county expressed concern that their efforts to address disproportionality would diminish when their Social Services Director - -who was highly committed to addressing disproportionality--retired. According to the National Conference of State Legislatures, six states have enacted state legislation to address disproportionality. These laws generally create commissions or task forces, require a study of the issue, or fund special projects to address disproportionality. For example, a Texas law required an analysis of data to determine whether child welfare enforcement actions were disproportionately initiated against any racial or ethnic group.[Footnote 69] In addition, some states included some discussion of African American disproportionality in their state child welfare plans. California, for example, pledged to meet the target of increasing the service provisions specifically for Native American and African American children. Finally, state child welfare officials, researchers, and other experts stressed the need to work across different social service systems because many of the factors that contribute to disproportionality lay outside the child welfare system. For example, one child welfare official we interviewed observed that efforts to address disproportionality in one system (e.g., child welfare) can be undone by lack of diligence in another (e.g., housing). Additionally, some state officials said that there was a need for collaboration among social service agencies, such as juvenile justice and education, because disproportionality in child welfare often results when families have not had their service and support needs met by other agencies. Although HHS Has Provided Assistance and Guidance, States Report That More Information and Technical Assistance Is Needed: HHS has made available technical assistance, guidance, and information to states on disproportionality at conferences, workshops and through various HHS Web sites. Since 2004, disproportionality and cultural competency have been discussed at training and technical assistance meetings attended by members of HHS's network of National Resource Centers, and since 2006 these issues have been a priority area for the network, according to HHS officials. Currently, HHS provides states and localities with information on disproportionality through various National Resource Center Web sites and the Children's Bureau Information Gateway Web site, such as links to literature examining various strategies and audio files of past teleconferences discussing disproportionality. HHS also provides guidance and information on promising approaches as well as technical assistance and training to improve states' efforts to find minority foster care and adoptive parents through its AdoptUsKids initiative and Web site. In 2003, HHS's Children's Bureau also published a study examining disproportionality[Footnote 70] and the Office of Planning, Research and Evaluation summarizes other published and unpublished research findings on disproportionality on its Web site. Although HHS does not require states to collect or report information on disproportionality, the agency has included state-based data on disproportionality in its Annual Child Welfare Outcomes Report to Congress. In addition, through an initiative known as the Culturally Competent Practice Knowledge Management Initiative, the agency is compiling an inventory of tools and best practices for addressing disproportionality. According to HHS officials, the agency plans to make this information available to consultants within its network of National Resource Centers to use in providing training and technical assistance to states and localities, but as of April 2007, HHS had not determined whether or how to make this information publicly available. As a whole, child welfare administrators we surveyed reported that in their view, their own states should be doing more to address disproportionality but added that having additional resources, including information on promising practices and technical assistance, would be useful in their efforts. Forty-two states reported that additional resources were needed to apply known strategies to reduce the disproportionality of African American children. All six states we visited were using funds from a private foundation, Casey Foundation, to support their initiatives. Similarly, 41 states reported that having information on best practices to address racial disproportionality would be at least moderately helpful to them. In responding to the survey, officials from one state noted that Casey Family Programs had developed helpful strategies to address the issue, and officials from another state noted that having a central federal repository to share information across states, including descriptions and evaluations of promising strategies, would help them more effectively address disproportionality. Twenty-five states also reported that receiving technical assistance from HHS in calculating disproportionality and tracking it over time would be useful.[Footnote 71] Some of these states volunteered through written comments that this additional assistance would be useful because state and local agencies have limited capacity to analyze or track disproportionality-related data. Nearly all of the states we visited had assistance from local universities or research institutes in analyzing data on disproportionality. California state child welfare officials told us that without the aid of a university researcher, they would not have the ability to help counties that lack the capacity to collect and analyze their data. At the time of our survey, eighteen states reported that they were not regularly analyzing or using data in their efforts to address disproportionality. Some child welfare officials and researchers we interviewed reported that the leadership and efforts made by the Department of Justice, Office of Juvenile Justice and Delinquency Prevention (OJJDP), to address the disproportionate representation of minorities in the juvenile justice system could serve as a model for child welfare. In response to similar issues with racial disproportionality, the Justice Department has overseen states' efforts in addressing disproportionality in the juvenile justice system, as mandated. To carry out its mission, OJJDP has established reporting requirements for states; provided guidance, technical assistance, and information on promising practices through a centralized location on the OJJDP Web site; and conducted regular conferences and training sessions for over a decade on the issue. According to a key official from OJJDP and a few state juvenile justice coordinators we interviewed, because of the legal mandate and federal funding provided over time, OJJDP's efforts have helped states implement strategies intended to reduce disproportionate minority contact in the juvenile justice system. As of 2005, nearly all eligible states and territories have devised plans to address disproportionality and regularly submit reports to OJJDP. States Reported That Some Current Federal Policies May Reduce the Disproportion of African American Children in Care, While Other Policies May Increase It: Federal policies that support services to families and adoption were generally considered helpful in reducing the proportion of African Americans in foster care, but policies that limit funds for prevention and legal guardianship were reported to have a negative effect, according to our review. Although it is difficult to isolate the effect of any one policy,[Footnote 72] many states reported that federal block grants that can be used to provide services to families help reduce disproportionality. At the same time, even more states reported that other policies constraining the use of federal child welfare funds work against this goal. States generally reported that policies promoting adoption-such as subsidies to families adopting children with special needs and the requirement to recruit minority adoptive parents-have been helpful, but wanted more support for legal guardianship. In particular, states considered the federal policy recognizing legal guardianship as helpful in enabling children to exit foster care, but policies limiting the use of federal funds to pay subsidies to guardians as a barrier. Federal policies that impose time frames on caseworkers for making permanency decisions may shorten the time children remain in care but may also impede states' ability to reunify children with their parents. Policies That Support Services to Families Were Considered Helpful, but Some Funding Policies Constrain States' Ability to Reduce Disproportionality: According to states we surveyed, having federal block grant funds available to provide services to families contributes to reducing the proportion of African American children in foster care compared to children of other races and ethnicities. However, policies that limit federal child welfare funds for preventive services or other purposes besides maintenance payments to foster care families are viewed as having a negative effect. (See fig. 13.) Figure 13: States' Views on Impact of Funding Policies on Disproportionality: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] In our survey, 23 states reported that the ability to use TANF block grant funds to provide parenting classes, substance abuse treatment programs, and payments to guardians who are relatives contributes to a reduction in the proportion of African American children in care in their states.[Footnote 73] Many officials and researchers we interviewed told us that having an adequate level of preventive services and family supports was particularly relevant for African American families living in poverty. However, as with all block grants, state officials determine the use of these funds and their program priorities. TANF funds used for child welfare in 2004 ranged from zero percent in eight states to 51 percent of Connecticut's total federal funds for such purposes, according to an Urban Institute report.[Footnote 74] Some officials from local child welfare agencies we interviewed also noted that because they did not have a steady source of funds for child welfare activities, it was difficult to plan for and provide preventive and family support services to these families. In a recent GAO report, states cited such services as the ones most in need of greater federal, state or local resources.[Footnote 75] Other policies constrain the amount of federal child welfare funds states can spend on services to support families, and states reported that these policies contributed to the disproportion of African American children in foster care. Of particular concern to 28 states were limitations on the use of federal funds under Title IV-B, which funds preventative and family support services. Under this part of the law, states are entitled to no more than their specified share of annual funding regardless of the number of families they serve in a year. These IV-B prevention funds can help divert children from foster care by providing services to their families and also help children exit foster care by providing supports to adoptive families and guardians. Yet the majority of federal funding for child welfare is distributed as payments for maintaining children already in foster care homes under another part of the law, Title IV-E. [Footnote 76] Twenty-five states we surveyed reported that the limited use of Title IV-E funds for other purposes besides making maintenance payments to foster care families, such as providing services to families, contribute to the proportion of African American children in care. According to California and Minnesota officials, because the majority of federal child welfare funds are used for foster care payments instead of preventive services, federal funding policies did not align with states' efforts to reduce the number of children entering foster care by serving at-risk children safely in their homes.[Footnote 77] Previous GAO work as well as other research has noted that federal child welfare funding favors reimbursement for foster care placements, while providing less support for services to prevent such placements.[Footnote 78] Every year since fiscal year 2004, the administration has proposed in its budget the creation of a Child Welfare Program Option under which states would have the option to receive federal foster care funds in the form of flexible grants, which they could use to fund a range of child welfare services and activities.[Footnote 79] This proposal has not been introduced as legislation. States Generally Viewed Federal Policies on Adoption as Helpful, but Other Policies Limit State Efforts to Find Alternate Placements or Reunify Families: In responding to our survey, states considered certain federal policies as helpful in reducing disproportionality, especially adoption policies and the recognition of guardianship. However, states viewed the lack of subsidies for guardianship and policies affecting the licensing of foster care and adoptive families as contributing to disproportionality. Views were mixed on federal policies that impose time frames for making permanency decisions. Adoption and Guardianship: Among federal policies that affect states' ability to find permanent homes for children, those that promote adoption were believed to reduce the proportion of African American children in foster care, according to our survey results. Although the recognition of guardianship as a placement option under federal law was also considered helpful, state and local officials reported that the lack of federal reimbursement for subsidies to guardians constrained their ability to place children in such arrangements. (See fig. 14.) Figure 14: States' Views on Impact of Adoption and Guardianship Policies on Disproportionality: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] In our survey, 22 states reported that the requirement to diligently recruit minority families contributes to a decrease in the proportion of African American children in care. According to officials from Illinois, New York, and North Carolina, the requirement to diligently recruit minority families has had a positive impact on moving African American children into permanent homes. For example, this requirement broadened the role and use of extended family as possible caregivers for children, according to an HHS survey of child welfare workers.[Footnote 80] State officials told us that it was a challenge to recruit a racially and ethnically diverse pool of potential foster and adoptive parents, as evidenced by the fact that more than half of states are not meeting HHS performance goals for recruitment.[Footnote 81] State officials noted the shortage of willing, appropriate, and qualified parents to adopt African American children, particularly older children, and researchers also cited a lack of resources among state and local agencies and federal guidance to implement new recruiting and training initiatives.[Footnote 82] Perhaps because of these challenges, 9 states in our survey reported that the policy requiring diligent recruitment had no effect on the proportion of African American children in care, and 15 states reported that they were unable to tell. Over the last 5 years, African American children as well as Native American children have consistently experienced lower rates of adoption than children of other races and ethnicities, according to HHS adoption data. (See fig. 15.) Figure 15: Adoption Rates for All Children, 2001 through 2005: [See PDF for image] Source: GAO analysis of HHS reported AFCARS data. Note: Adoption rates are calculated as total estimated count of adoptions with public agency involvement finalized during the year of interest divided by the number of children waiting to be adopted on the last day of the prior year. Children waiting to be adopted are defined by HHS as those children who have goal of adoption and/or have parental rights terminated, excluding children with termination of parental rights who are 16 and older and have the goal of emancipation. [End of figure] Providing adoption incentive payments to states generally helps reduce the proportion of African American children in care, according to our survey; however, these benefits may not be sustainable over time. In our survey, 20 states reported that these federal incentive payments provided to states for increasing adoptions contributes to reducing the proportion of African American children in care. A state official from Texas's child welfare agency told us that in 2005 the state received the highest adoption incentive payments among all states and that the number of African American children adopted has increased each year since 2004. However, states face challenges under this program because they must reach higher benchmarks each year to continue to earn adoption incentive payments. While the total number of adoptions nationally increased significantly in the late 1990s, since 2000 adoption rates have reached a plateau, according to HHS data and other research. Twenty states reported that the federal policy that provides subsidies to parents who adopt a child considered as having special needs contributes positively to reducing the proportion of African American children in foster care.[Footnote 83] Of African American children who were adopted from foster care in 2004 who states classified as having special needs, the child's race provided the basis for the classification in 20 percent of cases. In contrast, race was the basis for the classification of about 10 percent of Hispanic, Asian, and Native American adopted children who were determined to have special needs in that year. The federal policy encouraging race-neutral adoptions was believed to have less effect than other policies on the proportion of African American children in foster care.[Footnote 84] Intended to eliminate race-related barriers to adoption, MEPA-IEP prohibits foster care and adoption agencies that receive federal funds from delaying or denying placement decisions on the basis of race, color or national origin. Although 15 states reported that encouraging race-neutral adoptions would help reduce disproportionality, 18 states responded that this policy had no effect, and an additional 12 states reported that they were unable to tell. An HHS 2003 study of child welfare agencies, staff, and partner agencies noted that confusion and a general lack of knowledge regarding what the law allowed or prohibited hindered its implementation. In support of this finding, child welfare officials we spoke with in Illinois and Texas also noted that child welfare workers may misunderstand or fear that they are not complying with the law's prohibition. These officials stated that in some cases child welfare workers may be less likely to place African American children with relatives or in African American adoptive homes because they mistakenly believe that the law prohibits or discourages same-race adoptions. Other researchers and officials told us they opposed the law's intent and were concerned about the detrimental effects of placing children with parents of another race on a child's well being.[Footnote 85] Some officials we interviewed stated that race should be given first priority in placing African American children in families for care as is done for Native American children under the Indian Child Welfare Act of 1978 (ICWA).[Footnote 86] According to a judge we interviewed in North Carolina, the encouragement of race-neutral adoptions led in some cases to African American children being placed in cross-racial homes in which they felt disconnected from their heritage. In addition to adoption, many child welfare officials and researchers we interviewed considered legal guardianship a particularly important way to help African American children exit foster care. Legal guardianship was formally recognized under federal law as another option for placing children in permanent homes.[Footnote 87] Some African American families, especially relatives, are reluctant to adopt because they do not want to terminate the parental rights of the child's parent, according to officials and researchers we interviewed. Legal guardianship allows a household to establish a permanent home for a child without terminating the parental rights of the birth parents. Seventeen states in our survey reported that this federal policy was believed to help decrease the proportion of African American children in their states' foster care systems. In California and Illinois, subsidized guardianships have been found to reduce the number of children in foster care, including African American children. In California about 16,000 children exited the state foster care system between 2000 and 2005 through their kinship guardianship program, and about 43 percent of these children were African American, according to data from state officials. Based on the results of the Illinois waiver and other states with waivers, subsidized guardianships have also been found to be at least cost neutral. However, according to state child welfare directors we surveyed and interviewed, the lack of federal reimbursement for subsidies to guardians constrained states' ability to place African American children in guardianship arrangements. In many cases, families that could otherwise serve as guardians lack the financial stability to permanently care for children without support, according to officials and researchers we interviewed. However, unless states are one of the seven that have a current federal demonstration waiver for assisted guardianship or kinship permanency programs, states cannot use federal child welfare funds to provide subsidies to legal guardians.[Footnote 88] According to state and local child welfare officials, states would like to have more flexibility to use Title IV-E funds for supporting guardianship placements, as is done with adoption. In discussions with us, HHS officials stated that the Administration's proposed Child Welfare Program Option would provide states with the flexibility to use Title IV-E funds for the entire range of child welfare purposes, including assisted guardianship. However, although an HHS official said that guardianship was not considered as permanent as adoption, the results for the child have been found to be essentially the same. In the Illinois evaluation, guardianship and adoption both provide comparable levels of stability for the child and showed similar outcomes in terms of children's emotional and physical health. Licensing and Time Frames: Federal policies regarding licensing, such as those that limit reimbursement for costs associated with the use of unlicensed relative caregivers and require criminal background checks on prospective caregivers, contribute to the disproportionality of African American children in foster care, according to states we surveyed. To a lesser extent, state officials reported that federally mandated time frames determining a child's permanency plan and whether parental rights should be terminated also had a negative effect. (See fig. 16.) Figure 16: States' Views on Impact of Federal Policies on Licensing and Time Frames for Making Permanency Decisions on Disproportionality: [See PDF for image] Source: GAO analysis of state child welfare survey responses. [End of figure] States considered federal policies on the licensing and the use of relatives to provide foster care as increasing the proportion of African American children in foster care. In some cases, states permanently place children with unlicensed relatives who are neither adoptive parents nor guardians; however, states cannot claim federal reimbursement for such kinship care.[Footnote 89] In our survey, 20 states reported that this policy limiting reimbursement for costs associated with the use of unlicensed relative caregivers contributed to the disproportionality of African American children by hindering their ability to place children with relatives. According to researchers and state officials we interviewed, such policies have a disproportionate impact on African American children because they are more likely to live with unlicensed relatives. These relatives may be able to provide safe homes for children but may also be more likely have lower incomes and have difficulty meeting foster care licensing requirements, such as having a certain number of bedrooms. Eighteen states reported that federal policies requiring states to perform criminal background checks on prospective caregivers, including relatives, contributes to disproportionality, while other states reported that these policies had no effect. Among child welfare officials and others we interviewed, some were concerned that federal law requiring states to conduct fingerprinting checks for prospective parents or other types of background checks on all adults in the household may deter some African American relatives from stepping forward as caregivers. However, 16 states saw federal policy on criminal checks as having no effect. This may be in part due to the fact that most states have their own requirements regarding background checks that are similar to or more stringent than federal requirements.[Footnote 90] Until recently, states could opt out of federal requirements for criminal background checks on prospective foster care and adoptive parents, but that provision was eliminated by the recently enacted Adam Walsh Child Safety and Protection Act. For the eight states that opted out of the federal requirements, federal regulations require them to verify that safety considerations with respect to the prospective foster or adoptive parents have been addressed.[Footnote 91] Some officials were concerned that the federal policy would limit their ability under previous policy to place African American and other children with relatives and other families. California and New York officials told us that their alternative plans allow them the flexibility to make exemptions case-by-case for foster care, adoptive, or guardianship families, typically relatives, that have past convictions that would otherwise be automatically prohibited by federal law.[Footnote 92] Although such exemptions make up a comparatively small proportion of total placements for children,[Footnote 93] state and county officials in California told us that their inability to make these exemptions beginning October 2008--when the prohibition on states' ability to opt out of federal requirements goes into effect-- may have a disproportionate impact in the placement of African American children with relatives or other families who they consider safe and appropriate for children. State officials had mixed views on federal policies that impose time frames on permanency decisions that affect whether children will be reunified with their parents or placed in an alternative home.[Footnote 94] About a third of states reported that federal policy requiring that states adhere to certain time frames for initiating plans to place children in permanent arrangements and for terminating parental rights contributed to an increase in the proportion of African American children in care. Some state officials and researchers we interviewed said that these time frames were not reasonable for some African American parents who have complex problems, such as substance abuse and mental health issues, that require more time to resolve or if they have difficulty in accessing services.[Footnote 95] When parental rights are terminated, some children become "legal orphans" and remain in foster care longer than if parents had been given more time to complete their reunification plans.[Footnote 96] According to an HHS official's analysis of AFCARS data, the percentage of children who have had their parental rights terminated but who did not find a permanent home and ultimately emancipated out of the foster care system increased from 3.3 percent in 2000 to 6.7 percent in 2005.[Footnote 97] On the other hand, some child welfare officials reported that the ASFA time frames have been helpful in ensuring that children do not languish in care and have helped reduce the proportion of African American children in care. Conclusions: Issues surrounding the disproportionate representation of African- American children in foster care are pervasive, continuing, and complex. They cut across different points in the child welfare process- -from before entry to exit from foster care--and they affect nearly all states in this nation to varying degrees. In efforts to reduce African American representation in foster care, state and local child welfare agencies face numerous challenges. These challenges include ensuring that decisions to place a child in foster care are not influenced by bias or cultural misconceptions about families or communities, and that parent's difficulties in accessing support services do not prevent a child from returning home. Adding to these challenges is the fact that many supports and services are provided through multiple social service systems and require actions outside the responsibility of child welfare agencies, such as the ability to obtain timely substance abuse treatment for parents or the availability of affordable housing. To facilitate access to services, state and local agencies bear the primary responsibility for coordinating and administering these services. To some extent, federal policies on adoption have supported states' efforts to reduce the foster care population, but among policies aimed at reducing the number of minority children in foster care, many states experienced challenges recruiting sufficient minority families that reflect the foster care population. African American children have generally seen lower adoption rates than children of other races, and in recent years the adoption rate for all children has reached a plateau. States report being constrained by the lack of federal subsidies for legal guardianship similar to those provided for adoption. Many consider legal guardianship to be more reflective of the cultural values held by some African Americans and other families of color and better suited to the needs of African American and Hispanic families who want to permanently care for related children without necessarily adopting them. As a strategy, subsidizing guardianships has demonstrated its value in providing permanent families for children and in reducing the number of African American children in foster care. It may also be cost-effective, given the experiences of the states that implemented this strategy using federal waivers. This may therefore be the time to reconsider the current distinctions that provide subsidies for adoption but not for guardianship. The importance of collecting and analyzing data by race is considered a crucial first step for addressing racial disparity within child welfare and other systems. Yet some states and localities report a lack of capacity to collect or analyze data that would better allow them to identify the strategies that would be most useful in addressing the problems in their state. HHS provides assistance to states on data analysis and practices through its technical assistance network and related Web sites, although the agency lacks the directive and funding that Department of Justice officials said were instrumental to their efforts to analyze data by race and provide guidance on promising practices. In response to this directive, states that have identified disproportionality in juvenile justice as an issue have regularly submitted reports to OJJDP and have devised plans to address the issue. In child welfare, states identified as being in the forefront of efforts to address disproportionality are relying on private organizations to provide financial and technical assistance. In the absence of research-based evidence on strategies that work for addressing disproportionality, states are seeking out promising practices used in other states. Despite the steps that HHS has taken to disseminate information about these strategies, states report that they need further information and technical assistance to strengthen their current efforts in addressing disproportionality. Matter for Congressional Consideration: To assist states in increasing the number of homes available for the permanent placement of African American and other children from foster care, we suggest that Congress consider amending federal law to allow federal reimbursement for legal guardianship similar to that currently provided for adoption. Recommendation for Executive Action: To enhance states' ability to reduce the proportion of African American children in foster care, the Secretary of HHS should further assist states in understanding the nature and extent of disproportionality in their child welfare systems and in taking steps to address the issue. These actions should include: * Encouraging states to regularly track state and local data on the racial disproportionality of children in foster care and use these data to develop strategies that can better enable them to prevent children's entry into foster care and speed their exit into permanent homes. HHS should also encourage states to make increased use of HHS's National Resource Centers as a source of technical assistance on this issue. * Completing and making publicly available information on disproportionality that the agency is developing under its Culturally Competent Practice Knowledge Initiative so that states have easier access to tools and strategies useful for addressing the issue. Agency Comments and Our Response: We provided a draft of this report to HHS for review and comment. HHS's written comments are provided in appendix III of this report. Our draft report included a recommendation that HHS pursue specific measures to allow federal reimbursement for legal guardianship. In commenting on the draft report, HHS disagreed with the recommendation, stating that the administration had already proposed a broad restructuring of child welfare funding, known as the Child Welfare Program Option, which would allow states to use federal funds for legal guardianship. Under this proposed restructuring of child welfare, states could choose to remain under the current foster care funding structure or they could instead receive a capped grant for a period of 5 years. States choosing the grant option would have the flexibility to use these funds for a wide range of child welfare purposes, including subsidizing guardianships. The current adoption assistance program would remain the same under this proposal. However, HHS has presented this option in its budget proposals each year since 2004, but no legislation has been offered to date to authorize it. Moreover, if enacted, it is unknown how many states would choose a capped grant that would allow greater program flexibility instead of the current title IV- E foster care entitlement funding. In light of these factors, we suggest that Congress consider taking action to allow adoption assistance payments to be used for legal guardianship. Current evidence indicates that such a change could help states increase the number of permanent homes available for African American and other children in foster care. Furthermore, some states have demonstrated this change can be achieved without increasing program costs. We have changed the final report to delete our recommendation to HHS and to include instead this matter for congressional consideration. In response to our recommendation that HHS take certain actions to further assist states in understanding and addressing the nature and extent of racial disproportionality in their child welfare systems, HHS stated that these actions were consistent with their current technical assistance efforts to encourage and assist states in addressing racial disproportionality. For example, HHS cited the variety of technical assistance available to states in areas such as data analysis and cultural competency. However, HHS did not address the specific actions that we recommended related to encouraging states to regularly track and use child welfare data on racial disproportionality and completing and making publicly available the information on disproportionality that it is developing through its Culturally Competent Practice Knowledge Initiative. We continue to believe that it is important for HHS to take these actions to further equip states to address this complex issue. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from its issue date. At that time, copies of this report will be sent to the Secretary of HHS, relevant congressional committees, and other interested parties. We will also make copies available to others upon request. In addition, the report will be made available at no charge on GAO's Web site at http://www.gao.gov. Please contact me at (202) 512-7215 if you or your staff have any questions about this report. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix IV. Signed by: Denise M. Fantone, Acting Director: Education, Workforce, and Income Security Issues: [End of section] Appendix I: Objectives, Scope, and Methodology: Objectives: For this study, we were asked to analyze (1) the major factors that have been identified as influencing the proportion of African American children entering and remaining in foster care compared to children of other races and ethnicities; (2) the extent that states and localities have implemented strategies that appear promising in addressing African American representation in foster care; and (3) the ways in which key federal child welfare policies may have influenced African American representation in foster care. Although we focused on African American children in this report, we also noted points of similarity or difference with children of other races and ethnicities as appropriate. Scope and Methodology: Overall, to address these three objectives, we used multiple methodologies, including administering a state survey; conducting site visits; interviewing researchers and federal agency officials; conducting a literature review using various criteria; and analyzing federal legislation and policies. More specifically, we conducted a nationwide Web-based survey of state child welfare administrators in 50 states and the District of Columbia between November 2006 and January 2007. To obtain a more in-depth understanding of the issues, especially of any promising strategies to address disproportionality of African American children, we conducted site visits to California, Illinois, Minnesota, New York, and North Carolina, and in addition, conducted telephone interviews with state and local child welfare officials, service providers, and court officials in Texas. These states were selected in accordance with various criteria discussed below. To extend our understanding, we interviewed child welfare researchers identified by others as knowledgeable on issues of racial disproportionality in foster care as well as representatives from national child welfare organizations, such as the Casey Family Programs; the Child Welfare League of America; Black Administrators in Child Welfare, Inc; and the Center for the Study of Social Policy, on these matters. In addition, we participated in child welfare-related conferences with sessions relevant to these objectives. We also conducted an extensive literature review and analyzed published research on racial disproportionality in foster care and strategies used by states and others to address this issue, and selected the research for this review based on particular criteria described below. At the federal level, we interviewed HHS officials responsible for foster care programs and related data, as well as officials at the Department of Juvenile Justice, which is required by law to address racial disproportionality in the juvenile justice systems. Finally, we analyzed federal child welfare legislation, agency documentation, and policies relevant to foster care that may have an impact on racial disproportionality. We conducted our work between June 2006 and June 2007 in accordance with generally accepted government auditing standards. Web-Based Survey: To obtain state perspectives on our objectives and the relative priority state child welfare agencies place on the challenges they face, we conducted a Web-based survey of child welfare directors in the 50 states and the District of Columbia. The survey was conducted using a self-administered electronic questionnaire posted on the Web. We contacted directors via e-mail announcing the survey and sent follow-up e-mails and made telephone calls as well to encourage responses. The survey data were collected between November 2006 and January 2007. We received completed surveys from 47 states and the District of Columbia (a 92 percent response rate). The states of New Jersey, Maryland, and Rhode Island did not return completed surveys. To develop the survey questions, we relied on information gathered through interviews with researchers, professional associations, and our literature review (see criteria for selecting literature). In addition, in July 2006, we solicited comments from various researchers and other experts on elements used in our survey to ensure their completeness. These elements included a list of factors that contribute to, strategies to address, and federal policies that may affect the disproportionality of African American children in foster care. We received comments on these elements from the Center for the Study of Social Policy, the Black Administrators in Child Welfare, and Westat, and made modifications accordingly. We worked closely with social science survey specialists to develop and pretest the questionnaire. Because these were not sample surveys, there are no sampling errors. However, the practical difficulties of conducting any survey may introduce errors, commonly referred to as nonsampling errors. For example, differences in how a particular question is interpreted, in the sources of information that are available to respondents, or how the data are entered into a database can introduce unwanted variability into the survey results. We took steps in the development of the questionnaires, the data collection, and data analysis to minimize these nonsampling errors. For example, prior to administering the survey, we pretested the content and format of the questionnaire with five states to determine whether (1) the survey questions were clear, (2) the terms used were precise, (3) respondents were able to provide the information we were seeking, and (4) the questions were unbiased. We made changes to the content and format of the final questionnaire based on pretest results. Because these were Web-based surveys in which respondents entered their responses directly into our database, there was a reduced possibility of data entry error. We also performed computer analyses to identify inconsistencies in responses and other indications of error. In addition, an independent analyst verified that the computer programs used to analyze the data were written correctly. Site Visits: To obtain a more in-depth understanding of issues, we conducted site visits to California, Illinois, Minnesota, New York, and North Carolina. In addition we conducted telephone interviews with Texas state child welfare officials, a researcher a service provider, and a judge in Texas. When viewed as a group, the states we visited reflected diversity in geographic location, rates of African American representation in foster care, strategies and initiatives used to address this disproportion, and program administration (state administered and county administered). In addition, the states we selected collectively covered nearly one-third of children in foster care across the nation. During these visits, we interviewed state and local child welfare officials; juvenile court judges and other child welfare-related legal representatives, such as attorneys and public guardians; community service providers; and others involved in the child welfare systems, such as academic researchers. We also collected information, reports, and data on disproportionality and initiatives to address this issue from state and local child welfare agencies and others during these visits. We cannot generalize our findings beyond the states we visited. Literature Review of Published Research on Disproportionality: To learn more about the major factors, strategies, and federal policies influencing whether African American children enter and remain in foster care compared to children of other races and ethnicities, we conducted a literature review. The literature we reviewed included research articles we identified through databases, such as Lexis-Nexis, J-STOR, and the National Clearinghouse on the Child Abuse and Neglect Information. We used various search terms, such as disproportionality, African American, foster care, child welfare system, and over- representation in searching these databases. We also reviewed literature cited in these studies and those we found on Web sites related to child welfare and disproportionality, as well as literature recommended to us from our interviews. In addition, we conducted a more intensive review about 50 studies identified through these methods that focused on factors affecting entry and length of stay in foster care. For each selected study, we determined whether the study's findings were generally reliable. Two GAO social science analysts assessed each study's research methodology, including its research design, sampling frame, selection of measure, data quality, limitation, and analytic techniques for its methodological soundness and the validity of the results and conclusions that were drawn. Interviews with Researchers and Child Welfare Organizations: For all three objectives we also conducted interviews with academic researchers and other experts on disproportionality issues, such as child welfare-related organizations. We identified child welfare researchers for our interviews through our literature review and through recommendations from child welfare officials and stakeholders as knowledgeable on issues of racial disproportionality in foster care. For this study we interviewed academic researchers affiliated with the following universities and research centers: Chapin Hall at the University of Chicago, Children and Family Research Center at the University of Illinois School of Social Work, Jordan Institute for Families at the University of North Carolina, School of Social Welfare at the University of California at Berkeley, University of Minnesota School of Social Work, University of Texas School of Social Work, and Hunter College School of Social Work of the City University of New York. We also interviewed researchers and other staff at the following organizations: Black Administrators in Child Welfare, Casey Family Programs, Center for the Study of Social Policy, Child Welfare League of America, National Association of Public Child Welfare Administrators, National Council of Juvenile and Family Court Judges, Nestor Associates, and Westat. To obtain clarification on the findings of the National Incidence Surveys, we also interviewed the principal investigator of these studies for HHS, also at Westat. Adoption and Foster Care Analysis and Reporting System (AFCARS): To obtain children welfare data we requested and analyzed the U.S. Children's Bureau AFCARS data from the National Data Archive on Child Abuse and Neglect at Cornell University (NDACAN). AFCARS is a federal database that provides case level data on all children covered by Title IV-B and Title IV-E of the Social Security Act. On a bi-annual basis, all states submit data to the Children's Bureau concerning each child in foster care and each child who has been adopted under the authority of the state's child welfare agency. To confirm the reliability of the data, social science methodologists at GAO conducted electronic data testing, comparing our figures with HHS and others who have reported similar data. We also interviewed several officials with NDACAN and HHS who were responsible for the data. We found the data to be sufficiently reliable for the purposes of this report. In calculating the adoption rates reported in figure 15, we used estimates from HHS's Web site of the total estimated count of adoptions with public agency involvement finalized at year end divided by the number of children waiting to be adopted on the last day of the prior year. Children waiting to be adopted are defined by HHS as those children who have the case goal of adoption and/or have parental rights terminated.[Footnote 98] According to an HHS official familiar with AFCARS data, the HHS reported numbers of children waiting to be adopted may be imprecise because of variations among states and over time.[Footnote 99] Accordingly, we also calculated the adoption rate by using the number of children in foster care on the last day of the prior year as the denominator. Under both analyses, African American and Native American children had lower adoption rates between 2001 and 2005 than children of other races and ethnicities. We chose not to report the findings under the second method of analysis because the numbers of children in foster care on the last day of the prior year include many children who are not waiting to be adopted, such as children who have a case goal of reunification and later reunify with their parents, which would greatly underestimate adoption rates.[Footnote 100] [End of section] Appendix II: Disproportionality Indexes of Children in Foster Care by Race and State: Among researchers and others, disproportionality indexes or ratios are used to characterize the extent of disproportionality in a particular area, whether nationwide, within a state, or within a county or metropolitan area. Table 4 represents the proportion of African American, White, Hispanic, Asian, and Native American children in the foster care system in each state when compared to the overall population of each racial category of children in that state. (See table 4.) In the table, we present disproportionality indexes by state for children in foster care on the last day of fiscal year 2004. In this table, for example, an index number of below 1.00 indicates an under-representation of African American in foster care in a state compared to African American children's proportions in the general child population in that state, while a number above 1.00 indicates an over-representation of African American children in foster care in a state compared to African American children's proportions in the general child population in that state. This table also displays this indexing methodology for the other four racial and ethnic categories.[Footnote 101] Overall, the disproportionality index nationwide for African American children is 2.26, which means that African American children were over- represented in foster care nationally in 2004 at a rate of more than twice their proportions in the U.S. child population. In fiscal year 2004, a total of 36 states had disproportionality indexes of 2.0 or more, and 16 states had disproportionality indexes of 3.0 or more for the number of African American children in foster care at the end of the fiscal year (indexes of 2.0 or more are bolded in the table 4). Within states, disproportionality rates may vary considerably, as noted earlier. Table 4: Disproportionality Index by State of Children Ages 17 Years of Age or Under in Foster Care as of Last Day of Fiscal Year 2004: State: Alabama; African American children[A]: 1.52; White children[B]: 0.77; Hispanic children[C]: 0.63; Asian children[D]: 0.05; Native American children[E]: 0.54. State: Alaska; African American children[A]: 2.23; White children[B]: 0.48; Hispanic children[C]: 0.25; Asian children[D]: 0.04; Native American children[E]: 3.07. State: Arizona; African American children[A]: 2.36; White children[B]: 0.95; Hispanic children[C]: 0.95; Asian children[D]: 0.13; Native American children[E]: 0.39. State: Arkansas; African American children[A]: 1.39; White children[B]: 0.84; Hispanic children[C]: 0.64; Asian children[D]: 0.30; Native American children[E]: 0.05. State: California; African American children[A]: 4.05; White children[B]: 0.75; Hispanic children[C]: 0.89; Asian children[D]: 0.17; Native American children[E]: 1.80. State: Colorado; African American children[A]: 2.78; White children[B]: 0.78; Hispanic children[C]: 1.27; Asian children[D]: 0.28; Native American children[E]: 1.94. State: Connecticut; African American children[A]: 2.76; White children[B]: 0.49; Hispanic children[C]: 2.02; Asian children[D]: 0.08; Native American children[E]: 0.58. State: Delaware; African American children[A]: 2.30; White children[B]: 0.58; Hispanic children[C]: 0.83; Asian children[D]: 0.05; Native American children[E]: 0.51. State: District of Columbia; African American children[A]: 1.30; White children[B]: 0.01; Hispanic children[C]: 0.31; Asian children[D]: 0.17; Native American children[E]: 0.00. State: Florida; African American children[A]: 1.93; White children[B]: 0.89; Hispanic children[C]: 0.42; Asian children[D]: 0.12; Native American children[E]: 0.88. State: Georgia; African American children[A]: 1.44; White children[B]: 0.80; Hispanic children[C]: 0.54; Asian children[D]: 0.07; Native American children[E]: 0.12. State: Hawaii; African American children[A]: 0.44; White children[B]: 0.58; Hispanic children[C]: 0.19; Asian children[D]: 1.01; Native American children[E]: 1.80. State: Idaho; African American children[A]: 3.37; White children[B]: 0.89; Hispanic children[C]: 1.12; Asian children[D]: 0.24; Native American children[E]: 5.86. State: Illinois; African American children[A]: 3.48; White children[B]: 0.50; Hispanic children[C]: 0.29; Asian children[D]: 0.02; Native American children[E]: 0.53. State: Indiana; African American children[A]: 3.01; White children[B]: 0.72; Hispanic children[C]: 0.87; Asian children[D]: 0.09; Native American children[E]: 1.36. State: Iowa; African American children[A]: 4.45; White children[B]: 0.86; Hispanic children[C]: 0.91; Asian children[D]: 0.79; Native American children[E]: 5.41. State: Kansas; African American children[A]: 2.93; White children[B]: 0.88; Hispanic children[C]: 0.47; Asian children[D]: 0.16; Native American children[E]: 1.22. State: Kentucky; African American children[A]: 2.02; White children[B]: 0.87; Hispanic children[C]: 0.26; Asian children[D]: 0.19; Native American children[E]: 0.68. State: Louisiana; African American children[A]: 1.40; White children[B]: 0.76; Hispanic children[C]: 0.28; Asian children[D]: 0.12; Native American children[E]: 0.64. State: Maine; African American children[A]: 1.74; White children[B]: 0.88; Hispanic children[C]: 2.15; Asian children[D]: 0.43; Native American children[E]: 1.52. State: Maryland; African American children[A]: 2.27; White children[B]: 0.39; Hispanic children[C]: 0.23; Asian children[D]: 0.08; Native American children[E]: 0.78. State: Massachusetts; African American children[A]: 2.23; White children[B]: 0.67; Hispanic children[C]: 2.18; Asian children[D]: 0.39; Native American children[E]: 0.89. State: Michigan; African American children[A]: 2.90; White children[B]: 0.56; Hispanic children[C]: 0.78; Asian children[D]: 0.13; Native American children[E]: 1.83. State: Minnesota; African American children[A]: 3.63; White children[B]: 0.63; Hispanic children[C]: 1.39; Asian children[D]: 0.37; Native American children[E]: 7.31. State: Mississippi; African American children[A]: 1.10; White children[B]: 0.92; Hispanic children[C]: 0.73; Asian children[D]: 0.54; Native American children[E]: 0.14. State: Missouri; African American children[A]: 2.15; White children[B]: 0.84; Hispanic children[C]: 0.62; Asian children[D]: 0.13; Native American children[E]: 1.04. State: Montana; African American children[A]: 2.68; White children[B]: 0.63; Hispanic children[C]: 1.62; Asian children[D]: 0.56; Native American children[E]: 3.44. State: Nebraska; African American children[A]: 2.76; White children[B]: 0.80; Hispanic children[C]: 0.98; Asian children[D]: 0.27; Native American children[E]: 6.54. State: Nevada; African American children[A]: 2.67; White children[B]: 1.07; Hispanic children[C]: 0.53; Asian children[D]: 0.36; Native American children[E]: 0.39. State: New Hampshire; African American children[A]: 4.37; White children[B]: 0.91; Hispanic children[C]: 1.79; Asian children[D]: 0.15; Native American children[E]: 0.90. State: New Jersey; African American children[A]: 3.81; White children[B]: 0.41; Hispanic children[C]: 0.35; Asian children[D]: 0.03; Native American children[E]: 1.27. State: New Mexico; African American children[A]: 2.81; White children[B]: 1.02; Hispanic children[C]: 1.06; Asian children[D]: 0.20; Native American children[E]: 0.40. State: New York; African American children[A]: 2.63; White children[B]: 0.37; Hispanic children[C]: 0.97; Asian children[D]: 0.08; Native American children[E]: 0.90. State: North Carolina; African American children[A]: 1.67; White children[B]: 0.74; Hispanic children[C]: 0.79; Asian children[D]: 0.27; Native American children[E]: 1.53. State: North Dakota; African American children[A]: 3.26; White children[B]: 0.69; Hispanic children[C]: 1.90; Asian children[D]: 2.98; Native American children[E]: 3.09. State: Ohio; African American children[A]: 2.87; White children[B]: 0.65; Hispanic children[C]: 0.98; Asian children[D]: 0.08; Native American children[E]: 0.96. State: Oklahoma; African American children[A]: 1.76; White children[B]: 0.70; Hispanic children[C]: 1.10; Asian children[D]: 0.12; Native American children[E]: 1.12. State: Oregon; African American children[A]: 3.27; White children[B]: 0.75; Hispanic children[C]: 0.70; Asian children[D]: 0.24; Native American children[E]: 8.68. State: Pennsylvania; African American children[A]: 3.56; White children[B]: 0.53; Hispanic children[C]: 1.38; Asian children[D]: 0.21; Native American children[E]: 1.10. State: Rhode Island; African American children[A]: 2.68; White children[B]: 0.76; Hispanic children[C]: 1.14; Asian children[D]: 0.47; Native American children[E]: 2.11. State: South Carolina; African American children[A]: 1.43; White children[B]: 0.74; Hispanic children[C]: 0.69; Asian children[D]: 0.19; Native American children[E]: 0.48. State: South Dakota; African American children[A]: 1.67; White children[B]: 0.41; Hispanic children[C]: 2.33; Asian children[D]: 0.28; Native American children[E]: 3.71. State: Tennessee; African American children[A]: 1.52; White children[B]: 0.84; Hispanic children[C]: 0.91; Asian children[D]: 0.16; Native American children[E]: 0.36. State: Texas; African American children[A]: 2.02; White children[B]: 0.81; Hispanic children[C]: 0.89; Asian children[D]: 0.09; Native American children[E]: 0.73. State: Utah; African American children[A]: 6.06; White children[B]: 0.82; Hispanic children[C]: 1.63; Asian children[D]: 0.91; Native American children[E]: 3.97. State: Vermont; African American children[A]: 3.24; White children[B]: 1.01; Hispanic children[C]: 0.45; Asian children[D]: 0.13; Native American children[E]: 0.70. State: Virginia; African American children[A]: 1.89; White children[B]: 0.72; Hispanic children[C]: 0.76; Asian children[D]: 0.06; Native American children[E]: 0.28. State: Washington; African American children[A]: 3.07; White children[B]: 0.80; Hispanic children[C]: 1.03; Asian children[D]: 0.19; Native American children[E]: 4.99. State: West Virginia; African American children[A]: 2.04; White children[B]: 0.92; Hispanic children[C]: 1.19; Asian children[D]: 0.00; Native American children[E]: 0.19. State: Wisconsin; African American children[A]: 4.69; White children[B]: 0.54; Hispanic children[C]: 1.26; Asian children[D]: 0.34; Native American children[E]: 2.48. State: Wyoming; African American children[A]: 4.28; White children[B]: 0.96; Hispanic children[C]: 1.06; Asian children[D]: 0.17; Native American children[E]: 0.24. State: United States; African American children[A]: 2.26; White children[B]: 0.68; Hispanic children[C]: 0.87; Asian children[D]: 0.22; Native American children[E]: 2.25. Source: GAO analysis using Census Population Estimates from 2004 and Adoption and Foster Care Analysis and Reporting System (AFCARS) for fiscal year 2004, which was made available to us by the National Data Archive on Child Abuse and Neglect (NDACAN), Cornell University, Ithaca, New York. Children identified by the child welfare system with two or more racial categories were not included. Note: Children who were identified by the child welfare system with two or more racial categories or who were unknown were not presented in this table. [A] Children identified by the child welfare system as African American, non-Hispanic, and with only one race category. [B] Children identified by the child welfare system as White, non- Hispanic, and with only one race category. [C] Children identified by the child welfare system as having Hispanic origins; not a racial category. [D] Children identified by the child welfare system as Asian, which includes Hawaiian and Pacific Islander, non-Hispanic and with only one race category. [E] Children identified by the child welfare system as Native American, non-Hispanic, and with only one race category. [End of table] To derive each state's disproportionality index, for example, we divided the proportion of African American children in foster care (the number of African American children in foster care divided by the total number of children in child foster care) by the proportion of the African American children in the general population (the number of African American children in the population divided by the total number in the general population).[Footnote 102] According to HHS AFCARS data from fiscal year 2004, there were 498,981 children ages 17 years and under in foster care in the United States on September 30, 2004. Of these children, 162,991 were African American. Census population estimates for 2004 show there were 73,258,205 children 17 years old and under in the general population, of which 10,805,487 children were African American. Using the methodology described, we obtained a disproportionality index of 2.26 nationally for African American children in foster care at the end of the fiscal year for 2004. This methodology was used for the other four racial categories. [End of section] Appendix III: Comments from the Department of Health and Human Services: Department Of Health And Human Services: Office of the Assistant Secretary for Legislation: Washington, D.C. 20201: Jun 18 2007: Denise Fantone, Acting Director: Education, Workforce, and Income Security Issues: U.S. Government Accountability Office" Washington, DC 20548: Dear Ms. Fantone: Enclosed are the Department's comments on the U.S. Government Accountability Office's (GAO) draft report entitled, "African American Children in Foster Care: Additional HHS Assistance Needed to Help States Reduce the Proportion in Care " (GAO-07-816), before its publication. The Department appreciates the opportunity to comment on this draft before its publication. Sincerely, Signed for: Vincent J. Ventimiglia: Assistant Secretary for Legislation: Comments Of The Department Of Health And Human Services On The Government Accountability Office Draft Report Entitled. "African American Children In Foster Care: Additional HHS Assistance Needed To Help States Reduce The Proportion In Care" (GAO-07-816): GAO Recommendations: (1) To assist states in increasing the number of homes available for permanent placement of African American and other children from foster care, the Secretary of HHS should conduct a study to determine the feasibility of allowing adoption assistance payments to also be used for legal guardianship. Such a study should begin with the promising waiver results and examine whether subsidized guardianship can be implemented more broadly with similar results. Depending on the results, HHS should consider submitting a legislative proposal to reauthorize the use of federal demonstration waivers to subsidize legal guardianships, or alternatively, to amend current law to allow federal reimbursement for legal guardianship. (2) To enhance states' ability to reduce the proportion of African American children in foster care, the Secretary of HHS should further assist states in understanding the nature and extent of disproportionality in their child welfare systems and in taking steps to address the issue. These actions should include: * Encouraging states to regularly track state and local data on the racial. disproportionality of children in foster care and use the data to develop strategies that can better enable them to prevent children's entry into foster care and speed their exit into permanent homes. HHS should also encourage states to make increased use of HHS's National Resource Centers as a source of technical assistance on this issue. * Completing and making publicly available information on disproportionality that the agency is developing under its Culturally Competent Practice Knowledge Initiative so that states have easier access to tools and strategies useful for addressing the issue. HHS Comments: We do not concur with recommendation (1). As noted in the body of the report, the Administration has already identified in the Child Welfare Program Option a proposal to allow States the option to receive their foster care funding as a flexible grant for a period of five years or to maintain the program as it is currently funded. The option would provide States with the flexibility to develop a seamless child welfare system that supports a continuum of services to families in crisis and children at risk from the provision of prevention services to the provision of subsidies for guardianships. States that choose the grant option will be able to use the funds for foster care payments, prevention activities, permanency efforts (including searching for relatives and subsidized guardianships), case management, administrative activities, such as data analysis of the level of representation of specific populations, training for child welfare staff in cultural competence, and other such services related to child welfare activities. States would be able to develop innovative and effective systems for preventing child abuse and neglect, keeping families and children of all backgrounds safely together, and moving children toward adoption and permanency, quickly. Flexibility was built into the Child Welfare Program Option based on knowledge developed from the previous and existing Federal Title IV-E waivers. The grant would encompass Federal funding for the existing Title IV-E foster care maintenance payments program and the associated administrative costs, including Statewide Automated Child Welfare Information Systems and training. State allocations would be based on historical expenditures in consultation with the States. States that choose the option must maintain their existing level of financial investment in child welfare programs. The Administration is committed to addressing the fundamental flaws of the program through the Child Welfare Program Option, thereby providing States that take it the flexibility to address the causes of disproportional representation, creatively and actively. Recommendation (2) is consistent with the Children's Bureau current technical assistance efforts to encourage and assist States in addressing disproportionality through multiple avenues. Technical assistance is available to States in a variety of forms, including data analysis to determine the extent of concerns at the State and local levels, development of effective case practice models sensitive to various cultural populations, and design of service systems that can be individualized to the specific needs of children and families. Cultural competence in child welfare is a topic that has been discussed for many years without an identifiable national model or working solution for implementation in States. As part of its work with States, Tribes, localities, and courts, the Children's Bureau has created a Training and Technical Assistance (TTA) Network to assist States, Tribes, localities, and courts implement system changes necessary to produce better outcomes for children, youth, and families involved in the child welfare system. Over several years of discussions with the TTA Network, ongoing concern has been raised about the over-representation of children and families of color in the public child welfare system-a negative consequence when practice lacks cultural competence. The Children's Bureau has highlighted for each member of its technical assistance network the importance of developing and providing leadership activities to identify and classify issues and barriers related to culturally competent practices and disseminating positive results and strategies for implementation across the country. The Children's Bureau applauds GAO for recognizing the importance of these existing efforts and for GAO's support of ACF's continuing focus on reducing disproportional representation of specific populations both at entry into foster care and in reducing the longer stay while in foster care. By highlighting the availability of these services, GAO is assisting ACF in publicizing the availability of technical assistance in this area. Additionally, GAO reported that most States are using strategies to address the over-representation of specific populations in foster care, such as involving families in decision making, building community supports, and broadening the search for relatives to care for children. It should be noted that ACF reviews States and provides feedback on performance of these strategies as part of the Child and Family Services Reviews (CFSRs). ACF makes separate determinations regarding substantial conformity for each of the seven outcomes and each of the seven systemic factors reviewed. All States reviewed during the first round of CFSRs were found to be in substantial conformity on some of these areas and not in substantial conformity on others. The design of the reviews provides an opportunity for States to enter into program improvement plans to move towards improving performance in identified outcomes and systemic factors. It is by focusing States on long-term strategic and active program improvement that the reviews achieve the objective of enhancing service to children and families. The Children's Bureau has begun the second round of reviews and has developed specific technical assistance strategies to help States prepare for the CFSRs and maximize the benefit derived from the program improvement planning process. The National Child Welfare Resource Center for Organizational Improvement (NRCOI) has worked with other national resource centers to develop a CFSR training and technical assistance package. This material will help agencies successfully prepare for the second round of the CFSR, enhance leadership and management capacity, and achieve better outcomes through systemic change. Information on the technical assistance can be found on the NRCOI website at hyperlink, http://muskie.usm.maine.edu/helpkids/cfsrta.htm. The Children's Bureau also funds the Child Welfare Information Gateway, which has information available for readers who wish to learn more about disproportionality and cultural competence (see hyperlink, http://www.childwelfare.gov/systemwide/cultural/disporp/related.cfm). [End of section] Appendix IV: GAO Contact and Staff Acknowledgments: GAO Contact: Denise M. Fantone, Acting Director (202) 512-7215 or fantoned@gao.gov: Acknowledgments: Cindy Ayers (Assistant Director) and Deborah A. Signer (Analyst-in- Charge) managed all aspects of the assignment. Christopher T. Langford, Theresa Lo, and Kimberly Siegal made significant contributions to this report, in all aspects of the work. Interns Lisa McMillen and Ashley Gilbert also contributed with data collection and analysis. In addition, Charles Willson assisted in the message and report development; Jay Smale contributed to the initial design of the engagement and, together with Luann Moy, assessed studies for the literature review; William R. Chatlos provided technical assistance in the development and pre-testing of the Web-based survey; Carolyn Boyce provided technical support in survey research and statistical analysis; George Quinn provided data analytic support; and James Rebbe, Attorney, provided legal support. FOOTNOTES [1] Racial disproportionality refers to the extent that children of a certain race or ethnic group are over-or underrepresented in foster care relative to their proportion in the population. (See app. II for disproportionality rates for African American, White, Hispanic, Asian, and Native American children by state in fiscal year 2004.) [2] Children also enter foster care for other reasons, such as their parents' illness, death, or disability or because of the children's delinquent behavior and truancy. [3] We are using the term "policy" in this report to include federal laws, regulations, and informal agency guidance. [4] Native Americans are also overrepresented nationally, but some are affected by different child welfare laws and oversight authority than African Americans, making comparisons challenging. [5] For more detail on our methodology for conducting our work, see appendix I. [6] The National Incidence Study (NIS) is a congressionally mandated, periodic effort of the National Center on Child Abuse and Neglect to obtain information about the current incidence of child abuse and neglect in the United States. NIS-1 was published in 1981, NIS-2 was published in 1988, and NIS-3 was published in 1996. The NIS-3 findings are based on a nationally representative sample of over 5,600 professionals in 842 agencies serving 42 counties in the United States. [7] For disproportionality rates for African American, White, Hispanic, Asian, and Native American children by state in fiscal year 2004, see appendix II. [8] These data are based on our analysis of HHS's AFCARS data for 2004 and U.S. Census population estimates for that year. [9] Pub. L. No. 103-382, §§ 551-553. [10] Pub. L. No. 104-188, § 1808. [11] Pub. L. No. 105-89. [12] These convictions include situations in which the prospective foster or adoptive parent was, at any time, convicted of felony child abuse or neglect, spousal abuse, a crime involving children (including child pornography), or a crime involving violence (including rape, sexual assault, or homicide, but not including other physical assault or battery); or if the record check shows a felony conviction for physical assault, battery, or a drug-related offense that was committed in the last 5 years. [13] As of July 2006, eight states were using alternative background check processes approved by HHS instead of specified federal background check requirements for prospective foster care and adoptive parents: California, New York, Idaho, Oklahoma, Oregon, Massachusetts, Ohio, and Arizona. [14] Pub. L. No. 109-248. [15] Under prior law, the type of criminal record check was not specified. Under this act, states are required to perform a fingerprint- based check. [16] Urban Institute 2005 Child Welfare Survey reported in May 2006. This funding analysis is the most recent available that shows federal funding used specifically for child welfare. [17] States are entitled to Title IV-E reimbursement on behalf of children who would have been eligible for Aid to Families with Dependent Children (AFDC) (as AFDC existed on July 16, 1996), but for the fact that they were removed from the home of certain specified relatives. Although the AFDC program was replaced by the Temporary Assistance for Needy Families program in 1996, eligibility for Title IV- E payments remains tied to the income eligibility requirements of the now-defunct AFDC program. [18] The term "special needs" is used in a distinct way in Title IV-E programs. In order to be considered a child with special needs for the purpose of providing adoption assistance payments, states must determine that the child should not return home and have a factor or condition that would make the child difficult to place for adoption without such payments. States are provided discretion under federal law to determine what these factors or conditions are and may include age, membership in a sibling or minority group, or having a medical or developmental disability that would make placement difficult. There are additional eligibility requirements to obtain adoption assistance subsidies as well. [19] For further information on Title IV-B funds, see GAO, Child Welfare: Enhanced Federal Oversight of Title IV-B Could Provide States Additional Information to Improve Services, GAO 03-956 (Washington, D.C.: Sept. 12, 2003). [20] Data as reported by the Urban Institute in 2006, based on its 2005 survey of state child welfare agencies. This survey asked states to estimate their amount of funds spent on child welfare activities for state fiscal year 2004, and over 40 states provided data for federal, state and local funds. Federal funding included Title IV-E and IV-B, TANF, SSBG, Medicaid, and "other." SSBG includes funding transferred from TANF. Medicaid funds reported here include expenditures for transportation, rehabilitative services, targeted case management, and mental health services in residential treatment facilities and exclude expenditures for routine health care services for children in foster care. "Other federal funds" include Social Security Income and Survivor's Benefits. [21] These two states are California (Los Angeles County) and Florida. Florida began its waiver project in October 2006, while California's project has not yet begun. [22] The CFSR outcomes include protecting children from abuse and neglect, fostering permanency and stability in children's living conditions, preserving the continuity of children's family relationships and connections, enhancing families' capacity to provide for their children's needs, and ensuring that children receive appropriate services to meet their educational needs. [23] For additional information regarding HHS's oversight and states' implementation of the CFSR process, see GAO, Child and Family Services Reviews: Better Use of Data and Improved Guidance Could Enhance HHS's Oversight of State Performance, GAO-04-333 (Washington, D.C.: Apr. 20, 2004). [24] On a semiannual basis, all states submit data to HHS concerning all children in foster care for whom state child welfare agencies have responsibility for placement, care, or supervision and on children who are adopted under the auspices of the state's public child welfare agency. AFCARS also includes information on foster and adoptive parents. [25] The Information Gateway was formerly the National Clearinghouse on Child Abuse and Neglect Information and the National Adoption Information Clearinghouse. [26] Mandated reporters are individuals required by law to report cases of children's abuse and neglect. They are usually professionals who have frequent contact with children, such as health care workers, teachers, social workers, and law enforcers. [27] U.S. Census American Community Survey, 2005. The next highest percentages of families living below the poverty level were Hispanic and Native American families, both at 21 percent. [28] Hispanic families are less likely than African American families to be headed by single parents, a cultural difference that has been cited as a protective factor for Hispanic families that reduces the potential for children's removal from their families for neglect or abuse, despite similar rates of poverty. [29] A. Sedlak and D. Broadhurst, Executive Summary of the Third National Incidence Study of Child Abuse and Neglect. (U.S. Department of Health and Human Services, Washington, D.C.: 1996). [30] U.S. Census American Community Survey data from 2005. Females make up the majority of single parent households. [31] U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Results from the 2005 National Survey on Drug Use and Health: National Findings (HHS, SAMHSA, Office of Applied Studies, Rockville, MD: 2006). [32] All AFCARS data cited in this report comes from our analysis of the AFCARS database, which was made available to us by the National Data Archive on Child Abuse and Neglect , Cornell University, Ithaca, New York, unless otherwise stated. NDACAN's funding is provided by grant 90-CA-1667 from the Children's Bureau, U. S. Department of Health and Human Services. [33] This number was not associated with a particular racial or ethnic group. [34] U.S. Department of Health and Human Services, Children of Color in the Child Welfare System: Perspectives from the Child Welfare Community (HHS, Administration for Children and Families, Children's Bureau, Washington, D.C.: 2003). [35] Some child welfare officials also pointed out that lower income families may even be referred to the child welfare system in order to gain access to services. [36] U.S. Department of Health and Human Services, Child Maltreatment 2005 (HHS, Administration on Children, Youth and Families, Washington, D.C.: 2007). [37] Cited by Christianne Lind, "Developing and Supporting a Continuum of Child Welfare Services," (The Finance Project), 2004. The Finance Project is a non-profit organization that provides research, consulting, and technical assistant to public and private leaders on investing and making financial decisions with regard to children, families, and local communities. [38] Ann F. Garland et al., "Racial and Ethnic Variations in Mental Health Care Utilization Among Children in Foster Care," Children's Services: Social Policy, Research and Practice, 3(3): 133-146 (2000). [39] Astra Outley, Representation for Children and Parents in Dependency Proceedings (The Pew Commission on Children in Foster Care, May 2004). [40] Casey Family Programs is a foundation that works nationally with state and local child welfare agencies and families to provide direct services, as well as child welfare practice and policy suggestions. [41] For example: R. L. Hampton and E. Newberger, "Child Abuse Incidence and Reporting by Hospitals: Significance of Severity, Class and Race," American Journal of Public Health (75) 1: 56-60 (1985). For information on other studies, see R. Hill, Synthesis of Research on Disproportionality in Child Welfare: An Update (Casey-Center for the Study of Social Policy Alliance for Racial Equity in the Child Welfare System, 2006). [42] Barbara Needell et al., "Black Children and Foster Care Placement in California" (Children and Youth Services Review, 25(5/6): 393-408 (2003). [43] S. Zuravin et al., "Predictors of Child Protective Service Intake Decisions: Case Closure, Referral to Continuing Services, or Foster Care Placement," in P.A. Curtis, G. Dale, Jr., and J.C. Kendall (eds.), The Foster Care Crisis ( Lincoln, Nebr.: University of Nebraska, 1999), pp. 63-83). [44] Robert B. Hill, Synthesis of Research on Disproportionality in Child Welfare: An Update (Casey-CSSP Alliance for Racial Equity in the Child Welfare System, 2006). [45] According to statistics from the Bureau of Justice, African American men were over two times more likely to have been in prison on December 31, 2005, than Hispanic men and over six times more likely than White men. Female incarceration rates, though substantially lower than male incarceration rates, reveal similar racial and ethnic differences. African American neighborhoods with highly concentrated poverty tend to have both high levels of foster care involvement as well as involvement in the criminal justice system, according to child welfare officials we spoke with and research we reviewed. [46] According to our analysis of AFCARS fiscal year 2004 data, African American children are even more disproportionally represented in foster care at older ages than other children. [47] Brenda D. Smith, "After Parental Rights Are Terminated: Factors Associated with Exiting Foster Care," Children and Youth Services Review, 25(12): 965-985 (2003). [48] Hispanic children were also more likely to use kinship care than Whites, according to our analysis of AFCARS fiscal year 2004 data. [49] For example see: R. Barth, "Family Reunification," Child Welfare Research Review, 2 (1997): 109-122. [50] GAO, Foster Care: Kinship Care Quality and Permanency Issues, GAO- 99-32 (Washington, D.C.: May, 1999). [51] R. Cook et al., Unpublished Analysis of Kinship Care Data, (1998), quoted in U. S. Department of Health and Human Services (HHS), Administration for Children and Families, Children's Bureau, Report to the Congress on Kinship Foster Care: Part I: Research Review, Part II: Secretary's Report to the Congress (Washington, D.C.: 2000). [52] Kinship care refers to situations in which a relative or a close family friend provides shelter and cares for a child who is still in the foster care system. Legal guardianship refers to situations in which a relative or caregiver has taken permanent legal custody of the child without terminating parental rights; with legal guardianship, the child has exited foster care. [53] HHS, Children of Color in the Child Welfare System: Perspectives from the Child Welfare Community (HHS, Administration for Children and Families, Children's Bureau, Washington, D.C.: 2003). [54] Sponsored by the People's Institute for Survival and Beyond and promoted by Casey Family Programs. [55] According to academic researchers and others, structural or institutional racism is defined as the historical, cultural, political, ideological, and economic practices that have produced disparities and disproportionality between the different racial groups. Under this conceptualization of racism, the focus is not on individual behavior but instead on systems that distribute resources along lines by race or ethnicity. [56] Will Johnson, "Effectiveness of California's Child Welfare Structured Decision Making Model: A Prospective Study of the Validity of the California Family Risk Assessment," (California Child Welfare Structured Decision Making Project, for the California Department of Social Services, Feb. 2004). See also L. Anthony Loman and Gary L. Siegel, "An Evaluation of the Minnesota SDM Family Risk Assessment: Final Report," (Institute of Applied Research, St. Louis, Missouri, conducted for the Department of Human Services, Minnesota: Dec. 2004). [57] HHS, Children of Color in the Child Welfare System: Perspectives from the Child Welfare Community (HHS, Administration for Children and Families, Children's Bureau, Washington, D.C.: 2003). [58] See GAO, Child Welfare and Juvenile Justice: Federal Agencies Could Play a Stronger Role in Helping States Reduce the Number of Children Placed Solely to Obtain Mental Health Service, GAO-03-397 (Washington, D.C.: Aug. 2003), for gaps in access to mental health service and see GAO, HHS Actions Could Improve Coordination of Services and Monitoring of States' Independent Living Programs, GAO-05-25 (Washington, D.C.: Nov. 2004) for gaps in supportive services for children transitioning out of foster care. [59] In response to the Health Insurance Portability and Accountability Act of 1996 (Pub. L. No. 104-191), known as HIPPA, the Secretary of HHS issued regulations covering the use and disclosure of protected health information. The Family Education Rights and Privacy Act (Pub. L. No. 93-380) governs access to student education records. [60] This study was based on data from 14 counties. [61] A 2006 report by Generations United found that a total of 35 states and the District of Columbia were subsidizing legal guardianships. (See hyperlink, http://ipath.gu.org/documents/A0/GU- GeneralFactSheetJune.pdf.) Three of these states, Maryland, New Jersey, and Rhode Island, did not respond to our survey on disproportionality. Other states may subsidize guardianships, but not limit these subsidies to families involved in the state child welfare system. [62] This may be in part because relative foster care homes can be exempt from federal timeline requirements. [63] Jill Duerr Berrick, "When Children Cannot Remain Home: Foster Family Care and Kinship Care," The Future of Children: Protecting Children from Abuse and Neglect, 8, no.1 (1998). [64] In 2003, the University of Illinois and Westat conducted an evaluation of Illinois' subsidized guardianship program, the largest of all programs funded by Title IV-E waivers, reaching 6,800 guardians. The study used an experimental design with random assignment of families into treatment and control groups. The study found that Illinois' subsidized guardianship waiver resulted in fewer children remaining in long-term foster care with ongoing administrative oversight, that home stability increased, and that the withdrawal of regular administrative oversight and casework services from the families did not result in higher rates of indicated subsequent reports of abuse and neglect. [65] Illinois' guardianship program was funded through a federal Title IV-E waiver. [66] States like Alaska, Hawaii, Utah, and Montana all reported in their survey comments that they were concentrating their efforts on disproportionality of other populations such as Native Americans and Hispanics because the number of African Americans in their foster care systems was comparatively small. [67] Diligent recruitment of African American foster and adoptive parents might be considered a more targeted strategy. However, we grouped this under strategies to increase the number of permanent homes because states are required by law to engage in this activity, and it is not necessarily an indication of their efforts to target efforts to address disproportionality. [68] Called the Breakthrough Series Collaborative (BSC), this program encourages participating jurisdictions to develop child welfare systems that are free of structural racism and benefit all children by engaging with other jurisdictions, developing leaders and disseminating lessons learned. [69] When this report was completed, the Texas Health and Human Services Commission and the Department of Family and Protective Services, as required by the law (Tex. Fam. Code Ann. Sec. 264.2041) also devised a remediation plan to prevent racial or ethnic disparities and an evaluation of policies and procedures should the results indicate disparate treatment of racial or ethnic groups. [70] HHS, Children of Color in the Child Welfare System: Perspectives from the Child Welfare Community (HHS, Administration for Children and Families, Children's Bureau, Washington, D.C.: 2003). [71] Although 18 states in our survey believed that having reporting requirements on disproportionality rates in the CFSR would be useful, nearly as many responded that it would not be useful, and HHS officials told us that the CFSR process was governed by statute and that they could not add such a requirement. In addition, some child welfare officials we interviewed believed that outcomes should be recorded by race. These would be similar to educational outcome requirements under the No Child Left Behind Act. It requires states to create an accountability system of assessments, graduation rates, and other indicators. Schools have to make adequate yearly progress to a state- determined level of proficiency. [72] Because policies are often implemented around the same time and overlap, changes in outcomes, such as adoption rates, may reflect the influence of multiple policies. [73] In addition to block grant funds, federal funds for prevention and family support services under Title IV-B have been useful in states' efforts to address disproportionality, according to some officials we interviewed. [74] See Cynthia Andrews Scarcella, Roseanna Bess, Erica Hecht Zielewski, and Rob Geen, The Cost of Protecting Vulnerable Children V (Washington, D.C.: Urban Institute, 2006). According to the Urban Institute survey, states' allocation of TANF and SSBG block grants constituted 32 percent of federal child welfare spending in fiscal year 2004. [75] See GAO 07-75. [76] States may claim federal reimbursement under Title IV-E for every eligible child who is placed in a licensed foster care homes, and states may seek reimbursement from the federal government for specified percentages of these costs. Of total federal child welfare expenditures for 2004, Title IV-E claims represented about 50 percent, while Title IV-B funds represented about 5 percent, according to data from the Urban Institute. [77] Although many officials we interviewed indicated that more funds should be placed in prevention, family preservation and family support services, findings from large-scale evaluations conducted by HHS showed that provision of these services provided no or little effect in reducing out-of-home placement, maltreatment recurrence, or improved family functioning beyond what normal casework services achieved. See HHS, Evaluation of Family Preservation and Reunification Programs, Final Report, Volumes 1 and 2 (Dec. 2002). [78] See GAO-06-787T; GAO 07-75; Fostering the Future: Safety, Permanence and Well-Being for Children in Foster Care, Pew Commission on Children in Foster Care (May 2004); and C. Lind, Developing and Supporting a Continuum of Child Welfare Services, Welfare Information Network, The Finance Project (Dec. 2004). [79] HHS discussed the current funding structure and the administration's proposed child welfare program option in its publication, Federal Foster Care Financing: How and Why the Current Funding Structure Fails to Meet the Needs of the Child Welfare Field, ASPE Issue Brief, (U.S. Department of Health and Human Services, Assistant Secretary for Planning and Evaluation, Washington, D.C.: Aug. 2005). [80] HHS, Children of Color in the Child Welfare System: Perspectives from the Child Welfare Community (HHS, Administration for Children and Families, Children's Bureau, Washington, D.C.: 2003). [81] This is based on HHS CFSR results between 2001 and 2004. Challenges in recruiting are consistent with survey responses in an earlier GAO study as well. See GAO, Child Welfare: Improving Social Service program, Training, and Technical Assistance Information Would Help Address Long-standing Service-level and Workforce Challenges, GAO- 07-75 (Washington, D.C.: Oct. 6, 2006). [82] Lorelei B. Mitchell, Richard P. Barth, Rebecca Green, Ariana Wall, et al., "Child Welfare Reform in the United States: Findings from a Local Agency Survey," Child Welfare, 84, no. 1 (Jan.-Feb. 2005). [83] As noted earlier, federal policy allows states to classify children as special needs if they have characteristics that make them difficult to place with adoptive families without adoption assistance, such as belonging to a sibling or minority group, or having a disability. [84] Most African American children were adopted by African American parents from 2000 to 2004, according to our analysis of HHS data, and the percentage of African American children in foster care adopted by single or married African American parents remained relatively steady at around 70 percent. [85] This concern was also expressed in a previous GAO report. See GAO, Foster Care Implementation of the Multiethnic Placement Act Poses Difficult Challenges, GAO-98-204 (Washington, D.C.: Sept. 14, 1998). [86] Pub. L. No. 95-608. ICWA requires that efforts be made to place Native American children with relatives or tribal families, unless a good reason exists not to follow these placement preferences. [87] Prior to ASFA's enactment, children's options for exiting foster care included being reunified with their parents, adopted by a relative or nonrelative, or achieving emancipation from foster care when they reached a certain age, usually 18. [88] According to HHS officials, as of May 2007, seven states have active demonstrations testing the use of subsidized guardianship as an alternative permanency option, and one state has not yet implemented its guardianship demonstration. [89] In the preamble to its 2000 regulations implementing ASFA, HHS clarified that relative caregivers must be fully licensed and meet the same licensing requirements as nonkin in order for the state to receive IV-E reimbursements for those families. 65 Fed. Reg. 4020 Except for background checks related to criminal convictions, states determine their own licensing requirements for prospective foster care, adoptive, and guardian parents, but must meet national safety standards as overseen through the CFSR process. [90] For example, according to information from HHS's Children's Bureau, 24 states as of August 2006 require federal criminal record checks for prospective adoptive parents and 23 states require fingerprinting checks. [91] 45 C.F.R. 1356.30(e). [92] For example, officials in California told us that county child welfare officials made an exemption for an uncle who, 3 years ago, had a felony drug conviction for being under the influence of marijuana and alcohol. The county welfare agency determined that he had since rehabilitated and would, along with the aunt, make an appropriate caregiver of three children. [93] For example, state officials from California estimated that local child welfare agencies made about 360 exemptions in California between 2005 and 2006 for relative caregivers who otherwise would have been automatically disqualified as caregivers under federal law. These exemptions accounted for 2 to 3 percent of all placements that were approved for foster care, some of whom later became guardians or adoptive parents, according to officials. [94] ASFA shortened the time frame in which a permanency hearing must be held for children when they first enter foster care from 18 months to 12 months and required states to file a petition to terminate parental rights for children who have been in foster care for 15 of the past 22 months. States may exempt children from the requirement to terminate parental rights if a child is placed with a relative. [95] However, timelines were often extended because of parents' difficulties in accessing services they needed to comply with requirements for reunifying with their children. See GAO, Foster Care: Recent Legislation Helps States Focus on Finding Permanent Homes for Children, but Long-Standing Barriers Remain, GAO-02-585 (Washington, D.C.: June 28, 2002). [96] To mitigate this situation, California passed legislation in 2005 permitting a child who has not been adopted after the passage of 3 years from termination of parental rights to petition the juvenile court for reinstatement of parental rights (Cal. Stats., AB 519, Chap. 634). [97] Some of the increase also represented better data reporting, as cited in Penelope L. Maza, "Children Who Fall Through the Cracks," The Roundtable, National Child Welfare Research Center, 21, no. 1 (2007). [98] This excluded children with termination of parental rights who are 16 and older and have the goal of emancipation. [99] According to this HHS official, until the last few years, some states may have been under-reporting cases in which termination of parental rights has occurred, and states vary in how they determine adoption as a case goal for children in foster care. HHS has been revising how it generates this estimate over the years. [100] For example, in 2005, reunification was the case goal for 50 percent of children, and 54 percent of children who exited foster care that year were reunified with their parent or primary caretaker. [101] Another method, used by child welfare researchers at the University of California at Berkeley to characterize differences among populations, is called a "disparity index." This index compares the levels of disproportionality between various racial and ethnic groups. For example, the disparity index for New York's African American children when compared to White children is 7.11--more than double the state's disproportionality rate for African American children of 2.63, as shown in table 4. (OJJDP also uses a similar index, which it terms the "relative rate index," in analyzing disproportionate minority contact.) [102] The methodology we used for our disproportionality calculations is based on the University of California at Berkeley Center for Social Services Research's Foster Care Dynamics Disproportionality and Disparity Index Web site. URL: < hyperlink, http://cssr.berkeley.edu/CWSCMSreports/dynamics/disprop/> GAO's Mission: The Government Accountability Office, the audit, evaluation and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO's Web site (www.gao.gov). Each weekday, GAO posts newly released reports, testimony, and correspondence on its Web site. To have GAO e-mail you a list of newly posted products every afternoon, go to www.gao.gov and select "Subscribe to Updates." Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. Government Accountability Office 441 G Street NW, Room LM Washington, D.C. 20548: To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202) 512-6061: To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov Automated answering system: (800) 424-5454 or (202) 512-7470: Congressional Relations: Gloria Jarmon, Managing Director, JarmonG@gao.gov (202) 512-4400 U.S. Government Accountability Office, 441 G Street NW, Room 7125 Washington, D.C. 20548: Public Affairs: Paul Anderson, Managing Director, AndersonP1@gao.gov (202) 512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, D.C. 20548: