Interim Report to the Congress on the Feasibility of a National Child Abuse Registry

Interim Report to the Congress on the Feasibility of a National Child Abuse Registry

May 2009

By: Laura Radel

Office of the Assistant Secretary for Planning and Evaluation

---

About 45 states maintain child abuse registries, which are databases that allow child protective services investigators routine access to child abuse history information to inform better current investigations.  Most states also use their registries to conduct pre-employment background checks on persons who work with children, such as day care providers.  The Adam Walsh Child Protection and Safety Act of 2006 (P.L. 109-248) requires that HHS conduct a study to assess the feasibility of establishing a national child abuse and neglect registry and present the results in a Report to Congress.  This interim report describes key issues regarding the feasibility of establishing a national child abuse registry and concludes that implementation is not feasible under the statutory limitations of the authorizing legislation.  Further feasibility study activities are underway and will be described in a final report when results are available.

---

Contents

Executive Summary

Sections

• Background
  • The Adam Walsh Child Protection and Safety Act of 2006
  • Legislative History of Congressional Efforts to Establish a National Child Abuse Registry
  • Activities Comprising This Initial Feasibility Assessment
• Characteristics of State Child Abuse Registries
• Purpose and Utility of a National Child Abuse Registry
  • Intended Uses of a National Child Abuse Registry
  • Quantifying the Potential for Child Safety Improvements
• Acquisition and Management of Data
  • Obtaining Data for a National Registry
    • Provision of Data by State, Local and Tribal Agencies
    • Managing Data from Many Sources
    • Alternative Structures for a National, Electronic Child Abuse Registry
  • Standardization and Comparability of Registry Information
    • Determining Which Cases Should be Included in a National Child Abuse Registry
    • Identifying Perpetrators Accurately
    • Characterizing Maltreatment
  • Maintaining a Registry and Responding to Inquiries
    • Keeping Registry Data Secure
    • Restricting Access to a National Child Abuse Registry to Authorized Users
    • Responding Promptly to Queries
    • Determining a Registrys Relationship to Other Data Collection Efforts
    • Complying with the Privacy Act
  • Issues Regarding Participation of Indian Tribes in a National Child Abuse Registry
• Due Process Considerations
• Lessons from Data Systems Used for Criminal Background Checks
  • Compiling and Referencing Criminal History Records
• Benefits and Costs of a National Child Abuse Registry
  • Potential Benefits
  • Anticipated Costs
    • Federal Costs
    • Costs to Agencies Submitting Data to and Making Inquiries of a National Registry
    • Costs to Individuals Identified (Truly or Falsely) as Child Maltreatment Perpetrators
    • Nonfinancial Costs of a National Child Abuse Registry
    • Financing Startup and Ongoing Operations
• Conclusions
• Next Steps

References

Appendices

1. Section 633 of the Adam Walsh Child Protection and Safety Act (P.L. 109-248, 42 U.S.C. 16990)
2. Characteristics of State Central Child Abuse Registries
3. Data Elements Currently Collected in the Child File of the National Child Abuse and Neglect Data System

Endnotes

Executive Summary

Child Protective Services (CPS) systems in each state receive and investigate reports of child abuse and neglect, and intervene with families as necessary to protect children.  Often the first step in investigating a maltreatment allegation is to check whether the agency has had past contact with the family and, if so, the nature of past allegations and the outcomes of previous investigations.  Investigators do this because the familys past conduct toward the child represents a risk factor used to assess the childs current safety and future risk of harm.

Child abuse registries were established to enable local CPS agencies within a state to share information with each other about past maltreatment investigations.  In many states these registries are also used to conduct pre-employment screenings of persons applying to work with, or volunteer with, children or other vulnerable populations.  Approximately 40 to 45 states currently operate child abuse registries, though they vary substantially in their coverage and the information contained.  However, there is no system that would enable national checks of child abuse histories, and inquiries from one state to another are often time consuming and cumbersome. Occasional tragedies in which children have been killed or seriously injured after their families fled from one state to another during or following a child protection investigation have led some to believe strongly that a national child maltreatment registry is needed.

Section 633 of The Adam Walsh Child Protection and Safety Act of 2006 (the Act) requires the U.S. Department of Health and Human Services (HHS, the Department) to establish a national child abuse registry and to conduct a feasibility study regarding implementation issues that would need to be addressed.  The statute requires the Secretary to address the costs and benefits of a national child abuse registry, make recommendations regarding a due process procedure for a national child abuse registry and regarding data standards that should be considered in implementing such a registry.

This document is an interim report of the feasibility study described in the Act, and as directed by the conference report accompanying the Departments FY 2008 appropriations.  Congress has since directed that $500,000 from the Departments FY 2009 appropriations for child abuse discretionary activities be designated for additional work on the feasibility study.  HHS will conduct additional feasibility activities in the coming year and will prepare a final report to the Congress containing results of the additional work when findings are available.  Additional activities that will be conducted as part of the feasibility study are described in the next steps section at the end of this report.

The initial feasibility assessment described in this interim report was conducted internally by HHS staff and included a literature search; an analysis of the statutory language; a descriptive analysis of state child abuse registries; identification of related federal activities including activities authorized under the Child Abuse Prevention and Treatment Act (CAPTA) and Department of Justice programs through which criminal history information is shared; stakeholder discussions; an examination of the particular issues related to including in a national child abuse registry entries submitted by Indian tribes; consideration of interactions with the Privacy Act; an analysis of the due process issues raised by the creation and use of a national child abuse registry; and, finally, a consideration of benefits and costs involved in establishing and managing such a registry.  The initial assessment results in four conclusions:

1. Conclusion 1:  Potential benefits of a national child abuse registry are largely unknown.  There is no data available with which to quantify improvements in child safety that may result from the implementation of a national child abuse registry.  In particular, it is unknown how frequently perpetrators are substantiated for child maltreatment in multiple states. This interim report identifies and describes the major components of anticipated costs of implementing a national child abuse registry.  Key among these would be the costs of establishing secure electronic systems to protect the data from unauthorized use, and addressing procedural weaknesses in some jurisdictions CPS systems to assure the accuracy and reliability of information included in a national registry.  The gap in information regarding the frequency with which a national child abuse registry could be helpful to child maltreatment investigators will be a primary focus of additional feasibility study activities.
2. Conclusion 2:  A lack of incentives for participation could result in a database that includes little information and fails to fulfill its intent.  The submission of data to a national child abuse registry would be voluntary, creating the risk that HHS could create a database to which few jurisdictions would provide data, making it of little practical value.  Only if a national child abuse registry is constructed in a way that meets the needs of state and local child protection agencies and creates conditions under which they would be willing to provide the necessary data, could a national registry become a useful child protection tool.  Additional work to assess states likelihood of participating in a national child abuse registry as described in the Adam Walsh Act will be conducted as part of our further feasibility activities.
3. Conclusion 3:  Before implementation could begin, legislative change would be needed to permit the collection of sufficient information to accurately identify perpetrators.  The Adam Walsh Act limits identifying information in a national child abuse registry to the perpetrators name.  This statutory language must be changed before a national child abuse registry could be implemented.  Because many names are common, name cannot be the only field used to determine whether or not there is a match between the individual about whom an inquiry is made and a perpetrator listed in a national child abuse registry.  Even with additional identifying fields, however, high false positive and false negative rates must be anticipated.
4. Conclusion 4:  Clarification is required on several key issues that are ambiguous in the authorizing statute; these must be resolved either within HHS or by Congress before implementation could proceed.  Key among these is whether a national registry is to be used only for investigative inquiries or also for child abuse history checks related to employment and licensing purposes.  Employment/licensing checks are not mentioned explicitly in the Adam Walsh Act, but would be allowed under the existing statutory language regarding access to information in the national child abuse registry.  In many states, such employment checks far outnumber investigative uses of their child abuse registries.  Basic decisions are also needed regarding how to maintain restricted access and validate the identities of legitimate users.

With respect to a due process procedure, there can be no federal substitute for procedural protections at the state or local level.  HHS must not create a system that would second guess local investigations and determinations.  If a national child abuse registry is established, HHS believes the only feasible way to effectively and efficiently provide due process protections is to require that submitting jurisdictions certify that for cases submitted to the national registry, minimum due process protections were available to the perpetrator.  Whether or not employment related background checks remain a permissible use of the registry is an important factor in determining the due process protections that would be required.  Key due process issues include the level of evidence used to make substantiation decisions; whether individuals included in such a registry must be notified of their inclusion and the implications of being listed in the registry, and the strength of the hearing or appeal procedures in place at the local level through which substantiation decisions may be challenged.  It should be noted that strong due process protections could necessitate significant changes to existing CPS investigation processes in some states that could be costly to implement and may discourage participation in a national registry.

HHS is not yet ready to recommend specific due process protections that would be required.  Our initial work has identified issues that have been the subject of legal challenges to state child abuse registries.  As feasibility work continues, we will seek examples of states due process procedures related to their own child abuse registries in an effort to design appropriate federal due process protections.

This initial feasibility assessment has determined that there are very substantial challenges involved in establishing a national child abuse registry. Implementation is not feasible under the statutory limitations of the authorizing legislation.  It would be possible to overcome the statutory limitations and other challenges, but doing so may involve substantial costs and could be burdensome to the state and local child protective services systems a national child abuse registry is intended to help.  In the end, it is not clear whether or by how much child safety would be improved through a national database of child abuse perpetrators.  Future feasibility study activities will further assess the potential benefits of a national child abuse registry and explore ways of addressing a number of the challenges identified to date.

Background

Child Protective Services (CPS) systems in each state receive and investigate reports of child abuse and neglect, and intervene with families as necessary to protect children.  In 2007, nearly 2 million child maltreatment investigations nationally resulted in determinations that approximately 794,000 children were victims of abuse or neglect (U.S. Department of Health and Human Services [HHS], 2009).  Approximately 1,760 children died as a result of abuse or neglect that year.  Certain types of professionals, such as teachers, doctors and others who work with children, are required to report suspected child maltreatment.  Reports are also received from family members, neighbors and other concerned members of the public.  Child protective services investigations are usually conducted by local government agencies, either a child protective services agency (of which there are approximately 2,600 across the U.S.) or a law enforcement agency, depending on how the state has organized itself (HHS, 2003).  In some states, and for some types of cases, CPS and law enforcement may conduct joint investigations (Winterfeld and Sakagawa, 2003).

When child protective services investigators begin investigating an allegation of abuse or neglect, typically among their first actions is to check whether the agency has had past contact with the family and, if so, the nature of any past allegations and the outcomes of these previous investigations.  They do this because the familys past conduct toward the child represents a risk factor to be considered in assessing the childs current safety as well as future risk of harm.  Approximately 45 states maintain state child abuse registries which contain in a database the results of past child protective services investigations¹.  Typically the registries include information about the perpetrator, the victim, and the circumstances of the maltreatment at the time the investigation occurred.  Originally registries were separate from other administrative data systems maintained by the child protection agency.  Today, many are sub-functions of larger information systems used to manage the agencys administrative data.

Child abuse registries were established to enable local CPS agencies within a state to share information with each other about past maltreatment investigations.  In many states these registries are also used to conduct pre-employment screenings of persons applying to work with (or volunteer with) children or other vulnerable populations.  This often includes screening persons applying for licenses as child care providers.  Since 2006, states have been required to conduct child abuse history checks for all prospective foster and adoptive parents in all states in which a household member has lived in the past five years (HHS, 2006).  As will be described below, the state registries vary considerably in what cases are included and how they operate.  Originally, their primary purpose was to allow investigators routine access to child abuse history information in order to better inform current investigations.  More recently, in many states this investigatory purpose has been overtaken by the employment and licensing check function which may make up the bulk of inquiries to these registries.

While within-state checks of child abuse registries are a standard part of most child protective services investigations, there is no system that would enable national checks of child abuse histories, and inquiries from one state to another are often time consuming and cumbersome.  Occasional tragedies in which children have been killed or seriously injured after their families fled from one state to another during or following a child protection investigation have led some to believe strongly that a national child maltreatment registry is needed.  The principal argument for such a registry is a belief that CPS investigators lack of easy access to information about previous maltreatment history from other states impedes a full assessment of safety risks during child abuse and neglect investigations and prevents full knowledge of backgrounds of foster and adoptive parent applicants.

The Adam Walsh Child Protection and Safety Act of 2006

Section 633 of The Adam Walsh Child Protection and Safety Act of 2006 (the Act, P.L. 109-248, 42 U.S.C.  16990) requires the U.S. Department of Health and Human Services (HHS, the Department) to establish a national child abuse registry and to conduct a feasibility study regarding implementation issues that would need to be addressed.  This document is an interim report of the feasibility study described in the Act.  The full text of the statutory requirement appears at Appendix A.  Other sections of the Act also require child abuse history checks for all prospective foster and adoptive parents in any state in which they have lived in the previous five years.

Key features and limitations of the national child abuse registry, as described in the statute or that are important in their omission, are as follows:

• The Secretary of HHS shall collect in a central electronic registry information on persons reported as ... perpetrators of a substantiated case of child abuse or neglect.
• Information provided to the Secretary of HHS shall be coextensive with a state or Indian tribes equivalent electronic registry if it maintains one.  Data may be submitted by state or tribal government agencies, or, at state option, by local government agencies responsible for conducting child protection investigations.
• The registry shall contain case-specific identifying information that is limited to the name of the perpetrator and the nature of the substantiated case of child abuse or neglect.
• The law explicitly prohibits HHS from requiring states and Indian tribes to modify their existing registries or child maltreatment records in complying with the Act and may not require the establishment of registries by those jurisdictions that do not have them.
• The law provides no incentives for a state that maintains a child abuse registry to submit data to a national registry and there are no consequences for not doing so.
• The national registry HHS is to establish shall only be accessible to any Federal, State, Indian Tribe or local government entity, or any agent of such entities, that has a need for such information in order to carry out its responsibilities under law to protect children from abuse or neglect.

This Interim Report to the Congress describes the results of the Departments initial efforts to address the statutes requirement that the Secretary conduct a study to address the feasibility of establishing data collection standards for a national child abuse and neglect registry.

The law requires that the feasibility study address:

• The costs and benefits of such data collection standards;
• Data collection standards currently employed by each state, Indian tribe or political subdivision of a state;
• Data collection standards that should be considered to establish a model of promising practices; and
• A due process procedure for a national registry.

Legislative History of Congressional Efforts to Establish a National Child Abuse Registry

Almost as soon as states established child abuse registries in the late 1960s and early 1970s there was discussion of blending them into a national registry.  These discussions actually predate the 1974 Child Abuse Prevention and Treatment Act (CAPTA) which forms the basis for many current federal child abuse prevention and treatment activities.  An early study of state central registries suggested in 1978 that there is support from child welfare professionals for setting up a nationwide registry to expand the scope and jurisdiction of central registries and notes that an early draft of federal legislation (P.L. 93-247) [CAPTA] did incorporate provisions for a nationwide registry, but this feature was dropped prior to passage (Gibelman and Grant, 1978, pages 405 and 412).  Despite professional support for a national registry, the same researchers questioned at the time whether the existing registries were achieving their stated goals and whether their potential for abuse of individual civil rights was justified by their effect in protecting children from maltreatment.

The legislative proposal for a national child abuse registry that was ultimately included in the Adam Walsh Act dates back to 2004.  Originally a stand-alone bill first introduced as H.R. 5219 in the 108^th Congress, several versions of the provision were included in House and Senate bills in both the 108^th and 109^th Congresses.  The original proposal was introduced in response to a particular case in which a child had died and it became known only after the fact that the family had a previous history of child maltreatment investigations in a neighboring state.  What began as a stand-alone bill ultimately became part of the Adam Walsh Child Protection and Safety Act of 2006.

Typically the legislative history of a statute includes not only the text of the statute, but also transcripts of hearings, committee reports and other materials that explain and examine the provisions, establish congressional intent, and generally provide insights into sponsors thinking about the legislation enacted.  In the case of Section 633 on a national child abuse registry, little of this background exists.  No hearings were held on the provision in the years before its passage, either as part of the Adam Walsh Act or in consideration of earlier stand-alone versions of the provision.  The report of the Judiciary Committee responsible for the Act does not mention this provision, and while this provision would normally have fallen under the jurisdiction of the committees in the House and Senate with jurisdiction over child abuse programs rather than the Judiciary Committees, those committees² produced no formal statements on the subject as the Act proceeded through the legislative process. Several members made floor speeches about a national child abuse registry as the bill was considered on the House and Senate floor, some in favor and others expressing concerns with the concept³.  However, these statements contain little detail about congressional intent that would provide guidance to resolve ambiguities in the statute that will be described and discussed below.  In addition, the Congressional Budget Offices cost estimate for the bill that ultimately became the Adam Walsh Act (H.R. 4472 in the 109^th Congress) does not estimate costs of a national child abuse registry.

As the Adam Walsh Act was considered by Congress, discussion of the bill as a whole focused on protecting children from unrelated predatory adults and referred primarily to provisions in the Act establishing a sex offender registry as well as efforts to protect children from offenders who contact children via the internet.  With regard to child maltreatment perpetrators, however, because most are parents whose only victims are their own children, a national child abuse registry would be of limited use as a protection from unrelated predators who were the focus of public statements about the legislation.

Then-President Bush signaled support for the registry in his signing statement on the Adam Walsh Act, noting the bill I sign today will help prevent child abuse by creating a National Child Abuse Registry and requiring investigators to do background checks on adoptive and foster parents before they approve [sic] to take custody of a child.  By giving child protective service professionals in all 50 states access to this critical information, we will improve their ability to investigate child abuse cases and help ensure that the vulnerable children are not put into situations of abuse or neglect (White House press release July 27, 2006).

Activities Comprising This Initial Feasibility Assessment

Congress has not, to date, appropriated funds to implement a national child abuse registry.  However, at the direction of the Appropriations Committee (U.S. Congress, House Appropriations Committee, 2007), HHS has conducted an initial feasibility assessment using monies otherwise appropriated for General Department Management.  This interim report identifies and describes a number of issues that would need to be resolved before a national child abuse registry could be implemented in a way that improves child protection investigations and enhances child safety.  Congress has also directed that $500,000 from the Departments FY 2009 appropriations for child abuse discretionary activities be designated for additional work on the feasibility study.  HHS will conduct additional feasibility activities in the coming year and will prepare a final report to the Congress containing results of the additional work when findings are available.  Additional activities that will be conducted as part of the feasibility study are described in the next steps section at the end of this report.

This initial feasibility assesment has sought to:

• Understand existing state child abuse registries and how they operate;
• Identify, as best we are able to do so, congressional intent regarding the purposes of a national child abuse registry and determine how well a registry could fulfill those purposes;
• Describe a number of key issues that would need to be resolved before a national child abuse registry could be implemented, including:
  • Defining the scope of a national child abuse registry in terms of which cases would be included and what information would be collected about each case;
  • Identifying the sources of data that would be used to compile a national child abuse registry and examining issues regarding the availability, currency, validity and comparability of these data across sources as well as the utility of the available data;
  • Determining who would be authorized users of a national child abuse registry and devising how to secure the data and restrict registry access to authorized users and for allowable purposes;
  • Resolving what due process protections should be afforded to perpetrators regarding their inclusion in a national child abuse registry and how those could be provided most effectively;
• Recognize parallels with data systems used for criminal background checks and determine how those systems have addressed related issues; 
• Identify the costs and benefits of implementing a national child abuse registry.

This initial feasibility assessment was conducted internally by HHS staff and included the following activities:

Literature Search.  A search was conducted of academic literature and reports produced by government agencies and other groups to identify information about state child abuse registries as well as available information and statistics on how often interstate issues arise in child abuse and neglect cases.

Analysis of the Statutory Language.  Staff worked with the Departments Office of General Counsel to consider the best interpretation of statutory language within the pertinent section of the Adam Walsh Act and to identify ambiguities and restrictions in the statute as enacted that affect feasibility.

Descriptive Analysis of State Child Abuse Registries.  State laws and policy information were reviewed to describe the characteristics and use of existing state registries and to identify issues that would arise in merging information contained within them.  A number of state court decisions regarding the operation of state child abuse registries were also reviewed.

Identification of Related Federal Activities.  Discussions were held with staff of federal agencies that operate related types of registries and background check systems to consider the parallels and differences with such other systems.  These include various background check systems operated by the Department of Justice as well as the Department of Defenses child abuse registry.  Several related HHS activities were considered as well, including the Centers for Disease Control and Preventions registry of violent child deaths and the Administration for Children and Familys registry of child support cases.

Stakeholder Discussions.  Conversations were held with a wide range of stakeholders to identify implementation issues that would need to be resolved in moving forward with a national child abuse registry.

Examination of Issues with respect to Indian Tribes.  Discussions were held with organizations and persons knowledgeable about Indian tribes child protection systems to consider the particular issues that would arise with regard to Indian tribes participation in a national child abuse registry.

Consideration of Interactions with the Privacy Act.  Discussions were held with HHS experts on the Privacy Act to determine how requirements of that law could be best met during the implementation of a national child abuse registry and how those issues have been handled with respect to related activities in other agencies.  The Privacy Act governs how records are handled within any federal data system in which information is held on individuals that is retrieved by name or other personal identifiers.

Assessment of Due Process Considerations.  Staff researched the due process protections available with respect to state child maltreatment determinations and the listing of perpetrators in state child abuse registries.  Past attention to these issues in legal scholarship and state court decisions was also considered.

Identification of Cost Considerations.  Discussions with knowledgeable individuals were used to identify the categories of costs that would be involved in implementing and managing a national child abuse registry.  These include the costs to the federal government, to agencies submitting data and making inquiries, to individuals identified, truly or falsely, as child maltreatment perpetrators, as well as potential non-financial costs of such a registry.

Characteristics of State Child Abuse Registries

As with most aspects of family law, state statutes rather than federal ones govern the provision of child protective services.  Every state maintains its own definition of child abuse and neglect, determines the parameters for conducting child protection investigations, and specifies the conditions under which the state has the authority to intervene in parental conduct.  States provide most of the funding for child protective services activities.  However, limited federal funding is available to states to operate and improve their child protective services activities, primarily through programs authorized under the Child Abuse Prevention and Treatment Act (CAPTA).  Forty two states spent a combined $312 million in funding from the Social Services Block Grant (SSBG) on child protective services activities in 2006 (HHS, 2008).  SSBG is a broad legislative authority with few specific requirements.  CAPTA funding, in contrast, is more limited but is targeted specifically on child protection activities.  State assurances related to its formula grants to state child protection agencies (funded at approximately $27 million annually in recent years) act to promote a common framework for child protection activities nationally. Within the broad framework provided by CAPTA, however, a great deal of variation exists in how state child protective services activities are conducted in each state (HHS, 2003).  Federal involvement in tribal child protection services is governed by the Indian Child Protection and Family Violence Prevention Act, P.L. 101-690, administered by the Bureau of Indian Affairs within the Department of Interior in cooperation with the Indian Health Service within the Department of Health and Human Services.

State laws regarding child abuse registries have been compiled and analyzed by the Child Welfare Information Gateway, an information service funded by the Administration for Children and Families within HHS (2005a, 2005b, 2005c, 2005d, 2005e and 2005f).  For the purposes of this initial feasibility assessment, several features of these state laws were reviewed.  Descriptive information from several sources on the conduct of state child protective services investigations and management of state child abuse registries was also considered in order to determine the variations in registry features that would affect feasibility of a national registry.  Significant among these was the National Study of Child Protective Services Systems and Reform Efforts (HHS, 2003) that describes state child protective services policies, including those related to child abuse registries.  Appendix B is adapted from that study and provides a state-by-state summary of registry characteristics.

State child abuse registries were created during the 1960s and 1970s to track the history and patterns of past child protective services investigations so that information could be available to aid current investigations.  Federal law does not require that a state maintain a registry, however CAPTA provides that any state that has one expunge information on unsubstantiated cases in files available to the public or used to conduct employment background checks (42 USC 5106(b)(2)(A)(xii)).  Unsubstantiated cases are those for which investigators found insufficient evidence to determine that the alleged maltreatment occurred or to identify the perpetrator.  Substantiated cases are those in which investigators determine, according to state-specific guidelines and definitions, that there is sufficient evidence to confirm the allegation.  States vary in the terms they use for these determinations, and several states have an intermediate category (often labeled indicated) in which there is some evidence to support the allegation but insufficient evidence to meet criteria for substantiation.  States also vary in the level of proof an investigator must demonstrate before substantiating the maltreatment.

CAPTA also requires that states participating in the Basic State Grant Program certify to the federal government that they have in place methods to maintain the confidentiality of records regarding child maltreatment investigations and that limit disclosures to certain types of entities (42 USC 5106a(b)(2)(A)(viii)); that their laws or procedures require the disclosure of confidential information to federal, state or local government entities that have a need for such information to carry out child protection responsibilities (42 USC 5106a(b)(2)(A)(ix)); and that they require cooperation among state law enforcement officials, courts, and human services agencies in the investigation, assessment, prosecution and treatment of child abuse and neglect (42 USC 5106a(b)(2)(A)(xi)).

State Definitions of Abuse and Neglect.  What constitutes abuse or neglect varies somewhat among the states, though nearly all states include basic categories of physical abuse, neglect, sexual abuse, emotional or psychological maltreatment, and medical neglect among the types of maltreatment recognized.  Approximately half the states include both substantiated and unsubstantiated reports in their registries, though these states are required by federal law to restrict non-investigative disclosures to substantiated cases (section 106(b)(2)(A)(xii) of the Child Abuse Prevention and Treatment Act, 42 U.S.C. 5106a).  Most of the other states (23 in 2003) limit registry entries to substantiated cases, though not all use the substantiated terminology.  Several states further restrict what cases are included in their registries.  For instance, West Virginia includes only criminal convictions in its registry, South Carolina includes only cases in which the child abuse finding has been confirmed in court, and Missouri also includes only adjudicated cases.  Michigan includes in its registry only certain categories of child abuse and neglect (HHS, 2003).  The Adam Walsh Act specifies that a national child abuse registry would include only substantiated cases.

Types of Perpetrators Investigated by Child Protective Services.  State statutes typically limit the jurisdiction of CPS agencies to maltreatment committed by parents or other persons in a caretaking role with respect to a child.  Perpetrators who are strangers to the child, or who are known to the child but are not responsible for the childs care, would be investigated by law enforcement agencies rather than child protective services.  States vary as to whether investigations of maltreatment by nonparental caregivers such as day care providers, foster parents, or staff of residential treatment facilities is handled by the child protective services agency or by other authorities such as licensing boards or police departments.  According to recent national data, 80 percent of child maltreatment perpetrators in cases investigated by the child protective services agency were the parents of the victim and 60 percent had committed neglect, which was by far the most common type of maltreatment (HHS, 2009).

Existence of State Child Abuse Registries.  According to an analysis of state child abuse statutes, as of 2005, forty two states, the District of Columbia, Puerto Rico and two territories (American Samoa and Guam) had statutorily created child abuse registries.  No registries were, at that time, required by statute in Colorado, Kentucky, Maine, Minnesota, West Virginia and Wisconsin.  However, additional states may operate child abuse registries operated through policy only (Child Welfare Information Gateway, 2005c; Welfare Information Gateway, 2005d).  In addition, though not mentioned in the Adam Walsh Act, the Department of Defense operates a child abuse registry with respect to maltreatment that occurs at military installations worldwide⁴.  Of the states whose statutes do not require registries, four (Colorado, Kentucky, Maine and West Virginia) responded to a survey that was part of the 2003 National Study of Child Protective Services Systems and Reform Efforts with information that their states did operate child abuse registries.  An additional three states (Georgia, Florida, and Oklahoma) responded to the 2003 study that their states did not operate registries, though in some of these states administrative data systems served some of the functions of a registry by making historical information about investigations available to investigators.  Georgias registry was ruled unconstitutional by that states superior court in 1998 (HHS, 2008b).  As will be discussed in more detail later in this report, a number of state child abuse registries have been subject to legal challenge, and modifications to registry operations have been required by courts in several states.

Uses of State Child Abuse Registries.  All states with registries use the information within them to provide historical information about families contact with child protective services to its own investigators as background to inform current investigations.  In addition, as of 2003, 44 states reported using their child abuse registries also to conduct child abuse history screenings of prospective foster and adoptive parents, persons applying for employment or volunteer opportunities working with children or other vulnerable populations, and/or persons applying for licenses to operate child care programs (HHS, 2003).

Organization and Management of State Child Abuse Registries.  States vary in how they organize and administer their child abuse registries and in what cases and what information about each case is included in a registry. Some are operated by the states child protective services agency, while in other states, including California, the registry is maintained by the Attorney Generals office.  In some cases, registry files are maintained at the state level, while in others the states system merely references the existence of an original file at the local level.  Most registries were originally separate, stand-alone information systems (initially as paper files, only later computerized).  However, in many states these systems now operate as a function of a larger child protective services intake system, or as part of a State Automated Child Welfare Information System (SACWIS).  States that implement a SACWIS using federal funding under title IV-E of the Social Security Act are required to record investigation decisions in their systems and, by law, a SACWIS is required to support the reporting of data to report data to the National Child Abuse and Neglect Data System.  An interface with the states central child abuse registry, if it has one, is optional (HHS, no date).  Forty states and the District of Columbia have either implemented or are developing a SACWIS.

Lack of Adjudication of Most Cases Included in Registries.  It is important for a discussion of a national registry of child abuse perpetrators to recognize that few child abuse cases are adjudicated and that a substantiation decision does not mean there has been a court finding in the case.  In fact, data for 2007 indicate that nationally only nineteen percent of substantiated cases that year involved a court action or petition.  There was considerable range among the states in the frequency with which substantiation decisions were accompanied by a court action.  Four states report court action in less than five percent of cases, while another four states report court action in over half of cases.  Substantiation also does not necessarily imply that the child welfare agency thought the child was in ongoing danger or need of protection.  In 2007, nearly 40 percent of victims received no services following the investigation, while just 21 percent of children were placed in foster care (HHS, 2009).

Purpose and Utility of a National Child Abuse Registry

A key factor in examining the feasibility and potential operational issues regarding a national child abuse registry is to understand the purposes for which it is created and whether a registry would fulfill these purposes.  This section describes available information on these topics in an effort to determine:

• For what purposes could information in a national child abuse registry be disclosed?
• Would a national child abuse registry, as described in the Adam Walsh Act, significantly improve child safety?

Intended Uses of a National Child Abuse Registry

The Adam Walsh Child Protection and Safety Act does not specifically indicate the purposes for a national child abuse registry.  Purposes must therefore be inferred from the statutes provisions regarding what information is to be contained in the registry and who is allowed access to the information and under what circumstances.  The Act notes that the national child abuse registry is to include only the names of child maltreatment perpetrators and information about the nature of the maltreatment perpetrated.  Access to the information is restricted to government agencies, or their agents, whose mission includes protecting children.  The Act does not specify the specific circumstances or purposes for which permissible users may access information about an individual.  The Act also does not specify whether the registry would include information on cases substantiated prior to enactment or the establishment of a national registry.

The broad requirement that disclosures of information included in the registry may only be to government agencies whose mission includes protecting children strongly implies that one of the purposes of the information is to increase child safety by increasing investigators knowledge of past substantiated maltreatment allegations.  Information about maltreatment history helps investigators to assess risk of harm to a child and understand the level of services a family requires in order for a child to remain in the home safely.  National child maltreatment history checks may assist investigators to detect undisclosed substantiated maltreatment incidents in states other than the one investigating the current incident.

It is less clear whether Congress intends that the information also be used for background checks in licensing and employment situations.  Background checks of the registry for non-investigation purposes are not explicitly mentioned in the Act or in statements about the bill that were made on the House and Senate floor.  However, most states use their registries for some employment and licensing purposes, particularly to detect whether persons applying for jobs or volunteer positions working with children or other vulnerable populations have a history of child maltreatment, and in at least some states most inquiries to the state registry are for these non-investigation purposes.  It must be recognized, however, that the denial of employment brings a potential financial impact to individuals included in the registry (or misidentified as such due to false positive matches) that would expose a system to legal challenge in ways that investigative uses alone would not (Phillips, 1993).

As the law is currently drafted, HHS believes that employment checks against a national registry are a permitted use of the registry by an entity allowed access to it, i.e., a federal, state, local or tribal entity (or agent of such entities) with a need for the information in order to carry out responsibilities to protect children from abuse or neglect.  Because the purpose of such checks is to protect children or other vulnerable populations, such a use would be permitted.  Problems are likely to arise, however, if a national registry permitted information provided by a state to be used in ways that were not allowed within the state.  In addition, inadequate procedural protections for alleged perpetrators during investigations may also draw legal challenge to a national registry which uses the results of those investigations and, as a result, negative consequences, either personal or financial, follow for the individual.  For a more detailed discussion of this issue, see the section elsewhere in this report headed Due Process Considerations.

Quantifying the Potential for Child Safety Improvements

An obvious question in considering the benefits of a national child abuse registry is how often families have substantiated investigations in more than one state.  Of particular concern would be cases of serious or fatal maltreatment in which there were previous out-of-state incidents, the knowledge of which might have led to different decisions by CPS staff which might have prevented the injury.  Child abuse history checks using a national registry would provide most added benefit in cases in which the individual had one or more substantiated cases in another state but had neither past in-state substantiations (which would be revealed when checking the states own registry) nor out-of-state criminal convictions related to the past investigations (since convictions would be detected through existing criminal background checks).  For the purposes of this initial feasibility assessment, staff sought information regarding the frequency with which such cases occur.  However, there is no available data on children who have been victimized in multiple states or on the mobility of abusive or neglectful families.  Most experts consulted during this inquiry thought the number is likely to be relatively small, except possibly in metropolitan areas that span multiple states.  It should be noted that nearly all the anecdotal cases related to staff in support of a national registry related to the movement of families among nearby states.  The remainder of this section describes related information staff were able to identify.

Family Mobility.  Census data reveal that between 2006 and 2007, 1.6 percent of families with children age 0 to 17 moved between states.  This includes 0.5 percent whose moves were within the same census division (a several state area), 0.3 percent whose moves were outside the division but within the same census region, and 0.8 percent who moved to a different census region within the U.S.  Mobility for female-headed families with children was very close to those of families in general (calculations based on U.S. Census Bureau, 2008).

Maltreatment Recurrence.  Maltreatment recurrence in general has been studied, primarily using within-state data.  Reduction of maltreatment recurrence is an outcome measured in federal reviews of state child welfare systems.  2005 data indicate that 6.6 percent of children had a second substantiated (within-state) maltreatment report within six months of an initial report that year (HHS, 2008).  A study using a longer time horizon found that 5.2 percent of children substantiated for maltreatment had been revictimized within five months (that is, they had a second substantiated maltreatment report), 8.2 percent had been revictimized within one year, and 16.7 percent had been revictimized within five years (HHS, 2005).  Only substantiations within the same state could be detected using the data.

Current Practice Regarding Interstate Registry Checks.  Several state officials contacted during the course of research for this report were asked to characterize how frequently interstate issues arose during child maltreatment investigations.  None had specific data on this, but the impressionistic consensus was that such issues were fairly common, though not routine.  None were able to quantify the number of requests made for access to other states child abuse registries for investigative purposes, or the number of out-of-state requests received for access to their registries.  Most believed these issues came up with greatest frequency with respect to neighboring states, particularly where a metropolitan area may span multiple states.

Existing Interstate State Child Abuse Registry Checks of Prospective Foster and Adoptive Parents.  States have been required since 2006 (by another provision of the Adam Walsh Act) to conduct out-of-state child abuse history checks for prospective foster and adoptive parents in any state the individual has lived in within the past five years.  While no figures are available regarding the number of such checks conducted, state experience in getting information back from one another has varied.  While many states report no difficulties responding to such requests quickly, and in getting responses to inquiries, others have reported that responses may take weeks.  In addition, several states do not have the legal authority to respond to such inquiries (National Resource Center on Family Centered Practice and Permanency Planning, 2008).

Employment Background Checks for Persons Seeking Work with Children.  With respect to employment-related background checks, staff sought available information on the volume of checks conducted against state child abuse registries and criminal history systems for categories of persons that would likely undergo a federal child abuse registry check if such a use were allowed.  The following information includes only a few states, but provides some sense of the scale involved in such requests.  California in 2004 conducted 10,000 in-state child abuse registry checks for investigative purposes and 40,000 for non-investigative (primarily employment and licensing) purposes (California Office of the Attorney General, 2004).  Thus employment-related checks represented 80 percent of its registry workload.  In 2001, 19 states reported conducting 650,000 criminal background checks of prospective child care employees, the largest category of persons potentially subject to employment-related child abuse history checks (SEARCH, 2002).  To put these figures in perspective, in 2007, nearly 2 million child maltreatment investigations were conducted nationally; substantiated cases included approximately 859,000 perpetrators (HHS, 2009).

In summary, there is no available evidence of substantial cross-state child maltreatment patterns and the limited anecdotal information available involves primarily cases with a regional rather than a national character.  The volume of inquiries to a national child abuse registry anticipated would depend especially on whether employment-related inquiries were allowed in addition to investigative ones, and whether inquiries would be made routinely or only with respect to cases in which a family were new to the jurisdiction or there was otherwise reason to suspect a possible cross-state aspect to the case.  No research evidence or data were identified to indicate that a national child abuse registry would significantly improve child safety. However, the evidence base is nearly nonexistent and we cannot rule out a phenomenon that would not be apparent from existing information sources. Because of the paucity of available information on this important topic, further study of the frequency with which perpetrators have substantiated maltreatment reports in multiple states will be a priority for the next stage of feasibility work.

Acquisition and Management of Data

A number of critical issues would need to be examined in deciding how best to implement a national child abuse registry.  This section describes those that will be most important to consider as Congress and HHS decide how or whether to move forward to implement a registry or other steps to improve interstate sharing of information regarding child maltreatment perpetrators.  Key feasibility questions in this regard include:

• Would state, local and tribal child protection agencies provide data to HHS for this purpose voluntarily?  What assurances and/or incentives would they require before doing so?
• Could perpetrators be identified reliably in a national child abuse registry?
• How would access to registry information be securely restricted to authorized users making inquiries for legitimate purposes?
• Is a centralized electronic database as specified in the Adam Walsh Act the most effective approach to managing data for a national registry?

Obtaining Data for a National Registry

A key risk in establishing a national registry is that HHS may create a database to which few jurisdictions would provide data, making it of little practical value.  A national child abuse registry would be useful only if its coverage were sufficiently national in scope to instill confidence among users that known maltreatment perpetrators could be identified in it, replacing the need for bilateral state-to-state inquiries.

Provision of Data by State, Local and Tribal Agencies

Many states are likely to be reluctant to hand over sensitive information on families to the federal government.  Many do not currently have the clear legal authority to do so.  Others will be concerned about the security and confidentiality of information they provide.  Particularly since participation in a national child abuse registry is to be voluntary, it will be essential to work with agencies whose data would be included to build a registry into a tool they would value and the utility of which would outweigh the burdens involved in providing the information and concerns about its use and potential misuse.  While some states view a national registry as a potentially useful tool, making it a reality would take a collaborative development process to negotiate the content, transmission, storage, use and protections provided regarding the information.

In order to get an idea of the potential size of a database of child abuse perpetrators, staff sought available information on the size of state registry databases and the volume of inquiries made to state registries annually.  No systematic information was available on the topic, though figures were identified for two large states.  Published sources reveal that, as of 2004, California maintained 905,000 entries in its central registry on 810,000 individual perpetrators and added approximately 35,000 new reports each year.  California also reports querying the central registry in 10,000 investigative inquiries and 40,000 non-investigative inquires annually.  Of the non-investigative inquiries (primarily employment related child abuse history checks), California finds a 5% match rate (California Office of the Attorney General, 2004).  Illinois adds 14,000 new reports to its State Central Registry annually, according to a 2008 news story (Pawlaczyk and Hundsdorfer, 2008).  It is not clear whether Congress intends that a national child abuse registry include cases substantiated prior to a registrys establishment.  More systematic information on the volume of state child abuse registries will be sought in the next phase of the feasibility study.

For the purposes of this feasibility study, state laws were reviewed regarding to whom and in what situations information about child maltreatment may be released (as compiled by the Child Welfare Information Gateway, 2005b).  This review revealed that few states currently have the legal authority to provide data about perpetrators to a national registry, or to have their data searched for such information through a central, federal mechanism. Most states would need to modify their laws if they choose to participate in a national child abuse registry.  In addition, while some states laws will likely permit sharing of information with the federal government, the re-disclosure of that information to other parties in response to registry queries adds a further level of complexity to the authority required that would need to be addressed in many state laws before a national registry were feasible.

Managing Data from Many Sources

Obtaining data for a national child abuse registry is vastly complicated by the language in the Adam Walsh Act that allows political subdivisions of a state to supply data directly to the federal government for inclusion in the registry.  Presumably states with county operated child protective services systems would be most interested in this feature.  For HHS, however, this provision makes implementation much more difficult because it multiplies, potentially by many times, the number of agencies whose information systems must be compatible with a new federal database and from which data must be transferred securely.  In addition, these agencies are not ones with whom HHS has established organizational working arrangements.  Our existing child protection programs, with the exception of small competitive demonstration grants, all interface with state agencies rather than with local government units.  For states that have developed SACWIS systems, no child welfare data should reside in any county specific system.  Yet even in non-SACWIS states that are state supervised and county administered, requiring HHS to negotiate individually with local jurisdictions for the delivery of information to a national child abuse registry seems unnecessarily inefficient and costly.

Alternative Structures for a National, Electronic Child Abuse Registry

The Adam Walsh Act requires HHS to collect data about perpetrators in a central electronic registry and specifies that the information in the registry shall be supplied by States and Indian tribes, or, at the option of a State, by political subdivisions of such State, to the Secretary of Health and Human Services.  This construction implies that data would be contained in a federal database owned, managed and controlled by the federal government.  Input received from stakeholders and other agencies operating systems that manage records used for background checks on individuals suggest that HHS consider alternative structures if implementation proceeds.  Statutory change would be required before such alternative structures could be considered.

Some experts consulted for this report urged that HHS (and Congress) consider potential advantages of a pointer system in which rather than holding records in a federal database, a central registry instead provide a mechanism by which state and/or local records could be searched through a federally operated search mechanism, but in which the actual records would be maintained by states in a decentralized fashion.  In this way, HHS would facilitate inter-state information sharing, but would not itself hold the records.  The most important advantage of not holding records in a federal database is that it would limit federal responsibility for the accuracy of information that is not created by federal agencies and that cannot be validated at the federal level.  While such a system would not preclude the need to comply with the Privacy Acts requirements (described elsewhere in this report) that records in federal systems be reliable, accurate, timely and complete, a referral system would place considerably less detail in a federal records system and would place the national registry in the role of facilitator of information exchange rather than the definitive source of case details.  In a decentralized system, the state would maintain control over which records are made available to a search mechanism.  It would also be possible, either in a centralized or decentralized fashion, to design a system in which different levels of information were available in response to different types of inquiries.  For instance, investigative inquiries could return something different than employment-related inquiries.

Standardization and Comparability of Registry Information

Whether records are held in a federal database or the federal government constructs a mechanism to search records held by participating state, local or tribal agencies, the establishment of a national child abuse registry necessitates consideration of the extent to which records should or can be standardized to ensure comparability.  The relevant provision of the Adam Walsh Act states that this section shall not be construed to require a State, Indian tribe, or political subdivision of a State to modifyan equivalent register of cases of child abuse or neglect that it maintains[or] any other record relating to child abuse or neglect.  HHS could not, therefore, require any changes to what is in state systems, though it could set minimum standards about the types of records that would be included in a national registry and about the specific information that would need to be present before an entry were made available in response to queries.  To make such a system useful it will be important to work closely with state, local and tribal agencies expected to submit information to a registry to develop and encourage the voluntary adoption of data standards that would enable the accurate interpretation of information in registry records while including as many jurisdictions as possible.  Alternatively, Congress could require participation in the registry and penalize the failure to submit information to a national registry.  However, unless funding is provided to offset state and local costs of compliance, such mandates would be burdensome to CPS agencies.  Experts consulted during the development of this report cautioned that the content and quality of information in state registry records currently is variable regarding both the identifying information about the perpetrator and what fields are available to characterize the nature of the abuse and neglect.  These issues are discussed next, along with a discussion of how to define the universe of cases that should be considered for inclusion in a national child abuse registry.

Determining Which Cases Should be Included in a National Child Abuse Registry

The Adam Walsh Act specifies that only substantiated (that is, proven) cases be included in a national child abuse registry.  This specification is important since many states include also unsubstantiated cases in their own registries for use during investigations (but not for other purposes).  During consultations there was consensus that the limitation of a federal child abuse registry to substantiated cases, as is currently specified in the law, is essential.  However, some stakeholders suggested that HHS further limit cases in a number of ways, or at least ensure that equivalent types of cases are included from each jurisdiction.  For instance, decisions would need to be made as to whether child maltreatment perpetrated by caretakers other than parents would be included in the registry since such cases are investigated by child protective services in some states and not others.  The legislation also does not specify whether cases substantiated prior to the passage of the Adam Walsh Act or the establishment of a national registry would be included.  The records of juvenile perpetrators are typically provided more protection than adults in criminal cases, though congressional intent with respect to the inclusion of juvenile perpetrators in a child abuse registry is unclear.  In addition, a number of experts suggested that HHS consider how to handle cases in which the individual named as perpetrator was not directly responsible for the childs maltreatment.  Most typically this would include a woman abused by a spouse or partner whose maltreatment substantiation relates to failing to protect her child from the same abuser, or from witnessing her abuse.  The inclusion of such individuals in a registry that could be used to deny them future employment in fields working with children or vulnerable populations was seen as further victimizing an extremely vulnerable population.  However, such cases may not be readily identifiable in state data if a decision were made to exclude them from a national registry.

A number of experts consulted suggested that a registry should be limited to cases in which a minimum standard of due process protections, to be defined federally either in statute or regulation, had been afforded to the perpetrator.  The extent of these protections would depend in large part on whether a registry were to be used only for investigative purposes or also for non investigative purposes such as employment background checks.  Some went further, to suggest that only cases that had been adjudicated should be included in a national registry, further assuring that perpetrators rights to due process are addressed before a record is available to the federal government and authorized users through a national registry.  Such restrictions assure the rights of persons accused of child maltreatment and would make a federal child abuse registry less susceptible to the sorts of legal challenges that have been an issue with respect to a number of states registries and which are described elsewhere in this report.  Such restrictions, however, would either result in the exclusion of many states data, or would necessitate costly changes to strengthen investigatory procedures in many states.  Due process issues are described in more detail elsewhere in this report.

Identifying Perpetrators Accurately

As has been noted, the Act specifies that the only personal identifier that may be included in a national registry is the perpetrators name.  This limitation is extremely problematic because many names are too common to allow for confident identification.  It is not feasible to implement a child abuse registry using an individuals name as the only personal identifier.  Consider, for instance, if a state were conducting an investigation in which the perpetrators name were Michael Jones.  If that name were run through a national registry there might be hundreds of matches, with no way to differentiate whether any were the particular individual in question.  As an illustration, staff looked up several common names in an online phone directory.  We found 45 persons named Michael Jones in Los Angeles alone.  Similarly, there are 89 listings for persons named Linda Smith in Utah and 70 David Millers in Nebraska.  Names are simply too common to allow for definitive matches without additional information.

A number of experts were consulted about what identifying information would be needed to make a federal registry practical.  In addition, a number of states forms currently used to request registry checks were reviewed to determine what fields they ask requesters to specify.  States typically use a variety of fields to conduct searches of their registries.  At minimum, they usually require the individuals date of birth, and most also use the individuals sex, race, Social Security Number, and often drivers license number.  Last known address and names and ages of their children are also often used.  For a national system, identifying the state and local jurisdiction where the case originated would also be important.  In establishing a national registry it would be desirable to have more flexible authority to specify in regulation or other policy guidance the fields necessary for establishing the identity of the perpetrator.  However, if perpetrators Social Security Numbers are to be among the fields included, because their use is restricted for privacy and security purposes, explicit legislative authority for their collection and use would be needed.

Even with additional fields, making a positive identification without biometric information, such as fingerprints, is difficult.  It is for this reason that criminal justice databases have largely moved to using fingerprints in addition to name-based identification systems.  A study conducted for the Department of Justice in the late 1990s regarding the efficacy of name-based matches of criminal justice records found that name-based searches of criminal records (which typically also use date of birth as a match criterion) have very high error rates (SEARCH, 1999).  That study found 5.5 percent of queries resulted in a false positive (i.e. the query indicated the individual had a criminal history when he or she did not), and 11.7 percent of queries resulted in a false negative (that is, finding no criminal record when the individual had one).  High error rates for name-based matches of a national child abuse registry are also likely.  If error rates for a name-based child abuse registry prove similar to those found in criminal justice records, every 100,000 checks run against a federal registry would result in approximately 5,000 false positive results and 10,000 false negatives.  With name as the only identifier, false positive matches would be much more frequent.

A national child abuse registry could not be constructed without a legislative change to permit the inclusion of sufficient identifying information on perpetrators to allow a name-based match.  Even with broader identifying information, however, significant numbers of false positive and false negative results must be expected.  Moving toward fingerprint matches would provide better accuracy, however we are aware of no state that currently uses fingerprints to identify perpetrators in their child abuse registries.  Moving to such a system would be complex and expensive to implement.  It would also move the child protection system much farther away from its social service origins and further toward a criminal justice orientation, which would be quite controversial.

Characterizing Maltreatment

During the preparation of this report, staff talked to a number of experts and stakeholders about how state child abuse registries and other child abuse and neglect information systems characterize the nature of maltreatment incidents, the only information beyond the perpetrators name that could be included in a national child abuse registry, under the Adam Walsh Act.  These conversations indicated that information on the nature of maltreatment incidents is currently quite variable.  The National Child Abuse and Neglect Data System (NCANDS), which collects from states case level information on child maltreatment reports and investigations, characterizes maltreatment by dividing incidents into five major categories:  physical abuse, neglect, sexual abuse, emotional maltreatment, and medical neglect.  An other category captures less common maltreatment types, including forms that may not be considered maltreatment in all states.  Cases may involve more than one type.  No severity measures are included, though their value has been considered a number of times⁵.  Available measures were always deemed too burdensome for administrative purposes or unreliable when used in non-research settings (i.e. training and monitoring of raters is required for consistency and they are more time consuming to use than is typically possible in a field setting).

Experts consulted for this report noted that state child abuse registry entries often include check lists of characteristics associated with the maltreatment as well as a text field providing a narrative description of the case.  It would be possible to include some of this information in a federal database, but without requiring changes to state and local systems, the resulting entries would be quite inconsistent.  Federal guidance to states that have developed SACWIS systems have encouraged the use of discrete data fields to identify types of maltreatment (HHS, no date).  Stakeholders suggested that important information characterizing the maltreatment would also include the perpetrators relationship to the victim as well as the age and sex of the victim.

Standard child maltreatment definitions have been developed and validated for use in the Department of Defenses Family Advocacy Programs which address maltreatment within military families (Slep and Heyman, 2006).  However, since states have jurisdiction over most family law, it is unlikely the federal government would seek to impose conformity with respect to child maltreatment definitions, particularly based on definitions designed for use with military families rather than the general public.

The Centers for Disease Control and Prevention has developed standardized child maltreatment definitions suggested for use in public health surveillance (CDC, 2008).  These standards, however, are designed for public health purposes and include no items about perpetrators.  To the extent state systems move toward these definitions they do have the potential to improve the uniformity of how child maltreatment is classified in administrative data systems.  However, these definitions do not necessarily track states legal definitions of abuse and neglect, and moving toward uniformity using these definitions would be a long process necessitating significant changes in state and local data systems.  While some states may ultimately be interested in moving toward such uniformity, the Adam Walsh Act does not allow HHS to require changes to child maltreatment record keeping.  The definitions published by the CDC go into much greater detail regarding the nature of the child maltreatment incident than does NCANDS or than is now typically captured by states administrative data systems.  For instance, where NCANDS will record that a child suffered physical abuse, the CDC data elements include fields describing whether the maltreatment resulted in illness or injury, the nature of the injury, the mechanism of the injury, the location of the injury, and more.  Similarly, there are 11 data elements describing sexual abuse.  Moving toward these definitions would involve significant change to CPS agencies data collection practices.

Maintaining a Registry and Responding to Inquiries

Assuming data for a registry were successfully compiled, additional challenges await pertaining to keeping that data secure, restricting access to authorized users, and providing for the ongoing maintenance and update of the information included in the registry.  This section of the report addresses these issues.

Keeping Registry Data Secure

Because it would contain extremely sensitive personal data, it would be important to design a national child abuse registry in ways that incorporate a high level of security to assure the system is not susceptible to unauthorized disclosures.  Typically these security measures would include requirements for encrypting data, establishing firewalls, and enforcing strong user authentication.  Some systems require dedicated telecommunications lines for the transmission of inquiries and responses, equipment dedicated to the function and not used by unauthorized personnel; and that terminals used for queries are kept in locked offices accessible only to persons authorized to use the system.  Any reasonably secure system will involve significant up front and ongoing costs to both the federal government and to users for hardware and software purchases, and technical assistance to ensure its appropriate use.  More information on the costs and benefits of a national child abuse registry appear elsewhere in this report.

Restricting Access to a National Child Abuse Registry to Authorized Users

The sensitivity of the information in a national child abuse registry makes it essential that access to the information be restricted to those with a legitimate need for the information.  The Adam Walsh Act restricts access to any Federal, State, Indian tribe, or local government entity, or any agent of such entities, that has a need for such information in order to carry out its responsibilities under law to protect children from child abuse and neglect, and requires HHS to establish standards for the dissemination of information in the national registry of substantiated cases of child abuse or neglect that comply with CAPTA provisions that limit public disclosure of data on individuals while encouraging cooperation in prosecution and treatment provision.

While states have procedures of their own that restrict access of registry information to authorized users for legitimate child protection purposes, circumstances arise occasionally in which information is misused.  For instance, in June 2008 three employees of New Yorks state child welfare agency were arrested for using information from that states child abuse registry for personal purposes.  This included providing information to friends for use in disputes with estranged family members (Capital 9 News, 2008).  While incidents such as these are not common, they do illustrate the importance of carefully regulating and monitoring access to the very sensitive information included in a child abuse registry.

Prior to implementation, HHS would need to define further what it means for a user to be an entity carrying out responsibilities to protect children from abuse or neglect.  Most likely is that this term would be defined as it is in CAPTA (section 106(b)(2)(ix) and 45 CFR 1340.14(i)), meaning agencies with access to a national child abuse registry are those that, in theory, are currently permitted to receive child abuse history information from other CPS agencies.  In addition, procedures would need to be established to verify that requests for background checks are from authorized agencies and only for legitimate purposes, which also must be defined.  These would be among the most significant tasks involved in implementing a national child abuse registry.  HHS does not currently have established programmatic relationships with most agencies that investigate child maltreatment, only the 56 state and territorial agencies that receive funding under CAPTAs Basic State Grant Program.  By contrast there are approximately 2,600 local child protection agencies nationally with tens of thousands of investigators (HHS, 2003).  Other potential users include staff in the over 17,000 state and local law enforcement agencies nationally.  Monitoring state and local access directly would be a monumental task.  Department of Justice experience with this issue in relation to criminal history records is discussed elsewhere in this report.

Responding Promptly to Queries

A key issue for many in considering how useful a national child abuse registry would be is whether inquiries would result in prompt responses.  In order to be useful for investigative inquiries, most stakeholders suggested that they would prefer responses to be immediate, but would find them at least somewhat useful if received within a couple of days.  Longer response times were considered of less utility in investigative situations, since decisions on safety often must be made quickly.  Response times of up to a week were considered by most to be useful in some situations, such as in response to licensing or employment related inquiries.  Many states, though not all, have been able to meet such timelines with respect to recent requirements to conduct out of state child abuse registry checks of prospective foster and adoptive parents.  It is less clear that states could respond as quickly if the volume of interstate checks grew significantly.  Automating such searches via a national database would help speed up this process (if, as discussed above, identifications could be made reliably).  A national system that simply passed requests through to states but did not necessarily return responses promptly was viewed less favorably.

Related to the issue of the time it takes to respond to an inquiry is how much of a time lag there is in the data in a registry system.  That is, are the data updated in real time, or is there a lag of days, weeks, or months between the time a report is substantiated and the time it is available through a registry system?  The National Child Abuse and Neglect Data System currently receives annual updates from states, and the Adoption and Foster Care Analysis and Reporting System is updated semi-annually.  The Department of Defense updates its child maltreatment registry on a quarterly basis, though individual services information systems are updated more frequently⁶.  The longer the lag time, the greater would be the possibility that recent relevant reports on the perpetrator are missed.  Frequent updates, however, would add to the logistical challenges and expense of maintaining a national child abuse registry.

Determining a Registrys Relationship to Other Data Collection Efforts

Since 1990, the Childrens Bureau has collected annually from states information about abused and neglected children known to state child protection agencies.  The National Child Abuse and Neglect Data System (NCANDS), through which this information is collected, is a voluntary system that describes child maltreatment reports, investigations, victims, and perpetrators.  Forty nine states, the District of Columbia, and Puerto Rico provided at least some data to NCANDS in 2006, though not all states provide data on all topics covered by the system.  All information is provided to HHS by state government agencies.  No information is submitted directly by local government entities and no data is collected from Indian tribes.

Child-level data is submitted to HHS and compiled by a contractor into an annual report describing the outcomes of child protective services investigations nationally and by state.  The data is also used to calculate performance measures by which the Childrens Bureau monitors child welfare program performance.  Further, the data is available to researchers for analysis through the Childrens Bureaus National Child Abuse and Neglect Data Archive.

The information collected in NCANDS does not currently include identifying information on either child maltreatment victims or perpetrators and therefore cannot, in its current form, serve as a national child abuse registry.  However, in establishing such a registry one option would be to use NCANDS as a starting point.  A list of child-level data elements collected within NCANDS appears at Appendix C.  NCANDS includes elements that:

• describe the characteristics of each child abuse report (e.g. the date of the report, the category of person who reported the maltreatment, and the disposition of the report);
• provide demographics of the child;
• identify the type(s) of abuse or neglect in several broad categories;
• list whether any of several child risk factors and caregiver risk factors are present;
• detail whether any of a series of services were provided to the family;
• describe the demographics of the perpetrator(s);
• quantify maltreatment-related fatalities.

Comparing data currently collected by NCANDS with that specified in the Adam Walsh Act for inclusion in a national child abuse registry reveals there to be relatively little overlap between the two.  The Act specifies that information in a national registry be limited to the name of the perpetrator and the nature of the abuse and neglect.  Names are not currently included in NCANDS and the nature of the maltreatment is described only by maltreatment type, i.e. physical abuse, neglect, sexual abuse, emotional maltreatment, medical neglect or other.  NCANDS does describe the perpetrators relationship to the victim (i.e. mother, father, etc.), but contains no descriptive information on the maltreatment incident, no assessment of the severity of the maltreatment, or of any resulting injuries or harm to the child (except in the case of fatalities, which are noted).

Experts consulted during the preparation of this report believed that a national child abuse registry could be constructed as an add-on component to the existing NCANDS data system.  Several observed, however, that building a national child abuse registry into NCANDS would fundamentally alter the nature of the data system.  Currently the focus of NCANDS is on understanding the broad phenomenon of child maltreatment and trends in the reports to and determinations of maltreatment made by the child protective services system.  A focus on identifying individual perpetrators and using the information to perform background checks would fundamentally alter its purpose and orientation.  Were NCANDS to be used as the basis for a national registry, additional fields would need to be added with information identifying the perpetrator and to document additional characteristics of the maltreatment.  Other sections of this report on standardization and data comparability describe these issues in further detail.

Complying with the Privacy Act

The Privacy Act establishes rules by which federal agencies must manage records that include personally identifiable information about individuals that is retrieved by the individuals name or another personal identifier.  To the extent records that become part of a national child abuse registry are controlled, maintained, and, potentially, disclosed by a federal agency, those records would be subject to the Privacy Act.  To comply with the Privacy Act, HHS would need to develop procedures to protect the information and to limit its use to intended purposes.  These procedures include the steps HHS will take to ensure that records included in a national child abuse registry are accurate, relevant, timely and complete.  The Privacy Act further requires that a notice in the Federal Register describe:

• The types of individuals covered by the data system;
• The nature of the information to be included in a national child abuse registry;
• Procedures that HHS will implement for responding to individuals inquiries about whether the system contains records on them, as well as to allow individuals to access records about them and to amend their records to correct any part that is not accurate, relevant, timely or complete;
• Circumstances under which data will be disclosed, to whom, and for what purpose;
• How long information will be retained by the federal government; and
• Physical, technical, and administrative safeguards that will be used to protect the information in the data system.

Because this database would be used for civil law enforcement purposes, HHS may choose to exempt itself from Privacy Act requirements regarding certain provisions, including the requirement to provide records to the subject individual upon request in some, but not all, circumstances.  Even if such an exemption were asserted, however, HHS would still need to adhere to other Privacy Act mandates regarding the maintenance, storage, and disclosure of data.  Because the records to be included in a national child abuse registry would not originate with the federal government, ensuring their accuracy and completeness would be challenging.  The key question is likely to become how much verification at the federal level would be required before information in a national child abuse registry is disclosed to an authorized requester.

If a national child abuse registry is created, the Department  would be required to ensure that the records are maintained with such accuracy, relevance, timeliness, and completeness as is reasonably necessary to assure fairness to the individual in [a] determination about that individual. Congressional direction to establish a national child abuse registry would satisfy the relevance requirement.  However, if information included in the registry includes textual notes or other information appended to the record, each such entry would need to be reviewed to ensure that it did not include extraneous information in addition to that which was necessary. Timeliness would require a standard for automatic deletion of records after certain periods of time, unless they were the subject of litigation.

Assuring accuracy and completeness is more complex.  Case law is mixed regarding the level of verification required before highly sensitive information about individuals originating from state records is disclosed by the federal government.  It is possible that each time a record were disseminated HHS would be required first to verify with the state that submitted the record that the state does, in fact, have a record on the individual, that the registrys copy of the information is accurate and includes the entire record that is intended to be kept, that the state intended to file the record in the national child abuse registry, and that the record has not been withdrawn or changed since the time the state filed it with us.  However the accuracy and completeness tests are met, the federal government would likely receive requests from the subject of the record to amend the record.  If we did not promulgate an exemption rule, at the very least, HHS would need to file a notation in the record or a statement of objection from the subject, and then disclose the statement along with the record each time it is disseminated.  The Department would also need a process to address potential inaccuracies that come to our attention, unless a decision is made to exempt the system from this aspect of the Privacy Act.

More specificity regarding steps that would meet Privacy Act requirements depends on decisions regarding the scope of registry records and the parameters placed around disclosures of those records.  The required activities will certainly involve a significant level of effort at the federal level as well as burden to state, local and tribal agencies whose records become part of a national child abuse registry.  The section of this report addressing due process issues discusses related protections that would be required to assure the fairness of a national child abuse registry to individuals identified within it.

Issues Regarding Participation of Indian Tribes in a National Child Abuse Registry

Section 633 of The Adam Walsh Act specifies that a national child abuse registry should include data submitted by Indian tribes as well as that of state and local child protective services agencies.  Tribal child protection agencies are also permitted access to registry data.  This section of the report describes a number of particular issues that may affect implementation with respect to Indian tribes.  To identify feasibility issues for tribes, discussions were held with staff of the National Indian Child Welfare Association (NICWA) as well as with staff of the Bureau of Indian Affairs (BIA) within the U.S. Department of Interior (DOI), which administers child welfare activities authorized under the Indian Child Welfare Act (ICWA).  ACF staff familiar with tribal child welfare issues were also consulted.

Individuals and organizations consulted all believed a very few Indian tribes currently maintain their own child abuse registries, though none could identify any in particular that did so.

Rather than maintaining their own registries, a more common occurrence is that Indian tribes, particularly those that have agreements with states through which the Indian tribe participates in the IV-E foster care and adoption assistance programs, have access to data from the child abuse registry of the state(s) in which they are located.  Some may also submit data to state registries.  Approximately 86 title IV-E agreements currently exist between Indian tribes and states for access to child welfare program funding, some of which involve multiple Indian tribes⁷.  Many of these Indian tribes also participate in the Title IV-B Subpart 2 Promoting Safe and Stable Families Program.  In addition, approximately 200 Indian tribes have, or are currently developing, sex offender registries in response to other provisions of the Act.  In several instances, individuals consulted for this report suggested Indian tribes they thought had child abuse registries, but in each case further checking revealed these to be sex offender registries in which some offenders had committed crimes against children, rather child abuse registries as typically used by state child protective services agencies.

The issue of tribal child abuse registries has been discussed by Congress from time to time in the past.  In fact, the Indian Child Protection and Family Violence Act of 1990 (P.L. 101-630, 25 U.S.C. 3204) required that the BIA, in consultation with HHS and the Attorney General, conduct a feasibility study of a tribal child abuse registry and prepare a Report to Congress about the results of the study.  BIA and congressional staff contacted were unsuccessful in locating a copy of that report.  However, congressional records revealed several references to the findings of a feasibility study conducted in response to the requirement:

• In 1995 testimony before the Senate Indian Affairs Committee, the Assistant Secretary for Indian Affairs at the time described that [t]he feasibility for a central registry on child abuse in Indian country was completed in October, 1994, and noted that BIA is exploring the costs, benefits, and feasibility issues raised by recommendations for the establishment of a tribal child abuse registry (U.S. Congress, Senate Committee on Indian Affairs, 1995, p. 5). 
• In a 2006 letter to Senator John McCain, then ranking member of the Senate Committee on Indian Affairs, the director of the Indian Health Service observed that the DOI consulted with the HHS and the Attorney General in 1990 and determined that the establishment of a Central Registry was not feasible due to privacy and legal concerns" (U.S. Congress, Senate Committee on Indian Affairs, 2007, p. 15).
• And in 2005, the director of the BIA testified that "contrary to the provisions of Section 405 of the Act, a Central Registry has not been established.  A feasibility study for this registry was completed in 1994.  The study concluded that the main obstacles to implementing the Central Registry were overcoming due process concerns and developing a mechanism that would require tribal implementation.  To date, these obstacles remain a challenge.  A federal study to identify factors that are impeding the reduction of child abuse in Indian country would fulfill the intent of Section 405, as the legal and other obstacles to implementation of the Central Registry may never be overcome." (U.S. Congress, Senate Committee on Indian Affairs, 2007, pp. 18-19).

Tribal participation in a national child abuse registry would be made difficult by the lack of existing tribal registries.  While states that do not have registries typically do have other administrative data systems that contain information on perpetrators used for investigative purposes, Indian tribes may lack these other data systems as well.  Further, the telecommunications infrastructure needed to transmit and receive data through a national registry, such as secure, dedicated terminals or high speed Internet access, may not be available to all Indian tribes.

Several individuals consulted about tribal issues with respect to implementing a national child abuse registry pointed out that the establishment of child abuse registries is rarely mentioned in discussions of next steps in improving tribal child protection systems and that other activities would likely be viewed as having a greater impact on the safety of tribal children.  In particular, several stakeholders mentioned that of higher priority are improvements to risk and safety assessment practice, the development of software tailored to Indian tribes needs that would keep track of investigations and open cases, and improvements to the prosecution of individuals who have previously maltreated children.  Another issue mentioned was that the time and costs involved for Indian tribes to conduct criminal background checks through existing systems is already a problem, and some feared that potential user fees associated with a national child abuse registry would compound the issue.

Due Process Considerations

Because child abuse registries contain sensitive personal information about families which is sometimes used in ways that have far-reaching implications, the establishment of a national registry raises serious issues regarding the reliability of the information that would be included in a registry, to whom information may be released, and for what purposes the information may be used.  Further, particularly if information may be used to influence someones employment, constitutional issues of due process become important as well.  This section of the report addresses the following questions:

• What due process issues are raised by the establishment of a national child abuse registry?
• How do state child protective services systems address these issues?

Due process refers to the 5^th and 14^th Amendments to the U.S. Constitution, which prohibit the government from depriving any person of life liberty or property without due process of law. The monetary benefits of employment are considered property, making their denial based on information included in a child abuse registry subject to the due process clause (Philips, 1993; Moore, 1995).  If information is used for investigative purposes only, these issues are less relevant, though some court cases have also examined whether there are liberty interests involved if information from a registry is used to take away parental rights or deny adoption applications.

Courts in a number of states have reviewed these issues with respect to child abuse registries, generally examining whether adequate protections for individuals rights are available before an individuals name is included in a child abuse registry.  These protections include that a relatively strong standard of proof be required for substantiation; that the individual has been notified of the finding and its implications; and that adequate procedures exist for hearings or appeals with respect to investigators decisions to substantiate maltreatment.  Court decisions in a number of states have forced changes to state registries procedures in these areas, and the registry in Georgia was declared unconstitutional in 1998 because of such procedural inadequacies⁸.  The potential federal use of this information in ways not contemplated by state procedures further raises the significance of these issues.

The Adam Walsh Child Protection and Safety Act expressly requires that a feasibility study of a child abuse registry make recommendations and findings concerning a due process procedure for a national registry (section 633(g)(1)(D)).  A first step in considering a federal due process procedure was to understand what procedures exist at the state level and apply to the initial decision to substantiate a child maltreatment allegation and include the perpetrators name in a states child abuse registry.  These issues are examined in this section.  The chart at appendix B describes characteristics of child abuse registries in each state.  Features related to due process are summarized below.

Standard of Proof Used for Substantiation Decisions.  An important aspect of state child abuse registries that varies considerably is the standard of proof required to make an investigative determination that a case is substantiated, and thus that the perpetrator would potentially be included in a national registry.⁹  Twenty nine states report using relatively strong legal standards that include:

• The clear and convincing standard¹⁰ indicates that that the evidence provides a high probability that the allegation is true.  Only one state (Pennsylvania) uses this standard for substantiation decisions.
• Preponderance of the evidence is used in most civil court actions and means that the allegation is more likely true than not.  This is the most common legal standard used to substantiate child maltreatment cases, used by 27 states.

The remaining states and the District of Columbia, 22 in all, use weaker legal standards for making substantiation decisions.  These include either that there exists some credible evidence, material evidence, reasonable evidence, or probable cause to believe the allegation is true.  Decisions that a case is to be substantiated under these standards indicate an investigators belief that maltreatment occurred is reasonably grounded in fact and circumstance, though these decisions do not need to meet the more likely true than not mark required by the preponderance of the evidence standard.  Because many large states, including California, New York, Illinois, and Ohio, use these weaker legal standards, more than half of child maltreatment substantiation decisions annually are made using these relatively low standards of proof.

Notification.   State child abuse statutes typically include requirements for notifying various parties about the outcomes of investigations.  As of 2005, 41 states had requirements that perpetrators be notified of the investigation decision (Child Welfare Information Gateway 2005f).  Typically, as part of that notification the individual would be informed that their name will appear on the states child abuse registry.  If a national registry is implemented it will be important to consider whether the individual must also be notified that their information will be submitted to a national registry and of the implications of their inclusion.  Prospectively, it is likely that HHS would want to ensure states perform such notifications at the time a substantiation decision is made regarding any perpetrator whose name would be included in a national registry.  If a national registry is to contain cases substantiated by states prior to implementation, notification becomes much more difficult.  Typically, state registries maintain entries for many years after the substantiation, often at least until a victim reaches age 18.  Contact information on the perpetrator is not updated, however, and it would be difficult and expensive, if not impossible, to make such notifications retrospectively.

Appeal.  The state statutes series compiled by the Child Welfare Information Gateway (2005e, 2005f) describes statutorily established hearing or appeals procedures that can be used by perpetrators to challenge substantiation decisions.  Only half of states had appeals procedures that were laid out in state law.  Other sources indicate that 42 states have an appeal procedure, though it may be specified in policy rather than statute (HHS, 2003).  Many of the appeal procedures states employ are administrative, involving a review within the CPS agency, by CPS staff.  Far fewer include a judicial appeal step or a hearing before someone outside the child protective services agency.  Little information exists on the outcomes of such appeals with which to judge the overall defensibility of CPS agencies substantiation decisions.  A news report from 2004 reported that Agency officials say more than half of [Connecticut Department of Children and Families] child abuse or neglect citations get overturned on appeal. (Poitras, 2004).  A similar news story in Illinois found that the state lost 27 percent of nearly 11,500 appeals filed over a 5½ year period from 2002 to 2007 (Pawlaczyk and Hundsdorfer, 2008).

Expungement.  Procedures for eliminating from state registries information that is inaccurate or that has been overturned has been a perpetual weakness of registries since their inception.  While CAPTA requires that states have procedures for expunging records, procedures for doing so vary and may not be adhered to.  A 2004 report on Californias child abuse and neglect system cites an audit finding that central registry records were poorly maintained.  While the initial audit was confined to one large county, the Child Abuse and Neglect Reporting Act Task Force reported that it is possible that half of the 800,000 records which [the California Department of Justice] presently maintains in [the states Child Abuse Central Index] should be purged. (California Office of the Attorney General, 2004, p. 24).

Access to Information in State Registries.  State laws vary in the details of who may access information contained within central registries, for what purposes, and what information may be provided to those with access rights.  State laws were reviewed to assess, preliminarily, whether those statutes currently provide the authority states would need to provide data to a national registry.  CAPTA currently requires as a condition of receiving funds from the Basic State Grant Program that a states child protection laws include provisions to require a State to disclose confidential information to any Federal, State, or local government entity, or any agent of such entity, that has a need for such information in order to carry out its responsibilities under law to protect children from abuse or neglect (42 U.S.C. 5106(b)(2)(A)(ix)).  The CAPTA provision is most often invoked to permit disclosures of information about a particular case with a current protection issue, but is broad enough to permit this information to be shared with the national registry described in the Adam Walsh Act.  Regardless, our review indicates that state statutes in only 13 states explicitly permit registry information to be shared with federal agencies that need it.  Laws of 30 states make no mention of federal access, and quite a few are ambiguous about whether federal agencies are included among those who may access information (Child Welfare Information Gateway 2005b).

With respect to a due process procedure, there can be no federal substitute for procedural protections at the state or local level.  HHS must not create a system that would second guess local investigations and determinations.  If a national child abuse registry is established, HHS believes the only feasible way to effectively and efficiently provide due process protections is to require that submitting jurisdictions certify that for cases submitted to the national registry, minimum due process protections were available to the perpetrator.  Whether or not employment related background checks remain a permissible use of the registry is an important factor in determining the due process protections that would be required.  It should be noted that strong due process protections could necessitate significant changes to existing CPS investigation processes in some states that may be costly to implement and that could discourage participation in a national registry.

HHS is not yet ready to recommend specific due process protections that would be required as part of a national child abuse registry.  Our initial feasibility assessment has identified issues that have been the subject of legal challenges to state child abuse registries around the country.  As feasibility work continues, we will seek examples of states due process procedures related to their own registries in an effort to design appropriate federal due process protections for a national child abuse registry.

Lessons from Data Systems Used for Criminal Background Checks

Child abuse registries are used to conduct child abuse history checks in ways that are in some respects similar to how criminal history checks are used by law enforcement agencies.  There are parallels in how criminal history records are used both for law enforcement purposes (i.e. criminal investigations), and for noncriminal justice purposes (such as for employment screenings).  These similarities and the long track record of the U.S. Department of Justice (DOJ) in facilitating nationwide access to criminal history records led us to consider how DOJs related systems have addressed a number of the issues that would need to be considered in implementing a national child abuse registry.  This section addresses the following questions:

• How does the federal government facilitate the national exchange of criminal history records for criminal and noncriminal justice purposes?
• What lessons do those systems hold for a national child abuse registry and what key differences between criminal history records and child protective services records should be considered in thinking about these systems as models?

The FBIs Criminal Justice Information Services (CJIS) Division serves as the focal point and central repository for criminal justice information services in the FBI.  It is the FBI's largest division and is responsible for administering several programs, including the Integrated Automated Fingerprint Identification System (IAFIS), the national criminal history record index known as the Interstate Identification Index (III or "Triple I"), the National Crime Information Center (NCIC), (including files relating to wanted persons, civil protection orders, registered sex offenders, and missing persons), and the National Instant Criminal Background Check System (NICS) (which processes background checks on prospective purchasers of firearms from federal firearms licensees).

The Integrated Automated Fingerprint System (IAFIS) integrates fingerprint records that have been sent to the FBI by the states and territories and federal law enforcement agencies, all of which have established their own Automated Fingerprint Identification Systems (AFIS).  Fingerprint identification has been a major responsibility of the FBI since 1924 and fingerprints have been a key part of the FBIs national criminal history record system.  IAFIS provides automated fingerprint search capabilities, latent fingerprint searching capability, electronic image storage, and electronic exchange of fingerprints and responses 24 hours a day, 365 days a year.  Categories of fingerprints currently maintained by the FBI include:  criminal fingerprints, federal applicants and employees, United States military, aliens, and those submitted to the FBI by persons desiring to have their fingerprints placed on record with the FBI for personal identification purposes.

The Interstate Identification Index (III) segment of IAFIS is the national system designed to provide automated criminal history record information.  The III is an index-pointer system that allows for the exchange of criminal history records ("rap sheets").  The III stores the criminal history records of federal offenders and those offenders submitted by participating and non-participating III states.  All information in the III is supported by fingerprint submissions.  Under the III, the FBI maintains an index of persons arrested for felonies or misdemeanors under either state or federal law.  The index includes identification data such as name, birth date, race, and sex, etc., and federal and state identification numbers.  As of October 2008, 50 states and the District of Columbia were participating in III.

The National Crime Information Center (NCIC) is a computerized database of documented criminal justice information available to virtually every law enforcement agency nationwide, 24 hours a day, 365 days a year.  The NCIC has been in operation since 1967, with the goal of assisting law enforcement in apprehending fugitives and locating stolen property.  This goal has since expanded to include locating missing persons and further protecting law enforcement personnel and the public.

The National Sex Offender Registry (NSOR) is part of the NCIC database.  The NSOR includes records for each sex offender and any other person required to register in a jurisdictions sex offender registry. Information in the NSOR is law enforcement in nature and is accessible by federal, tribal, state, and local criminal justice agencies.  In addition, the U.S. Department of Justice manages the Dru Sjodin National Sex Offender Public Website which provides the means for the public to access all jurisdictions publicly available sex offender registry records through a single website.

The National Instant Criminal Background Check System (NICS) is used to determine the eligibility of prospective firearm transferees in accordance with federal and state law.  Firearms dealers initiate background checks through the NICS by telephonically contacting either a state-designated agency or the FBI NICS Section's Contracted Call Centers, or through an Internet-based electronic access.  Based on data specific to FBI-generated NICS checks, in most cases, a dealer receives an immediate proceed or deny response; however, when a valid match yields incomplete information (e.g., a missing disposition), a delay occurs and the NICS is provided three business days to allow for further research.  In 2006, over ten million (state and federal initiated) background checks were submitted through the NICS.

Compiling and Referencing Criminal History Records

A child abuse registry would essentially entail a search of administrative records regarding persons named during CPS investigations as child maltreatment perpetrators.  This section examines how criminal history records differ from child abuse history records and how the III handles issues that are likely to be important in establishing a national child abuse registry.

The Department of Justice has spent many years assisting states to upgrade the quality of their criminal history records and systems for sharing them securely in order to achieve the level of information access currently available to state and local law enforcement agencies.  The National Criminal History Improvement Program between 1995 and 2005 awarded states and territories $495 million to improve the quality, timeliness, and immediate accessibility of criminal history records and related information.  As of January, 2006, over 63 million records were included among the files accessed for the National Instant Criminal Background Check System.  The bulk of these (56 million) were Interstate Identification Index files as described above.  The National Sex Offender Registry at that time included approximately 414,000 records (Ramker, 2006).

National Criminal History Record Search State Records in State Repositories.  The criminal history records searched, with some exceptions, are those maintained by state criminal history record repositories.  The III system does not access records directly from local agencies.  This is in contrast to requirements of the Adam Walsh Act regarding a child abuse registry, which specifies that local jurisdictions may submit records directly to a federal system.

Criminal History Records are Maintained by States.  The Federal Bureau of Investigation once maintained records in a central system, however since the 1980s it has moved to a decentralized system in which most criminal history records are held in state criminal history record repositories and are searched through a national database that allows law enforcement access to records from across the nation.  Described as an index-pointer system (U.S. Department of Justice, 2001), the FBI maintains a master name index which directs an inquiry to one or more state agencies which maintain a matching record.  State records are provided to the inquiring agency through The International Justice and Public Safety Information Sharing Network (Nlets), a secure telecommunications system.  States are responsible for the accuracy of their criminal history records.

Criminal History Records Pertain Primarily to Arrests.  Most of the arrest information provided in criminal history record checks are related to felonies or serious misdemeanors.  Arrest records are provided to law enforcement agencies for criminal justice purposes.  Use of criminal history record information for noncriminal justice purposes is more sensitive and requires the establishment of positive identification through fingerprint processing.  In addition to the federal government, twenty seven states have ratified the National Crime Prevention and Privacy Compact (Compact) Act of 1998.  The Compact established an electronic information sharing system among the federal government and the states to exchange criminal history records for noncriminal justice purposes authorized by federal or state law.¹¹

Criminal History Records are Often Publicly Accessible.  Historically, information about convictions has been available widely and is not subject to the confidentiality protections afforded in child maltreatment cases.  A 2001 survey revealed that in 29 states anyone could obtain criminal history for another person on request (SEARCH 2001).  Thus in these uses the federal government is facilitating access to information but not providing authority for access to information that would not have otherwise been available to the user from the state holding the original record.  Access to data regarding arrests rather than convictions (more akin to unsubstantiated maltreatment investigations) is more restricted.  However, even though criminal records are often publicly available at the local level, courts have found that in compiling records centrally the federal government negates their practical obscurity and must require more substantial privacy safeguards than would have been available in their original form (Davis, 2003).

Positive Identification is Accomplished through Fingerprints.  Nearly all justice system background checks now use fingerprints as the primary means of identification.  Name-based checks are not considered sufficiently reliable to assure positive identification.  Even with fingerprint identification, federal criminal records systems typically include as many identifiers as possible to ensure accurate matches.  Social Security Numbers, drivers license numbers, aliases, addresses, and other fields allow for more confident identification of individuals in these systems.  Few such fields are typically available in child abuse registry records and, as currently written, the Adam Walsh Act prohibits a federal child abuse registry from containing identifying information beyond the perpetrators name.

Victims are Not Identified in Criminal History Records.  While adult criminal perpetrators are not typically entitled to privacy with respect to criminal records, juvenile perpetrators and crime victims are not identified in criminal records accessible for background checks.  For instance, the National Sex Offenders Registry contains no identifying information on victims.  In child abuse registries, however, names and other information about perpetrators children (usually the victims) are among the key fields often used to verify the individuals identity.  Even if not named in a national child abuse registry, some victims will be identifiable because once you list John Doe of Anytown, Anystate was substantiated on a particular date for neglect of his three children ages X, Y and Z, even without names those children are potentially identifiable.  The Adam Walsh Act requires that the national child abuse registry include information on the nature of the substantiated case of abuse or neglect. Careful consideration is needed about how much identifying information about victims should be included in a national child abuse registry and released to registry users.

The Number of Noncriminal Records Relevant to Firearms Purchases Submitted to the NICS Continues to Increase.  The NICS is a name-based system in which the name and descriptive information of a prospective firearms transferee is searched through the III (e.g., criminal history records), the NCIC (e.g., protection orders, warrants), and the NICS Index.  The NICS Index maintains records, voluntarily contributed by local, state, and federal agencies, of  persons who are federally disqualified the transfer of a firearm, such as those specific to prohibiting mental health information or substance abuse.  Although the number of records maintained in the NICS Index continues to increase, many qualifying records have not yet been contributed.  For instance, at the end of 2005, the NICS Index maintained fewer than 1,000 disqualifying records specific to controlled substance abuse.  Recent legislation as a result of the NICS Improvement Amendments Act of 2007 (PL 110-180) establishes incentives and penalties to encourage states to include this information in the system.

While broader in scope than a child abuse registry, Department of Justice experience managing access to criminal records provides insights relevant to establishing procedures for accessing information on perpetrators of child maltreatment.  The Department of Justice has years of experience monitoring allowable access to criminal records systems maintained by or accessed through the FBIs Criminal Justice Information Services (CJIS) Division.  In order to protect the integrity of the information the FBI has established a secure telecommunications network through which records are exchanged.  States may access records through the CJIS systems only after establishing user agreements with the Department of Justice that specify strict guidelines about the circumstances under which records may be accessed and used.  Regular audits are conducted by FBI personnel to assure that states use remains in compliance with the user agreements.  States must maintain computer equipment and telecommunications lines or channels in ways that meet the FBIs security standards.  Computer terminals used to access the FBIs system must be in a secure, locked location within the agencys facility, with physical access only by authorized personnel.  There are requirements also for the personnel who have access to information and the records and audit trail that must be maintained on each data request¹².  One small state whose child protective services agency has established access to NCIC information reports that getting their systems certified as compliant with the FBIs requirements took years and cost over a million dollars to accomplish¹³.

The FBI controls access to the NCIC at the state level.  The access of local personnel is handled by the state and must go through a state telecommunications network.  Allowing direct access of local staff to a national child abuse registry, as seems contemplated in the Adam Walsh Act, would be a more complex undertaking that would be very difficult, and potentially costly, to accomplish in a secure fashion.

Benefits and Costs of a National Child Abuse Registry

The Adam Walsh Act requires that HHSs feasibility study of a national child abuse registry address the costs and benefits of such a registry.  Available information about costs and benefits is provided below.  Costs will depend to a large extent on a number of decisions that have not been made regarding direction and scope of a registry.  The discussion below addresses key cost questions including:

• What potential benefits could result from the implementation of a national child abuse registry?
• What categories of costs would be involved in establishing a national child abuse registry, including direct costs of establishing and maintaining a registry and costs that would be incurred by others as a system is implemented?

Where available, information about costs of related efforts is described to inform readers about the possible magnitude or range of costs involved.  Because actual costs of a national child abuse registry will depend on significant decisions about form and scope that have not yet been made, detailed cost estimates have not been prepared.  The text box that follows this cost discussion identifies a number of unresolved questions that would significantly affect system costs.

Potential Benefits

The primary benefit of a national child abuse registry would be investigators improved knowledge of any historical pattern of abuse or neglect regarding an alleged perpetrator.  This information would help an investigator establish patterns of behavior and determine which party is the more credible as conflicting information is gathered during an investigation.  Information about past incidents can help child welfare agencies better judge potential current risk to the child and what level of protection may be required.  However, no information is available to quantify safety improvements or avoided maltreatment that might result from improved information about past substantiated maltreatment in other states.  If a registry is used also for employment-related background checks, presumably there would also be some avoided child abuse and neglect as a result of denying employment to unsuitable persons.  Few data are available to quantify such benefits, however.  A California report on its registry indicated that in 2003, five percent of non-investigative inquiries resulted in positive matches; no similar figure on investigative inquiries was provided (California Office of the Attorney General, 2004).

It should be noted that benefits of a national child abuse registry could be realized only if many or most states participate by providing data on perpetrators to a registry and only if costs and procedures involved in performing a registry check do not discourage its use.

Because the near total lack of information about how frequently child maltreatment perpetrators offend in multiple states prevents us from assessing the potential benefits of a national child abuse registry, obtaining better information on this key issue will be a primary focus of the next phase of the feasibility study.

Anticipated Costs

Costs of a national child abuse registry include the federal costs of operating a registry, costs to states of providing information through a registry, and costs to individuals upon whom child abuse history checks are conducted.  These are described below.

Federal Costs

• Federal Start-up Costs.  Initial expenses would include the costs involved in designing and establishing a national child abuse registry.  These costs would include a systems review of existing state databases; a collaborative process to engage states and Indian tribes in the design of a federal registry; programming and other tasks involved in establishing the system once decisions have been made regarding its features.  As a rough indication of magnitude of these costs, the establishment of the AdoptUsKids web site cost $4 million across fiscal years 2000 and 2001.  AdoptUsKids.org is a tool for connecting foster and adoptive families with waiting children throughout the United States.  It contains information submitted by states about children available for adoption and may be searched by families potentially interested in adopting a child.  While substantially different in function, federal costs involved to compile initial records for a national child abuse registry would likely be an effort of similar scope to that endeavor.  However, hardware and telecommunications costs related to ensuring the security of data exchanged are likely to be substantial for a national child abuse registry and beyond what has been required for AdoptUsKids.
• Technical Assistance and Training Costs. These would include activities associated with informing state, local and tribal staff about procedures for submitting data to the national registry and for submitting queries to the registry, as well as operating a help desk for users.  Costs in this category will vary depending on the number of agencies submitting data (particularly if local submissions are permitted) and whether access is handled through centralized access mechanisms in each state (for instance allowing access only through the states own registry), or whether inquiries are to come directly from local investigators (of whom there are many and among whom there is a high turnover rate).  There will be significant initial costs in this category as well as ongoing costs of training new staff to use the system.  A system with high levels of security is likely to have additional costs relating to assuring that security measures are implemented properly by users permitted to query the system. 
• Ongoing Operational Costs.  These would include the expenses involved in maintaining a registry and responding to queries.  Key here would be the costs of validating the access rights of allowable system users, as well as any compliance activities that may be needed to ensure registry information is used only as intended, and that information provided through the registry meets Privacy Act standards for ensuring records are accurate, relevant, timely and complete.  There will also be federal costs related to responding to challenges to information contained in the registry or to false positive identifications, and responding to any legal challenges to the registry itself.  In addition, costs of periodic resubmissions and updating outdated data must also be considered.  As a general indication of magnitude, the Department of Defenses child abuse registry has annual operational costs of about $1.75 million.¹⁴ However, it should be noted that the access issues for the DODs registry are significantly less challenging than those under consideration here since most users of the DOD registry are DOD employees or contractors.  Reliable identification of perpetrators is less challenging for the military registry as well because most perpetrators are also military personnel.  The expenses involved with addressing security issues to provide electronic access to registry information and verify authorized users will be substantial.  Specifics have not been estimated here because there are too many unanswered questions regarding how a registry would be structured.

Costs to Agencies Submitting Data to and Making Inquiries of a National Registry

• Costs of Exchanging Information Securely.  States and other entities submitting data to a national child abuse registry would need the computer equipment and telecommunications infrastructure to support secure access to the information.  Specifics would depend on the security requirements established and the number of individuals and organizations provided access to the registry.
• Costs of Meeting Minimum Procedural Protections.  As described elsewhere in this report, it may be desirable to guarantee minimum due process protections have been met at the time a report is substantiated before the information about a perpetrator is included in a national registry.  State or tribal agencies that do not currently operate according to whatever minimum standards are established would have costs associated with adjusting their systems to provide these procedural protections.  For instance, adjustments may be needed in procedures for providing notification to individuals whose names would be included in a registry, and for improving hearing or appeals procedures through which individuals may challenge investigation findings.

Costs to Individuals Identified (Truly or Falsely) as Child Maltreatment Perpetrators

• Consequences of Positive Identification.  When a query to a national child abuse registry identifies the individual as having a previous substantiated child maltreatment report, that identification may have consequences for the identified individual.  In some cases an investigators elevated assessment of risk to a child will result in different interventions with respect to a family under investigation.  If the query was related to an individuals fitness to become a foster or adoptive parent, the individuals application may be denied.  If the query is related to prospective employment or volunteer work, the individual may be denied the job for which they have applied.  These consequences will be incurred both in the case of true positives and in cases of false positive identifications that, as described elsewhere in this report, must be expected with a name-based registry.

Nonfinancial Costs of a National Child Abuse Registry

• Further Movement of Child Protective Services Away from a Social Work Focus.  Several stakeholders contacted to discuss issues related to a national child abuse registry noted that such a registry would further accentuate the movement of the child welfare field away from a helping, social work orientation toward the families being investigated, and more toward a criminal justice, punishment-oriented focus.  Some saw this trend as a cost to families, though it cannot be quantified.
• Costs of Foregone Prevention and Remediation.  Child protection agencies in some states (11 states as of 2003) may not provide services to families unless an investigation has substantiated a maltreatment allegation (HHS, 2003).  In additional states, families with substantiated cases may be prioritized for services.  If families become more likely to resist substantiation or to appeal findings because of additional negative implications of a child maltreatment finding, this route to service provision may be lost and families may forego needed services, raising risks to children.

Financing Startup and Ongoing Operations

If implementation of a national child abuse registry moves forward, whatever its format and content, funding will be necessary to pay for both start-up costs and ongoing operations.  These costs could be covered in a number of ways.  Startup costs would require congressional appropriations. Ongoing operational costs could be funded either through annual congressional appropriations, or through user fees charged either per query or on a subscription basis to entities seeking access to the data.  If the activity is to be funded by user fees, additional legislative authority would be needed if HHS is to retain for the purposes of operating a national child abuse registry any fees charged to users.  While HHS has current authority to charge fees for services, any fees collected in the absence of retention authority would go to the U.S. Treasury rather than being retained to support the establishment and maintenance of the activity for which they are charged (OMB Circular A-25, Section 9).

Whether a national child abuse registry could be supported by user fees depends, in part, on how high the fees would be and whether it is expected that child protection agencies would use such checks routinely or whether they would be conducted in relatively exceptional circumstances when it is suspected that out of state findings might exist or might affect the safety assessment for a child.  While there are user fees involved in criminal background checks for foster and adoptive parents, and some states charge each other for out-of-state child abuse registry checks for these individuals, those costs are generally passed along to the applicant rather than being borne by the child protection agency.  Any fees for investigative checks (i.e. those conducted as part of a child maltreatment investigation), could not be passed along to the subject of the investigation, however, and would need to be incorporated into agency budgets.

Another consideration here would be whether the additional cost and hassle factor of another fee-based background check would discourage families from considering foster or adoptive parenting of children in state custody.  While no assessment has been made regarding fees that would be necessary to support a national child abuse registry, information about fees charged for other types of background checks was sought.  The FBI currently charges $15.25 - $30.25, depending on the method of submission (electronic or manual) for a national fingerprint-based criminal history record check for noncriminal justice purposes.  A number of state child abuse registries and criminal records depositories charge similar fees¹⁵.

Conclusions

The preceding sections of this report have outlined a number of key issues regarding the feasibility of establishing a national child abuse registry.  This section draws several conclusions based on the issues raised.

Conclusion 1:  Potential benefits of a national child abuse registry are largely unknown.

There is no data available with which to quantify improvements in child safety that may result from the implementation of a national child abuse registry.  In particular, it is unknown how frequently perpetrators have been substantiated for child maltreatment in multiple states.  This interim report identifies and describes the major components of anticipated costs of implementing a national child abuse registry.  Key among these would be the costs of establishing secure electronic systems to protect the data from unauthorized use, and addressing procedural weaknesses in some jurisdictions CPS systems to assure the accuracy and reliability of information included in a national registry.  The gap in information with which to determine the frequency with which a national child abuse registry would be helpful to child maltreatment investigators will be the primary focus of future feasibility study activities.

Conclusion 2:  A lack of incentives for participation could result in a database that includes little information and does not fulfill its intent.

The Adam Walsh Child Protection and Safety Act calls upon HHS to establish a national child abuse registry.  However, it is unclear whether states or Indian tribes would be willing to provide data to such a registry.  The law contains neither incentives to encourage participation nor consequences for declining.  No funding is provided to offset the costs of providing or accessing the data.  In addition, many states lack authority in their current laws to provide this data to the federal government.  The most likely scenario is that HHS would create a database to which few jurisdictions would submit data, with the result that the registry could not serve its purpose.  Before a registry could be implemented successfully, the active engagement of state officials would be necessary to determine their information needs, as well as their concerns regarding the protection of information they would provide to a national registry.  Extensive discussions would be needed to develop a consensus regarding the maintenance and control of sensitive information about families that is created at the state and local levels but to which access would be controlled at the federal level.  Only if a national registry is constructed in a way that meets the needs of state and local child protection agencies and creates conditions under which they would be willing to provide the necessary data, could a registry become a useful child protection tool.  Further work to assess states likelihood of participating voluntarily in a national child abuse registry will be conducted as part of our further study of feasibility issues.

Conclusion 3:  Before implementation of a national child abuse registry could begin, legislative change would be needed to permit the collection of sufficient information to accurately identify perpetrators. 

The Adam Walsh Child Protection and Safety Act limits identifying information in a national child abuse registry to the perpetrators name.  This statutory language must be changed before a national child abuse registry could be implemented.  Because many names are very common, name cannot be the only field used to determine whether or not there is a match between the individual about whom an inquiry is made and a perpetrator listed in a national child abuse registry.  Additional identifying data needed would include, at minimum, the perpetrators date of birth and the address of the perpetrator at the time of the substantiated maltreatment incident.  However, it is preferable that any statutory language be flexible in allowing the Department to define necessary identifying information in regulation or other policy guidance.  Even with additional identifying fields, however, as has been described above, name-based matches have high false positive and false negative rates that must be anticipated.  Procedures would be needed for ruling out false positive matches without negative consequences to those so identified.  If this issue is not addressed well, false positives related to non-investigatory inquiries could overwhelm local staff responding to information returned from inquiries to a national child abuse registry.

Conclusion 4: Clarification is required on several key issues that are ambiguous in the authorizing statute; these must be resolved either within HHS or by Congress before implementation could proceed. 

This feasibility study has identified a number of important issues, some of them inter-related, that would need to be resolved and that would influence the form and function of a national child abuse registry.  Key among these is whether a national registry is to be used only for investigative inquiries or also for child abuse history checks related to employment and licensing purposes.  In many states, such employment checks far outnumber investigative uses of their child abuse registries.  While such inquiries are not mentioned explicitly in the Act and were not discussed during congressional debate, HHS has determined that they are permitted by the authorizing statute.  The use of a registry for non-investigatory inquiries affects the volume of inquires expected and the potential consequences for individuals and families of positive results and especially of false positive identifications.  In turn, the increased consequences of being identified in a national registry if employment checks are permitted may necessitate more elaborate due process protections for individuals before their listing in a national child abuse registry.  Allowing registry inquiries related to employment and licensing also raises issues with respect to states that do not allow their own registries to be used for such purposes.

The volume of inquiries would also depend on whether it is expected (or required) that a national child abuse registry check be completed routinely during child protection investigations, or whether checks are instead to be used primarily in situations in which the investigator is aware of potential interstate issues.  This factor will interact with decisions about how a registry is financed and in particular whether fees are charged for access.  If fees are charged and inquiries are expected on a routine basis, costs to agencies could become prohibitive.

Basic decisions are also needed regarding how to define who would have access to the registry and how to maintain restricted access and validate the identities of legitimate users given the large numbers of CPS and law enforcement agencies with whom HHS does not currently have a direct relationship, as well as high turnover rates among staff.  If a registry is implemented, HHS would strongly prefer to limit the entry of data to state and tribal entities.  Allowing sub-state jurisdictions to submit data directly would significantly complicate the operation of a national child abuse registry and would raise both implementation expenses and ongoing operational costs.  Doing so would also undermine the Departments longstanding intention that each states SACWIS would serve as the comprehensive system for managing child welfare data and for submitting child welfare data to HHS.

Next Steps

This initial feasibility assessment has established that implementation of a national child abuse registry is not feasible under the statutory limitations of the Adam Walsh Child Protection and Safety Act.  A number of the specific ambiguities and limitations of the statute could be addressed legislatively.  However, solutions to other issues may involve substantial costs and be burdensome to the state and local child protective services systems a national child abuse registry is intended to help.  In the end, it is not clear whether child safety would be improved substantially by a national database of child abuse perpetrators.

HHS will conduct additional feasibility activities in the coming year and will prepare a final report to the Congress containing results of the additional work when findings are available.  The next phase of the feasibility study will seek to address a number of issues that could not be resolved in this initial feasibility assessment, and will include the following activities:

• An effort to determine how frequently child maltreatment perpetrators offend in multiple states.  The lack of information on this topic prevents us from assessing the potential benefits of a national child abuse registry.
• A review of the data systems comprising state child maltreatment registries.  Such a review will enable the Department to respond to the statutes requirement that we identify data standards for a national child abuse registry.
• Further contacts with states to assess their interest in participating in a national child abuse registry and to identify in more detail factors that would hinder participation.  A better understanding of these issues will allow us to determine whether a voluntary registry without incentives for participation will have sufficient coverage to make it a useful child protection tool.

State child abuse registries play an important role in maltreatment investigations.  This initial feasibility assessment has revealed significant challenges to making a national child abuse registry similarly useful.  We look forward to working with the Congress to resolve the issues identified and to determine how to best facilitate the interstate exchange of child maltreatment information in order to protect children from maltreatment.

References

Capitol 9 News, OCFS employees accused of breaching confidential child abuse register accessed June 7, 2008 at: http://capitalnews9.com/Default.aspx?ArID=117523

California Office of the Attorney General (2004).  Child Abuse and Neglect Reporting Act Task Force Report.  Sacramento, California:  Office of the Attorney General.  Online at: http://www.ag.ca.gov/publications/childabuse.pdf

Child Welfare Information Gateway (2005a).  Disclosure of Confidential Child Abuse and Neglect Records.  Washington, DC:  U.S. Department of Health and Human Services, Administration for Children and Families.  Online at: http://www.childwelfare.gov/systemwide/laws_policies/statutes/confide.cfm

Child Welfare Information Gateway (2005b).  Disclosure of Confidential Child Abuse and Neglect Records:  Summary of State Laws.  Washington, DC:  U.S. Department of Health and Human Services, Administration for Children and Families.  Online at: http://www.childwelfare.gov/systemwide/laws_policies/statutes/confideall.pdf

Child Welfare Information Gateway (2005c).  Establishment and Maintenance of Central Registries for Child Abuse Reports.  Washington, DC:  U.S. Department of Health and Human Services, Administration for Children and Families.  Online at: http://www.childwelfare.gov/systemwide/laws_policies/statutes/centreg.cfm

Child Welfare Information Gateway (2005d).  Establishment and Maintenance of Central Registries for Child Abuse Reports:  Summary of State Laws.  Washington, DC:  U.S. Department of Health and Human Services, Administration for Children and Families.  Online at: http://www.childwelfare.gov/systemwide/laws_policies/statutes/centregall.pdf

Child Welfare Information Gateway (2005e).  Review and Expunction of Central Registries and Reporting Records.  Washington, DC:  U.S. Department of Health and Human Services, Administration for Children and Families.  Online at:  http://www.childwelfare.gov/systemwide/laws_policies/statutes/registry.cfm

Child Welfare Information Gateway (2005f).  Review and Expunction of Central Registries and Reporting Records:  Summary of State Laws.  Washington, DC:  U.S. Department of Health and Human Services, Administration for Children and Families.  Online at: http://www.childwelfare.gov/systemwide/laws_policies/statutes/registryall.pdf

Davis, C.N. (2003).  Electronic access to information and the privacy paradox.  Social Science Computer Review, 21(1):15-25.

Drake, B., Jonson-Reid, M, Way, I and Chung, S (2003). Substantiation and recidivism.  Child Maltreatment 8:248-260.

Dunhem, R. and Oppenheim, E., (2002).  Understanding Criminal Records Checks:  Ensuring the Safety of Children Placed with Foster and Adoptive Parents.  Washington, DC:  The Association of Administrators of the Interstate Compact on Adoption and Medical Assistance.

Flango, V.E., (1991).  Can central registries improve substantiation rates in child abuse and neglect cases? Child Abuse and Neglect 15:403-413.

Gibelman, M and Grant, S (1978).  The uses and misuses of central registries in child protective services.  Child Welfare 57(7):405-413.

Fluke, J.D., Shusterman, G.R., Hollinshead, D.M. and Yuan, Y.T., (2008).  Longitudinal analysis of repeated child abuse reporting and victimization:  multistate analysis of associated factors.  Child Maltreatment 13:76-88.

Leeb, R.T., Paulozzi, L., Melanson, C., Simon, T., Arias, I. (2008).  Child Maltreatment Surveillance:  Uniform Definitions for Public Health and Recommended Data Elements, Version 1.0.  Atlanta, Georgia:  Centers for Disease Control and Prevention, National Center for Industry Prevention and Control.

Miller, J.L. and Bissell, M. (2007).  Adam Walsh Child Protection and Safety Act of 2006:  Issues for Child Welfare Agencies.  East Greenwich, Rhode Island:  Childfocus Partners. Online at: http://www.childfocuspartners.com/images/AdamWalsh.final.pdf

Moore, J.D. (1995).  Charting a course between Scylla and Charybdis:  Child abuse registries and procedural due process.  North Carolina Law Review 73:2063.

National Resource Center for Family Centered Practice and Permanency Planning (2008).  State Child Abuse Registries.  Accessed online August 26, 2008 at: http://www.hunter.cuny.edu/socwork/nrcfcpp/downloads/policy-issues/State_Child_Abuse_Registries.pdf

Pawlaczyk, G and Hundsdorfer, B (2008). Guilty until proven innocent:  Parents accused of child abuse by DCFS fight to clear their names.  Belleview News-Democrat, May 5, 2008.

Philips, Michael R. (1993).  The Constitutionality of Employer-Accessible Child Abuse Registries:  Due Process Implications of Governmental Occupational Blacklisting, in Michigan Law Review, 92(1): 139-194).

Poitras, Colin (2004).  DCF Blacklist Followed Child Care Workers.  In the Hartford (Connecticut) Courant December 5, 2004.

Ramker, GF (2006).  Improving Criminal History Records for Background Checks, 2005.  Washington, DC:  U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Sabotta. E.E. and Davis, R.L., (1992).  Fatality after Report to a Child Abuse Registry in Washington State, 1973-1986.  in Child Abuse and Neglect 16:627-635.

SEARCH, The National Consortium for Justice Information and Statistics (1999).  Interstate Identification Index Name Check Efficacy: Report of the National Task Force to the Attorney General.  Sacramento, California:  SEARCH.

SEARCH, The National Consortium for Justice Information and Statistics (2001).  Survey of States that Provide Some Level of Open Access to Their Criminal History Records. Sacramento, California:  SEARCH. Online at: http://www.search.org/files/pdf/open_records_survey.pdf

SEARCH, The National Consortium for Justice Information and Statistics (2002).  Survey on Issues Related to Background Checks on Individuals Who Work in Day Care Facilities.  Sacramento, California:  SEARCH. Online at: http://www.search.org/files/pdf/ia_survey_results.pdf

Slep, A.M.S. and Heyman, P.E. (2006).  Creating and Field-Testing Child Maltreatment Definitions:  Improving the Reliability of Substantiation Decisions.  Child Maltreatment 11:3:217-236.

U.S. Census Bureau (2008).  Current Population Survey, Geographical Mobility:  2006 to 2007, Detailed Tables, Table 8.  General Mobility of Family Householders, by Type of Household, Age of Householder, and Presence and Age of Own Children Under 18: 2006 to 2007.  Online at: http://www.census.gov/population/socdemo/migration/cps2007/tab08-1.xls

U.S. Congress, House Appropriations Committee (2007). House Appropriations Committee Print on Consolidated Appropriations Act, 2008.  P.L. 110-161, [language regarding this study appears on page 163].

U.S. Congress, Senate Committee on Indian Affairs (2007). Amending the Indian Child Protection and Family Violence Prevention Act to Identify and Remove Barriers to Reducing Child Abuse, to Provide for Examinations of Certain Children, and for Other Purposes, report on S. 398, 110th Cong., 1st sess., April 10, 2007, S.Rept. 110-45. Washington: Government Printing Office.

U.S. Congress, Senate Committee on Indian Affairs (1995). To Reauthorize Appropriations for Certain Programs under the Indian Child Protection and Family Violence Prevention Act, and for Other Purposes, report on S. 441, 104th Cong., 1st sess., April 18, 1995, S.Rept. 104-53.  Washington, DC:  Government Printing Office.

U.S. Department of Health and Human Services, Administration for Children and Families/Childrens Bureau and Office of the Assistant Secretary for Planning and Evaluation (2003).  National Study of Child Protective Services and Reform Efforts:  Review of State CPS Policy.  Washington, D.C.:  U.S. Department of Health and Human Services.  Online at: http://aspe.hhs.gov/basic-report/national-study-child-protective-services-systems-and-reform-efforts-2

U.S. Department of Health and Human Services, Administration for Children and Families (2006).  ACYF-CB-IM-06-04.

U.S. Department of Health and Human Services, Administration for Children, Youth and Families (2009).  Child Maltreatment 2007.  Washington, DC:  U.S. Government Printing Office. http://www.acf.hhs.gov/programs/cb/pubs/cm07/

U.S. Department of Health and Human Services, Administration for Children and Families (2008b).  Child Welfare Outcomes 2002-2005.  Washington:  DC:  U.S. Department of Health and Human Services.

U.S. Department of Health and Human Services, Administration for Children and Families (no date).  SACWIS Assessment Review Guide Requirements Overview.  Online at: http://www.acf.hhs.gov/programs/cb/systems/sacwis/ppt_assess_final/sld001.htm

U.S. Department of Health and Human Services, Administration for Children and Families, Office of Community Services (no date). Social Services Block Grant Program Annual Report 2006.   Online at: http://www.acf.hhs.gov/programs/ocs/ssbg/annrpt/2006/index.html

U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation (2007).  Rereporting and Recurrence of Child Maltreatment:  Findings from NCANDS.  Washington, DC:  U.S. Department of Health and Human Services.  pdf-report/rereporting-and-recurrence-child-maltreatment

U.S. Department of Interior, Bureau of Indian Affairs (2006).  Indian Child Welfare Handbook.  Washington, DC:  Bureau of Indian Affairs.  http://www.doi.gov/bia/docs/Indian_child_welfare_handbook_InsidePages.pdf

U.S. Department of Justice, Bureau of Justice Statistics (2001).  Use and Management of Criminal History Record Information:  A Comprehensive Report, 2001 Update.  Washington, DC:  U.S. Department of Justice.

U.S. Department of Justice, Federal Bureau of Investigations (2006).  National Instant Criminal Background Check System:  Operations 2005.  Washington, DC:  U.S. Departement of Justice.

White House (2006).  President Signs H.R. 4472, the Adam Walsh Child Protection and Safety Act of 2006.  Online at http://www.whitehouse.gov/news/releases/2006/07/20060727-6.html.

Winterfeld, A.P and Sakagawa, T (2003). Investigation Models for Child Abuse and Neglect Collaboration with Law Enforcement.  Englewood, Colorado:  American Humane Association.

Whiting, L. (1977). The central registry for child abuse cases:  rethinking the basic assumptions.  Child Welfare 56(1):761-7.

Selected Court Cases Regarding Child Abuse Registries

Mildred Jamison, et al. v. State of Missouri, Department of Social Services, Division of Family Services, 218 S.W.3^rd 399 (Missouri 2007).

Petition of James J. Preisendorfer, New Hampshire Division for Children, Youth and Families, 719 A.2d 590 (New Hampshire 1998).

State of Georgia et al. v. Jackson, 496 SE2d 912 (Georgia 1998).

Hunt v. Indiana Family and Social Services Admin., slip opinion, 2007 WL 2349629 (S.D. Ind. 2007).

Appendix A: Section 633 of the Adam Walsh Child Protection and Safety Act (P.L. 109-248, 42 U.S.C. 16990)

SEC. 633. NATIONAL REGISTRY OF SUBSTANTIATED CASES OF CHILD ABUSE.

(a) In General- The Secretary of Health and Human Services, in consultation with the Attorney General, shall create a national registry of substantiated cases of child abuse or neglect. (b) Information-

(1) COLLECTION- The information in the registry described in subsection (a) shall be supplied by States and Indian tribes, or, at the option of a State, by political subdivisions of such State, to the Secretary of Health and Human Services. (2) TYPE OF INFORMATION- The registry described in subsection (a) shall collect in a central electronic registry information on persons reported to a State, Indian tribe, or political subdivision of a State as perpetrators of a substantiated case of child abuse or neglect.

(c) Scope of Information-

(1) IN GENERAL-

(A) TREATMENT OF REPORTS- The information to be provided to the Secretary of Health and Human Services under this section shall relate to substantiated reports of child abuse or neglect. (B) EXCEPTION- If a State, Indian tribe, or political subdivision of a State has an electronic register of cases of child abuse or neglect equivalent to the registry established under this section that it maintains pursuant to a requirement or authorization under any other provision of law, the information provided to the Secretary of Health and Human Services under this section shall be coextensive with that in such register.

(2) FORM- Information provided to the Secretary of Health and Human Services under this section--

(A) shall be in a standardized electronic form determined by the Secretary of Health and Human Services; and (B) shall contain case-specific identifying information that is limited to the name of the perpetrator and the nature of the substantiated case of child abuse or neglect, and that complies with clauses (viii) and (ix) of section 106(b)(2)(A) of the Child Abuse Prevention and Treatment Act (42 U.S.C. 5106(b)(2)(A) (viii) and (ix)).

(d) Construction- This section shall not be construed to require a State, Indian tribe, or political subdivision of a State to modify--

(1) an equivalent register of cases of child abuse or neglect that it maintains pursuant to a requirement or authorization under any other provision of law; or (2) any other record relating to child abuse or neglect, regardless of whether the report of abuse or neglect was substantiated, unsubstantiated, or determined to be unfounded.

(e) Accessibility- Information contained in the national registry shall only be accessible to any Federal, State, Indian tribe, or local government entity, or any agent of such entities, that has a need for such information in order to carry out its responsibilities under law to protect children from child abuse and neglect. (f) Dissemination- The Secretary of Health and Human Services shall establish standards for the dissemination of information in the national registry of substantiated cases of child abuse or neglect. Such standards shall comply with clauses (viii) and (ix) of section 106(b)(2)(A) of the Child Abuse Prevention and Treatment Act (42 U.S.C. 5106(b)(2)(A) (viii) and (ix)). (g) Study-

(1) IN GENERAL- The Secretary of Health and Human Services shall conduct a study on the feasibility of establishing data collection standards for a national child abuse and neglect registry with recommendations and findings concerning--

(A) costs and benefits of such data collection standards; (B) data collection standards currently employed by each State, Indian tribe, or political subdivision of a State; (C) data collection standards that should be considered to establish a model of promising practices; and (D) a due process procedure for a national registry.

(2) REPORT- Not later than 1 year after the date of enactment of this Act, the Secretary of Homeland Security shall submit to the Committees on the Judiciary in the House of Representatives and the United States Senate and the Senate Committee on Health, Education, Labor and Pensions and the House Committee on Education and the Workforce a report containing the recommendations and findings of the study on data collection standards for a national child abuse registry authorized under this subsection. (3) AUTHORIZATION OF APPROPRIATIONS- There is authorized to be appropriated $500,000 for the period of fiscal years 2006 and 2007 to carry out the study required by this subsection.

[ Go to Contents ]

Appendix B: Characteristics of State Central Child Abuse Registries¹⁶

Characteristics of State Central Child Abuse Registries

State	Central Registry Required	Reports Included	Level of Evidence	Expungement Rules	Due Process and Notification17
AK	No only the client database which includes information about alleged perpetrator.	All investigation reports but not reports of harm.	Reasonable		Not required to notify perpetrator of disposition decision.
AL	Yes	All reports excluding cases with no CAN, but case is opened.	Preponderance	In cases not indicated, after 5 years if no subsequent reports. Alleged perpetrator must request.	Appeal made to Attorney Generals office. Alleged perpetrators working with children are entitled to court hearing, those not working with children entitled to administrative review (through the department). County department and perpetrator notified of expungement. Confidentiality provisions for unsubstantiated cases. State notifies alleged perpetrator of disposition decision.
AR	Yes	All investigated reports and any record of screened out reports.	Preponderance	Unsubstantiated shall be expunged promptly. Department may keep information on unsubstantiated reports for risk assessment but may not disclose except as specified by law. True reports shall be maintained.	Appeal is available, source not specified. Department must notify reporter of whether or not an investigation has been conducted and services offered within 10 days. Perpetrator notified of disposition decision.
AZ	Yes	Substantiated reports after the appeal process.	Probable Cause	Substantiated reports are maintained in registry for 25 years from date of report.	Appeal made to Office of Administrative Hearings. Other provisions: internal review by Protective Services Review Team. Perpetrator notified of disposition decision and placement on central registry, but only after appeal process for substantiated cases.
CA	Yes	All reports except unfounded	Credible	For inconclusive or unsubstantiated reports after 10 years with no subsequent report or victim over age 18 may request name be removed.	Notify reporter of disposition. Upon request will notify an individual of their registry status.
CO	Yes	Confirmed reports	Preponderance	Sealed 10 years after childs 18th birthday. Director of registry may seal or expunge for good cause and with written notification to subject of report. Minor reports must be expunged after 6 months unless there is a Dependency and Neglect or criminal filing.	Subject notified of placement on registry, right to appeal, and expungement. If the subject is acquitted but not expunged, the Central Registry must notify the subject and also of right to appeal. Policy states that appeals are made to the Central Registry Director. Per statute, subject has right to a fair hearing before an administrative law judge. County may object to acquittals. Department has burden of proof.
CT	Yes	Confirmed reports	Reasonable Cause	After childs 18th birthday records are sealed, then expunged after 7 years.	Reporter notified of status of investigation. Appeals are made to Regional Administrator or the Director of the hotline. If subject desires, he or she may appeal to administrative hearing process of the State and then to courts. Perpetrator notified of disposition decision.
DC	Yes	Substantiated, unsubstantiated maintained separately without perpetrator identification. All reports are entered; only supported reports will remain in registry.	Credible	18th birthday of child if no suspicion or evidence that younger sibling being abused, or 5 years after end of services whichever is first.	Appeal made to Child and Family Service Agency Office of Fair Hearings. Perpetrator notified of disposition decision.
FL	No refers to hotline which keeps reports for 7 years or youngest child is age 18.	All reports	Substantiated-preponderance	Until 7 years after last entry to record or child turns 18. Unfounded and No Findings cases may be expunged if requested but must consider childs safety and likelihood of returning to system By request of subject of report or childs parent. No open cases can be expunged.	Appeal not available for alleged perpetrator a childs due process. Perpetrator notified of disposition decision.
GA	No - Ruled unconstitutional and replaced by Protective Services Data System (PSDT).		Preponderance	PSDT recording cancelled if review determines allegation should be unsubstantiated.	May request either panel review or administrative review but not both. Neither will be scheduled if decision on allegations made or pending in juvenile or superior court court is recourse. Not required to notify perpetrator of disposition decision.
HI	Yes	All reports	Reasonable, foreseeable risk	If petition dismissed by court or unsubstantiated shall be expunged within in 3 years. Ruled out, frivolous or bad faith allegations must be expunged within 60 days.	Appeal made to Department of Human Services. Appeal must be made within 90 days from which client is notified. Perpetrator notified of disposition decision.
IA	Yes	Founded on registry. Assessment data shall not be placed on registry. Department may develop rules for maintenance of data not placed on registry.	Preponderance	For founded cases data is sealed 10 years after last reported incident unless good cause is shown. Sealed data is expunged after 8 years unless case involves abuse, then kept for 30 years. Lack of preponderance that abuse has occurred.	Appeal made to DHS Appeals Section. Other provisions: record check, evaluation and court review, and registry review (informal process). Subject may file written Statement within 6 months of notification to request correction of data or findings. Department provides evidentiary hearing. Subject may appeal finding of hearing in district court. Perpetrator notified of disposition decision.
ID	Yes	Substantiated reports	Preponderance	Report is expunged no less than 5 years after Department has closed the case. As of 3/02, when a person has successfully appealed.	Appeal made to Division administrator of Family and Community Services and regional program manager review appeals. Other provisions: Division Administrator works with Regional Program Manager for review. Perpetrator notified of disposition decision.
IL	Yes	Indicated and unfounded under certain conditions (if requested to track false reporting, serious physical abuse, sexual abuse or child death).	Credible	Unfounded expunged forthwith but information may be made available to CPS units when investigating a subsequent report or to subject of report if requested within 60 days.	If subject requests removal within 60 days and department refuses or does not act within 30 days, subject has right to hearing. Hearing conducted by Director or his designee. Department has burden of proof. Appeals are made to an appeals unit. Perpetrator notified of disposition decision.
IN	Yes	Substantiated; informal adjustment made after 180 days if family fails to comply with services.	Credible	Expunged within 10 days if hearing officer finds it to be unsubstantiated, if court determines no CAN, criminal charges dismissed or result in not guilty verdict, No later than 6 months after name entered for failure to participate in a services agreement, not later than 20 years after court determines child in need of services or when victim reaches age 24. Expunged immediately for administrative or clerical error; 20 years or until victim reaches 24 years of age (if court adjudicated), or 180 days for unsubstantiated cases.	Administrative hearing may be requested within 30 days after notified of substantiated. Conducted by Administrative hearing officer. Department has burden of proof. Appeal made to Judicial Court. Perpetrator notified of placement on central registry. Not required to notify perpetrator of disposition decision.
KS	Yes	Validated reports	Clear and Convincing	After 3 years from most recent incident or if new information found or circumstances change.	Appeals made to Secretary. Perpetrator notified of disposition decision and of placement.
KY	Yes	Not specified in process of changing criteria.	Preponderance		Appeals made to Office of Performance Enhancement, Quality Initiatives Branch. Other provisions: CAPTA, local resolution hearings, service complaints. Perpetrator notified of disposition decision.
LA	Yes	Valid Reports	Reasonable	If unable to locate client or client is uncooperative, report is expunged after 3 years. Justified maintained until child is 18 or 5 years have passed since findings, whichever is longer. Fatality investigations with a valid finding maintained for 20 years.	Appeals made to court. If report is recorded as justified and no petition is subsequently filed alleging the child is in need of care, subject may file a written motion for correction in parish court where finding was made. Perpetrator notified of disposition decision.
MA	Yes	Supported cases. All reports unless determined allegation invalid (frivolous or absolute determination that no CA/N).	Reasonable	If unsubstantiated expunged after 1 year. If substantiated when child reaches age 18 or 1 year after termination of services, whichever is last.	Appeals may be made and fair hearing conducted by hearing officer. Other provisions: Superior Court, grievance procedure (e.g., against social worker, decision). Perpetrator notified of disposition decision.
MD	Yes	Indicated reports	Preponderance	Unsubstantiated within 5 years if no further reports, if ruled out within 120 days if no further reports, after 7 years for substantiated if no further entries for individual.	Appeal to office of Administrative Hearings, or to circuit court. Alleged perpetrator notified within 30 days of finding of substantiated or unsubstantiated of finding and opportunity to appeal. Perpetrator notified of disposition decision.
ME	Yes	Substantiated reports	Preponderance	Unsubstantiated expunged after 18 months with no additional report, may be retained for 5 years for Medicaid audits but stored separately. If a finding is overturned on appeal, report is expunged immediately.	Appeals go to Director, QA staff reviews, overturns are reviewed by Director or QA Supervisor. May appeal to superior court. Perpetrator notified of disposition decision.
MI	Yes	Substantiated before July 1, 1999. Afterward Category I or II (child not safe and in need of services) or perpetrators who cause serious harm.	Preponderance	If unsubstantiated (no timeframe given). If substantiated remains until subject is dead. Records can be expunged if child is safe and services dont need to be monitored, or if a case is downgraded to a less serious level.	Perpetrator must be notified within 30 days of right to request expungement or an administrative hearing if refused. Hearing is before hearing officer appointed by the department and requires preponderance of evidence. May hold a re-hearing upon new evidence or misapplication of the law. Appeal to local office, administrative hearing by local FIA office that conducted investigation. Other provision: only alleged perpetrator or alleged perpetrators attorney can request can request expungement. Perpetrator notified of disposition decision.
MN	No		Preponderance	No determination of maltreatment or need for CPS services records must be maintained for 4 years. If determination of maltreatment or need for CPS services must be maintained for 10 years after final entry.	Perpetrator notified of disposition decision.
MO	Yes	Cases that are court adjudicated, show probable cause for CAN.	Preponderance	Records are retained indefinitely. After 5 years if insufficient evidence of abuse and neglect is found, after 10 years if unable to locate subject may petition for removal after 1 year. Substantiated cases may be retained indefinitely.	Perpetrator and parents notified of findings within 90 days. Have 60 days to request reversal. May seek administrative review by child abuse and neglect review board. Standard is probable cause. If unsatisfied may request judicial review in county of residence within 60 days. Appeals made to county office. Perpetrator notified of disposition decision.
MS	Yes	All credible evidence reports	Credible	If subsequent information indicates credible evidence did not exist or decided as result of fair hearing process.	Subject may request a fair hearing within 10 days. Hearing conducted by department. Subject may be represented by an attorney. Appeal to MS Division of Family and Childrens Services Protection Unit. Other provisions: Fair hearing, attorney representation, due process provided by CAPTA. Perpetrator notified of disposition decision when evidence is found.
MT	Yes	All reports	Preponderance	Within 30 days of finding that report is unfounded.	Subject may appeal to the division administrator, then may request Fair Hearing conducted by the departments hearing officer. Perpetrator notified of disposition decision.
NC	Yes	All reports	Preponderance		Perpetrator notified of disposition decision.
ND	Yes	Services required reports	Some credible evidence	10 years from date of decision	Statute directs department to hold appeal hearings and adopt rules for doing so. Policy states that appeals made to Office of Administrative Hearings. Appeal made to DHS. Office of Administrative Hearings - conducts hearings. Perpetrator notified of disposition decision.
NE	Yes	Substantiated, petition to be filed. Inconclusive or unfounded.	Preponderance	Unfounded expunged forthwith. Any record at any time if good cause and upon notice to subjects. Individuals may request expungement.	Subject may request expungement at any time subsequent to finding. Appeal made to departments legal division and local court. Subject has right to fair hearing within department conducted by department head or designee. Department has burden of proof. These decisions may be appealed under the Administrative Procedure Act. Perpetrator notified of disposition decision.
NH	Yes	Founded reports	Preponderance	Founded retained for 7 years. Unfounded at-risk reported maintained for 3 years. There is a proposal to expunge after 7 years this would become statute and policy.	Fair hearing officer through Administrative Appeals Unit within DHHS. Other provisions: Administrative fair hearing. Perpetrators are notified and given opportunity to appeal. Not required to notify perpetrator of disposition decision.
NJ	Yes	Founded reports	Preponderance	Unfounded must be expunged (no timeframe given). Name may be removed on successful appeal through Divisions internal dispute procedures or court hearing.	Initial appeal made to local division office. Internal administrative and court hearings. Subject has right to representation, to bring witnesses, review records, ask questions and submit written statements. Perpetrator notified of disposition decision.
NM	Yes	Substantiated reports	Credible	An individual can request a review, the results of a review are noted in the record, but persons name will not be removed.	The results of any substantiated investigation which is not the subject of a court action may be reviewed through the Departments administrative review process. Not required to notify perpetrator of disposition decision.
NV	Yes	Substantiated		Unless credible evidence of CA/N must be expunged at conclusion of investigation or no later than 60 days after report filed, whichever comes first. Substantiated reports sealed no later than 10 years after child turns 18.
NY	Yes	All indicated reports	Credible	Unfounded or successfully appealed reports are sealed except for department, court or law enforcement use in subsequent investigations. All expunged 10 years after 18th birthday of youngest child named in report. Record may be expunged at any time by the Office of Children and Family Services if subject presents clear and convincing evidence that affirmatively refutes the allegation.	Subject may request amendment of record up to 90 days after notified of completion of investigation. If commissioner does not amend report within 90 days subject may request fair hearing held by department or designated agency. If denied may then request a court hearing. Perpetrator notified of disposition decision.
OH	Yes	All reports	Credible	Expungement time frames are based on disposition or case resolution. No risk resolution is expunged after 3 months, low risk = 6 months, low, moderate risk = 1 year, moderate risk = 5 years, moderate, high risk = 10 years. Substantiated reports are expunged 10 years from date of disposition, indicated reports are expunged 5 years from date of disposition and unsubstantiated reports are expunged 3 months from the date of disposition unless subsequent reports are received. In the event that subsequent reports are received, reports are linked and maintained in accordance with longest retention timeframe.	Within 3 days of completion of the assessment/investigation, the Public Children Services Agency (PCSA) shall notify the alleged perpetrator in writing of the case disposition. Administrative appeal made to, and grievance review by, the PCSA. No appeal at State level. Written copies of grievance process must be given within 3 working days of request. Perpetrator notified of disposition decision.
OK	No, but information system handles some functions.	All reports except finding of reasonable parental discipline.	Credible	Expunged only by order of the court unless other State or federal law specifies otherwise. Finding of reasonable parental discipline also expunged.	Alleged perpetrator may appeal finding and placement of name on the information system. Administrative Review conducted at State office level. Perpetrator notified of disposition decision.
OR	None for perpetrators, but registry for victims.	Founded reports	Reasonable		Perpetrator notified of disposition decision.
PA	Yes	Determined, founded, or indicated reports	Substantial evidence or clear and convincing/beyond reasonable doubt.	Unfounded are expunged 120 days after year of the report (except for cases accepted for services = 1 year plus 120 days after closure). Founded and Indicated cases are expunged when subject child reaches the age of 23. Perpetrators with date of birth or social security number information are kept indefinitely. Secretary may expunge at any time for good cause. If report is unfounded but subjects found to need services arranged by the county, the county may retain the record but identify it as unfounded.	Person named as perpetrator may request amendment or expungement within 45 days of notification. Appeals made to Secretary of the department. If refused or not acted upon in 30 days, subject has right to hearing before the departments Bureau of Hearing and Appeals. Burden of proof is on the agency. Other provisions: Three levels 1: administrative review and OCYF, 2: hearing with bureau of hearings and appeals, 3: court system. Perpetrator notified of disposition decision.
RI	Yes	All reports	Preponderance	Unfounded destroyed 3 years. Indicated cases never expunged, except if appeal is in favor of alleged perpetrator.	Appeal decided by administrative hearing officer. Can appeal to family court and as high as supreme court. Perpetrator notified of disposition decision.
SC	Yes	Only court ordered perpetrators.	Preponderance	Information identifying alleged perpetrator must be removed immediately upon determination of unfounded. Category II and III unfounded record may be retained for 1 year. Other information must be destroyed 7 years from date services are terminated. Department may maintain indicated report without information identifying a perpetrator.	Appeal to family court. Other provision: CPS Appeals Committee or family court. Perpetrator notified of disposition decision.
SD	Yes	Substantiated reports	Preponderance	Unsubstantiated must be expunged after 3 years if no subsequent reports. Substantiated may be removed after 7 years if individual requests a hearing and can prove by preponderance of the evidence that information should be removed. (Court adjudicated perpetrator may not request information removal.) Individual may request that inaccurate information be removed.	Appeal to department through informal review process. Appeal process occurs before name placed on Central Registry. Perpetrator is notified 30 days before name goes into registry, during which time they can begin appeal process. Subject may request amendment or expungement in writing within 30 days of notice of substantiated. If denied or department does not act within 30 days may request an administrative hearing. Decision made by department but is subject to judicial review. Perpetrator notified of disposition decision.
TN	Yes	All reports	Material Evidence	When a defendant is found not guilty of severe child abuse or sexual abuse.	Appeal made to Commissioners office. Other provisions: when a persons employment or volunteer status is affected. Not required to notify perpetrator of disposition decision.
TX	Yes	Disposition of reason to believe or person designated as the perpetrator.	Preponderance	Retained until 18th birthday of youngest child in the investigation or 5 years after case is closed whichever is first. If case involves removal, case is not expunged.	Alleged perpetrator may request administrative review during investigation and a hearing regarding the departments decision to release information. Appeals made to TDPRS administrator not involved in the case. Other provision: Other appeals go to PRS Ombudsman and administrative law judge. Perpetrator notified of disposition decision.
UT	Yes	Confirmed reports	Reasonable		Subject may challenge finding within 30 days of notification. Division may approve or deny. If Division requests it or fails to act within 30 days Office of Administrative Hearings and to district court holds adjudicative proceedings. Perpetrator notified of disposition decision.
VA	Yes	Central Registry for founded. Unfounded are kept in separate data system accessible only to department and local departments.	Preponderance	Unfounded are maintained for 1 year in SACWIS; not kept in registry. Date of report if no subsequent reports. Department may keep them for up to 2 more years if requested by subject. Records may be purged immediately if court orders that civil action has determined bad faith. Founded level 1 expunged 18 years past date of complaint, level 2 expunged 7 years after date of complaint and level 3, 3 years after complaint.	Appeal goes to local department; can be filed with State if appeal is upheld at local level. Other provisions: hearing with State hearing officer or circuit court. Perpetrator notified of disposition decision.
VT	Yes	Substantiated	Reasonable	Information entered on individuals under age 10 expunged upon 18th birthday. Name of alleged perpetrator for unsubstantiated cases destroyed if no court proceeding brought within 1 year, kept indefinitely for substantiated cases of notification. All records destroyed when youngest child in the case reaches the age of 18.	Person may apply at any time for review. Three levels: review by district director, review by commissioner and review by Human Services Board that holds a fair hearing. Perpetrator notified of disposition decision.
WA	No perpetrators name recorded on CAMIS information system.	All investigations, log of screened out cases	Preponderance	At the end of 6 years from receipt, an unfounded report shall be purged unless there has been a subsequent report.	Person identified as the perpetrator shall be notified by certified mail with return receipt. Person named as perpetrator in founded report has the right to request review and amendment of finding within 20 days of receipt of written notification. The appeal is initially filed with the area administrator, then they can appeal to administrative law judge and then to superior court.
WI	No	All reports	Preponderance		Perpetrator notified of disposition decision.
WV	Yes	Only cases with criminal convictions	Preponderance	All shall be destroyed six years following their preparation unless there are pending proceedings. N/A (since all are criminal convictions).	No appeal process because part of criminal process. Not required to notify perpetrator of disposition decision.
WY	Yes	Founded, under investigation, except when alleged perpetrator is a minor.	Credible	Data error, new evidence, change of findings due to administrative review, fair hearing, or district court appeal, rehabilitation is demonstrated as determined by panel, death of alleged perpetrator. Within 6 months, reports classified as under investigation must be classified as founded or unfounded. Unfounded must be expunged. Founded may be expunged if error shown, new evidence, successful appeal, rehabilitation is shown as determined by a panel appointed by Director, allegations substantiated at low risk or death of perpetrator.	Appeal may be made initially to District Manager who attempts to resolve dispute. From there they go to an administrative hearing. Hearing Officer is provided by the department. Perpetrators may also provide a written statement for the file during investigation and up to 20 days after disposition. Perpetrator notified of disposition decision.

[ Go to Contents ]

Appendix C: Data Elements Currently Collected in the Child File of the National Child Abuse and Neglect Data System¹⁸

[ Go to Contents ]

---

Endnotes

¹ While usually referred to as child abuse registries, these databases usually include information on a variety of child maltreatment types, including physical abuse, neglect, sexual abuse, emotional maltreatment, and medical neglect.  Nationally in 2007, 59 percent of child maltreatment victims experienced neglect, 11 percent were physically abused, 8 percent were sexually abused, 4 percent were psychologically maltreated, 1 percent were medically neglected, and 13 percent experienced multiple forms of maltreatment. (U.S. Department of Health and Human Services, Administration on Children, Youth and Families, 2009). Throughout this report these registries are referred to as child abuse registries to conform with statutory language in the Adam Walsh Child Protection and Safety Act and common usage in the field.

² The Senate Committee on Health, Education, Labor and Pensions and the House Committee on Education and the Workforce.

³ For instance, statements by Mr. Kennedy, Mr. Kyl, Mr. Leahy and Mr. Frist appear in the Congressional Record for July 20, 2006, beginning at: http://frwebgate.access.gpo.gov/cgi-bin/getpage.cgi?position=all&page=S8023&dbname=2006_record

⁴ The DoDs registry is known as the Child Maltreatment and Domestic Abuse Incident Reporting System, commonly referred to as the Family Advocacy Program Central Registry.

⁵ Personal communication with Ying-Ying Yuan, Walter R. McDonald and Associates, April 1, 2008.

⁶ Personal communication with David Lloyd, Director, Family Advocacy Program, Department of Defense, April 9, 2008.

⁷ The programmatic landscape of tribal child welfare programs is likely to change as a result of the recent Fostering Connections to Success and Increasing Adoptions Act of 2008, P.L. 110-351, that now allows tribes to receive title IV-E foster care and adoption assistance funds directly rather than through agreements with states.

⁸ Legal decisions from Georgia, Illinois, Indiana, Missouri, New Hampshire and New York appear in the references section.  In most of these cases low standards of evidence were the primary issue and courts demanded that a preponderance of the evidence standard be used.

⁹ While several sources have reported information on legal standards used by states in child protection investigations, data presented here are the most recent available, from Child Maltreatment 2006 (HHS 2008).

¹⁰ Descriptions of legal standards are adapted from Miriam Websters Dictionary of Law, 1996.

¹¹ Information on the National Crime Prevention and Privacy Compact is available at: http://www.fbi.gov/hq/cjisd/web%20page/pdf/cctrifold.pdf)

¹² These requirements are summarized in the NCICs Privacy Act Notice, published an 64 FR 52343.

¹³ Personal Communication with Vin McAteer, Rhode Island Division of Child Protective Services, April 15, 2008.

¹⁴ Personal communication, David Lloyd, Director of Family Advocacy Programs, Department of Defense, April 9, 2008.

¹⁵ A 2000 survey of criminal records fees is available at: http://www.search.org/files/pdf/new_web_version.pdf; and information on child abuse registry procedures including some states that charge fees are described at: http://www.hunter.cuny.edu/socwork/nrcfcpp/downloads/policy-issues/State_Child_Abuse_Registries.pdf.

¹⁶ Note: Most information in this table is drawn directly from the National Study of Child Protective Services Systems and Reform Efforts (HHS, 2003).  However, the column on level of evidence is taken from more recent information available in Child Maltreatment 2006 (HHS, 2008)

¹⁷ Notifications listed only if mentioned specifically as required in statute or policy.  Most states allow access to Central Registry information to perpetrators, parents/guardians, or reporters.  These are not listed here unless the state laws require notification.

¹⁸ The table displays the data element names and field positions for the Child File. Three columns are listed in this table, the first indicates the field position of the data element, the second column indicates the name of the data element, and the third column indicates the abbreviated name of the data element.