This Report to the Congress responds to a requirement in the Adam Walsh Child Protection and Safety Act of 2006 that directs the Department of Health and Human Services (HHS) to study the feasibility of a national registry of child maltreatment perpetrators (also known as a national child abuse registry). It follows up on an interim report issued in 2009. The feasibility study included several components. We surveyed key informants in the states to gain a better understanding of the content and operations of state registries and other repositories of information on child maltreatment perpetrators, and current practices with respect to interstate inquiries regarding alleged child maltreatment perpetrators. We also sought states' input as to their interest in participating in a voluntary national registry and the benefits and barriers they see related to their participation. Further, we reviewed relevant court cases in which aspects of states' child abuse registries have been challenged. This review informs our consideration of a due process procedure for a national registry, as required by the Adam Walsh Act. Finally, through a prevalence study we sought information to better quantify the potential benefits of a national registry of child maltreatment perpetrators, in particular by estimating the number of perpetrators nationally who have been substantiated as perpetrators of child abuse or neglect in multiple states. It is primarily in detecting such perpetrators that a national registry would have advantages beyond those of existing single state registries.
Several important challenges became clear as we examined the key issues in implementing a national registry of child maltreatment perpetrators. Among these were whether perpetrators could be accurately identified in a national registry; challenges to states' participation in a voluntary registry; and whether the information in common across existing state registries would be sufficient to produce the safety benefits sought.
State child welfare staff responding to our survey believed that the most significant potential benefit of a functional registry of child maltreatment perpetrators would be to save time on the part of workers who request maltreatment history information from other states. However, this benefit could only be realized if most States voluntarily participate in a registry, making state-to-state inquiries unnecessary. Among the 38 states responding to our survey, 26 reported that current laws or policies definitely or possibly prohibit their participation in a national registry. States are also concerned about the potential costs of participating, and some doubted a registry could provide sufficient information to be useful for child maltreatment investigations.
States' current child abuse registries contain relatively little detail on perpetrators, so a national registry would be likely to provide only a statement that "A person with the same name and birth date as [the subject] was substantiated for X type of maltreatment in Y State on Z date." For anything beyond that, the investigator would need to contact the state in which the matching record was found to inquire whether further details were available. If the primary use of a national registry is to conduct employment background checks, the additional workload related to addressing follow-up inquiries in these cases could negate any potential efficiency gain related to child maltreatment investigations. Only about one-third of states responding to our survey provide information from their state child abuse registries in response to out-of-state inquiries for employment background checks.
Our rough estimate of the federal costs of a child maltreatment perpetrators registry includes $4 million for initial development and $4 million to $6 million in annual costs to maintain the registry and respond to inquiries. These estimates do not include states' costs to participate.
Following from the feasibility study's findings, HHS concludes the following:
- Current statutory limits to the information that could be contained in a national registry would prevent the accurate identification of child maltreatment perpetrators.
The Adam Walsh Act limits the identifying information about perpetrators that could be included in a registry to their name. Because many names are common, this limitation would cause a registry to produce very high rates of inaccurate, false positive matches. That is, the vast majority of all matches produced by a registry would falsely identify an individual as a child abuser because there is someone else in the database with the same or a similar name. More accurate matches would require additional identifying information regarding perpetrators. For instance, including perpetrators' sex and date of birth would significantly decrease the rate of inaccurate matches. In addition, most states, though not all, also have information on the Social Security numbers of perpetrators, the inclusion of which could further improve the accuracy of a national registry. However, even with this additional information, an indeterminate number of false positive identifications will still occur.
Additional steps could further improve the accuracy of identification within a registry, though these steps go beyond current practice in existing single state registries. For instance, fingerprint identification is used in the databases used to produce most criminal background checks. However, no state currently includes fingerprints in their child abuse and neglect data systems and adding such information to states' child maltreatment databases would be quite costly.
- Under current law, the predominant use of a national registry would be for employment background checks not explicitly mentioned in the statute.
Congressional debate about a national registry of child maltreatment perpetrators centered on the use of such a registry during child maltreatment investigations. While employment background checks are never mentioned in the law, as currently authorized we expect such inquiries to be the predominant use of a national registry. In states that use their existing registries for employment background checks, those inquiries far outnumber requests for information to be used in child abuse investigations. In addition, assuming the registry were used regularly for employment background checks in cases where single state checks are currently conducted, resolving questions about the accuracy of disputed matches could divert state staff resources and attention away from the investigatory functions that a national registry was promoted as improving.
- If a national registry would be used for employment background checks, due process requirements for a national registry will need to be stronger than those in place in a number of states.
Given the wide variations in state practice regarding existing due process protections for individuals entered into state child maltreatment registries, there are serious, legitimate concerns about using a national registry of child maltreatment perpetrators for conducting employment background checks. The Adam Walsh Act requires that this report discuss the due process requirements that would be necessary for a national registry. As is described more fully in the research volume that accompanies this report (see Appendix 1), there is considerable case law, in some respects conflicting, regarding the Constitutionally-required due process protections that are necessary for state child abuse registries that are used for employment background checks and that would therefore be required of a national registry intended to perform such a function. In addition, this is currently an area of active litigation and legislation. Since our 2009 interim report, at least 7 states have seen legislative or court consideration of due process issues regarding their state child abuse registries.
Federal requirements would need to be designed if a registry were actually implemented, taking into account the pertinent case law at the time of implementation. But as we suggested in our Interim Report to the Congress, we believe the only practical way to handle due process issues for a federal registry would be to establish minimum standards that a state would need to certify as having been met in the particular case before a name is added to the national registry. It would be extremely impractical for the Federal Government to put in place additional protections that had not been provided at the state level at the time of the state's original determination of an individual's status as a child maltreatment perpetrator.
Our current thinking is that minimum standards for due process in a national registry that is to be used for employment-related inquiries would include: (1) that the substantiation decision used a legal standard at least as strong as preponderance of the evidence (that is, that it is more likely than not that the maltreatment occurred and that the individual designated as the perpetrator was responsible); (2) that the perpetrator had the opportunity to challenge his or her designation as a perpetrator and that such challenges are resolved in a timely fashion; and (3) that the perpetrator was notified of their inclusion on the state's registry of maltreatment perpetrators and informed about the implications of their inclusion. While each of these protections is current practice in more than half of states, under these due process requirements some states would need to make changes in their current investigation practices before it could place perpetrators' names in a national registry.
- A national registry of child maltreatment perpetrators would provide limited information for child maltreatment investigations beyond what is already available from existing single state registries.
Our study found no evidence of a widespread phenomenon of child maltreatment perpetrators who offend in multiple states. Our prevalence study revealed that 1.5 percent of persons identified as child maltreatment perpetrators in 2009 (an estimated 7,852 individuals) had any substantiated maltreatment incidents in another state within the preceding five years. The vast majority of those who did have an incident in another state had a single additional substantiation in a single additional state, most often for neglect. Exceedingly few had multiple incidents that would suggest a pattern of predatory behavior. In the 22 States that participated in our prevalence study there were 345 individuals who had more than two matches and just 44 individuals who had a substantiated child maltreatment investigation in more than one other state over a five year period. In addition, just one half of one percent of child maltreatment deaths in states participating in the study was attributed to a perpetrator who had a substantiated maltreatment report in another state (4 in total). In contrast, states' existing child maltreatment registries typically report repeated substantiated maltreatment incidents (within the same state and over a similar five-year time period) in roughly 17 percent of cases and there were over nine hundred child deaths in study states that were attributed to perpetrators who did not have substantiated maltreatment investigations in other states.
- A lack of participation in a voluntary registry system could prevent a registry from fulfilling its intent.
The best designed database of information about perpetrators will not be helpful for child maltreatment investigations if states do not populate it with information on perpetrators. There are numerous issues that may inhibit states' participation in a national registry. Chief among these is states' costs to participate. This includes the technology costs of establishing secure systems with which to exchange information with a federal registry, working with the Federal Government around issues of establishing systems to verify the identities of legitimate users of the registry, and, potentially, altering their investigation and appeals procedures to conform to federal due process standards.
Encouraging state participation could take several forms. In particular, states would be much more likely to participate in a federal registry if the Federal Government provides funding to help cover implementation costs. Alternatively, funds for related federally funded programs could be conditioned on participation in a national registry. However, should states choose to forgo such other funding, an attempt to leverage participation could undermine rather than bolster states' child protection activities.
In sum, a number of steps must be taken if a national registry of child maltreatment perpetrators were to be implemented in a way that could accurately identify perpetrators, protect individuals with common names from being falsely identified, protect the rights of those identified as perpetrators, and secure the voluntary participation of most states. Accomplishing this, a national registry could then provide limited information in response to inquiries, most likely a statement that "A person with the same name and birth date as [the subject] was substantiated for X type of maltreatment in Y state on Z date."
A decision to move forward with implementation should consider whether a national registry of child maltreatment perpetrators would successfully realize the child safety benefits that were anticipated in the discussion of this provision that surrounded the passage of the Adam Walsh Child Protection and Safety Act. In addition, particularly given current budget realities, an implementation decision should consider whether this or alternative child safety investments would be most effective in promoting the well-being of vulnerable children.