States report to us that a national registry would be of value only if it included comprehensive, timely and accurate data, and that responses to inquires were provided promptly. A great many things would be needed to realize that vision.
In particular, states report that in order to be of use, most states would need to participate early on and responses to registry inquiries would need to be prompt. Convincing a critical mass of states to participate quickly may require incentives to states for participation, such as providing funds to offset states' costs of getting started. That fewer than half of states provided the limited data needed for the prevalence study does not provide confidence that most states would voluntarily participate in a national registry if implementation proceeds. It must also be recognized that many states report definite or potential legal barriers to participation. Among the 36 states responding to our survey, 10 states reported that current laws or policies definitely prohibit their participation in a national registry and another 16 states reported possible statutory or policy barriers. A national registry could not operate effectively until state laws and policies that prevent the sharing of registry information with the Federal Government are amended.
A second challenge is that a national registry would need to establish a reliable way to restrict registry access to authorized users. Because the database would contain extremely sensitive personal data, it would be important to prevent unauthorized disclosures. Yet, as described in the interim report, verifying that inquiries are from legitimate sources could be a significant task. Our survey of key informants revealed extremely weak security in how states currently process inquiries to their existing child abuse registries. Several states have no procedures in place to verify the identity of persons making inquiries to their state registries and most require only that inquiries be made in writing on agency letterhead, which in the computer era may be easily simulated. Only a few have more stringent verification procedures, and these may be time consuming as well as unreliable. For instance, one state reports checking whether someone with the requestor's name is listed in the telephone listing of their reported employer and may call a third party at the employer's agency to verify the request. The procedures currently in place in states would not suffice for a federal database of this sensitivity, and establishing reliable verification procedures could be a time intensive and expensive task. The Interim Report discussed the more secure (and more expensive) ways in which this issue is typically handled by the Justice Department for criminal background checks. These include requirements upon local agencies for secure networks and computers dedicated only to accessing federal databases and maintained in locked rooms accessible only to authorized personnel, as well as auditable records documenting the reasons behind each inquiry submitted to the database. Such procedures would add significantly to the costs of implementation.
A third major challenge is establishing a due process procedure that would stand up to legal scrutiny without making the participation of many states infeasible. As described in Appendix D to the research report, courts with jurisdiction over a number of states have required procedural standards that are more stringent than those employed in many other states on issues including standards of proof used to substantiate the maltreatment, appeals procedures, and notification requirements. If federal requirements are set as high as some courts have required, states with less stringent standards would be excluded unless they invest significant effort in strengthening their procedures. On the other hand, setting lower requirements than what some federal and state courts have required would leave a federal registry vulnerable to legal challenge and might prevent the participation of states operating under case law that requires higher standards. It is not clear that there is an alternative that could satisfy both pressures. The most significant issue here is that 20 states currently substantiate child protection investigations using standards of proof less strong than preponderance of the evidence (which requires that the evidence shows it is more likely than not that the maltreatment occurred and that the identified perpetrator was responsible). While rulings are mixed on the topic, courts in multiple states, including the Second Federal Circuit and the Supreme Courts of Missouri and New Hampshire, have ruled that such weak standards of proof are not sufficient for the purposes of including an individual in a registry of maltreatment perpetrators, particularly if such a registry is used for employment background checks. Appendix E to the research report includes a table describing existing due process requirements associated with states' databases of information on child maltreatment perpetrators.
States participating in our survey of key informants identified important barriers to their potential participation in a national registry. The most frequent barriers mentioned were differences in definitions, findings, due process and rules for expunging old or overturned cases between (and even within) states (this barrier was mentioned by 22 states), that participating would require costly changes to their information technology systems (15 states), and that participating would require staff resources that are scarce (13 states). A number of states also mentioned that they believed a national registry of maltreatment perpetrators would be able to provide too little information to be useful and that follow-up phone calls would be needed to interpret what little information could be provided through a registry. The full research report describes in more detail what states told us about barriers to participation.