Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research Matthew T. Walton, PhD, MSSW*, Maik Schutze, MHS*, Shawndaya Thrasher, MSW, MA*, Taylor Johnston, BA* Introduction children placed in out-of-home care was more than double for the children of mothers with During Federal Fiscal Year 2018, the Adop- KEY CONSIDERATIONS: tion and Foster Care Analysis and Reporting mental illness than for other children involved with Child Protective Services (CPS).3 The risk • The Family First Prevention System reported that there were more than of removal from the home was also found to Services Act of 2018 was passed half a million children served at some point by be notably higher for the children of mothers to afford states the means to more state foster care systems.1 This population of with anxiety disorders in that study. Econom- meaningfully address some of the children has risen by nearly 40,000 since 2014 ic hardships – especially those that result in most challenging problems in child – a 5.74 percent increase in just that four-year welfare practice. homelessness and housing instability – have period. There is more than one explanation also been discovered to increase the risk of a for this concerning trend, but researchers and • This child welfare reform initiative family's involvement with CPS.6 state agencies have overwhelmingly pointed relies heavily on research to inform practice, and the data collected by to a common trio of problems that contribute The Health and Economic Burden of Child state administrative data platforms is to the heightened risk of removing children Maltreatment in the United States an exceptional source of information from their families: 1) the rise in prevalence of Children placed in foster care suffer from for program evaluation analyses. harmful substance use; 2) the incapacitating higher rates of medical and mental health effects of severe mental illness and; 3) house- conditions than their peers who have not • States that are interested in hold economic hardships. experienced foster care placement or severe implementing the Family First Prevention Services Act should explore maltreatment. Specifically, there is evidence While the opioid epidemic in the U.S. may whether their current state law allows these children exhibit a litany of health have exacerbated these problems in recent this data to be used for the purposes problems that are highly associated with their years, they are not new. Child welfare authori- of research and evaluation. exposure to adverse childhood experiences ties have long known of this trio's detrimental – such as higher rates of smoking, cardiovas- effects on families, and evidence of these cular disease, and suicide later in their lives.7-10 associations has been documented by several "Prevention is the authors.2-4 For example, analyses using data- Incident mental health problems seem to be daughter of intelligence." sets from the Centers for Disease Control and especially acute and costly for children who experience foster care placement instability Prevention (CDC) and U.S. Children's Bureau – Sir Walter Raleigh (i.e., multiple unplanned placements, disrup- produced an estimate that each 10 percent in- tions in case plans, etc.).11 Tragically, from crease in drug overdose deaths in a typical US 2003-2016, these health conditions contrib- * Kentucky Cabinet for Health and Family Services, county was associated with a 4.4 percent in- Office of Health Data and Analytics, Division of Analytics uted to mortality rates that were significantly crease in foster care entries.5 Similarly, when it 275 E. Main Street, Frankfort, KY 40601 higher for children in foster care than the Corresponding author: Matthew Walton; comes to mental illness, analyses of state level general U.S. child population.12 matthew.walton@ky.gov data in Missouri found that the proportion of This project was supported by the Kentucky Cabinet for Health and Family Services. The content of this brief is solely the re- sponsibility of the authors and does not necessarily represent the official views of the Cabinet for Health and Family Services. Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research Just as maltreatment and out-of-home care placement are dis- in terms of spending these dollars on health and social services ruptive and painful for individual children and their families, that have been determined, through published research evidence, they have been shown to adversely affect the communities where to be likely to prevent the need for foster care placements. In other families reside. Abuse and neglect broadly inhibit a child's capac- words, services that can be targeted for the most at-risk families, ity to thrive and realize their full human potential, which in turn and utilized to keep the home environment safe enough to allow decreases their likelihood for wellness and economic security over the child to remain there. By law, these so-named 'prevention ser- their lifespan – effects that can ripple through a society. When vices' must fall into at least one of the following set of categories considering the associated risks of costly illness, criminal justice outlined by Congress: involvement, educational disruption and lost productivity, CDC researchers estimated that child maltreatment accounts for $428 1. Mental health and substance abuse prevention and treatment billion in annual economic burden to the U.S. population (in 2018 services, dollars).13 2. In-home parent skill-based programs, and The central premise of Family First is that 3. Kinship navigator services.17 efforts to avoid such family disruptions are While states have been offering such services to families for warranted, and that too many children have decades, before the passage of Family First they were typically re- quired to wait until after a child had been removed from the home been removed from their parents before before they could initiate them. While there is a certain logic to sufficient preventive measures have been this approach (i.e., make sure children are safe while their parents attempted. are receiving help), national reunification data highlight how diffi- cult it is to safely bring children back to family environments they Parenting young children is challenging. Parenting young were removed from. According to the Casey Family Programs: children while coping with depression and barely making ends (1) 49% of children have a case plan that involves reunification meet is even more so. While the spectrum of maltreatment that with their parents; (2) Black and Native American children are children endure in this country ranges in terms of its severity and considerably less likely to be reunified with their parents and; (3) harm, data from the U.S. Children's Bureau finds that ~60% of children who are removed before their first birthday are much less cases are substantiated for neglect only.14 Because child neglect likely to be reunified than older children.18 is often a result of overwhelmed or under-resourced parents, it follows that there are many families whose circumstances can be It is now widely acknowledged that out-of-home care place- improved with appropriate intervention and support. ment itself is a significant traumatizing event for a child – even in cases where it is ultimately in the child's best interest.19 The Any meaningful policy that aims to improve the lives of children central premise of Family First is that efforts to avoid such family necessarily involves fortifying the capacity of their caregivers. This disruptions are warranted, and that too many children have been issue brief will outline how the federal government has crafted removed from their parents before sufficient preventive measures such a policy framework with the passage of the Family First have been attempted. This has often been the result of a dearth Prevention Services Act. It will specifically discuss this initiative of available or accessible services for families involved with CPS. in terms of the emphasis that the Act places on child welfare In such instances, the child welfare professionals serving these research to build new evidence, and how administrative data families are left to choose between a set of options were none of collected by state child welfare agencies can be used to facilitate their choices are ideal. It is traumatic to remove a child from their this research to discover which interventions can effectively keep parent, but there are also serious risks associated with leaving children safe. children in homes where severe maltreatment has occurred. How should CPS workers decide which homes are safe enough to leave The Family First Prevention Services Act of 2018 a child in? Which programs and services should states choose to Communities have called for the resources to curb their child keep children safe and keep families together? welfare problems for years. Ultimately, the collective harms of maltreatment and out-of-home care placement led community The Role of Child Welfare Research in Family First leaders to petition their political representatives to act on behalf Family First places a great deal of emphasis on states providing of these vulnerable children. In response to these calls to action services and treatments that are both safe and effective. To ensure from their constituents, Congress passed the Family First Pre- this, Family First created the Title IV-E Prevention Services Clear- vention Services Act (Family First), which was signed into law inghouse in 2018 to conduct systematic reviews of child welfare in February of 2018.15,16 Principally, Family First reforms federal services research, and then rank services on a scale in terms of funding mechanisms found in sections of Title IV of the Social the strength of the empirical support for their use with families in Security Act related to public child welfare system operations the field. This evidence hierarchy categorizes services into one of (e.g., Title IV-E, Title IV-B). It affords the states greater flexibility three conditions: 'promising' (the minimum allowable standard 2 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research in order to fund with Title IV-E dollars), 'supported' (services In many cases, these programs have developed support from their with demonstrated benefits compared to an untreated group at 6 communities and have gained the faith of the CPS workforce. months after discharge), or 'well supported' (those services with Consider an example. If a program falls below the 'promising' des- the most established efficacy – typically through randomized con- ignation and the state wishes to continue providing it as they im- trolled trials). States are financially incentivized to provide ser- plement Family First, that state is left with three choices: (1) dis- vices that have been designated 'well supported' by the Title IV-E continue the service; (2) fund the service through a non-Family Prevention Services Clearinghouse.* Table 1 outlines requirements First source, or; (3) build the evidence base around their program put forth by Family First that services must meet in order to be (and ultimately demonstrate that it meets the minimum criteria to considered for the Clearinghouse's systematic review process. receive funds under Family First). Assuming the state is unwilling to abandon their program and unable to secure non-Family First Importantly, Family First requires that a given service must have funding, they will need to navigate the aforementioned challenges been evaluated by a study that, ''(bb) was carried out in a usual of such evaluations. This brief will proceed to offer insights that care or practice setting." – criteria that are outlined both in the may be useful for states in such a position. federal law itself as well as the handbook released by the Title IV-E Prevention Services Clearinghouse.17,20 Table 1 – General Practice Requirements for Family First Evidence-Based Prevention Services These study design requirements create a challenge for states in their efforts to conduct rigorous evaluations of child welfare (I) The practice has a book, manual, or other available writings that services. Evaluations at the size and scope necessary to meet these specify the components of the practice protocol and describe standards are often expensive, complex, and time consuming. Of how to administer the practice. particular importance, states must consider the most method- (II) There is no empirical basis suggesting that, compared to its ologically appropriate and ethically sound manner of identifying likely benefits, the practice constitutes a risk of harm to those untreated comparison groups to enable valid treatment effect estimates. This is especially true when the handbook issued by the receiving it. Title IV-E Clearinghouse requires that, "Comparison groups must (III) If multiple outcome studies have been conducted, the overall be "no or minimal intervention" or "treatment as usual" groups."20 weight of evidence supports the benefits of the practice. Understandably, many state child welfare administrators are not comfortable randomly assigning high-risk families to a "no or (IV) Outcome measures are reliable and valid, and are administrat- minimal intervention" condition. ed consistently and accurately across all those receiving the practice. Evaluations at the size and scope necessary (V) There is no case data suggesting a risk of harm that was proba- to meet these standards are often expensive, bly caused by the treatment and that was severe or frequent. complex, and time consuming. Family First Prevention Services Act of 2018, § 111. Foster Care Prevention Services and Programs. Admittedly, a thorough discussion of the balance between rigorous research methodology, professional ethics, and providing the high- est quality services to families in need is beyond beyond this brief's State Child Welfare Information Systems and their Value for Research scope. Suffice it to say, the researchers charged with performing these One solution that states have used to address these research program evaluations have to think creatively and work within the in- challenges is to make administrative data from their Comprehen- escapable limitations of trying to learn as much as possible while keep- sive Child Welfare Information Systems (CCWIS) available to ing any risk to children and families below an acceptable threshold. researchers for program evaluation.21,22 Indeed, the U.S. Depart- Regardless of the ultimate approach taken, one thing is certain – this ment of Health and Human Services Administration for Children type of research requires large amounts of data if it is to be done well. and Families issued an information memorandum in 2013 en- couraging state child welfare agencies to make this data available If The Evidence-Building Process Is So Challeng- for evaluations. They claimed that doing so would further the ing, What Should States Do? purposes of providing the field with more practice and policy research, produce more theoretical and empirical studies on While Family First creates new opportunities for states, the evi- government child welfare programming, and provide high-quality dence standards that programs must meet in order to be funded insights to inform decision-making.23 through Family First pose a dilemma for many states. Many public child welfare agencies offer programs and services that fall below the 'well-supported' designation (or even 'promising') * The passage of the Family First Transition and Support Act presents a caveat. Several original funding provisions have been delayed to provide states with – this may simply be because they have yet to be the subject of a additional flexibility to accommodate Family First implementation. peer-reviewed article that documents how effective they truly are. 3 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research This approach of using CCWIS administrative data for research (xi) A person, agency, or organization engaged in a bonafide purposes radically reduces the costs associated with primary research or evaluation project, but without information identi- data collection, stores longitudinal data at an unrivaled scale, and fying individuals named in a report or record, unless having documents multiple outcomes of interest to the child welfare field that information open for review is essential to the research (out of home care placement, substantiated allegations of abuse or or evaluation, the appropriate State official gives prior written neglect, number of days a child spent in out of home care, etc.). approval, and the child, through his/her representative as cited Furthermore, sharing this data allows researchers to maintain a in paragraph (i) of this section, gives permission to release the "light touch" – allowing them to conduct their work while avoid- information. ing undue interference in clinical operations in the field. There are several examples of prevention services whose Clearinghouse How the States Have Used This Statutory Guidance ratings were bolstered or informed by studies that used these There are a number of reasons to share data related to child abuse state administrative data sources for their analyses; such as the and neglect. For example, states allow data sharing for the pur- Homebuilders® Intensive Family Preservation Services program,24 poses of care coordination for child victims31, forensic investiga- the SafeCare program,25 and the Nurse-Family Partnership pro- tion, law enforcement activities, and pre-employment screening gram.26,27 for childcare workers.32 The authors performed a review of state laws that grant explicit permission to share child welfare data for Disclosure of Confidential Child Welfare Case Re- the purposes of research or evaluation. This review found that cords: Statutory Concerns state legislatures have primarily used the CAPTA language above While this data is of high value for researchers, there are im- to outline the terms of such data use. There are 29 states that portant legal and ethical matters to consider when sharing and expressly describe disclosure of their child protective services case using it for analyses.28 Unlike Protected Health Information data to those engaged in research. The remaining 23 states have (PHI) – which has clear Federal provisions for research outlined no such language (the sum of 29 + 23 is inclusive of Washington, in HIPAA – privacy regulations related to child welfare case data DC and Puerto Rico, who also report NCANDS data to the fed- are largely written and administered at the state level. As a result, eral government). Figure 1 describes where each state falls with these regulations vary significantly across states in terms of whom regard to their legislative/statutory language related to use of their is permitted access, and under what conditions.29 Moreover, each data for this type of research. state child welfare agency makes its own set of choices about what This dichotomy should not be interpreted to mean that states information to record in its CCWIS system, the financial and per- without such explicit language in their state laws cannot or do not sonnel resources to devote towards maintenance and upgrades to use their data for research purposes. Some states, such as Dela- that system, and how available to make the records to researchers ware, allow for their executive branch agency to set the terms for and program evaluators. data sharing. Specifically, the language used includes: The federal government helps to establish data conventions, "To protect the privacy of the family and the child named in a oversees a research repository of state child welfare data,30 and report, the division shall establish guidelines concerning the provides example statutory language for sharing and protecting disclosure of information concerning the abuse and neglect that data. Specifically, amendments made in 1988 to the Child involving a child. The division may require persons to make Abuse Prevention and Treatment Act (CAPTA) established annu- written requests for access to records maintained by the divi- al reporting of state data through the National Child Abuse and sion."32 Neglect Data System (NCANDS). Similar amendments to Title IV-E of the Social Security Act in 1986 established the Adoption It is important to consider the possibility that the last time many and Foster Care Analysis and Reporting System (AFCARS) – a state legislatures have written or re-visited their statutes on disclo- data collection tool to specifically monitor the features of children sure of this data was before it was stored in the kinds of digital sys- placed in out of home care or adopted. These systems provide tems used today. The capacity of modern servers and the comput- nationally representative information, as well as standardize the ing power now available may warrant new legislative conversations fields or variables that states capture in their own systems. Espe- about how to use this data to inform state child welfare decisions. cially pertinent to this brief, CAPTA also provides model statuto- Given the new landscape ushered in by Family First legislation, ry language for states to adopt in order to guide data sharing for and the emphasis it places on supporting programmatic decisions research and evaluation purposes. This includes the following: with research, the opportunity is ripe for states to explore how their current laws help or hinder the evaluation process. (2) If a State chooses to, it may authorize by statute disclosure to any or all of the following persons and agencies, under limita- tions and procedures the State determines: 4 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research Figure 1 - States with Statutory Language Inclusive of Sharing Child Welfare Data for Research Purposes WA VT MT ME ND NH OR MN MA ID WI NY SD WY RI MI CT IA PA NV NE NJ IL OH DE UT IN CA CO WV MD VA DC KS MO KY NC TN AZ OK NM AR SC State Does Not Have Statutory MS AL GA Language That Allows for Data Sharing for Research Purposes TX LA State Does Have Statutory Language That Allows for Data FL Sharing for Research Purposes AK HI PR Timeline of Milestone Federal Child Welfare Policy Activities 1974 1988 1997 2006 2018 Child Abuse CAPTA Amendments Create Adoption and Safe The Children's Bureau Family First Prevention Prevention and the National Child Abuse Families Act (ASFA) Launches the Child Services Act (FFPSA) Child Abuse Treatment Act and Neglect Data System Welfare Information Prevention (CAPTA) and (NCANDS); National Data Gateway Treatment Act Archive on Child Abuse and (CAPTA) Neglect (NDACAN) established 5 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research Data Linking and Prospects for Future Research on brokers, encrypted computing hardware, and strong password proto- Child Safety, Permanency, and Wellbeing cols for anyone with rights to access information. Finally, investigators that see the promise of using CCWIS data for research and evaluation CCWIS data can produce worthwhile insights as a sole source of and wish to expand its access for these purposes would be wise to information, but becomes exponentially more functional when it approach their cause with the sensitivity and care of a diplomat. In is linked to other administrative data sources. Linking allows for most instances, administrative data was not collected for the purposes enhancing the quality of studies and extending the frame of their of research, has no institutional history of being used that way, and analyses. For example, pairing child welfare case information with requires new, and occasionally uncomfortable, interagency lines of Temporary Assistance for Needy Families (TANF) data would allow communication – especially in cases where data linking is concerned. a researcher to: a) gauge baseline equivalence in poverty between A patient, steadfast attitude combined with a clear and compelling treatment groups; b) test the adjuvant effects of anti-poverty programs vision for the benefits of this kind of collaboration are the hallmarks of in child welfare practice and; c) identify whether ancillary benefits success in this endeavor. are realized from certain child welfare interventions (i.e., improved household economic circumstances). In light of the many lessons Conclusion learned from the Adverse Childhood Experiences (ACE) studies7-9 With the passage of Family First, the federal government has and the line of research spawned by knowledge of the role of the social asserted that the ideal environment for children to grow up in is determinants of health, there are countless untold stories to be found with their family, in their own homes. It is a great tragedy when in states' housing records, corrections data, or Medicaid claims from this is untenable, yet Congress has now equipped the States with CPS-involved families. Indeed, this is a trail that has already begun to new tools to prevent out of home care placements from hap- be blazed by intrepid scholars. Table 3 offers examples of published pening for more families. This is, at its core, a hopeful assertion studies that illustrate the power of linking state administrative data to – that circumstances can change and families can heal. Decades learn about the experiences of children and their families. of research have provided insights about which services work, Research Ethics and Data Governance for which families, and under which conditions. More studies This capacity for knowledge building should be considered with the are needed to guide the country if this prevention-focused era utmost care and respect for the privacy of the citizens from whom the of child welfare practice is to be fully realized. The keys that will data was originally collected. Researchers would be wise to continu- unlock many of these future insights will be found in state capitols ally remember that this information is often produced as the result of across the country. The choices made there will weave together extreme trauma, and should therefore be treated with great sensitivity the story of the Family First Prevention Services Act. Which and a protective posture. The best practices that have emerged from storytellers are invited to the table – as well as the richness of their the fields of data science, academic research, and data governance narrative – depends on these choices. States may choose to view should be utilized in these instances; such as submitting proposals to their data as a symbol of faith entrusted to them by their citizens, Institutional Review Boards, de-identifying/anonymizing data when something to be put to work and used to grow their understand- possible, and executing rigorous data use agreements to clarify the ing; or they may choose to bury it in the ground. States would terms of data sharing and establish mutual expectations for its protec- be wise to view the passage of Family First as an opportunity to tion. Further protections on this data could involve the use of honest decide which course to take. Table 3 - Examples of Child Welfare Research Using Linked Data Article Data Sources Research Insight Florence et al. (2013)33 •The National Survey of Child and Adolescent Well-Being Approximately 9% of national Medicaid spending on chil- (NSCAW) dren each year is attributable to child abuse and neglect. •Medicaid Analytic Extract (MAX) Ghertner et al. (2018)5 •Healthcare Cost & Utilization Project (HCUP) County-level hospitalizations and overdose deaths re- lated to illicit substance use are significantly associated •National Child Abuse & Neglect Data System (NCANDS) with rates of child maltreatment reports and foster care •Adoption & Foster Care Analysis & Reporting System entries in that county. (AFCARS) Raghavan et al. •National Survey of Child and Adolescent Well-Being The greater the severity of maltreatment a Medicaid-en- (2016)34 (NSCAW) rolled child experiences, the higher the odds that child has of receiving a prescription for psychiatric medication. •Medicaid Analytic Extract (MAX) 6 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research References 19. Howard, K., Martin, A., Berlin, L. J., & Brooks-Gunn, J. (2011). Early mother– child separation, parenting, and child well-being in Early Head Start fami- 1. US Department of Health and Human Services. (2019). The AFCARS Report, lies. Attachment & Human Development, 13(1), 5-26. Federal Fiscal Year 2018. Retrieved on June 10, 2020. 20. Wilson, S. J., Price, C. S., Kerns, S. E. U., Dastrup, S. D., & Brown, S. R. 2. Young, N. K., Boles, S. M., & Otero, C. (2007). Parental substance use (2019). Title IV-E Prevention Services Clearinghouse Handbook of Standards disorders and child maltreatment: Overlap, gaps, and opportunities. Child and Procedures, version 1.0, OPRE Report # 2019-56, Washington, DC: Office Maltreatment, 12(2), 137-149. of Planning, Research, and Evaluation, Administration for Children and Fami- lies, U.S. Department of Health and Human Services. 3. Kohl, P. L., Jonson-Reid, M., & Drake, B. (2011). Maternal mental illness and the safety and stability of maltreated children. Child Abuse & Neglect, 35(5), 21. Witte, S. (2020). Case file analyses in child protection research: Review of 309-318. methodological challenges and development of a framework. Children and Youth Services Review, 108, 104551. 4. Pelton, L. H. (2015). The continuing role of material factors in child maltreat- ment and placement. Child Abuse & Neglect, 41, 30-39. 22. Kum, H. C., Stewart, C. J., Rose, R. A., & Duncan, D. F. (2015). Using big data for evidence based governance in child welfare. Children and Youth Services 5. Ghertner, R., Waters, A., Radel, L., & Crouse, G. (2018). The role of substance Review, 58, 127-136. use in child welfare caseloads. Children and Youth Services Review, 90, 83-93. 23. United States Administration for Children and Families. (September 12, 2013). 6. Culhane, J. F., Webb, D., Grim, S., & Metraux, S. (2003). Prevalence of child ACYF-CB-IM-13-02: Encouraging Child Welfare Agencies to share data with welfare services involvement among homeless and low-income mothers: A various Children's Bureau discretionary grant projects, or related federal- five-year birth cohort study. J. Soc. & Soc. Welfare, 30, 79. ly-funded initiatives. 7. Waehrer, G. M., Miller, T. R., Silverio Marques, S. C., Oh, D. L., & Burke 24. Kirk, R. S., & Griffith, D. P. (2004). Intensive family preservation services: Harris, N. (2020). Disease burden of adverse childhood experiences across 14 Demonstrating placement prevention using event history analysis. Social states. PloS One, 15(1), e0226134. Work Research, 28(1), 5-16. 8. Hughes, K., Bellis, M. A., Hardcastle, K. A., Sethi, D., Butchart, A., Mikton, C., 25. Gershater-Molko, R. M., Lutzker, J. R., & Wesch, D. (2002). Using recidivism ... & Dunne, M. P. (2017). The effect of multiple adverse childhood experiences data to evaluate project safecare: Teaching bonding, safety, and health care on health: a systematic review and meta-analysis. The Lancet Public Health, skills to parents. Child Maltreatment, 7(3), 277-285. 2(8), e356-e366 26. Olds, D. L., Henderson, C. R., Chamberlin, R., & Tatelbaum, R. (1986). 9. Afifi, T. O., MacMillan, H. L., Boyle, M., Cheung, K., Taillieu, T., Turner, S., Preventing child abuse and neglect: a randomized trial of nurse home visita- & Sareen, J. (2016). Child abuse and physical health in adulthood. Health tion. Pediatrics, 78(1), 65-78. Reports, 27, 10–18. 27. Matone, M., Kellom, K., Griffis, H., Quarshie, W., Faerber, J., Gierlach, P., ... & 10. Turney, K., & Wildeman, C. (2016). Mental and physical health of children in Cronholm, P. F. (2018). A mixed methods evaluation of early childhood abuse foster care. Pediatrics, 138(5). prevention within evidence-based home visiting programs. Maternal and Child Health Journal, 22(1), 79-91. 11. Rubin, D. M., Alessandrini, E. A., Feudtner, C., Mandell, D. S., Localio, A. R., & Hadley, T. (2004). Placement stability and mental health costs for children 28. Barclay, C. (2013). When the Need to Know Outweighs Privacy: Granting in foster care. Pediatrics, 113(5), 1336-1341. Access to Child Welfare Records in the Fifty States. Child. Legal Rts. J., 34, 175. 12. Chaiyachati, B. H., Wood, J. N., Mitra, N., & Chaiyachati, K. H. (2020). All- 29. Child Welfare Information Gateway. (2017). Disclosure of confidential child Cause mortality among children in the US foster care system, 2003-2016. abuse and neglect records. Washington, DC: U.S. Department of Health and JAMA Pediatrics. Human Services, Children's Bureau. 13. Peterson, C., Florence, C., & Klevens, J. (2018). The economic burden of child 30. National Data Archive on Child Abuse and Neglect. https://www.ndacan.acf. maltreatment in the United States, 2015. Child Abuse & Neglect, 86, 178–183. hhs.gov/. 14. U.S. Department of Health and Human Services, Administration for Children 31. Greiner, M. V., Beal, S. J., Dexheimer, J. W., Divekar, P., Patel, V., & Hall, E. and Families, Administration on Children, Youth, and Families, Children's S. (2019). Improving Information Sharing for Youth in Foster Care. Pediat- Bureau. (2020). Child Maltreatment 2018. rics, 144(2). 15. Lindell, K. U., Sorenson, C. K., & Mangold, S. V. (2020). The Family First 32. Delaware Annotate Code, Title 16, § 906. Prevention Services Act: A New Era of Child Welfare Reform. Public Health Reports, 135(2), 282-286. 33. Florence, C., Brown, D. S., Fang, X., & Thompson, H. F. (2013). Health care costs associated with child maltreatment: impact on Medicaid. Pediat- 16. Villalpando, F. (2019). Family First Prevention Services Act: An Overhaul of rics, 132(2), 312-318. National Child Welfare Policies. Child Legal Rights Journal, 39, 283. 34. Raghavan, R., Brown, D. S., Allaire, B. T., Ross, R. E., & Landsverk, J. (2016). 17. Family First Prevention Services Act of 2018, § 111 - Foster Care Prevention Associations between magnitude of child maltreatment and Medicaid expen- Services and Programs. ditures for psychotropic medications. Psychiatric Services, 67(8), 916-919. 18. Casey Family Programs (2011). Breakthrough Series Collaborative: Timely Permanency through Reunification. 7 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research Appendix – State Statutory Language Related to Sharing Child Welfare Data for the Purposes of Research for Each U.S. State Does State Have State Explicit Language Law or Statute Language Related to Research & Evaluation Involving Research? AL Code (5) For use by any person engaged in bona fide research who is authorized to have access Alabama Yes to such information by the Commissioner of the Department of Human Resources; or § 26-14-8 Alaska Stat. §§ 47.17.040; Alaska No No direct reference to research or evaluation. 47.10.093 AZ Rev. Stat. 2. DCS information to a person who is conducting bona fide research, the results of which Arizona Yes § 8-807 might provide DCS information that is beneficial in improving the department. (6) (A) A person, agency, or organization engaged in a bona fide research or evaluation project having value as determined by the Department of Human Services and the Department of Arkansas State Police in future planning for programs for maltreated children or in developing policy directions. AR Code Arkansas Yes (B) However, any confidential information provided for a research or evaluation project § 12-18-909 under this subdivision (g)(6) shall not be redisclosed. (C) However, if a research or evaluation project results in the publication of related material, confidential information provided for a research or evaluation project under this subdivision (g)(6) shall not be disclosed CA Penal Code California No No direct reference to research or evaluation. § 11167.5 (o) A person, agency, or organization engaged in a bona fide research or evaluation project, but without information identifying individuals named in a report, unless having said identi- CO Rev. Stat. fying information open for review is essential to the research and evaluation, in which case Colorado Yes § 19-1-307 the executive director of the state department of human services shall give prior written approval and the child through a legal representative shall give permission to release the identifying information (10) An individual conducting bona fide research, provided no information identifying the Ann. Stat. §§ 17a-28; subject of the record is disclosed unless (A) such information is essential to the purpose Connecticut Yes 17a-101k of the research; and (B) the department has given written approval for the use of such information Delaware No Ann. Code Tit. 16, § 906 No direct reference to research or evaluation. (i) Any person authorized by the department who is engaged in the use of such records or information for bona fide research, statistical, or audit purposes. Such individual or entity shall enter into a privacy and security agreement with the department and shall comply Florida Yes Ann. Stat. § 39.202 with all laws and rules governing the use of such records and information for research and statistical purposes. Information identifying the subjects of such records or information shall be treated as confidential by the researcher and shall not be released in any form. Ann. Code §§ 49-5-41; b)(1) ...individuals who are engaged in legitimate research for educational, scientific, or Georgia Yes 49-5-185 public purposes and who comply with the provisions of this subsection.. Hawaii No Rev. Stat. § 350-1.4 No direct reference to research or evaluation. Ann. Code §§ 16-1626; Idaho No No direct reference to research or evaluation. 16-1629 Comp. Stat. Ch. 325, § (10) Any person authorized by the Director, in writing, for audit or bona fide research Illinois Yes 5/11.1 purposes Ann. Stat. §§ 31-33-18-1; Indiana No No direct reference to research or evaluation. 31-33-26-7 e. Others as follows, but only with respect to report data and disposition data for cases of founded child abuse subject to placement in the registry pursuant to section 232.71D: (1) To a person conducting bona fide research on child abuse, but without data identify- Iowa Yes Ann. Stat. § 235A.15 ing individuals named in a child abuse report, unless having that data open to review is essential to the research or evaluation and the authorized registry officials give prior written approval and the child, the child's guardian or guardian ad litem and the person named in a report as having abused a child give permission to release the data. 8 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research Does State Have State Explicit Language Law or Statute Language Related to Research & Evaluation Involving Research? Kansas No Ann. Stat. § 38-2209 No direct reference to research or evaluation. Kentucky No Rev. Stat. § 620.050 No direct reference to research or evaluation. (5)(a) The department may release the information described in Paragraph (1), except names and any other identifying information, to a professional person or professor or grad- Louisiana Yes Rev. Stat. § 46:56 uate student of a college or university who is engaged in bona fide professional, academic, or scholarly research in the field of child welfare services or to a duly authorized person conducting an audit of the department. F. Any person engaged in bona fide research, provided that no personally identifying information is made available, unless it is essential to the researcher and the commissioner Maine Yes Rev. Stat. Tit. 22, § 4008 or the commissioner's designee gives prior approval. If the researcher desires to contact a subject of a record, the subject's consent shall be obtained by the department prior to the contact. Family Law § 5-707; Hum. Maryland No No direct reference to research or evaluation. Serv. Code § 1-202 Massachu- Ann. Laws Ch. 119, §§ No No direct reference to research or evaluation. setts 51E & 51F (i) A person, agency, or organization engaged in a bona fide research or evaluation project. The person, agency, or organization shall not release information identifying a person named in the report or record unless that person's written consent is obtained. The person, agency, or organization shall not conduct a personal interview with a family without the family's prior consent and shall not disclose information that would identify the child or the Michigan Yes Comp. Laws § 722.627 child's family or other identifying information. The department director may authorize the release of information to a person, agency, or organization described in this subdivision if the release contributes to the purposes of this act and the person, agency, or organization has appropriate controls to maintain the confidentiality of personally identifying information for a person named in a report or record made under this act. Ann. Stat. § 626.556, Minnesota No No direct reference to research or evaluation. Subd. 11 (e) Any person engaged in a bona fide research purpose, provided that no information Ann. Code §§ 43-21-257; identifying the subject of the records shall be made available to the researcher unless it is Mississippi Yes 43-21-261 absolutely essential to the research purpose and the judge gives prior written approval, and the child, through his or her representative, gives permission to release the information. (7) Any person engaged in a bona fide research purpose, with the permission of the director; provided, however, that no information identifying the child named in the report as a victim or the reporters shall be made available to the researcher, unless the identifying information is essential to the research or evaluation and the child named in the report as a victim or, if the child is less than eighteen years of age, through the child's parent, or guardian provides written permission; Missouri Yes Ann. Stat. § 210.150 (13) Any person who is a tenure-track or full-time research faculty member at an accredit- ed institution of higher education engaged in scholarly research, with the permission of the director. Prior to the release of any identifying information, the director shall require the re- searcher to present a plan for maintaining the confidentiality of the identifying information. The researcher shall be prohibited from releasing the identifying information of individual cases. (j) a person, agency, or organization that is engaged in a bona fide research or evaluation Montana Yes Ann. Code § 41-3-205 project and that is authorized by the department to conduct the research or evaluation; (5) Any person engaged in bona fide research or auditing. No information identifying the Nebraska Yes Ann. Stat. § 28-726 subjects of the report of child abuse or neglect shall be made available to the researcher or auditor. (g) A person engaged in bona fide research or an audit, but information identifying the Nevada Yes Rev. Stat. § 432B.280 subjects of a report must not be made available to the person (e) Access to case records by a person conducting a bona fide research or evaluation project, provided that no information identifying the subject of the record shall be disclosed New Hamp- Yes Rev. Stat. § 170-G:8-a unless such information is essential to the purpose of the research, each person identified shire in the record or an authorized representative has authorized such disclosure in writing, and the department has granted its approval in writing. 9 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research Does State Have State Explicit Language Law or Statute Language Related to Research & Evaluation Involving Research? d. The department may release the records and reports referred to in subsection a. of this section to any person engaged in a bona fide research purpose, provided, however, that no names or other information identifying persons named in the report shall be made available New Jersey Yes Ann. Stat. § 9:6-8.10a to the researcher unless it is absolutely essential to the research purpose and provided further that the approval of the Commissioner of Children and Families or his designee shall first have been obtained. New Mexico No Ann. Stat. § 32A-4-33 No direct reference to research or evaluation. (h) any person engaged in a bona fide research purpose provided, however, that no infor- mation identifying the subjects of the report or other persons named in the report shall be New York Yes Soc. Serv. Law § 422 made available to the researcher unless it is absolutely essential to the research purpose and the department gives prior approval. North Car- No Gen. Stat. § 7B-2901 No direct reference to research or evaluation. olina g. A person engaged in a bona fide research purpose approved by the department's institutional review board; provided, however, that no individually identifiable information as North Dakota Yes Cent. Code § 50-25.1-11 defined in section 50-06-15 is made available to the researcher unless the information is absolutely essential to the research purpose and the department gives prior approval. Rev. Code §§ 2151.421; Ohio No No direct reference to research or evaluation. 2151.423 6. Any person or agency for research purposes, if all of the following conditions are met: a. The person or agency conducting the research is employed by the State of Oklahoma or is under contract with the state and is authorized by the Department to conduct the research, and Ann. Stat. Tit. 10A, § Oklahoma Yes 1-6-103 b. The person or agency conducting the research ensures that all documents containing identifying information are maintained in secure locations and access to the documents by unauthorized persons is prohibited; that no identifying information is included in documents generated from the research conducted; and that all identifying information is deleted from documents used in the research when the research is completed (3) The Department of Human Services may make reports and records compiled under the Oregon Yes Rev. Stat. § 419B.035 provisions of ORS 419B.010 to 419B.050 available to… or for research when the Director of Human Services gives prior written approval. Pennsylvania No Cons. Stat. Tit. 23, § 6340 No direct reference to research or evaluation. (e) Any person conducting bona fide data research. The Secretary or his or her designee shall evaluate and determine whether or not written authorization should be given. Said person shall not be given information related to the identity of the informant or informants, Puerto Rico Yes Ann. Laws Tit. 8, § 446f the victim of abuse, or the subject of the report. The criteria for compliance with the provi- sions of this subsection shall be established by the Secretary. The confidentiality provisions contained in this chapter shall also extend to the research work described herein. Rhode Island No Gen. Laws § 42-72-8 No direct reference to research or evaluation. South Car- (17) any person engaged in bona fide research with the written permission of the state Yes Ann. Code § 63-7-1990 olina director or the director's designee, subject to limitations the state director may impose South Dakota No Ann. Stat. § 26-8A-13 No direct reference to research or evaluation. (4) Any person engaged in bona fide research or audit purposes. However, no information identifying the subjects of the report shall be made available to the researcher unless such information is absolutely essential to the research purpose, suitable provision is made to maintain the confidentiality of the data and the department has given written approval Ann. Code §§ 37-1-612; Tennessee Yes 37-5-107 (f) The department shall adopt such rules as may be necessary to carry out the following purposes: …, and for cooperation with scientific and governmental research on child abuse and neglect. Texas No Fam. Code § 261.201 No direct reference to research or evaluation. 10 Privacy, Policy, & Permanency: The Family First Prevention Services Act and the use of Administrative Data for Child Welfare Research Does State Have State Explicit Language Law or Statute Language Related to Research & Evaluation Involving Research? Ann. Code §§ 62A-4a-412; (i) a person engaged in bona fide research, when approved by the director of the division, if Utah Yes 63G-2-202(10) the information does not include names and addresses Ann. Stat. Tit. 33, §§ 4916; Vermont No No direct reference to research or evaluation. 4921 Ann. Code §§ 63.2-1514; Virginia No No direct reference to research or evaluation. 63.2-1515; 63.2-1503 Washington No Rev. Code § 13.50.100 No direct reference to research or evaluation. The staff which maintains the Child Protection Register may release information from said Washington, Register for research and evaluation only upon an order of the Superior Court of the District Yes Ann. Code § 4-1302.03 DC of Columbia; provided, however, that no information identifying the persons named in a report shall be made available to the researcher or evaluator. West Virginia No Ann. Code § 49-5-101 No direct reference to research or evaluation. 12. A person engaged in bona fide research, with the permission of the department. Wisconsin Yes Ann. Stat. § 48.981 Information identifying subjects and reporters may not be disclosed to the researcher. Wyoming No Ann. Stat. § 14-3-214 No direct reference to research or evaluation. 11