[Federal Register: January 11, 2008 (Volume 73, Number 8)]
[Proposed Rules]               
[Page 2081-2142]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr11ja08-13]                         


[[Page 2081]]

-----------------------------------------------------------------------

Part II





Department of Health and Human Services





-----------------------------------------------------------------------



Administration for Children and Families



-----------------------------------------------------------------------



45 CFR Part 1355



Adoption and Foster Care Analysis and Reporting System; Proposed Rule


[[Page 2082]]


-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Administration for Children and Families

45 CFR Part 1355

RIN 0970-AC23

 
Adoption and Foster Care Analysis and Reporting System

AGENCY: Administration on Children, Youth and Families (ACYF), 
Administration for Children and Families (ACF), Department of Health 
and Human Services (DHHS).

ACTION: Notice of proposed rulemaking (NPRM).

-----------------------------------------------------------------------

SUMMARY: The Administration for Children and Families (ACF) is 
proposing to amend the Adoption and Foster Care Analysis and Reporting 
System (AFCARS) regulations at 45 CFR 1355.40 and the appendices to 
part 1355 to modify the requirements for States to collect and report 
data to ACF on children in out-of-home care and in subsidized adoption 
or guardianship arrangements with the State. This proposed rule also 
implements the AFCARS penalty requirements of the Adoption Promotion 
Act of 2003 (Pub. L. 108-145).

DATES: In order to be considered, we must receive written comments on 
this notice of proposed rulemaking on or before March 11, 2008.

ADDRESSES: Interested persons are invited to submit written comments 
regarding this proposed rule via regular postal mail to Kathleen 
McHugh, Director, Division of Policy, Children's Bureau, Administration 
on Children, Youth and Families, Administration for Children and 
Families, 1250 Maryland Avenue, SW., Suite 800, Washington, DC 20024. 
Please be aware that mail sent to us may take an additional 3-4 days to 
process due to changes in mail handling resulting from the anthrax 
crisis of October 2001. If you choose to use an express, overnight, or 
other special delivery method, please ensure first that they are able 
to deliver to the above address. You may also transmit comments 
electronically via e-mail to CBComments@acf.hhs.gov or via the Internet 
at: http://www.regulations.gov. We urge you to submit comments 

electronically to ensure they are received in a timely manner. Please 
be sure to include identifying information on any correspondence. To 
download an electronic version of the rule, you should access http://www.regulations.gov/.
 Comments will be available for public inspection 

Monday through Friday 8:30 a.m. to 5 p.m. at the above address by 
contacting Miranda Lynch at (202) 205-8138.
    Comments that concern information collection requirements must be 
sent to the Office of Management and Budget at the address listed in 
the Paperwork Reduction Act section of this preamble. A copy of these 
comments also may be sent to the Department representative listed 
above.

FOR FURTHER INFORMATION CONTACT: Kathleen McHugh, Director of Policy, 
Children's Bureau, Administration on Children, Youth and Families, 
(202) 401-5789 or by e-mail at kmchugh@acf.hhs.gov. Do not e-mail 
comments on the Notice of Proposed Rulemaking to this address.

SUPPLEMENTARY INFORMATION: The preamble to this notice of proposed 
rulemaking is organized as follows:

I. Background on Foster Care and Adoption Data Collection
II. Consultation and Regulation Development
III. Overview of Major Revisions to AFCARS
IV. Section-by-Section Discussion of NPRM
V. Impact Analysis
VI. List of Subjects

I. Background on Foster Care and Adoption Data Collection

    In 1982, the Department, through a grant to the American Public 
Human Services Association (formerly the American Public Welfare 
Association), implemented the Voluntary Cooperative Information System 
(VCIS) to collect aggregate information annually about children in 
foster care and special needs adoption from State child welfare 
agencies. While some States reported data to VCIS, by 1986, Congress 
and other stakeholders recognized that there were a number of 
weaknesses in VCIS. Namely, VCIS was criticized for intermittent 
reporting by the States; the use of a variety of reporting periods; a 
lack of common definitions for data elements; a lack of timeliness of 
the data, poor data quality, and the collection of aggregate data which 
had limited analytic utility.
    As a result of these and other concerns, the President signed 
Public Law 99-509 on October 21, 1986, which in part added section 479 
to title IV-E of the Social Security Act (the Act). Section 479 of the 
Act describes the series of steps that the Department of Health and 
Human Services (DHHS) was required to take to establish a national data 
collection system for adoption and foster care. We were required to 
develop a system that avoids unnecessary diversion of resources from 
agencies responsible for adoption and foster care and assures that the 
data collected is reliable and consistent over time and across 
jurisdictions through the use of uniform definitions and methodologies. 
Furthermore, the law required the system to provide comprehensive 
national information on the demographic characteristics of adopted and 
foster children and their parents (biological, foster and/or adoptive 
parents); the status of the foster care population (including the 
number of children in foster care, length of placement, type of 
placement, availability for adoption, and goals for ending or 
continuing foster care); the number and characteristics of children 
placed in or removed from foster care; children adopted or with respect 
to whom adoptions have been terminated; children placed in foster care 
outside the State which has placement and care responsibility; and, the 
extent and nature of assistance provided by Federal, State and local 
adoption and foster care programs and the characteristics of the 
children to whom such assistance is provided.
    The President signed into law the Omnibus Budget Reconciliation Act 
of 1993 (Pub. L. 103-66) on August 19, 1993. Public Law 103-66 provides 
States with the opportunity to obtain title IV-E funds to plan, design, 
develop, and implement a Statewide Automated Child Welfare Information 
System (SACWIS). On December 22, 1993, ACF published final rules to 
establish the AFCARS and implement SACWIS.
    In the AFCARS final rule we required States to submit certain data 
to us on a semi-annual basis about children in foster care and 
adoptions that involve the State agency. The rule required States that 
chose to develop a SACWIS to ensure that their system could report 
information to AFCARS. We also set forth data standards that each State 
must meet to be considered in compliance with the AFCARS requirements.
    States were required to report the first AFCARS data to us for FY 
1995. However, it was not until FY 1998, when we implemented AFCARS 
financial penalties for a State not submitting data or submitting data 
of poor quality that the data became stable enough for ACF and others 
to use for a wide variety of purposes.
    The President signed the Adoption and Safe Families Act of 1997 
(Pub. L. 105-89) in November 1997, which required the use of AFCARS 
data for two specific activities: The calculation of Adoption Incentive 
Payments (section 473A of the Act) and the Child Welfare Outcomes 
Annual Report (section 479A of the Act). Since that

[[Page 2083]]

time, data from AFCARS also has been used to provide samples for the 
Child and Family Services Reviews (CFSR) and title IV-E reviews; to 
develop outcome and performance measures for the CFSR, the Office of 
Management and Budget's Program Assessment and Rating Tool (PART) and 
the Government Performance and Results Act (GPRA); to calculate State 
allocations for the Chafee Foster Care Independence Program (section 
477 of the Act); to generate short- and long-term budget projections; 
to conduct trend analyses for short- and long-term program planning; 
and to respond to requests for information from the Congress, other 
Federal agencies, States, media and the public about children in foster 
care and children being adopted.
    Due to a settlement of several States' appeals of AFCARS penalties, 
ACF discontinued withholding Federal funds for a State's failure to 
comply with AFCARS requirements in January 2002 (see ACYF-CB-IM-02-03). 
However, late in 2003 the President signed the Adoption Promotion Act 
of 2003 (Pub. L. 108-145), which required ACF to institute specific 
financial penalties for a State's noncompliance with AFCARS 
requirements. We notified States in ACYF-CB-IM-04-04 issued on Feb. 17, 
2004, that we will not assess penalties until we issue revised final 
AFCARS regulations, the subject of this proposed rule.

II. Consultation and Regulation Development

    In the preamble to the AFCARS final regulation issued in 1993, we 
indicated that we would revisit the regulations to assess how we may 
improve AFCARS (58 FR 67917). This proposed rule is the culmination of 
that process. We undertook an intensive review of every aspect of 
AFCARS in developing the proposals in this NPRM. We analyzed the types 
of technical assistance requested by and provided to States, our 
findings from AFCARS assessment reviews, and reports from the past 
several years issued by the Government Accounting Office (GAO) and the 
Department's Office of the Inspector General (OIG) on AFCARS-related 
issues.
    ACF also consulted with the public through a variety of focus 
groups and a Federal Register notice (68 FR 22386, April 28, 2003) 
seeking comments. More than 80 people participated in the focus groups, 
and over 40 individuals and groups submitted written comments in 
response to the Federal Register announcement. Thirty-two States, 15 
national organizations and 20 interested members of the public provided 
comments through one or more of these mechanisms.
    During consultation we solicited feedback on:
     The specific strengths of AFCARS;
     The specific weaknesses of AFCARS or suggestions for areas 
of improvement, including ideas about how the suggested improvement 
could be made and how the Federal government could facilitate the 
changes;
     Data elements currently in AFCARS that could be deleted 
and any elements that should be added;
     Strategies to improve data quality for AFCARS, including 
the use of incentives; and
     How the AFCARS data files are structured and submitted.
    Many stakeholders recognized that AFCARS has considerable strengths 
that include, but are not limited to: The ability to produce timely 
reports that estimate the number of children in foster care and those 
being adopted; the ability to support in-depth analyses of case-level 
data; and the ability to generate information that had not been 
anticipated when AFCARS was established.
    However, commenters also noted that expansion of the use of AFCARS 
data has highlighted areas that need improvement. For example, there 
are substantive gaps in the areas covered by the current data elements 
such as information about adoption disruptions, the placement 
experiences of sibling groups, the demographics and assistance provided 
to children under adoption assistance agreements, where children are 
placed when they are placed out-of-State, and the identification of the 
different populations served by child welfare agencies (e.g. children 
in out-of-home care due primarily to their involvement with juvenile 
justice or their need for mental health services). In particular, 
stakeholders point out that data from AFCARS is insufficient to support 
expanded analysis of data for the CFSRs and other performance measures. 
Many commenters also believe that we need to refine some of the 
definitions of AFCARS data elements and their response categories (e.g. 
expand reasons for exit), and how these and other changes in data 
elements might be facilitated in the future. In addition to the need 
for new and refined data elements, stakeholders noted that the data 
structure of AFCARS may need to be revised to take advantage of 
advances in information technology and/or to make possible the 
utilization of a wider variety of analytical techniques.
    The section-by-section summary provides more discussion on how 
specific comments factored into our proposal.

III. Overview of Major Revisions to AFCARS

    In this NPRM we are focusing our improvements on five general 
areas: Restructuring the data to capture more information over time; 
expanding the reporting populations; capturing greater detail on 
children in out-of-home care; improving the quality of data; and 
eliminating unnecessary data and inefficiencies in the data submission 
process.

Restructuring Data

    We propose that AFCARS data support longitudinal data analysis by 
capturing more comprehensive information on a child's experiences in a 
State's foster care system. The existing AFCARS requires that States 
report some living arrangement, provider, and permanency information 
relative to the child's most recent experiences in his/her most recent 
foster care episode only. We propose instead, that States collect and 
report information on: (1) The timing and circumstances of each of the 
child's removals from home and placements in out-of-home care, (2) the 
timing and type of each permanency plan decision (e.g., reunification 
or adoption) made for a child, (3) the time span and nature of each 
living arrangement the child experiences while in foster care, (4) 
details on each foster family home provider, if applicable, and (5) the 
timing and circumstances of each of the child's exits from out-of-home 
care.

Expanding Reporting Populations

    We propose to expand the foster care reporting population to 
include, generally, all children who have been placed away from their 
parents or legal guardians for whom the State title IV-B/IV-E agency 
has placement and care responsibility. In doing so, we are also 
renaming the reporting population as the ``out-of-home care reporting 
population.'' This reporting population includes children who are in 
living arrangements that are not traditionally considered foster care 
under our title IV-B and IV-E program rules. Children who are under the 
placement and care responsibility of the State agency and are placed in 
juvenile justice facilities and other living arrangements which are 
non-reimbursable under title IV-E such as psychiatric treatment 
facilities are included in the revised AFCARS out-of-home care 
reporting population. In the existing regulation, children who were in 
juvenile justice facilities and other facilities not traditionally 
considered

[[Page 2084]]

foster care were included in AFCARS in limited circumstances. We also 
have expanded our reporting population to include children who are the 
subject of a guardianship subsidy agreement, whereas these children are 
not currently reported to AFCARS.

Capturing Greater Detail

    We have added and clarified a number of elements so States may 
provide us with greater detail on the demographics and circumstances of 
children in out-of-home care. These changes are designed to permit 
enhanced analysis of the factors that may affect a child's permanency 
and well-being and include:
     New elements that allow us to identify certain populations 
of children who are dealing with issues other than child maltreatment, 
such as children who are involved in the juvenile justice system prior 
to and during their out-of-home care stay and those who are out of 
their own homes to obtain mental health services;
     New elements for States to update information on the 
circumstances affecting the child and family during the child's out-of-
home care stay;
     New elements that allow us to identify where more than one 
family member is in out-of-home care, such as sibling groups and minor 
parents who have their children with them in out-of-home care;
     New elements to better describe the household composition 
of the homes from which children are removed and the location and type 
of living arrangements in which children are placed by the State 
agency;
     Elements that tell us about a child's well-being including 
new elements on immunizations and educational performance as well as 
clarified elements on children's health, behavioral and mental health 
conditions;
     Revised and new elements that enhance our understanding of 
domestic and intercountry adoptions, prior adoptions and adoption 
disruptions, displacements and dissolutions; and,
     Revised and new elements designed to better track State 
and Federal financial support of foster care, adoption subsidies, 
adoption nonrecurring costs and guardianships.

Improving Data Quality

    We propose to improve AFCARS data quality in several ways. First, 
we propose to clarify many existing element descriptions that 
stakeholders informed us were problematic. Second, we propose to 
strengthen our assessment and identification of errors within a State's 
data file. In particular, we are proposing to develop cross-file checks 
to identify defaults and other faulty programming that result in skewed 
data across a State's entire data file. Finally, we propose to 
implement penalties for States that do not meet our file and data 
quality standards for AFCARS consistent with section 474(f) of the Act.

Eliminating Unnecessary Features

    We propose to eliminate a number of features in the AFCARS 
regulation that are no longer useful to us or the States. We propose to 
dispose of State reporting of summary adoption and foster care files, 
merge most currently reported adoption information into the foster care 
data file and take technical submission requirements out of the 
regulation.
    These major changes to AFCARS along with all other features of the 
proposed database are detailed in the section-by-section discussion 
below.

IV. Section-by-Section Discussion of NPRM

    The reader should note that the proposed regulations will replace 
in their entirety the existing AFCARS regulations at 45 CFR 1355.40 and 
the appendices to part 1355. Although we are retaining certain 
requirements of the existing AFCARS, such requirements are often set 
forth in different and new sections or paragraphs in this proposed 
rule.

1355.40 Scope of the Adoption and Foster Care Analysis and Reporting 
System

    In section 1355.40 we propose a scope statement for AFCARS. The 
proposed scope statement explains which entities must report data to 
ACF and the data that those entities must report.
Section 1355.40(a)
    In paragraph (a), we propose that all State agencies that 
administer titles IV-B and IV-E of the Act collect and report 
information to AFCARS. This is consistent with the existing scope of 
AFCARS and our legislative authority in section 479 of the Act. 
Currently, all States, the District of Columbia and Puerto Rico operate 
title IV-B and IV-E programs.
Section 1355.40(b)
    In paragraph (b), we describe the scope of the AFCARS requirements. 
We propose that a State collect and submit to us, on a semi-annual 
basis, information on a child's experiences in out-of-home care and 
information on children under adoption assistance and guardianship 
subsidy agreements.
    The scope of the proposed requirements is broader than the current 
AFCARS in three significant ways. First, the scope of the AFCARS out-
of-home care reporting population, currently known as the ``foster 
care'' reporting population has changed to include, generally, all 
children who are living away from their parents or legal guardians for 
whom the State agency has placement and care responsibility. Currently, 
the AFCARS foster care reporting population focuses primarily on 
children in foster care settings as defined by the title IV-B and IV-E 
programs only. Second, we are expanding the scope of certain 
information to include a child's entire historical and current 
experience in out-of-home care so that we can establish a more 
comprehensive and longitudinal database. Currently State agencies 
report to AFCARS limited information on a child's most recent and first 
foster care episode during the report period. Finally, we propose that 
States report on children involved in adoption agreements and 
guardianship subsidy arrangements on an ongoing basis. At the present 
time, State agencies report to AFCARS information on finalized 
adoptions in which the State agency was involved at the point of 
finalization only. In large part, we are expanding the scope of AFCARS 
data in response to overwhelming support for doing so from stakeholders 
and to meet our program needs. The full extent of these proposed 
changes is explained further in subsequent sections on the reporting 
population and data elements.
    A few commenters suggested that ACF also consider expanding the 
scope of AFCARS to require State agencies to collect and report 
detailed information on children who receive child welfare services in 
their own homes. We believe that requiring States to report data on 
these activities to AFCARS exceeds our existing legislative authority 
in section 479 of the Act. Even so, we wish to note that AFCARS is not 
the sole data-related activity in child welfare that ACF manages. 
Through the National Child Abuse and Neglect Data System (NCANDS), 
States voluntarily provide us with data on child maltreatment and the 
extent to which the State child protective services agency provides 
services. We encourage State agencies to use the same unique person 
identifiers in AFCARS and NCANDS so that we can understand to what 
extent children receive prevention services before they must enter out-
of-home care. In addition, we have proposed a mandatory reporting 
system under the Chafee Foster Care Independence

[[Page 2085]]

Program (section 477 of the Act) which, in part, will require States to 
submit detailed information on the independent living services they 
provide to youth who are in foster care, or who have aged out of foster 
care (see 71 FR 40346). In that NPRM we propose to require States to 
use the same unique person identifier (child case or record number) for 
reporting a child's independent living services as they do for AFCARS. 
We believe, therefore, that we have adequate provisions for States to 
report on how they serve our nation's most vulnerable children and 
families without exceeding our legislative authority for AFCARS.
Section 1355.40(c)
    In paragraph (c) we define the scope of out-of-home care for AFCARS 
purposes which serves as a basis for the out-of-home care reporting 
population. ``Out-of-home care'' refers to children who have been 
placed away from their parents or legal guardians for a period of 24 
hours or more and for whom the State title IV-B/IV-E agency has 
placement and care responsibility, regardless of the child's living 
arrangement. This is different than our programmatic definition of 
foster care in 45 CFR 1355.20, and thus the scope of the current AFCARS 
foster care reporting population (see 45 CFR 1355.40(a)(2) and appendix 
A to part 1355, section II) in a number of ways. The most significant 
difference between the two terms is that the proposed AFCARS definition 
of out-of-home care will include children who are placed away from 
their parents for whom the State title IV-B/IV-E agency has placement 
and care authority, irrespective of their living arrangement. This 
stands in contrast to the foster care definition used for the title IV-
B and IV-E programs in 45 CFR 1355.20 and policy in the Child Welfare 
Policy Manual Section, which incorporates traditional foster care 
settings only (e.g., foster family homes, child care institution and 
group homes).
    We believe it is essential to develop a definition of out-of-home 
care for the purpose of data reporting distinct from the definition of 
foster care for the Federal child welfare programs, to meet their 
separate goals. The programmatic definition of foster care is for the 
purposes of describing the population for whom States must meet Federal 
child welfare requirements for safety, permanency and well-being as 
described in titles IV-B and IV-E of the Act and 45 CFR 1355, 1356 and 
1357. Nothing in this proposal changes to whom the Federal child 
protection requirements apply. AFCARS, on the other hand, has as one of 
its central goals as described in section 479 of the Act, the ability 
to provide comprehensive national information on the dynamics of 
children in the foster care system, including ``the status of the 
foster care population (including the number of children in foster 
care, length of placement, type of placement, availability for 
adoption, and goals for ending or continuing foster care),'' and ``the 
number and characteristics of children placed in or removed from foster 
care.'' Our experience with AFCARS is that the existing data on the 
number of children in foster care, the length of placements, and the 
characteristics of children as they move in and exit foster care is 
incomplete and often misleading without additional information about 
when children move from those out-of-home care living arrangements that 
are within the scope of foster care to detention facilities, 
psychiatric hospitals, assessment centers, and other facilities that 
are outside the scope of foster care. Particularly, as we have 
conducted AFCARS assessment reviews and CFSRs in many States, we have 
been challenged in pinpointing the scope of each State's foster care 
system and therefore, whether certain Federal child welfare 
requirements apply. By defining the AFCARS out-of-home care reporting 
population broadly, along with more specifically defining the type of 
living arrangements and circumstances of a child's stay in out-of-home 
care we believe that we can better track how and why children enter 
foster care, understand the dynamics of State foster care systems, and 
distinguish the subpopulation for whom State child welfare agencies are 
accountable to meet the Federal child protection requirements (section 
422(b)(8)(A) of the Act).
    We have specified in this proposed regulation that for AFCARS, we 
are seeking information on children who are under the placement and 
care of the State agency and away from their parents for 24 hours or 
more. This timeframe has not changed. However, the timeframe was noted 
in an appendix to the regulation rather than in the regulation text 
itself. We see no reason to include children in AFCARS who have been 
out of their homes for fewer than 24 hours.
    The proposed regulatory definition of out-of-home care also 
clarifies that the term refers to children who are considered minors 
according to the State's age of majority. This proposal is consistent 
with existing AFCARS policy (Child Welfare Policy Manual 1.3) and our 
regulatory definition of children at 45 CFR 1357.10(c) for the programs 
under title IV-B of the Act. We understand that most States consider 
young people up to age 18 as children. Several States, however, 
consider older youth (i.e., up to age 21) who are in their placement 
and care responsibility as minors.

1355.41 Reporting Populations

    We propose to add a new section 1355.41 on reporting populations to 
this part.
Section 1355.41(a) Out-of-Home Care Reporting Population
    In paragraph (a), we propose a new out-of-home care reporting 
population which identifies children States must include in an AFCARS 
out-of-home care data file. In general, we propose that State agencies 
must report information to AFCARS consistent with the AFCARS out-of-
home definition; that is, all minor children who have been placed away 
from their parents or legal guardians for a period of 24 hours or more 
and for whom the State title IV-B/IV-E agency has placement and care 
responsibility.
    In subparagraphs (a)(1)(i) through (a)(1)(iv), we propose to 
expound on which children are included in the reporting population. 
Although some of the children described in these subparagraphs are 
covered implicitly in the reporting population as generally stated in 
paragraph (a)(1), the subcategories provide more detail on the scope of 
the reporting population.
    In subparagraph (a)(1)(i), we propose to clarify that the reporting 
population is inclusive of any child who is under the placement and 
care responsibility of another public agency that has an agreement 
under section 472(a)(2)(B) of the Act with the title IV-B/IV-E agency 
for the payment of foster care maintenance payments on the child's 
behalf. This provision is consistent with existing AFCARS regulations 
that define the foster care reporting population (Appendix A to 45 CFR 
1355, Section II). Typically, State agencies enter these agreements 
with Indian tribes, and separate juvenile justice agencies or mental 
health agencies in order for the State to claim title IV-E on behalf of 
children who are otherwise eligible for the foster care maintenance 
payments program. These other public agencies do not submit information 
on children in the reporting population to ACF separately from the 
title IV-B/IV-E State agency. Rather, this information must be a part 
of the title IV-B/IV-E State agency's AFCARS submission.
    In subparagraph (a)(1)(ii), we propose to codify existing policy 
that a State

[[Page 2086]]

continue to collect and report information to AFCARS for as long as the 
State is making title IV-E foster care maintenance payments on the 
child's behalf, regardless of the State's age of majority (Child 
Welfare Policy Manual 1.3 2). Under the title IV-E program, 
the State is permitted to make foster care maintenance payments for 
young people who have attained 18 years of age, but not yet 19 years of 
age, who are full-time students expected to complete their secondary 
schooling or equivalent training before reaching age 19 (Child Welfare 
Policy Manual 8.3A.2 1). We acknowledge that this condition 
may require the State to report data beyond the State's age of majority 
as described in 1355.40(c). However, this provision is necessary to 
allow us to track the extent of assistance and the characteristics of 
all children for whom State agencies make Federal foster care 
maintenance payments consistent with section 479(c)(3)(D) of the Act.
    In subparagraph (a)(1)(iii), we propose to include in the out-of-
home care reporting population a child under the State agency's 
placement and care responsibility who is in any living arrangement, 
regardless of whether that living arrangement is a traditional foster 
care setting. We explain that States are to include children in out-of-
home care who are placed in settings such as detention facilities, 
psychiatric or other hospitals, and jails, but this is not an all-
inclusive list. The specified facilities have been raised most 
frequently in questions by State agencies because some youth may 
transition in and out of traditional foster care settings and these 
facilities. We want to clarify explicitly that a child who is in a 
living arrangement that is not a traditional foster care setting is a 
part of the AFCARS out-of-home care reporting population if the child 
is away from his parents or legal guardians while under the State title 
IV-B/IV-E agency's placement and care, even if the child remains in 
that setting for the entire report period. We understand that, in 
practice, most State agencies may not have included these children in 
the AFCARS foster care population to date, since our current policy 
does not require this reporting. Our current policy requires only that 
a State report a child who moves from a traditional foster care 
placement to a juvenile justice placement, as long as the State intends 
to return the child to foster care (Child Welfare Policy Manual 1.3 
12).
    As discussed previously, we believe that it is beneficial to compel 
State agencies to collect and report information to us on an ongoing 
basis when the child is under the State agency's placement and care 
responsibility away from his parents or legal guardians, regardless of 
the setting. We believe that doing so will allow us to follow a child 
through the various out-of-home placement settings that are connected 
closely to the foster care system but may not be managed by the State 
child welfare agency directly. Including these settings will permit 
States and ACF to complete longitudinal analyses of children's out-of-
home care experiences, as advocated by States and others in the field. 
In addition, we believe that requiring State agencies to submit 
information on a child's entire experience while under the placement 
and care responsibility of the State, rather than having to generate 
information based on identifying select types of settings, will be less 
burdensome. We welcome comment on this proposal.
    The reader should note that although the State will report all 
children placed away from their parents and legal guardians under its 
placement and care authority regardless of the child's living 
arrangements, States and ACF will be able to identify children who are 
in the narrower definition of foster care as defined by our program 
rules. This is because we are proposing to better categorize a child's 
living arrangements in the data elements. We will, therefore, be able 
to select samples for reviews or other analyses that look at foster 
care as used in the title IV-B and IV-E programs separately from other 
living arrangements.
    In subparagraph (a)(1)(iv), we require that a State continue 
reporting a child to AFCARS who is missing or has run away, is 
attending camp or on vacation, or is visiting with his immediate or 
extended family. In these situations, the child remains in out-of-home 
care under the agency's placement and care responsibility. These 
situations do not represent a State agency's need to move the child.
    Finally, in paragraph (a)(2) we propose that the State discontinue 
reporting a child to AFCARS if the State agency's placement and care 
authority ends (or is discharged), if the State agency returns the 
child home to his or her parents or legal guardians, or the child 
reaches the age of majority unless such a child continues to receive 
title IV-E foster care maintenance payments. The child has exited the 
reporting population for AFCARS purposes and has completed an out-of-
home care episode in these circumstances. This provision is, in part, a 
departure from the existing regulation. Many States over the years and 
during consultation have highlighted the need for more definitive 
guidance on when the child should be considered to have exited the 
AFCARS reporting population. States have pointed out that when a child 
leaves the AFCARS reporting population is of critical importance in 
defining consistently the length of time a child stays in foster care, 
as well as re-entries into foster care, for the CFSRs and other Federal 
child welfare outcome measures.
    We propose to continue State reporting of information until the 
child is no longer under the agency's placement and care responsibility 
because we are interested in understanding the child's entire out-of-
home experience. Children who are legally discharged from the State 
agency's placement and care responsibility have always been considered 
to have exited foster care under the existing AFCARS requirements. This 
would include children who may remain away from their parents or legal 
guardians but whose placement and care responsibility are transferred 
to another agency with no connection to the State agency.
    However, we propose for the first time that children who are 
returned home to their parents or guardians be excluded from the AFCARS 
reporting population. Previous policy suggested that a State report to 
AFCARS children who were returned home and supervised by the State 
agency in an after-care status for a period of six months, unless a 
court order indicated another time period (Child Welfare Policy Manual 
1.2B.7 7 and 1.3 11). Because we do not have a 
specific response option for States to report children in an after-care 
status in the existing AFCARS, we have instructed States to report the 
child on a trial home visit. There is, however, a distinction between a 
child who is visiting home, whether to stay connected to his or her 
family or to try reunification, and a child who the State agency has 
returned home. We agree with the States that contend that even though a 
State may continue to have some ongoing role in supervising or 
monitoring the child in his home, the child is no longer in out-of-home 
care for all practical purposes, but is at home. Furthermore, some 
State courts do not discharge a State's placement and care 
responsibility routinely, or in a timely fashion; sometimes this event 
occurs months after a child is in his or her own home. We concur that 
children in these situations should not be considered to be part of the 
AFCARS out-of-home care reporting population so as not to distort a 
child's length of

[[Page 2087]]

stay in care. We welcome comments on this proposal.
    We also want to clarify here that the proposed out-of-home care 
reporting population does not include those children who are under the 
State agency's ``supervision'' authority, unlike the current 
regulation. We found the reference to supervision to be problematic 
because we never defined the term ``supervision'' further in AFCARS 
regulations or policy. Thus States have questioned whether the existing 
reporting population includes children in a variety of settings for 
whom the State agency has only a legal duty to supervise with no 
concurrent placement and care responsibility. We wish to be clear that 
children who are receiving services only in the homes of their parent 
or legal guardian(s) and children who may be placed away from their 
parents or legal guardians but for whom the State title IV-B/IV-E 
agency has no placement and care responsibility are not a part of the 
proposed AFCARS out-of-home care reporting population.
Section 1355.41(b) Adoption Assistance and Guardianship Subsidy 
Reporting Population
    In subparagraph (b)(1), we propose that the State include 
information on all children for whom there is either a title IV-E 
adoption assistance agreement or a State adoption assistance agreement 
in effect during the report period. This includes children in a pre-
adoptive living arrangement. Children under such adoption agreements 
are a part of the reporting population regardless of whether a 
financial subsidy is paid on the child's behalf. We believe that 
requiring State agencies to collect and report information on these 
populations is necessary since there is no reliable information on 
these populations other than State claims data for Federal adoption 
funds, which have substantial analytical limitations.
    As a result of successful adoption initiatives, some States now 
have more children receiving adoption assistance than receiving foster 
care maintenance payments. With the increased activity in adoption and 
the corresponding outlays for the program, there has been an increase 
in requests for information about the population from the Congress, 
States, the media, and other sources. There also is a growing need at 
the Federal level for information to use for planning and budget 
projection purposes.
    Children who are in out-of-home care and who are the subject of a 
title IV-E adoption assistance agreement are likely to show up in both 
the out-of-home care and adoption assistance subsidy files until the 
point of the finalization of the adoption. In part, this is because 
sections 473 and 475(3) of the Act require States to enter into title 
IV-E adoption assistance agreements with adoptive parents prior to the 
finalization of a child's adoption, during which time the child may 
remain in out-of-home care. This may be true of children under State 
adoption assistance agreements as well, depending on State 
requirements. However, we believe we need this duplication of data in 
order to get complete information on the child's out-of-home care and 
adoption assistance experiences. Since we understand that the time 
between when an adoption assistance agreement becomes effective and the 
finalization of the child's adoption is relatively short, we expect 
such duplication to be limited. We welcome comments on this proposal.
    In subparagraph (b)(2), we seek information on children on whose 
behalf a subsidy is paid pursuant to a guardianship agreement with the 
State agency because we are interested in providing a national picture 
of children in these arrangements for the first time. We are not 
proposing that States include in the reporting population children who 
may be the subject of a guardianship or guardianship agreement in which 
a financial subsidy is not paid to the child's guardian. We believe 
that non-subsidized guardianships are a small portion of the 
guardianship arrangements in which State agencies are involved, that 
States maintain little information on them and there exists no 
compelling interest for ACF to require States to report information on 
these arrangements.
    States provide guardianship subsidies to a legal guardian for the 
care and support of a child who may be at risk of entering foster care 
or who may have otherwise remained in foster care. Although there is no 
Federal requirement or entitlement funding for States to provide 
guardianship subsidies, we understand that more than half of the States 
provide these supports to encourage greater permanency for children for 
whom adoption and reunification have been ruled out.
    States have established subsidized guardianship programs using 
State and local funds and funds from the Temporary Assistance for Needy 
Families Program. Seven States have obtained a child welfare 
demonstration waiver pursuant to section 1130 of the Act to test the 
effectiveness of a subsidized guardianship program for children in 
foster care. The demonstration waivers provide States with greater 
flexibility to use title IV-B and title IV-E funds for services that 
can facilitate improved safety, permanency and well-being for children. 
(Our authority to permit States to conduct new waivers expired in March 
2006). Our proposed reporting population includes children in any 
subsidized guardianship arrangement regardless of the source of 
funding.

1355.42 Data Reporting Requirements

    We propose to add a new section 1355.42 on data reporting 
requirements, including the report periods for the data files, general 
provisions for collecting and submitting the out-of-home care and 
adoption assistance and guardianship subsidy files, and record 
retention rules to comply with AFCARS requirements.
Section 1355.42(a) Report Periods and Deadlines
    In paragraph (a), we propose that each State submit an out-of-home 
care data file and an adoption assistance and guardianship subsidy data 
file to ACF on children in the reporting populations on a semi-annual 
basis. The report periods extend from April 1 to September 30 and from 
October 1 to March 31 of each Federal fiscal year. These report periods 
are the same as in the existing AFCARS.
    Several stakeholders suggested that we consider moving to annual, 
or even less frequent reporting, rather than semi-annual reporting of 
AFCARS data. Many commenters were concerned about the perceived 
complications of ACF compiling an annual file from two semi-annual 
submissions for the purposes of the CFSRs and the annual outcomes 
report to Congress. We want to assure States that we are able to create 
an annual file. We believe that some States' concerns about compiling 
an annual file were related to their inability to replicate the 
information from ACF precisely. ACF has recently started using a 
readily-available software program. The logic associated with this 
software's de-duplication function is readily transferable to other 
software packages; therefore, States will be able to replicate the 
annual files more easily. Finally, we expect that the new requirements 
proposed here for using a permanent and encrypted person identification 
number (see proposed 45 CFR 1355.43(a)(4), 1355.43(a)(5) and 
1355.44(a)(3) in this NPRM) will aid both our own and States' ability 
to create annual files.
    Further, we believe that an annual submission would hamper our 
ability to provide timely data and analysis to stakeholders and would 
not meet our needs. A six-month submission process

[[Page 2088]]

is preferable because AFCARS is now linked inextricably to a number of 
ACF priorities and legislative requirements, including the CFSRs and 
title IV-E eligibility reviews. For example, most States are monitoring 
their progress in achieving the steps of their CFSR program improvement 
plans on a quarterly basis. Because States submit AFCARS twice a year, 
we can provide States with their results on the statewide data 
indicators every six months for comparison. A move to annual 
submissions would mean that a State would not be able to use AFCARS 
data to see how it has improved as timely. Annual data would add six 
additional months to the time it would take ACF to verify whether a 
State has achieved the agreed upon amount of improvement for a CFSR 
program improvement plan. Also, annual AFCARS submissions would mean 
that our period under review for the CFSR onsite review would need to 
be extended and we could not review States as frequently because they 
are tied to the AFCARS report period. Finally, the title IV-E 
eligibility reviews require that we select a sample of children who 
received foster care maintenance payments during a six-month period 
that coincides with the State's most recent AFCARS submission (45 CFR 
1356.71). In formulating the title IV-E reviews, we chose a recent six-
month AFCARS period specifically so that we would review recent cases 
of children in foster care.
    We also propose in paragraph (a) that State agencies submit their 
data files to us within 15 calendar days of the end of the report 
period. If this date falls on a weekend, the State must submit their 
files by the end of the following Monday. This is a change from the 
current AFCARS, which allows a 45-day period in which State agencies 
may prepare their data files for submittal to ACF. Although some 
stakeholders requested more time to prepare their files, we believe 
that the shorter time frame is workable and will also better meet State 
and Federal needs for data.
    As mentioned earlier, AFCARS data is used extensively in a number 
of ACF priorities and requirements, including the Child and Family 
Service Reviews. If ACF receives the data a month earlier than we do 
now, we will be better able to analyze the data for use in CFSR data 
profiles and program improvement plans. Also, since adoption incentive 
funds are tied to how well States perform in increasing their adoptions 
as seen in the AFCARS data, we can award adoption incentive funds to 
States sooner.
    The vast improvements in automation in the field of child welfare 
strengthen our belief that a State can prepare data files within 15 
days. Now States can record and verify data in a more timely fashion 
than when the original AFCARS regulation was issued. Finally, we have 
provided significant technical assistance to States to encourage 
ongoing quality assurance checks on the data recorded in their 
information systems. We believe that State agencies will be able to 
meet this shorter time frame for submitting data with continued and 
routine use of our data quality utilities. We welcome comment on the 
shorter submission time frame.
    Finally, in paragraph (a) we require that State agencies submit 
their data to us in two separate data files. Currently, State agencies 
must submit four data files (Appendices A and B to 45 CFR 1355): (1) A 
detailed foster care file with information on a child in foster care 
during the report period; (2) a detailed adoption file with information 
on all children adopted during the report period in whose adoption the 
State agency has some involvement; (3) a foster care summary file in 
which the State indicates the total number of foster care records and 
the age distribution of children in those records; and, (4) an adoption 
summary file in which the State indicates the total number of adoption 
records and the age distribution of the children adopted.
    We propose to eliminate the existing foster care and adoption 
summary files because they are no longer necessary. ACF originally 
intended to use the summary files to verify the completeness of a 
State's data submissions and to ensure that the file was not corrupted 
during transmission. The summary files also were to serve as a quick 
count of the number of children in foster care and those being adopted. 
However, because the summary files contain aggregate data, the number 
of children who entered, were discharged, were adopted, were served or 
were in care on a specific day cannot be determined. Further, we are 
able to use new technology that is better able to verify the 
completeness of a State's data submission without requiring the State 
to generate summary files.
    The proposed out-of-home care data file contains the majority of 
information that State agencies report to us currently in the detailed 
foster care and adoption data files. We propose to discontinue the 
submission of voluntary adoption data and eliminate the separate 
adoption data file. Rather, children who are adopted from out-of-home 
care will be included in the out-of-home care data file, and children 
for whom the State agency has been involved in their adoption by 
entering into an adoption assistance agreement will be included in the 
adoption assistance and guardianship subsidy data file (some children 
will be reported in both files). The current separate adoption data 
file was developed originally to permit State agencies to submit data 
on all adoptions (inclusive of private, independent, or international 
adoptions in which the State agency was not involved) without the data 
appearing erroneous due to duplicated information that may have 
resulted from States' obtaining the data from a variety of sources. For 
example, had States obtained their data on all adoptions from court 
records and incorporated that data into the foster care data file, 
public agency adoptions would have been duplicated. This strategy was 
based on the premise that State agencies would voluntarily submit data 
on adoptions outside of the public agency. However, just a few States 
have submitted non-public agency adoption data consistently, making the 
information unusable.
Section 1355.42(b) Out-of-Home Care Data File
    In paragraph (b), we provide instructions on how the State must 
report the out-of-home care information under the proposed 45 CFR 
1355.43.
    Specifically, in paragraph (b)(1), we propose that a State provide 
us with the most recent information for the elements regarding general 
information, child information, and parent or legal guardian 
information (proposed 45 CFR 1355.43(a), (b) and (c)). This means that 
in each file submission we are seeking current, point-in-time data for 
these elements similar to the time frame for most elements in the 
existing AFCARS. This information is largely demographic in nature, and 
tends to remain static over a six-month report period or even longer. 
For example, information on the child's parent, such as race, ethnicity 
and date of birth, does not change over the course of a report period.
    In paragraph (b)(2), we propose that a State submit recent and 
historical information for the elements regarding removal information, 
living arrangements and provider information, permanency plans and 
ongoing circumstances, general exit information, and exit to adoption 
information (proposed 45 CFR 1355.43(d), (e), (f), (g) and (h), 
respectively). This information is required, unless the exception 
described below applies. This means that for every file submission, we 
are seeking information on the child's lifetime experience while in 
out-of-home care in the State's placement and

[[Page 2089]]

care responsibility as described through the reporting of these 
elements. This will allow ACF to develop a comprehensive picture of a 
child's lifetime experience with entries, living arrangements, 
permanency plans, and exits while in the State's child welfare system. 
This is in contrast to the existing AFCARS, which requires that a State 
submit certain detailed information on the child's current foster care 
episode and current placement setting only.
    We want to be clear that we propose that State agencies submit 
recent and historical information pertaining to removals, permanency 
plans and circumstances, living arrangements and exits every report 
period rather than updates on children who were in out-of-home care 
before or who remain in out-of-home care from one report period to the 
next. This is because we have learned from the existing structure of 
AFCARS that gaps in information raise numerous questions about the 
child's experiences and make the data more difficult to analyze. Part 
of our goal in developing this proposed regulation is to eliminate 
features of the existing AFCARS that hinder the collection of reliable, 
quality data. If we were to ask State agencies to report only changes 
in the child's living arrangements, permanency goals, entry into or 
exit from out-of-home care we would not have a way to verify that the 
child's experiences have, in fact, remained the same. We also believe 
that this approach is less burdensome on States. Although sending a 
child's history involves submitting more data to us than an update as 
children exit and re-enter out-of-home care and their living 
arrangements and permanency plans change, we believe that it is less 
complicated and therefore requires fewer State resources than the 
alternative. In other words, sending a child's full history requires 
the State to submit all the information it has on these elements, 
rather than figure out a way to cull out only the information that has 
changed each report period.
    We propose to get more comprehensive data for certain elements in 
response to our own need for data and in response to stakeholders' 
requests that ACF consider how to move AFCARS towards gathering some 
longitudinal information. Many States noted that they already have this 
capability. A number of States also asserted that the breadth of this 
information allowed them to conduct more sophisticated analysis on what 
happens to a child, or groups of children in foster care. Further, 
States and other stakeholders saw this type of information as critical 
to the CFSRs. In particular, stakeholders believe that longitudinal 
data would better inform CFSR measures such as time in foster care, 
foster care re-entries and the stability of foster care placements. For 
example, once we have information on all out-of-home care episodes a 
child experiences, we can potentially analyze data to determine whether 
children entering out-of-home care for the first time after a certain 
point in time have more positive outcomes than those who entered out-
of-home care earlier. Also, we can potentially use the data to improve 
upon our placement stability measure by not only analyzing the number 
of placements that a child experiences in an episode, but the type of 
placements as well. Further, with the richness of data that 
longitudinal information can provide, ACF and States can be better 
informed in developing and implementing program improvement plans to 
address compliance issues raised during a CFSR. In light of the results 
of the first round of CFSRs and the challenges that are ahead for 
States in implementing changes to their child welfare systems, we find 
the potential to have improved data for use in the CFSR and other 
priorities a compelling reason for proposing these changes. We welcome 
comments on this approach.
    We chose to propose gathering comprehensive data on removals, 
permanency plans and ongoing circumstances, living arrangements and 
exits after considering whether a more limited approach to developing 
longitudinal data would meet our needs for data, as well as those of 
the States. The limited option would require a State to submit detailed 
removal, permanency plan, living arrangement and exit information on 
the child's four most recent out-of-home care episodes. We also 
considered requiring detailed living arrangement information on the 
child's four most recent living arrangements only. Under this option, 
the State would inform us how many total removals and living 
arrangements the child had experienced. We considered four out-of-home 
care episodes because our analysis of existing AFCARS data suggests 
that the vast majority (approximately 99 percent) of children in the 
existing foster care reporting population have no more than four foster 
care episodes. This option would allow us to capture almost all foster 
care episodes without requiring State agencies to submit extensive 
histories on children. We similarly thought that limiting the number of 
living arrangements that State agencies would report to AFCARS would 
minimize the burden of this approach.
    Ultimately, we decided that this more narrow approach was not 
sufficient. One problem with a limited longitudinal database was that 
we would have no information on the children who present some of the 
more significant challenges to States. Children who experience high 
numbers of multiple living arrangements or frequently enter and exit 
out-of-home care are some of the nation's most vulnerable children. 
Furthermore, these children often require States to expend more of 
their resources to address their problems.
    In paragraph (b)(3), we propose an exception to the requirement to 
report complete information on all out-of-home care episodes for 
children in the reporting population. The exception applies to those 
children who had an out-of-home care episode prior to the effective 
date of the forthcoming final rule. Specifically, the exception applies 
to: (1) Children who are in out-of-home care on the effective date who 
also had a prior episode before the final rule goes into effect, and 
(2) children who enter out-of-home care after the effective date who 
had a prior episode before the final rule goes into effect. For such 
children, we are proposing that the State report the child's removal 
dates, exit dates and exit reasons (1355.43(d)(1), (g)(1), and (g)(3) 
respectively) for each out-of-home care episode that occurred before 
the final rule effective date. The exception does not apply to a 
child's ``open'' or ongoing episode that coincides with the effective 
date of the final rule; for such children we propose that a State 
report all information described in paragraphs (b)(1) and (b)(2) during 
that ongoing out-of-home care episode. As time passes after the final 
rule goes into effect, this provision will apply to a diminishing 
number of children who are in the out-of-home care reporting 
population.
    We propose this exception to the general rule to report complete 
information in order to strike a balance between our desire for recent 
and historical information on all children in out-of-home care in 
accordance with the proposed new AFCARS elements with the challenge 
that some State agencies may face in gathering this information for a 
child's previous contacts with the State child welfare system before 
these new rules go into effect. We chose to have State agencies report 
at least the child's prior removal and exit dates and exit reasons, 
because we believe these elements are most critical to our ability to 
construct certain cohorts of children for analysis in the CFSRs and 
other outcome-based activities. Further, States

[[Page 2090]]

currently collect this information in the normal course of their 
casework activities for children in foster care and report some 
information for these elements under the existing AFCARS.
    Our expectation is that for children who experience an out-of-home 
care episode prior to the implementation of the proposed new AFCARS, 
States will report more than the minimum information required by the 
exception. We expect, but do not require, States to provide as much 
information as they have in their case files and information systems on 
the child's out-of-home care episodes that occur before the effective 
date of the final rule and at least as much information as they report 
currently under the existing AFCARS. States that do not provide this 
additional information will not be penalized. States that provide it 
with errors will not be penalized either. From our review of States 
with a SACWIS, we have found that many States are collecting 
comprehensive information and information that pertains to the proposed 
new elements. Therefore, we believe that it is reasonable to expect 
States to provide us with information on the new elements regarding 
prior episodes even in the absence of a mandate. In fact, we considered 
establishing different exceptions to the requirement to report 
comprehensive information for those States that have an operational 
SACWIS versus those that do not because we believe that the type of 
information they are able to collect and report is more complete and 
robust than other States. Even so, since this is the first time we are 
requiring certain information in AFCARS, we believe that we must allow 
all States an equal opportunity to collect the proposed information for 
children who already are known to the State.
    Finally, we acknowledge that even though we propose that States 
report a child's removal and exit dates and exit reasons of the out-of-
home care episodes that occur prior to the final rule effective date, 
this limited information will be newly required for some children in 
certain circumstances. In particular, since we propose to expand the 
reporting population to include children who are in out-of-home care 
settings that are not considered foster care under our program rules, 
States have not consistently reported removal and exit dates and exit 
reasons for AFCARS purposes. Further, since the existing AFCARS 
requires that States report the date of first and latest removal and 
exit reason for the most recent foster care episode in a six-month 
period, some children may have interim removal dates and exit dates and 
reasons that States currently are not reporting to us. We still 
believe, however, that while this proposed reporting may be newly 
required, States generally have this information as a matter of course 
in their own information systems and this requirement would not produce 
an undue burden. We welcome comment on this provision.
Section 1355.42(c) Adoption Assistance and Guardianship Subsidy Data 
File
    In paragraph (c), we propose that the State submit recent, point-
in-time information for all elements in this data file. This 
information is needed only at a given point in the report period 
because it is static over time. For example, adoption subsidies may 
remain the same over many years or for the duration of the adoption 
assistance agreement, unless the parent requests a change in the amount 
of the subsidy, or the child's circumstances change.
Section 1355.42(d) Reporting Missing Information
    In paragraph (d), we propose how the State must report missing 
information. If the State agency fails to collect the information for 
an element, the State agency must report the element as blank or 
missing. The State agency may not develop program codes that default or 
map information that caseworkers did not collect or enter into the 
State's information system to a valid response option. This is the case 
even when there may be a response option for an element that allows the 
State to indicate that the information has not yet been determined or 
is unknown. This provision is consistent with ACF's longstanding 
practice; however, States have pointed out that there is no official 
guidance on this issue. Therefore, we wish to state unequivocally that 
this practice of defaulting is not permitted.
    For example, we propose that the State indicate the specific 
permanency plan for a child or indicate that the permanency plan has 
not yet been determined for the child. If the State's information 
system is programmed in a way to allow the worker to select various 
plans (i.e., adoption, reunification, etc.) or not input the 
information at all (i.e., leave the information blank), the State 
agency may not report to ACF the child's plan as ``not yet 
determined,'' when the State does not have any information. Rather, the 
State may only report that the plan is ``not yet determined'' if the 
State has programmed its information system in a way that allows the 
worker to select that he/she has actually not yet determined the plan.
Section 1355.42(e) Electronic Submission
    In paragraph (e) of this section we propose that States submit 
their data files to ACF electronically, consistent with ACF's 
specifications. States currently submit their data files to us 
electronically; however, we are removing from the regulation a number 
of technical specifications that detail how States must submit their 
files electronically (see appendix C to part 1355). Instead, we will 
issue technical requirements and specifications through official ACF 
policy subsequent to our issuance of the final rule. We have learned 
through our experience with the existing AFCARS that it is prudent not 
to regulate the technical specifications for transmitting data. As 
technology changes, we must be able to keep pace with the most current, 
practical and efficient transmission methods that will meet State and 
Federal needs.
    We are particularly interested in exploring new technologies due to 
the enactment of the E-Government Act of 2002 (Pub. L. 107-347). This 
law focuses the Federal government on using improved internet-based 
technology to make it easier for State or local governments and 
citizens to interact with the Federal government. One internet-based 
technology that we are exploring for AFCARS is the use of Extensible 
Mark-Up Language (XML). XML is a text-based format that allows entities 
to describe, deliver and exchange data among a range of applications, 
provided that the sender and receiver have agreed in advance on the 
data definitions. We believe that XML has several benefits to States 
and ACF, including:
     Enabling the integration and collation of any data and 
information irrespective of storage environment or document type;
     Facilitating data interchange independent of the operating 
system and hardware; and,
     Allowing new data elements to be added readily with 
minimal changes to the data file format.
    We recognize that some States already have implemented the use of 
XML to transfer data, while others may have encountered some barriers 
to doing so.
Section 1355.42(f) Record Retention
    In paragraph (f), we propose that States retain records for as long 
as necessary to comply with the AFCARS reporting requirements. In 
particular, we are making Departmental record retention rules in 45 CFR 
92.42(b) and

[[Page 2091]]

(c) inapplicable to AFCARS. These Departmental record retention rules 
require States to retain financial and programmatic records, supporting 
documents, and statistical records related to Federal programs and 
requirements for a period of three years. Because we are seeking 
comprehensive data on children in out-of-home care, including 
information on their prior experiences with the child welfare system, a 
three-year retention period is insufficient.
    Practically, this means the State must keep applicable records 
until the child reaches the age of majority in the State, or else is no 
longer of an age to be in the reporting populations. This is because we 
propose that a State keep a child's identification number consistent 
over time and indicate the child's entire history with the child 
welfare system. Since a child's information is likely to be contained 
in an automated information system and relatively simple to archive, we 
believe these record retention rules are reasonable.

1355.43 Out-Of-Home Care Data File Elements

    We propose to add a new section 1355.43 providing all elements for 
the out-of-home care data file. Under this section, each element is 
described in detail and the acceptable response options are also 
defined. (Attachment A to the preamble contains a quick reference of 
all the out-of-home care elements.) We propose that the State agency 
must collect and report the information described in these elements for 
each child in the out-of-home care reporting population.
Section 1355.43(a) General Information
    In paragraph (a) of this section we propose that States collect and 
report general information that identifies the reporting State and the 
child in out-of-home care.
    State. In paragraph (a)(1), we propose that the State responsible 
for reporting the child identify itself using the numeric two-digit 
State Federal Information Processing Standards (FIPS) code. We use the 
FIPS code because it is a standard issued by the National Institute of 
Standards and Technology (NIST) to ensure uniform identification of 
geographic entities through all Federal government agencies. The 
requirement for the State to identify itself is not new (see appendix A 
to part 1355, section II, I.A); however, the existing regulation 
incorrectly requests that the State use the alphabetic U.S. Postal 
Service abbreviation rather than the FIPS code. We corrected this 
mistake in policy (Child Welfare Policy Manual 1.2A.3 1 and 
1.2B.2 4), but are now codifying it in regulation.
    Report date. In paragraph (a)(2), we propose that a State continue 
to indicate the report period date (see appendix A to part 1355, 
section II, I.B). Specifically, States are to report to us the last 
month and year that corresponds with the end of the report period, 
which will always be either March or September of any given year.
    Local agency. In paragraph (a)(3), we propose that the State report 
to us the local agency that has responsibility for the child using a 
five-digit FIPS code. The local agency must be the county or a county 
equivalent unit which has responsibility for the child. The information 
requested is the same as in the existing AFCARS regulations (see 
appendix A to part 1355, Section II, I.C). However, consistent with 
existing policy we want to emphasize that we are interested in the 
location of the agency that has responsibility of the child, and not 
the county where the child is residing (Child Welfare Policy Manual 
1.2B.2 3).
    Child record number. In paragraph (a)(4), we propose that the State 
report the child's record number, which is a unique person 
identification number, as an encrypted number. The person 
identification number must remain the same for the child until the age 
of majority, no matter where the child lives while in the State's 
placement and care responsibility and across all report periods and 
episodes of out-of-home care. If the child was previously adopted in 
the State, however, the State may provide a new record number for the 
child. The State must apply and retain the same encryption routine or 
method for the person identification number across all report periods. 
The State's encryption methodology must meet all ACF standards that we 
prescribe through technical bulletins or policy.
    This is a revised element in that we are no longer allowing the use 
of sequential numbers for AFCARS and propose rules for encryption and 
consistent numbers (see appendix A to part 1355, section II, I.D). The 
changes to this element are based on findings from AFCARS reviews and 
technical assistance which indicate that some States use different 
identification numbers or change key or seed numbers for the same 
child. One issue that has been identified in some States that do not 
have a SACWIS is that the child's record number may change if the child 
moves within the State. We are concerned about a State's ability to 
track a child's complete out-of-home care experience in the State when 
they do not use the same identification number, so we propose that 
States discontinue this practice.
    Further, we believe that States share our desire to understand the 
entire experience of a child with the State's child welfare system. 
Numerous commenters on the Federal Register notice suggested keeping a 
child's identification number consistent through his or her child 
welfare experience. That is why we also have required States to use the 
same single person identification number for reporting a youth to the 
National Youth in Transition Database and encouraged States to use the 
same number for reporting a child to the National Child Abuse and 
Neglect Data System (NCANDS).
    Encryption will ensure that the child's identity will remain 
confidential. Encryption is one of a number of different methodologies 
that a State can use to code confidential information. However, we are 
requiring encryption as opposed to other methods of coding confidential 
data, like sequential numbering, because it is secure and easier than 
other methods to cross-reference files for identification at a later 
date. For example, encryption protects a child's sensitive information 
by masking the State or local agency's person identification number 
from Federal staff, researchers or other persons who may come into 
contact with the data the State submits to ACF. In practice, a State 
encrypts a record number by introducing a seed or key number into a 
mathematical formula to code the numbers. The State reveals the 
original person identification number by re-introducing the same seed 
or key number to reverse the mathematical formula, a process known as 
decryption. The State ensures confidentiality by keeping the 
mathematical formula secure and limiting access to the formula to 
authorized persons only.
    Encryption also is more efficient than some other methods because 
the State need only safeguard the seed or key number, not a whole list 
of numbers, which crosswalk between the masked identification number 
and the real record number. Furthermore, the vast majority of States 
use encryption methods already in reporting information to AFCARS. The 
few States that do not use encryption currently have indicated to ACF 
that they intend to use encryption in the near future. We believe, 
therefore, that requiring an encryption method will impose a minimal 
burden on States.
    Finally, we have created an exception to the general requirement 
that a child's

[[Page 2092]]

record number remains the same until the age of majority. The exception 
is for a child who re-enters out-of-home care following an adoption. We 
are making this exception in recognition of some State policies to 
change identifiers for children when they are adopted after being in 
out-of-home care. Regardless of a change in the child's record number, 
the State must report the child's entire child welfare experience.
    Family Record Number. In paragraph (a)(5), we propose for the first 
time that the State report a unique and encrypted family record number 
that is associated with the child. Provided the child's family remains 
the same during the child's out-of-home care and any subsequent out-of-
home care episodes, this number must remain the same regardless of 
where the child or family resides. However, should the child's family 
change due to adoption we propose that the State submit the adoptive 
family's record number.
    Although we have not requested this information before in AFCARS, 
we believe that all States use a family number or equivalent in their 
case management systems to identify the family in which the child in 
foster care is a member. We propose to collect this information 
primarily to aid in the identification of sibling groups, which we 
describe in greater detail in section 1355.43(b)(11).
Section 1355.43(b) Child Information
    In paragraph (b) we propose that States collect and report various 
characteristics of the child in the out-of-home care reporting 
population.
    Child's date of birth. In paragraph (b)(1), we propose to continue 
to require States to report the child's date of birth (see appendix A 
to part 1355, section II, II.A). The only change that we made in the 
proposed definition is to no longer instruct States to report an 
abandoned child's date of birth as the 15th of the month. During AFCARS 
assessment reviews, we found that many States were not aware of this 
instruction or that workers were reluctant to enter an unknown birth 
date as the 15th of the month. Moreover, we have come to realize that 
the State child welfare agency is often able to establish or estimate 
an abandoned child's date of birth quickly by consulting with health 
officials and/or records. Therefore, we are requiring that the State 
always provide the child's actual or estimated date of birth.
    Child's gender. In paragraph (b)(2), we propose that States report 
information on the child's gender, consistent with the existing 
regulation (see appendix A to part 1355, section II, II.B).
    Child's race. In paragraph (b)(3) we propose to continue to require 
information on the race of the child (see appendix A to part 1355, 
section II, II.C). The racial categories of American Indian or Alaska 
Native, Asian, Black or African American, Native Hawaiian or Other 
Pacific Islander and White listed in proposed subparagraphs (b)(3)(i) 
through (b)(3)(v) are consistent with the Office of Management and 
Budget's (OMB) standards for collecting information on race. (See OMB's 
Provisional Guidance of the Implementation of the 1997 Standards for 
Federal Data on Race and Ethnicity, at http://www.whitehouse.gov/omb/inforeg/re_guidance2000update.pdf
 for more information.) Each racial 

category is a separate data element to represent the fact that the OMB 
standards require States to allow an individual to identify with more 
than one race. Consistent with the OMB standards, self-reporting or 
self-identification is the preferred method for collecting data on race 
and ethnicity. This means that the State is to allow the child, if age 
appropriate, or the child's parent(s) to determine race.
    If the child's race is unknown, the State is to so indicate in 
subparagraph (b)(3)(vi). A child's race can be categorized as unknown 
only if a child or his parents do not actually know the child's race. 
The fact that the State agency has not asked the child or parent for 
the child's race is not an acceptable use of the unknown response 
option. Further, it is acceptable for the child to identify that he or 
she is multi-racial, but does not know one of those races. In such 
cases, the State must indicate the racial classifications that apply 
and also indicate that a race is unknown. If the child is abandoned, 
the State must so indicate in subparagraph (b)(3)(vii). We have 
provided a definition of abandoned so that we are clear that it is to 
be used in very restrictive circumstances and not any time a parent may 
be temporarily unavailable. If a child or young person who was 
abandoned as an infant identifies as being of a certain race or 
multiple races, the State must indicate the applicable race(s), rather 
than abandoned. Finally, in the situation in which the child or child's 
parent declines to identify any race, the State must so indicate in 
subparagraph (b)(3)(viii).
    Child's Hispanic or Latino ethnicity. In paragraph (b)(4), we 
propose that a State report the Hispanic or Latino ethnicity of the 
child. Similar to race, these definitions are consistent with the OMB 
race and ethnicity standards and should be self-reported by the child 
or his parent. Also, the State may report whether the child's ethnicity 
is unknown because the parent or child does not know the information, 
whether the child is abandoned, or that the child or parent has 
declined to provide this information.
    In the elements in paragraph (b)(5) and its subparagraphs, we 
propose for the first time that the State report the child's use of 
language. We propose to collect this information because we believe 
language is an important characteristic of a child that may aid the 
State in delivering services to him or her. Further, those children who 
do not speak English or who communicate through sign language may face 
particular challenges in a State's child welfare system. If we collect 
this information we will be able to analyze the data to see if language 
used has an effect on a child's experience in foster care. We believe 
that having this information will be a greater benefit to ACF and the 
States than the relatively low burden on caseworkers in collecting the 
data. We welcome comments on this new element.
    Child's language. In paragraph (b)(5), we propose that the State 
indicate whether the child is verbal, pre-verbal or non-verbal. We are 
defining verbal to include the use of any language, whether it be a 
spoken language or other communication, such as sign language. A child 
who is pre-verbal is one who is too young to use language. A non-verbal 
child is a child who is of an appropriate age to use language but 
appears unable or incapable of using language. The child may be non-
verbal due to a significant developmental delay or severe deprivation 
of exposure to language. We believe that we must capture a child's 
ability to be verbal along with the specific languages the child uses 
to be able to analyze this characteristic correctly.
    Languages used. In subparagraph (b)(5)(i), we require that the 
State indicate all the languages that a child uses, if appropriate. We 
have provided several response options that reflect the most common 
languages used in the United States. However, the State is to indicate 
any other language(s) the child uses that is not in that list. For a 
child who uses sign language, the State is to indicate that the child 
uses sign language in addition to any other language (e.g., English or 
Spanish) used.
    Language preference. In subparagraph (b)(5)(ii), we propose that 
the State indicate the language with which the child has the greatest 
facility if the child uses more than one language. For children who are 
bilingual or multilingual with an equal facility in those languages, 
the State may indicate all that apply.

[[Page 2093]]

    We considered requesting information on the child's primary 
language only, but found this terminology problematic for individuals 
who may be bilingual or multilingual. We also considered whether we 
should ask which language the child used in his/her home, but found 
that construction equally problematic for multilingual families. We 
believe that allowing the State to identify the languages used by the 
child and the ones in which the child has the greatest facility is the 
most straightforward way of gathering the information we desire.
    Health, behavioral or mental health conditions. In paragraph 
(b)(6), we propose to continue to require States to report information 
on whether a child has been diagnosed with a health, behavioral or 
mental health condition, with some modifications (see appendix A to 
part 1355, section II, II.D). Information pertaining to the health 
characteristics of a child is important in understanding the length of 
time children remain in care, their placement needs, and, in general, 
the needs of children being served by the agency. We believe that by 
collecting this information in AFCARS, we can better support the CFSR 
in gathering information on children's well-being. Further, requiring 
this information is consistent with the provision in section 475(1)(C) 
of the Act for the State to have a case plan that includes the child's 
health records and known medical problems.
    We propose to continue to require that the State indicate diagnoses 
made by a qualified professional only as determined by the State. A 
qualified professional may be a doctor, psychiatrist, or, if applicable 
in the State, a licensed clinical psychologist or social worker. We 
make this distinction as a means to gather information on medically 
diagnosed conditions rather than conditions that may be observed by a 
caseworker to determine the most appropriate placement or referrals for 
a child. Additionally, this data element will provide ACF with 
information on whether children in out-of-home care have received a 
clinical assessment for the diagnosed conditions.
    The proposed language also expands upon the types of conditions in 
the existing regulation. We chose to expand the list of conditions 
because we learned through AFCARS and SACWIS reviews and providing 
technical assistance that States had difficulty matching children's 
actual diagnoses with the existing AFCARS categories. We believe that 
this has caused data on children's health conditions to be 
underreported in the past. We developed the new AFCARS categories based 
on the International Classification of Diseases (ICD) and the 
Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition 
(DSM IV). We separated some conditions that are grouped together in one 
category in either the ICD or the DSM IV in order for the information 
to better meet our needs. We tried to create categories that 
distinguish conditions that may be more medically/physically based, 
education-related, or mental/emotional in nature.
    Specifically, we propose to continue to collect information on 
whether a child is visually or hearing impaired but have made the two 
into separate response options because the needs of these two groups 
are distinct. We continue to gather information on mental/emotional 
disorders but have narrowed the definition to those types that are more 
severe or prolonged in nature. We have broken out the previous category 
by adding childhood disorders and anxiety disorders. The DSM IV 
categorizes learning disabilities under ``disorders usually first 
diagnosed in infancy, childhood, or adolescence.'' We believe, since 
this condition relates to educational achievement, that it should be 
separated from the other conditions listed in ``childhood disorders.'' 
Also, we propose to add categories related to drug and substance abuse 
separately in order to distinguish these disorders from other 
behaviors. Finally, we have added the specific category ``developmental 
disability'' to reflect the definition in section 102(8) of the 
Developmental Disabilities Assistance and Bill of Rights Act of 2000 
(Pub. L. 106-402).
    We also propose to change the title of these elements from 
``disabilities'' to ``health, behavioral or mental health conditions.'' 
Our intent for collecting this information is to gather data on the 
problems, disorders, and behaviors of the children in out-of-home care, 
rather than pinpoint children whose conditions meet a narrow 
construction of disability. Also, since what is considered a disability 
can vary for Federal or State programs, insurance purposes, or other 
benefits, we chose to use a more general term.
    Finally, we want to be clear that States must report information 
known prior to the child's current out-of-home care episode. It is 
likely that some of the diagnosed conditions will not be corrected or 
cured in a short period of time. Therefore, if a child re-enters out-
of-home care, the State must report the previously known diagnosis if 
it is still applicable. This principle also applies to a child newly 
entering out-of-home care who has a known diagnosed condition. For 
instance, a child may have been born with a congenital defect and is 
undergoing treatment (or not) for the problem. If the State agency is 
aware and has obtained a medical summary, then this information should 
be recorded and reported to AFCARS.
    Current immunizations. In paragraph (b)(7), we propose for the 
first time that a State indicate whether the child's immunizations are 
current as of the end of the report period. A State agency is to 
indicate whether the child's immunizations are current, or the State 
agency may indicate that it has not yet determined the status of the 
child's immunizations because it has not compiled or obtained the 
child's immunization records. If a child is too young to be immunized 
at the time of reporting, i.e., the child is a newborn, the State may 
indicate that the child's immunizations are current. For the purposes 
of AFCARS, we are requiring that States determine whether immunizations 
are up-to-date in accordance with the Recommended Childhood and 
Adolescent Immunization Schedule (available from the Centers for 
Disease Control (CDC)) in consultation with the child's practitioner.
    We are seeking this information because we are interested in 
gathering data that will allow us to understand more about a child's 
well-being while in out-of-home care. Further, this information is 
readily available to States in most cases since it is a required part 
of a foster child's case plan (section 475(1)(C)(v) of the Act).
    Educational Performance. In paragraph (b)(8), we propose for the 
first time that a State report information on whether the child has 
repeated grades in school (in subparagraph (b)(8)(i)) and the number of 
repeated grades (in subparagraph (b)(8)(ii)). In subparagraph 
(b)(8)(ii), the State must consider each time a child repeats a grade 
separately. For example, if a child remained in the tenth grade for 
three school years, the State must report the number of grades repeated 
as two.
    We have chosen grade level performance as a proposed new data 
element in an effort to learn more about a child's well-being while in 
out-of-home care. A recent study of students in Illinois indicated that 
children in foster care are more likely to be behind in their grade 
level performance than students who have not experienced a removal from 
home (Chapin Hall, Educational Experiences of Children in Out-Of-Home 
Care, 2004). We believe that grade level performance is an appropriate 
indicator of educational performance because it is used

[[Page 2094]]

consistently across the country, is appropriate for all school-age 
children, and relatively simple for a State agency to collect and 
report. Further, we believe that this element is consistent with the 
statutory requirement for States to compile information on the child's 
grade level performance while in foster care (section 475(1)(C)(ii) of 
the Act).
    Special education. In paragraph (b)(9), we propose to collect 
information for the first time about whether the child received special 
education instruction during the report period. The term ``special 
education,'' as defined in 20 U.S.C. 1401(29), means specifically 
designed instruction, at no cost to parents, to meet the unique needs 
of a child with a disability. States are to indicate whether the child 
received special education during the report period, or indicate that 
the State agency has not yet determined whether the child is receiving 
special education. We are specifically requesting that States indicate 
whether the child actually receives special education instruction, 
rather than whether the child needs or has been referred for special 
education services. We believe that data on whether the child actually 
receives special education will be more reliable than information on 
eligibility for such services and this information will be simpler for 
States to obtain.
    We propose to collect this information because of our interest in 
monitoring the well-being of children in the out-of-home care reporting 
population and our desire to provide a more comprehensive picture of 
the needs of children. We also believe that gathering this information 
is consistent with the case plan requirements in section 475(1)(C) of 
the Act.
    Prior adoption. In paragraph (b)(10), we propose to continue the 
requirement for the State agency to report whether the child has 
experienced a prior finalized adoption (see appendix A to part 1355, 
section II, II.E). We clarify in the proposed regulation text that we 
are interested in whether the child has experienced a finalized 
adoption prior to the current out-of-home care episode as opposed to an 
adoption that occurs during the current out-of-home care episode. We 
also are clarifying that the State is to include any type of prior 
adoption in this element, regardless of whether the adoption was 
public, private, independent, or an intercountry adoption. Many 
commenters on the Federal Register notice expressed a desire for 
continuing and expanding the information we collect on prior adoptions 
to better determine the extent to which children in out-of-home care 
are involved in dissolved adoptions where the adoptive parents' rights 
are terminated and displaced adoptions where the child enters out-of-
home care after a finalized adoption.
    Prior adoption date. In paragraph (b)(10)(i), we propose for the 
first time that a State report the finalization date of the child's 
prior adoption. In the case of an intercountry adoption, the child's 
parents may have gone through a readoption process in the State where 
they reside. While in many cases this process is optional for a child 
whose adoption was finalized in the originating country, we understand 
that there are some States that require the child to be readopted in 
his/her State of residence. In such cases, we are requiring that the 
State provide the date that the adoption is considered final in 
accordance with the State's laws on readoption.
    In the existing AFCARS, we ask the State to report the child's age 
range at the time of the prior finalized adoption (appendix B to part 
1355 section II, II.E). This information, however, was insufficient to 
determine accurately when the child was previously adopted. Thus, we 
propose that the State report the actual finalization date to allow us 
to determine how much time has elapsed between the child's previous 
adoption and his or her current out-of-home care stay.
    Prior adoption type. In paragraph (b)(10)(ii), we seek information 
for the first time on the type of adoption the child experienced 
previously. In this element, States must distinguish between a prior 
adoption that occurs out of the reporting State's foster care system, 
another State's foster care system, an intercountry adoption, or 
another type of private or independent adoption. Commenters on the 
Federal Register notice believed that an element of this nature would 
be useful in informing our understanding of dissolved and displaced 
adoptions.
    We define intercountry adoptions as those that occur in another 
country, or those adoptions that are finalized in the United States 
after the foreign child has been brought into the country for the 
purposes of adoption. Another country in this case means any country 
outside of our definition of a State for title IV-B in 45 CFR 1355.20. 
We seek this information primarily in response to the requirements of 
the Intercountry Adoption Act (IAA) of 2000 (Pub. L. 106-279). The IAA 
added section 422(b)(14) to the Social Security Act and requires that a 
State collect and report certain information on children who are 
adopted from other countries and who enter State custody as a result of 
the disruption of a placement for adoption or the dissolution of that 
adoption. This information will allow us to compile the number of 
children and permanency plans for children involved in dissolved 
adoptions and from where such children originated.
    Prior adoption location. In subparagraph (b)(10)(iii), we propose 
that a State submit the FIPS code which corresponds with the State or 
country in which the child was previously adopted, if applicable. This 
also is a new element. We propose to collect this information so that 
we can calculate accurately the dissolution and displacement rates for 
both the State in which the child was adopted and the State in which 
the displacement or dissolution occurred. Further, collecting 
information on the actual country of the prior adoption will inform our 
understanding of intercountry adoptions that require the intervention 
by State public child welfare agencies consistent with the IAA.
    Number of siblings living with the child at removal. In paragraph 
(b)(11), we propose for the first time that the State report the total 
number of siblings living with the child at the time of the child's 
removal from home, if any. These siblings may be biological, legal or 
by marriage but cannot be adults according to the State's age of 
majority. The State is not to include the child who is the subject of 
the report (i.e., the child whose record number is reported for the 
element in paragraph (b)(4)) in this count.
    We wish to be clear that States must report only the number of the 
child's siblings who were living with the child at removal and not the 
total number of siblings of the child. This includes all siblings 
living with the child at removal, whether the sibling relationship is 
biological, legal or by marriage. We are making this distinction 
because it is more useful for us to know the number of sisters and 
brothers who lived with the child rather than the sum total of all 
siblings regardless of where they lived. Since we are interested in 
understanding the dynamics of sibling groups for permanency planning 
purposes, we do not believe it is necessary for the State also to 
report information on a child's brothers or sisters who are not present 
in the home and for whom the parent/legal guardian may not be 
responsible.
    The reason that we require States to report this information is 
because we want to get an accurate count of the number of siblings in 
out-of-home care who were actually living together at one time prior to 
the entry of the child into out-of-home care. We need this element 
specifically so that we can understand

[[Page 2095]]

when the number of siblings in out-of-home care is different from the 
number of siblings who were living together at removal. For example, it 
is possible that the mother could give birth to an infant who is 
removed from home after the reported child enters out-of-home care, 
thereby increasing the count of the number of children in out-of-home 
care, but not the number of siblings in the AFCARS population when the 
child was removed. This has implications for the child's permanency 
plan and State agency expectations for placing siblings together.
    We propose this element, along with the family identification 
number (discussed previously) and the number of siblings placed 
together (described later), in order to get information on sibling 
groups for a variety of reasons. Good practice dictates that, where 
possible and in the best interests of the child, siblings in out-of-
home placements should be placed together. However, we also know that 
addressing the needs of sibling groups provides agencies with special 
challenges. The data that we propose to collect, among other things, 
will provide us with extremely useful information about siblings. For 
example, this data will allow us to analyze how being a part of a 
sibling group involved with the child welfare agency affects the 
timeliness and success of reunification. Furthermore, it is especially 
important to know about sibling groups for adoption purposes, since we 
know that many children placed into out-of-home care are later placed 
for adoption. In addition, most States use ``sibling'' groups as one of 
the special needs categories for providing adoption subsidies. We 
understand that this is one of the most difficult groups of children 
for whom States must find adoptive homes.
    Many Federal Register commenters agree that we need to modify 
AFCARS to obtain information on siblings. Commenters believe that such 
data will allow States to track sibling groups that are placed together 
or apart; analyze how well agencies preserve sibling attachments, as 
well as determine and implement services that specifically address the 
needs of sibling groups. Typically, States have this information in 
case files, but it is not yet an established practice for all States to 
track this information in their case management systems. We found 
through the CFSRs that a State can lose track of a child's siblings. We 
believe that requiring States to report sibling groups through AFCARS 
will decrease the frequency of this happening.
    Finally, requiring sibling information in AFCARS will be useful for 
the CFSRs. In the CFSR, we rate States on several items that relate to 
this issue, such as preserving family connections, visiting between 
children in foster care and their families, and relative placements. As 
States enter program improvement plans (PIPs) to improve these areas, 
it will be helpful to have this data in AFCARS to be able to identify 
where the problems are and track progress over time. We also rate the 
safety and well being items on all children in the family, regardless 
of whether the case is a foster care case.
    Minor parent. In paragraph (b)(12), we propose that the State 
collect and report the number of children either fathered or borne by 
the young person in the State's AFCARS report. If the young person has 
no children, the State must indicate zero. States are to report the 
total of all children of the young parent, irrespective of whether or 
not such children live with their parent.
    Commenters requested an element of this nature and we feel it is 
important for us to have improved data about the characteristics of 
young people in out-of-home care. This information can allow us to 
analyze the extent to which having children affects a youth's 
permanency plan. This data element also will be used in conjunction 
with a subsequent data element in 45 CFR 1355.43(e)(9) about the 
population of young people in out-of-home care who have children for 
whom they are responsible and are living with them. The combination of 
information in the two elements will allow us to determine the number 
of young people in out-of-home care who have children, and the extent 
to which those young people are responsible for the care of their 
children.
    Child financial and medical assistance. In paragraph (b)(13), we 
propose that a State report for the first time the type of financial 
and medical assistance that the child received during the current six-
month report period. The State is to indicate whether the child 
receives benefits under title XVI of the Act (including SSI), the 
State's Medicaid program including under title XIX waivers or 
demonstrations, the State's Children's Health Insurance Program (SCHIP) 
including under title XXI waivers or demonstrations, a State adoption 
subsidy, a State foster care payment, child support, other financial 
assistance or no financial assistance.
    While there are elements in the existing AFCARS that require States 
to report the sources of Federal support for the child, this element is 
different in that it focuses on a variety of financial and medical 
assistance rather than just Federal support. The statute at section 
479(c)(3)(D) of the Act requires that we collect national information 
on ``the extent and nature of assistance provided by Federal, State, 
and local adoption and foster care programs.'' As such, we believe that 
expanding the scope of our financial and medical assistance elements to 
gather more information on assistance for the child is required by law. 
This proposed element, in conjunction with the following element on 
receipt of title IV-E foster care maintenance payments and elements in 
the living arrangement section of this NPRM (1355.43(e)), will allow us 
to gather more information on the kinds of financial and medical 
assistance that support children in out-of-home care.
    Title IV-E foster care during report period. In paragraph (b)(14), 
we propose a new element for the State to report specifically whether 
the child received a title IV-E foster care maintenance payment during 
the current report period. The State is to respond affirmatively that 
the child has received a title IV-E foster care maintenance payment 
only if one was paid on the child's behalf during the current six-month 
report period, or the child is eligible for the program in accordance 
with section 472(a) of the Act and the State will claim Federal 
reimbursement under title IV-E for the child's foster care maintenance 
payment.
    This element is used primarily to extract the title IV-E foster 
care eligibility review samples. Currently, the title IV-E foster care 
eligibility review sample is drawn from an existing AFCARS element that 
requires States to identify foster care maintenance payments as one of 
many Federal sources of support for the child. We have learned through 
technical assistance and AFCARS assessment reviews, however, that 
States often report this element incorrectly. A common mistake with the 
existing element involves the State indicating that the child is 
receiving title IV-E foster care maintenance payments when the child 
has met some title IV-E eligibility requirements (e.g., AFDC 
eligibility) but not all. We wish to isolate this element so that we 
can clearly define it and improve the sample selection process for the 
title IV-E foster care eligibility reviews.
Section 1355.43(c) Parent or Legal Guardian Information
    In paragraph (c), we are seeking demographic information on the 
child's parent(s) or legal guardian(s).
    Year of birth of parent(s) or legal guardian(s). In paragraphs 
(c)(1) and

[[Page 2096]]

(c)(2), we propose that the State collect and report to AFCARS the 
birth year of the child's parents or legal guardians. This information 
is sought on the child's parent or legal guardians regardless of with 
whom the child is living at the time of removal from home. If the State 
cannot obtain this information because the child is abandoned, the 
State must so indicate. This information differs from the existing 
AFCARS in that we currently request the year of birth of the child's 
caretakers from whom he or she was removed (see appendix A to part 
1355, section II, VII.B). The information collected under the existing 
regulation does not clearly indicate whether the child's caretaker was 
the parent, legal guardian, or some other person who was temporarily 
taking care of the child at the time that the child was removed from 
home. Because of this lack of clarity, our ability to analyze the 
existing data is limited.
    We believe that focusing the proposed elements on the child's 
parents or legal guardians is more consistent with the statutory 
mandate to collect demographic information on the biological and 
adoptive parents of children in foster care (section 479(c)(3)(A) of 
the Act). By expanding our requirement to gather the year of birth of 
all legal parents (i.e., inclusive of biological parents, adoptive 
parents and stepparents) or the child's legal guardian, we believe we 
are better meeting the intent of the statute to understand the 
characteristics of persons who are legally responsible for children who 
must enter foster care.
    Mother married at time of the child's birth. In paragraph (c)(3), 
we propose that a State report to us whether the child's biological 
mother was a married person at the time the child was born. This 
element is similar to one that States collect currently, except that in 
the existing element we require that a State provide this information 
only for children who are adopted (see appendix B to part 1355, section 
II, IV.B). We believe that this information is better suited for the 
out-of-home care reporting population as a whole. According to 
comments, some stakeholders believed this information was unnecessary 
while others believed it should be expanded to be reported for the 
entire out-of-home care reporting population. We chose to expand the 
reporting of this element for a few reasons. First, we understand from 
AFCARS assessment reviews that many States already collect this 
information when a child enters out-of-home care rather than at the 
point of adoption, so broadening the scope of this requirement should 
not increase the burden on States. Second, from our analysis of the 
existing data on whether the child's mother was married at the time of 
the child's birth, we have found that the marriage rates in our 
population are lower than the national average. According to the 
National Center for Health Statistics, 34% of births are to unmarried 
women compared to over half of the births of children adopted from 
public foster care systems. One of the priorities of this 
administration is to promote healthy marriages, in part, because 
researchers have found many benefits for children and youth who are 
raised by parents in healthy marriages. In that context, we are 
interested in gathering data that may help us assess if a mother's 
marital status at the time of the child's birth is a factor in a 
child's child welfare experience. This collection also is consistent 
with the statutory mandate to collect demographic information under 
section 479(c)(3)(A) of the Act.
    Termination of parental rights petition. In paragraphs (c)(4) and 
(c)(6), we seek new information on the date that a petition to 
terminate parental rights (TPR) was filed against the child's parents. 
This information will provide us with data we can use to evaluate how 
States are complying with the requirement in section 475(5)(E) of the 
Act to file a petition to terminate the parental rights of certain 
children in foster care. Further, this information, in conjunction with 
information collected on final dates of TPR and adoption, will help us 
determine how long it takes for permanency to be achieved for children 
who are adopted.
    Termination of parental rights. In proposed paragraphs (c)(5) and 
(c)(7), we continue the existing requirement for States to collect and 
report data on the date that parental rights are terminated for each 
parent (see appendix A to part 1355, section II, VIII).
    For all data elements related to the termination of parental 
rights, we propose to clarify that we are seeking information on a 
child's putative father, if applicable. A putative father is a person 
who is alleged to be the father of a child, or who claims to be the 
father of a child, at a time when there may not be enough evidence or 
information available to determine if that is correct. For the current 
AFCARS we have fielded questions on whether States should provide 
information on putative fathers. Since States must terminate the 
parental rights of any putative fathers to ensure that a child legally 
is free adoption, we want to be clear that we are interested in this 
information as well.
    Finally, we would like to note that we propose to eliminate the 
existing element on the family structure of the child's caretakers from 
whom the child was removed (see appendix A to part 1355, section II, 
VII.A). We concur with several commenters to the Federal Register 
notice that this information is not useful as currently constructed. 
However, we have proposed alternative elements in paragraph (d) that we 
believe will give us better insight into the composition of the child's 
household at the time of removal.
Section 1355.43(d) Removal Information
    In paragraph (d) we propose that the State submit information 
related to the child's removal from home and the assumption of 
responsibility by the State agency for placement and care of the child. 
We request that for any child in the reporting population, the State 
submit removal information regarding every occasion that the child is 
removed from home until the child has reached the age of majority. This 
is a significant change from the existing AFCARS, where we require 
detailed removal information on the child's most recent removal only.
    The major reason for making this change is that we will be able to 
analyze more accurately the frequency and circumstances surrounding a 
child's entry into out-of home care. As pointed out earlier, many 
States and other stakeholders have indicated that longitudinal data 
that permits the examination of entry, exit, permanency plan and living 
arrangement information is critical to the CFSR process and other 
efforts to measure outcomes.
    Date of child's removal. In paragraph (d)(1), we propose that the 
State collect and report the date or dates on which the child was 
removed from his or her parents or legal guardians and placed under the 
placement and care responsibility of the State title IV-B/IV-E agency. 
This proposed element differs from the existing AFCARS, which asks for 
the dates of the child's first removal and latest removal from home for 
the purpose of placement in a foster care setting (see appendix A to 
part 1355, section II, III.A). The proposed element requires the State 
to report all removal dates in one element and clarifies which dates 
the State must report in certain circumstances.
    In many cases the date of the child's removal will be when the 
child is removed physically from his home and placed directly into out-
of-home care. However, for a child who was already away from his 
parents at the time the

[[Page 2097]]

State child welfare agency receives placement and care responsibility 
(i.e., in the case of a runaway, constructive removal, or transfer of 
placement and care responsibility from a separate public agency), the 
State agency must report the date when it receives placement and care 
responsibility rather than the date of physical removal. Further, if 
the child was in out-of-home care previously and returned home with 
continued State agency placement and care responsibility (which must be 
reported as an exit in accordance with our proposed reporting 
population), the date of the child's removal is the date of the new 
removal from the child's home.
    A major reason why we are proposing that States report all removal 
dates is so that we can accurately analyze a child's repeat foster care 
re-entry rate for CFSR purposes, particularly any associated length of 
time to re-entry. Currently, we are able to measure a child's re-entry 
rate using AFCARS information, but this information has limitations. 
For example, the current AFCARS does not allow us to analyze the 
child's entire detailed history of removals. Furthermore, by requiring 
that the State title IV-B/IV-E agency provide us with all of the dates 
in a child's entire removal history, rather than only the first and 
current removal dates, we can identify trends that might assist States 
in better understanding their data and making program improvements as 
needed. Without the entire history, we are unable to determine, for 
example, the effects of States' program improvement planning efforts on 
repeat entries into foster care, the duration of all episodes of foster 
care, and the outcomes of a child's stay in foster care.
    We do not believe that the changes to the removal date will be an 
additional burden on States because we understand that most, if not all 
States, have this information in their existing information systems. In 
fact, this proposal may ease State burden such that the State can 
simply transmit all of its removal date information, rather than 
separating out which dates to report for AFCARS purposes only. We 
welcome comments on this proposal.
    Removal transaction date. In paragraph (d)(2), we propose that the 
State title IV-B/IV-E agency continue to report the date that the State 
agency entered the child's removal date into the State's information 
system (see appendix A to part 1355, section II, III.A). This 
transaction date must accompany every removal date. This must be a 
computer-generated, non-modifiable date. To be timely, the date must be 
entered within 15 days of the child's removal from his/her parent and 
placement under the agency's responsibility.
    Although this is a significant change in the time frame for the 
State to enter the date of a child's removal, we have found that States 
report more accurate, high quality data when the transaction date is 
entered into the information system close in time to the event that it 
describes. This is our ultimate goal with this proposed change; to have 
accurate dates of removal for all children reported. A child's removal 
date is one of the most critical data elements in the AFCARS, as it is 
the anchor date for calculating certain CFSR outcome measures and is 
necessary for other purposes as well.
    Some commenters to the Federal Register notice suggested that 
entering the transaction date should be secondary to ensuring child 
safety. We agree that child safety is paramount, and understand the 
competing demands placed on child welfare workers. However, we have not 
changed our position that States must enter the child's removal date 
into the State's information system in a timely manner. Further, 
information from our analysis of AFCARS data submitted for the FY 2003 
and FY 2004 report periods indicate that three-fourths of the cases are 
entered within 15 days of the child's removal. Therefore, we do not 
believe that this proposed change will be a significant departure from 
State practice in most instances. We welcome comments on this proposed 
change.
    Environment at removal. In paragraph (d)(3), we propose that the 
State agency report if the child was living in a household or in 
another environment at the time of each removal. This is a new element. 
We propose that States report whether the child was living in a 
household or another environment (e.g., the child has run away) so 
that, in conjunction with the two subsequent elements on household 
composition and biological parents' marital status, we can learn more 
about the child's home or situation prior to entering out-of-home care. 
The existing AFCARS requires a State to report the family structure of 
the child's caretakers at removal. We have found this information to be 
insufficient for our analytical needs as it does not provide 
information about with whom the child was living, if anyone, or 
identify family relationships specifically. We believe that more detail 
about the child's environment at removal will inform our analysis of 
how children come into out-of-home care and their child outcomes.
    Household composition at removal. In paragraph (d)(4) and its 
subparagraphs (d)(4)(i) through (xi), we propose for the first time 
that the State report all adults in the child's household with whom the 
child was living at the time of each removal. We propose that States 
identify the composition of the child's household if the child was 
actually removed from a home environment as identified in the previous 
element. States may identify parents, grandparents, other relatives, a 
paramour of a parent or caretaker, other non-relatives, adult siblings, 
or other non-related caretakers, by indicating how many of each 
category of persons was in the home. For example, if the child was 
living with the biological mother and stepfather at removal, the State 
would indicate that there was one biological parent, one stepparent, 
and indicate a zero for all other persons.
    We propose to require that States report this information because 
we want to gather as much information as is practical about a child's 
life at the time of removal to conduct various analyses relating to 
under what circumstances and with whom children are living before they 
enter out-of-home care. We are aware that some children who are legally 
removed from their parents do not live with them at the point of 
removal, or are also cared for by another adult. Some may be living 
informally with relatives or neighbors. In short, having this 
information will enrich what we know about children who enter out-of-
home care.
    We have been careful to clarify in our description of a non-related 
caretaker that States report information on only those persons who have 
assumed responsibility for the day-to-day care of the child. This is 
because we are interested in collecting information on those persons 
who have an ongoing caretaking role for the child as opposed to those 
who may have temporary physical possession of the child. We believe it 
serves little analytic purpose to gather information on persons who are 
not part of the child's household prior to the child's entry into out-
of-home care. For example, there may be a situation where a parent 
leaves the child with a babysitter or neighbor for the day but has not 
returned a couple of days later, at which point the babysitter or 
neighbor contacts the child welfare agency. In such a situation, the 
babysitter or neighbor has not assumed responsibility for that child 
and the State must report information on the persons in the child's 
household instead. We welcome comments on this element.
    Biological parents' marital status. In paragraph (d)(5), we propose 
that the State report the marital relationship

[[Page 2098]]

between the child's biological parents if the child was removed from at 
least one biological parent. We propose that the State report whether 
the biological parents are married to each other and whether they are 
living together at the time of the child's removal. We also have a 
category for a deceased biological parent that should be used 
regardless of the parents' marital status at the time of the parent's 
death. We are proposing this element because, as noted earlier, we are 
interested in the role that marriage plays in positive child outcomes, 
particularly as it relates to the child's biological parents.
    Manner of removal. In paragraph (d)(6), we propose that the State 
title IV-B/IV-E agency continue to collect and report on the State's 
authority to remove the child from home for each removal (see appendix 
A to part 1355, section II, IV.A). We have made no changes to the 
information that is reported, except that it must be reported for every 
removal the child experiences. Specifically, the State title IV-B/IV-E 
agency is to indicate whether the State's authority for removing the 
child from home for each removal was based on a court order or a 
voluntary placement agreement. If this is not yet determined, the State 
must so indicate and update the record to reflect the manner of removal 
once it is known. We continue to envision that the ``not yet 
determined'' category will happen in short-term cases only since 
establishing the appropriate legal authority to remove a child from 
home is an initial and critical State agency responsibility.
    We considered making changes to this section in an attempt to 
distinguish court orders that are for the placement of children into 
the agency's responsibility for dependency reasons and those that are 
for juvenile justice agency involvement reasons. Because State practice 
with regard to this issue is so varied, we do not think that there is a 
single way to categorize court orders. Therefore, we propose changes to 
the elements related to child and family circumstances at removal and 
juvenile justice involvement to gather information on children with 
juvenile justice agency involvement.
    Child and family circumstances at removal. In paragraph (d)(7), we 
propose to collect data about the circumstances surrounding the child 
and family at the time of the child's removal from home. While 
currently we collect information on the circumstances associated with a 
child's most recent removal (see appendix A to part 1355, section II, 
IV.B for all response options), we propose in this element to require 
this information for every removal and expand the list of 
circumstances, among other things, to include juvenile justice 
information.
    We do not characterize these circumstances as the reasons for or 
causes of removal, although certainly some of these factors may have 
been the sole basis for the removal. Consistent with the existing 
AFCARS, we propose that the State agency only include information in 
this element that it has gathered about the child, the child's family 
and circumstances at the time the agency removes the child from home. 
As the State investigates and works with a family, the agency may learn 
of other factors or underlying issues that could have contributed to or 
necessitated removal. But we are not seeking that information here. 
Rather, we propose additional elements to capture circumstances that 
may arise during the course of the child's stay in out-of-home care as 
discussed later in the permanency and exit sections of the NPRM. In 
this element, we wish to understand, in a comprehensive manner, what is 
occurring in a child's life at the time of removal. Therefore, we 
propose to retain the current feature of AFCARS to require that the 
State indicate all of the circumstances associated with a child's 
removal. We have had concerns with the practice in some State agencies 
of reporting only the primary reason associated with the child's 
removal, leaving out important information about other relevant 
circumstances. We want to emphasize here that the State must report all 
of the circumstances at the time of the child's removal. Below, we 
explain all the response options for this element.
    Juvenile Justice. We propose two new response options for 
circumstances at removal that are juvenile justice related. Currently, 
in AFCARS, the circumstances associated with the child's removal do not 
include the child's involvement, if any, with the juvenile justice 
system. Consequently, we have not been able to identify which children 
begin their out-of-home care experience with alleged or adjudicated 
delinquent or status offenses. As indicated earlier, we have heard 
through a variety of sources, including comments on the Federal 
Register notice and the CFSRs, that it is important to clarify the 
characteristics of the reporting population so that we will be able to 
analyze the differences in various CFSR and other outcome measures.
    Specifically, we propose that a State report whether the child is 
alleged or found to be a status offender at removal. We propose to 
define status offenses as those that are specific to juveniles, 
including but not limited to, running away from home, underage alcohol 
violations and truancy. We propose that the State title IV-B/IV-E 
agency report a child status offen