Foster Care: Health Needs of Many Young Children Are Unknown and Unmet
(Letter Report, 05/26/95, GAO/HEHS-95-114).

Pursuant to a congressional request, GAO provided information on the
health-related services needed and received by young children in foster
care, focusing on: (1) the relationship between the receipt of
health-related services and foster care placements with relatives versus
placements with nonrelatives; and (2) how responsible agencies are
ensuring that foster children are receiving needed health-related
services.

GAO found that: (1) despite foster care agency regulations requiring
comprehensive routine health care, an estimated 12 percent of young
foster children receive no routine health care, 34 percent receive no
immunizations, and 32 percent have some identified health needs that are
not met; (2) an estimated 78 percent of young foster children are at
high risk for human immunodeficiency virus (HIV) as a result of parental
drug abuse, yet only about 9 percent of foster children are tested for
HIV; (3) young foster children placed with relatives receive fewer
health-related services than children placed with nonrelative foster
parents, possibly since relative caregivers receive less monitoring and
assistance from caseworkers; (4) the number of young children placed
with relatives increased 379 percent between 1986 and 1991, which
resulted in the lower likelihood of these children receiving services
associated with kinship care; and (5) the Department of Health and Human
Services has not designated any technical assistance to assist states
with health-related programs for foster children and does not audit
states' compliance with health-related safeguards for foster children.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-95-114
     TITLE:  Foster Care: Health Needs of Many Young Children Are 
             Unknown and Unmet
      DATE:  05/26/95
   SUBJECT:  Acquired immunodeficiency syndrome
             Foster children
             Health care services
             Federal/state relations
             Child custody
             State-administered programs
             Disease detection or diagnosis
             Drug abuse
             Parents
             Immunization services
IDENTIFIER:  AFDC
             Aid to Families with Dependent Children Program
             Los Angeles County (CA)
             New York (NY)
             Philadelphia County (PA)
             Medicaid Program
             New York
             California
             Pennsylvania
             AIDS
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


Cover
================================================================ COVER


Report to the Ranking Minority Member, Subcommittee on Human
Resources, Committee on Ways and Means, House of Representatives

May 1995

FOSTER CARE - HEALTH NEEDS OF MANY
YOUNG CHILDREN ARE UNKNOWN AND
UNMET

GAO/HEHS-95-114

Services for Young Foster Children


Abbreviations
=============================================================== ABBREV

  ACF - Administration for Children and Families
  AFDC - Aid to Families With Dependent Children
  AIDS - acquired immunodeficiency syndrome
  CDC - Centers for Disease Control and Prevention
  EPSDT - Early and Periodic Screening, Diagnosis, and Treatment
  HHS - Department of Health and Human Services
  HIV - human immunodeficiency virus

Letter
=============================================================== LETTER


B-259319

May 26, 1995

The Honorable Harold E.  Ford
Ranking Minority Member
Subcommittee on Human Resources
Committee on Ways and Means
House of Representatives

Dear Mr.  Ford: 

Foster children are among the most vulnerable individuals in the
welfare population.  As a group, they are sicker than homeless
children and children living in the poorest sections of inner cities. 
Of particular concern is the health of young foster children since
conditions left untreated during the first 3 years of life can
influence functioning into adulthood and impede a child's ability to
become self-sufficient later in life.  Yet, little comprehensive
information is available about the provision of health-related
services to meet the needs of young foster children. 

Last year, we reported that the population of young foster
children--those 36 months of age and younger--changed significantly
between the late 1980s and the early 1990s.\1 The average monthly
number of children in foster care nationwide increased 53 percent,
from 280,000 to 429,000, during this period.  The total foster care
population in the three states reviewed increased 66 percent between
1986 and 1991, while the number of young foster children more than
doubled--increasing by 110 percent.  In addition, we found that a
greater proportion of young children entered the system because of
some form of neglect; came from families where at least one of the
parents was abusing drugs; had serious health-related problems; and
were at risk for future problems as a result of prenatal drug
exposure. 

Understanding the ability of state child welfare agencies to meet the
needs of foster children is critical as policymakers consider
restructuring federal welfare policies and responsibilities.  Federal
monies are currently used to assist states with the cost of foster
care.  Legislation being considered by the 104th Congress would give
the states even greater responsibility for foster children through
block grants. 

Our earlier work responded to your request that we compare and
contrast the population sizes and distinctive characteristics of
young foster children between 1986 and 1991.  This report responds to
the remaining issues in that request regarding the service needs of
young foster children.  Specifically, this report provides
information on (1) the health-related services needed and received by
young children in foster care, (2) the relationship between the
receipt of health-related services and foster care placements with
relatives versus placements with nonrelatives, and (3) what
responsible agencies are doing to ensure that these children are
receiving needed health-related services. 

To develop this information we reviewed foster care programs in
California, New York, and Pennsylvania, the states with the largest
average monthly foster care populations in 1991.  In addition, we
analyzed random samples of case files from Los Angeles County, New
York City,\2 and Philadelphia County from a combined population of
22,755 young foster children.  These locations cared for a
substantial portion of each state's young foster children:  44
percent in California, 81 percent in New York, and 29 percent in
Pennsylvania.  We analyzed electronic databases as provided to us by
state and county officials to select our samples and determine the
number of children placed with relatives and nonrelatives.  Our scope
and methodology are discussed further in appendix I. 


--------------------
\1 We reported on young foster children in Los Angeles County, New
York City, and Philadelphia County.  See Foster Care:  Parental Drug
Abuse Has Alarming Impact on Young Children (GAO/HEHS-94-89, Apr.  4,
1994). 

\2 New York City comprises five boroughs and is treated in the state
database as a county.  In this report, we refer to it as a county. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

Our work indicates that a significant proportion of young foster
children did not receive critical health-related services in the
three locations reviewed--Los Angeles County, New York City, and
Philadelphia County.  Despite state and county foster care agency
regulations requiring comprehensive routine health care,\3 an
estimated 12 percent of young foster children received no routine
health care, 34 percent received no immunizations, and 32 percent had
at least some identified health needs that were not met. 
Furthermore, an estimated 78 percent of young foster children were at
high risk for human immunodeficiency virus (HIV) as a result of
parental drug abuse, yet only an estimated 9 percent of young foster
children were tested for it.  Early identification of HIV-infected
children begins with HIV risk assessment.  Without early
identification, HIV-infected children with mild or no symptoms cannot
receive the early medical care that is known to be effective with
young children. 

We also found that young foster children placed with relatives
received fewer health-related services of all kinds than children
placed with nonrelative foster parents.  Other research indicates
that relative caregivers often receive less monitoring and assistance
from caseworkers.  For California and New York--the states where
placement data were available--the number of young children placed
with relatives increased by 379 percent between 1986 and 1991, while
the number of young children placed with nonrelative foster parents
increased by 54 percent.  Consequently, because a larger number of
foster children were placed with relatives in 1991 than in 1986,
substantially more children were subjected to the lower likelihood of
receiving services associated with kinship care. 

Local foster care agencies continue to grapple with designing
programs to meet the health-related service needs of children.  In
the locations reviewed, agencies have revised health-related foster
care regulations and modified their programs in efforts to improve
the delivery of health care to foster children.  Although the
Department of Health and Human Services (HHS) recently increased its
technical assistance to states by contracting for National Resource
Centers,\4 none is designated to assist states with health-related
programs for foster children.  Furthermore, while HHS audits states
for compliance with federally mandated safeguards for foster
children, these audits omit review of compliance with health-related
safeguards.  Given the importance of health care during the first 3
years of life, an improved response to the health needs of this
vulnerable population is vital. 


--------------------
\3 County foster care regulations state the frequency for required,
comprehensive medical examinations for well children.  We refer to
this type of examination as comprehensive routine health care. 

\4 The National Resource Centers assist HHS' Administration for
Children and Families staff in responding to states' questions and
provide free technical assistance to states. 


   BACKGROUND
------------------------------------------------------------ Letter :2

Responsibility for providing care and services to foster children is
shared by federal, state, and county governments, with HHS having
responsibility for oversight of federal foster care programs.  The
Administration for Children and Families (ACF) within HHS helps the
states to develop plans required under title IV-B of the Social
Security Act; reviews and approves those plans; conducts audits to
certify states' compliance with the safeguards for foster children,
thereby making states eligible for additional federal funds; and
allocates funds to states, among other duties. 

The Social Security Act of 1935 was amended twice to include
safeguards for foster children.  The Adoption Assistance and Child
Welfare Act of 1980 (P.L.  96-272) added most of these
safeguards--such as requirements that the case file contain a plan
for appropriate care and services, as determined by state and local
foster care policies; periodic court or administrative reviews; and a
reunification program to return children to their parents.  This act
also authorized HHS to provide technical assistance to aid states in
developing programs to meet the requirements of the law. 
Furthermore, the Omnibus Budget Reconciliation Act of 1989 (P.L. 
101-239) added other safeguards to the Social Security Act, including
a requirement to maintain health records for foster children. 

A combination of federal, state, and county funds may be used to
provide services to young foster children.  States may participate in
federal programs authorized by the Social Security Act such as title
IV-B, matching grants for various child welfare services; title IV-E,
an uncapped entitlement for a portion of the maintenance of foster
children who are eligible under the Aid to Families With Dependent
Children (AFDC) program; title XIX, Medicaid, an entitlement for a
portion of medical services; or title XX, block grants for a wide
array of social services for children.  In addition, the Education of
the Handicapped Act, part H, authorizes grants to states for early
intervention programs for handicapped infants and toddlers. 

Except for federal title IV-E expenditures, data were unavailable to
estimate federal, state, and county expenditures for services for
foster children.  In the last 10 years, federal title IV-E
expenditures for the administration and maintenance of AFDC-eligible
foster children increased from about $546 million in 1985 to an
estimated $2.9 billion in 1995.  When foster children do not meet
title IV-E eligibility for federal funding, states must bear the full
cost for maintaining these children.  However, some states pass at
least a portion of these costs to their counties. 


   YOUNG FOSTER CHILDREN IN THE
   LOCATIONS REVIEWED DID NOT
   ALWAYS RECEIVE NEEDED
   HEALTH-RELATED SERVICES
------------------------------------------------------------ Letter :3

All young children need routine, comprehensive medical monitoring,
treatment for minor illnesses, and immunizations to grow up healthy. 
In the three locations reviewed, state and county regulations require
that children in foster care receive periodic medical examinations
and treatment.\5 Research indicates that children at risk for serious
health problems as a result of prenatal drug exposure often need
additional assessments and specialized care.  Child development
experts generally agree that health care is particularly important
during the first 36 months of life as language, motor, psychological,
and social skills develop.  Conditions left untreated during the
first 3 years of life can influence functioning into adulthood. 

Some young foster children in the locations we reviewed did not
receive even the most basic health service--required routine care. 
In addition, many children had identified health-related needs that
were not met, including the need for specialized services.  Foster
care agencies refer foster parents to community-based programs and
practitioners, rather than providing the services directly.  Foster
children in the locations reviewed are eligible for Medicaid to cover
the cost of these health-related community-based services. 


--------------------
\5 Regulations include a requirement that children receive an initial
examination when they enter foster care.  Los Angeles County and New
York City require an examination within 30 days of entry into foster
care; Philadelphia County requires an examination within 60 days. 
Children who received a comprehensive examination within 90 days
before entering foster care are exempt from this requirement. 


      ROUTINE HEALTH CARE MAY NOT
      BE ENSURED
---------------------------------------------------------- Letter :3.1

Despite state and county foster care regulations, comprehensive
routine health care for young foster children may not be ensured. 
Specifically, an estimated 12 percent of the children received no
routine health care, and 34 percent received no immunizations in the
three locations reviewed.  Furthermore, case files at all three
locations did not reflect the exact nature or extent of what services
were provided in many cases.  Thus, children we noted as having
received routine medical care may have received as little care as one
visit with a physician for treatment of a minor illness rather than
comprehensive or ongoing medical care.  (See table II.1 in app.  II.)

While Early and Periodic Screening, Diagnosis, and Treatment (EPSDT)
services are one way to ensure that children receive comprehensive
medical examinations, only an estimated 1 percent of the young foster
children in the locations reviewed received EPSDT services.  EPSDT
services are specific, comprehensive medical examinations and
follow-up treatment that states must offer to Medicaid-eligible
children.  EPSDT examinations can serve as an effective safeguard of
a child's overall health and development and as a gateway to other
health-related services.  (See table II.1 in app.  II.)

Children with no known health problems were less likely to receive
routine care than children who were at risk for or had serious health
problems.\6 For the locations reviewed, an estimated 28 percent of
the children with no known serious health problems did not receive
any health-related services.  By comparison, only 6 percent of
children who were at high risk for serious health problems because of
prenatal drug exposure and 2 percent of children with serious
physical health problems did not receive any health-related services. 
Without routine health care, children with no known health problems
are not monitored to identify and treat health and developmental
problems as they occur.  (See table II.2 in app.  II.)


--------------------
\6 Serious health problems of children in our review include fetal
alcohol syndrome, low birth weight, cardiac defects or heart
problems, HIV or acquired immunodeficiency syndrome (AIDS), and
developmental delays. 


      SPECIALIZED HEALTH NEEDS OF
      YOUNG FOSTER CHILDREN WERE
      UNMET
---------------------------------------------------------- Letter :3.2

In addition to routine health care, young foster children need many
specialized health-related services.  As we previously reported, an
estimated 58 percent of young foster children in the three locations
reviewed had serious physical health problems, and 62 percent were at
high risk for serious health problems as a result of prenatal drug
exposure.  Many of these children may need health-related services
and treatment beyond those needed by the average child.  (See fig.  1
and table II.3 in
app.  II.)

   Figure 1:  Specialized Services
   Received in Three Counties

   (See figure in printed
   edition.)

\a Includes treatment for asthma, syphilis, seizures, and kidney
problems. 

\b Includes blood, laboratory, and radiology. 

\c Includes developmental, psychological, and cardiological. 

\d Includes care for HIV, pneumonia, and failure to thrive, as well
as surgery. 

\e Includes apnea monitors, infant stimulation, and speech therapy. 

\f Includes therapeutic day care and Head Start services. 

Source:  Case file review. 

Although young foster children received a wide variety of services
from health care providers, many children had identified
health-related needs that were not met.  Based on information
collected from case files, we matched the health-related needs
identified and the services received for each child and estimated
that one-third of the children in the locations reviewed had some
identified needs that were not met.\7 These unmet needs included
pulmonary and speech therapy; psychotherapy; developmental
assessments; infant stimulation services; cardiological, urological,
and neurological examinations; and testing for sickle cell anemia,
syphilis, and HIV.  (See fig.  2 and table II.4 in app.  II.)

   Figure 2:  Extent to Which
   Identified Needs Were Met in
   Three Counties

   (See figure in printed
   edition.)

Note:  Point estimates do not total 100 percent because of rounding
and records lacking data on identified needs. 

\a The point estimates for the three locations varied widely in these
two categories. 

Source:  Case file review. 

Of those children with no identified health-related needs, about
one-half in each location received no routine health care, and less
than one-half received a toxicology screen at birth to detect recent
prenatal drug exposure.  Thus, many of these children may have had
health-related needs that were not identified and, consequently, were
not met. 


--------------------
\7 The number of identified needs per child ranged from 1 to 14 and
averaged approximately 5. 


   HIV-INFECTED CHILDREN NEED TO
   BE IDENTIFIED EARLY, BUT RISK
   ASSESSMENTS OFTEN DID NOT TAKE
   PLACE
------------------------------------------------------------ Letter :4

One particularly critical health need of young foster children is HIV
risk assessment\8 because most young children in the locations
reviewed are at risk for the infection as a result of parental drug
abuse.  Without risk assessment, a child's HIV status may not be
determined early because HIV-infected children can remain
asymptomatic or exhibit only minimal signs of infection for years. 
Recent medical advances in early identification and treatment can
enhance and prolong the lives of these children.  Early
identification is also critical because HIV-infected children should
receive modified immunizations to prevent adverse reactions, and
their exposure to infectious illnesses such as measles or chicken
pox--which are particularly hazardous to these children--should be
minimized. 

While state laws, and the county policies based on them, do not
prohibit HIV testing or the disclosure of test results, some can
hamper HIV testing and disclosure.  State laws and county foster care
policies, where they exist, vary widely.  In some locations,
including the three reviewed, these laws and policies impede HIV
testing and disclosure by specifying the risk factors that must be
present in order to request HIV testing; who has the authority to
consent to testing; and to whom HIV test results can be disclosed. 
For example, for the 36 states with HIV testing policies as of 1992,
one-half of the foster care agencies in the states with testing
policies may not have authority to consent to an HIV test for a
child, even when the child was identified by the agency as being at
high risk for HIV. 


--------------------
\8 An HIV risk assessment compares a child's family history and
health condition against the foster care agency's specified HIV risk
factors to make an informed determination about whether a child
should be tested for HIV. 


      THE NUMBER OF HIV-INFECTED
      CHILDREN IS LARGELY UNKNOWN
      AS FEW CHILDREN ARE TESTED
---------------------------------------------------------- Letter :4.1

Foster care agencies in the locations reviewed do not know the full
extent of their caseloads that is at high risk for HIV since no
mechanism exists to ensure that all young foster children are
assessed for HIV risk.  While two of the three locations we reviewed
currently have some HIV risk assessment requirements, one location
did not require risk assessments for all foster children until
recently, and the other has not implemented clear assessment
procedures.  HIV risk has long been associated with intravenous drug
use, but more recent research has established an equally strong link
between HIV risk and the lifestyle of nonintravenous cocaine and
crack users.  Using New York City's current HIV risk factors, which
include nonintravenous drug use, we assessed our 1991 population of
young foster children on the basis of one risk factor, parental drug
abuse.  Accordingly, we estimated that at least 78 percent of the
children in the three locations reviewed were at high risk for HIV. 

We estimated that only 9 percent of the young foster children in the
locations reviewed were actually tested for HIV, despite the large
proportion at high risk and statistics indicating that these are
locations with a high incidence of HIV.  The American Academy of
Pediatrics Task Force on Pediatric AIDS recommends HIV testing for
all foster children with high-risk factors or in areas with a high
incidence of HIV to facilitate appropriate medical treatment and
follow-up.  We estimated the HIV infection rate for children born in
1993 and found that the three states reviewed ranked 2nd, 14th, and
26th, based on national data from the Centers for Disease Control and
Prevention (CDC) on blind HIV testing of newborns.  (See table II.5
in app.  II.)

Few data are available on the number of foster children infected with
HIV.  One study reported that the number of foster children in New
York City known to be born to HIV-infected mothers increased 26
percent from 1991 to 1993.  While data were not available for
California and Pennsylvania, 1988 research found that Los Angeles had
the fastest growing rate of AIDS cases in the nation.  Furthermore,
anecdotal evidence suggests that in the Los Angeles area, and the
west coast in general, most AIDS in women is due to heterosexual
contact.  However, according to experts on pediatric AIDS, foster
care agencies do not commonly recognize a history of high-risk
heterosexual contact as sufficient grounds for HIV testing. 


   CHILDREN PLACED WITH RELATIVES
   IN LOS ANGELES COUNTY AND NEW
   YORK CITY WERE LESS LIKELY TO
   RECEIVE HEALTH-RELATED SERVICES
------------------------------------------------------------ Letter :5

Young children placed exclusively with relatives--known as kinship
care--were less likely to receive health-related services than
children placed exclusively with nonrelatives--known as traditional
foster care.\9 Specifically, children placed in kinship care were
nearly three times as likely as those placed in traditional foster
care to have received no routine health care.  Moreover, these
children were less likely to receive health-related services of all
kinds.  Since studies indicate that children in kinship care remain
in foster care longer, and they receive a lower level of service, the
likelihood is greater that these children will go without needed
services for longer periods.  (See fig.  3 and table II.6 in app. 
II.)

   Figure 3:  Services Received by
   Placement Type in Los Angeles
   County and New York City

   (See figure in printed
   edition.)

\a Consists of varying amounts of medical practitioner care and/or
EPSDT examinations. 

\b Includes children who received at least one immunization and
excludes children who were under 90 days of age. 

\c Differences are statistically significant at the 90-percent
confidence level. 

\d Includes treatment for asthma, syphilis, seizures, and kidney
problems. 

\e Includes blood, laboratory, and radiology. 

\f Includes developmental, psychological, and cardiological. 

\g Includes care for HIV, pneumonia, and failure to thrive, as well
as surgery. 

\h Includes apnea monitors, infant stimulation, and speech therapy. 

\i Includes therapeutic day care and Head Start services. 

\j Consists of children placed exclusively in traditional foster
care. 

\k Consists of children placed exclusively in kinship care. 

Young children placed in kinship care in the two locations reviewed
were also an estimated three times more likely than those placed in
traditional foster care to be at risk for future problems because of
prenatal drug exposure.\10 Furthermore, because drug-exposed children
are more likely to be at risk for HIV and developmental delays, the
need for health-related services for children in kinship care is even
more critical.  Yet, only 11 percent of children placed exclusively
in kinship care received specialized examinations, such as
developmental evaluations; whereas, 42 percent of those placed
exclusively in traditional foster care received specialized
examinations.  (See table II.6 in app.  II.)

While we did not determine why children in kinship care received less
health-related care, or compare other aspects of care by placement
type, we reviewed key studies on kinship care.  Research found that
foster care agencies treat kinship care placements and traditional
foster care placements differently.  Studies indicate that
caseworkers generally provide less monitoring and assistance to
kinship care placements.  Some states have policies requiring less
frequent caseworker visitations to kinship care homes, although these
homes are more likely to be unlicensed.  For example, a 1992 HHS
study found that in 30 states, children may be placed in kinship care
homes whether or not the homes meet minimum standards designed to
ensure the safety and suitability of foster homes and foster parents. 
Mandatory orientation and training for foster parents are the most
frequently waived licensing requirements for kinship caregivers.  A
1994 Child Welfare League of America report on kinship care found few
studies that focused on either kinship care providers or the children
in their care.  These studies, which were limited in scope, provided
little information regarding the advantages of different types of
placement. 


--------------------
\9 We used only Los Angeles County and New York City data in our
analysis of county case file data regarding kinship and traditional
foster care placements.  Because the sample for Philadelphia County
contained only one child who was placed exclusively in kinship care,
we eliminated that location from this analysis. 

\10 This analysis excluded children who were at risk for serious
health problems and also had serious health conditions.  When these
children were included in the analysis, young children placed
exclusively in kinship care were about as likely as those placed
exclusively in traditional foster care to be at risk for future
problems as a result of prenatal drug exposure. 


      KINSHIP CARE INCREASED
      DRAMATICALLY
---------------------------------------------------------- Letter :5.1

Analysis of the California and New York state databases showed the
number of children of all ages in kinship care increased by over 350
percent between 1986 and 1991, and this percentage increase was even
higher for young foster children, at 379 percent.\11 (See fig.  4 and
table II.7 in app.  II.)

   Figure 4:  Increase in Kinship
   and Traditional Foster Care in
   California and New York Between
   1986 and 1991

   (See figure in printed
   edition.)

Note:  Counts represent all children in foster care at the end of the
calendar years. 

\a Part of the increase in kinship care placements is due to a
lawsuit filed in the New York Supreme Court, Eugene F.  v.  Gross,
which sought to require New York City to follow regulations to
formally include children who are placed with relatives in the foster
care caseload and make them eligible for services. 

Source:  State electronic databases. 

The dramatic increase of children in kinship care between 1986 and
1991 resulted in nearly equal numbers of placements in kinship and
traditional foster care in the three counties reviewed.  We estimated
that 49 percent of the young children had been placed in kinship care
at some time during the 1991 review period, while 53 percent had been
placed in traditional foster care.  (See table II.8 in app.  II.)

Some studies contend that the increase in kinship placements may have
been due, at least initially, to a shortage of traditional foster
homes.  Other studies posit that this increase may be the result of
state and county interpretations of the Adoption Assistance and Child
Welfare Act of 1980 as implying a preference for relative placements. 
In recognizing that foster care would continue to be a necessary
child welfare service, this act required states to place children in
the "least restrictive (most family-like) setting available," which
has been interpreted by many states as implying a preference for
kinship care.  As of 1992, 44 states commonly placed foster children
in kinship care, and 29 states had policies in place requiring foster
care agencies to give preference to relatives of foster children. 


--------------------
\11 Pennsylvania does not have a statewide foster care database. 


   HEALTH-RELATED NEEDS OF FOSTER
   CHILDREN OFTEN GO UNMET DESPITE
   AGENCY EFFORTS
------------------------------------------------------------ Letter :6

The foster care agencies reviewed struggle to ensure that the
health-related needs of children in their care are met.  About
one-third of all states, including the three reviewed, have
established only broad guidelines within which counties administer
foster care programs.  Thus, counties in these states develop and
implement programs with considerable autonomy, which results in a
variety of approaches being used. 

County foster care agencies in the locations reviewed have altered
their health-related policies, regulations, and programs in efforts
to improve the agencies' ability to meet the health needs of foster
children.  For example, one of the foster care agencies we reviewed
continues to develop and implement recordkeeping systems in an
attempt to improve its ability to ensure that foster children receive
needed services.  The agency is currently implementing its third
variation of a medical recordkeeping system in recent years. 
Implementation of the first two versions was unsuccessful, and the
third was too recently implemented for us to determine its success. 
However, because the third version is substantially similar to its
predecessors, its likelihood of success is limited. 

Other efforts by this agency have focused on establishing medical
clinics for foster children.  It established a comprehensive
assessment center at the county-run children's emergency shelter, but
that effort appears to have met with only limited success.  Medical
staff at the center told us that it is seldom used by foster children
of any age who reside outside the shelter.  The foster care agency is
also supporting the development of an ambitious and complex system of
multidisciplinary assessment and medical clinics for foster children. 
This most recent effort, while promising, depends on factors largely
outside the control of the foster care agency, such as the continuing
involvement of the academia-based physicians who proposed the system
and the viability of a complex design for funding services. 
Furthermore, the system needs strong support from within the agency
and procedures that direct foster children to the new system
if--unlike the current assessment center--it is to reach even a
sustainable level of utilization. 

Recognizing that states need assistance in improving their child
welfare programs, including foster care, ACF recently increased its
technical assistance efforts.  Within the past year, HHS contracted
for 10 National Resource Centers to assist ACF staff in responding to
states' questions and to provide free technical assistance to states. 
Each resource center specializes in a child welfare issue such as
permanency planning, abandoned infants, or special needs adoption. 
However, none of the resource centers is designated to help states
with ensuring health-related services for foster children. 

ACF also audits states to certify states' compliance with the
safeguards for foster children specified in the Social Security Act. 
However, the audits do not examine compliance with all safeguards. 
The safeguards include a requirement that case files of foster
children contain up-to-date health-related information, such as
records of immunizations and a child's health conditions.  We
previously reported that HHS did not audit states on their compliance
with all required safeguards for foster children, and we recommended
that it expand its audits to include all safeguards.\12

In March 1995, ACF officials confirmed that their audits still do not
examine whether states are complying with the health-related
safeguards.  An HHS determination that a state has passed its
compliance audit entitles the state to receive the full federal child
welfare funding available by law.  However, since HHS does not audit
for compliance with the health-related safeguards, states have no
federal financial incentive to comply with them.  ACF plans to
include these safeguards in future audits, according to the same
officials. 


--------------------
\12 Foster Care:  Incomplete Implementation of the Reforms and
Unknown Effectiveness (GAO/PEMD-89-17, Aug.  14, 1989). 


   CONCLUSIONS
------------------------------------------------------------ Letter :7

Important health-related needs, including routine medical
examinations and various specialized services, remained unmet for
nearly one-third of the young foster children in the locations
reviewed.  Additionally, most young foster children in the locations
reviewed were at high risk for HIV as a result of parental drug
abuse, yet few children were actually tested for the infection. 
Furthermore, those in kinship care were less likely than those placed
in traditional foster care to receive needed health-related services. 

Despite federal safeguards for foster children, as well as
regulations of responsible agencies to ensure adequate health care
for foster children, agencies continue to struggle to meet the
complex health needs of young children.  Federal efforts to help
states design and implement effective foster care health programs
have been extremely limited, as evidenced by the lack of both ACF
audits and technical assistance to states on health-related issues. 
Our work confirms our earlier recommendation that ACF audits be
expanded to include all foster care safeguards.  We continue to
believe that ACF should take this action. 

Finally, foster care agencies have been slow to respond to one
critical health need--HIV risk assessment--which is the first step in
identifying HIV-infected children so that they can receive
appropriate and timely health care.  Yet, even if all foster children
were systematically assessed, HIV testing of high-risk children can
still be hampered by state laws and county policies.  Finally, while
we do not know why children in kinship care generally receive fewer
health-related services than children in traditional care, research
indicates that kinship caregivers receive less monitoring and
assistance from foster care agencies than traditional foster
caregivers. 

These findings are particularly disturbing given the vulnerable
nature of the population of young foster children.  Whether the
federal government retains the foster care program in its current
form or creates block grants to the states, these issues warrant
attention. 


   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :8

We provided HHS as well as the cognizant social services agencies of
the three states and locations reviewed with the opportunity to
comment on a draft of this report.  We received comments\13 from the
state of New York, New York City, and Los Angeles County. 
Philadelphia County responded that it could not comment on the
specifics of the report because of a pending lawsuit.  However, it
indicated a few general concerns.  We did not receive comments from
HHS, the state of California, or the state of Pennsylvania. 

One aspect of our report was commented on by three respondents.  New
York State, Los Angeles County, and Philadelphia County expressed
concern about the age of our data.  While our report is based on 1991
data, those were the most current data available when the study
began.  To ensure the continuing usefulness of the data and other
aspects of our study, we continued to monitor the locations reviewed
through spring 1995 to determine if any changes in policies and
programs had occurred that could substantially alter our conclusions. 
While some promising changes have occurred, either the locations that
provided comments to our draft provided us with no data to support
their assertions that the delivery of services has improved or it is
too early to determine the impact of the changes.  For example, New
York City commented that it is implementing a state early care
intervention program and has trained staff in the use of the program. 
However, it is too early to judge the impact of this new effort. 

Another aspect of our report was commented on by two respondents. 
New York State and Philadelphia County questioned the appropriateness
of combining the results of our analysis of cases across the three
locations reviewed.  As we stated in the report, we determined that
the conclusions drawn from our analysis were similar for each
location with two exceptions:  Philadelphia County was dropped from
analyses of kinship care, and data depicted in figure 2 included two
categories where the results varied widely by location.  With these
exceptions, the results were sufficiently consistent across all three
locations that we do not believe that presenting the aggregate
results unfairly portrays the situations in any of the locations. 


--------------------
\13 We received comments from the New York State Department of Social
Services, the New York City Human Resources Administration, the Los
Angeles County Department of Children and Family Services, and the
Philadelphia Department of Human Services. 


      THE STATE OF NEW YORK AND
      NEW YORK CITY COMMENTS
---------------------------------------------------------- Letter :8.1

The state of New York questioned the adequacy of the sample size.  We
arrived at our sample size using accepted statistical procedures that
gave us an adequate level of precision at the 95-percent confidence
level to support our findings.  Our detailed methodology is presented
in appendix I and the confidence intervals are presented in appendix
II. 

The state also expressed doubts about the accuracy of several of our
statistical findings, conveying its belief that an ongoing state
study regarding foster care medical services will produce different
results.  It believes that its ongoing study will produce a more
favorable picture of its ability to meet the needs of foster
children.  However, we cannot evaluate this opinion because the state
did not provide us with any results from this study.  Furthermore,
the state provided little information on the methodology being
employed, and we do not know whether the state plans to conduct such
analyses as would make it possible to compare their results with
ours. 

New York City raised different issues related to our methodology.  It
questioned whether inadequate caseworker recordkeeping provided an
incomplete depiction of the health-related services received by young
foster children.  Before beginning the case file review, to test the
feasibility of using this method, we reviewed the case files of a
small sample of children and then requested the foster care agencies
in the three locations reviewed to provide information on those same
children from all possible sources, including service providers and
foster parents.  In general, we found that the additional information
provided from these other sources did not change the conclusions we
had reached on the basis of our case file review regarding the level
of services these children received.  On this basis, we concluded
that the information in the case files would be sufficient for our
analytical purposes. 

New York State agrees with the importance of risk assessment for HIV
and agrees that it does not know the full extent of HIV-infected
children in foster care.  However, it disagrees that this occurs in
New York City because of a lack of a mechanism to carry out risk
assessments.  Furthermore, the state asserted that more children are
now being assessed and tested for HIV as a result of changes in its
policies.  We agree that this state has the most comprehensive
policies on risk assessment of the three locations reviewed.  For
this reason, we used a portion of their risk assessment policies as
criteria in one analysis.  However, the large gap we reported between
the number of children who were at risk for HIV, based on one New
York City risk criterion, and the number actually tested indicated
that the mechanisms in place did not ensure that their procedures
were consistently carried out.  While it is possible that recent
changes in New York State or New York City policies may have improved
the ability to identify HIV-infected children, state officials
pointed out that they have not been able to formally implement
regulations that would put their latest policy changes in place
because of a state moratorium on regulatory action. 

New York State agrees that children placed in kinship care were less
likely to receive services than children placed in traditional foster
care, and this finding was confirmed by the state's own study. 
However, it disagrees with the inclusion in the report of data on the
growth in kinship care because it asserts that such data do not
reflect the proportion of children actually in kinship care and
traditional foster care.  We agree that it is useful to understand
the proportion of children in different types of care, and this
information is included in the report.  However, we believe that
presenting data on the large growth in kinship care placements
between 1986 and 1991 is also useful to understand that the
utilization of kinship care has changed significantly since the
mid-1980s. 

New York State agrees with our conclusion that periodic reviews for
compliance with federal standards are appropriate.  It also made
technical comments on our characterizations of county versus state
regulations and of New York City as a county, and our description of
HIV testing policies.  On the basis of these comments, we modified
the report as appropriate. 


      LOS ANGELES COUNTY COMMENTS
---------------------------------------------------------- Letter :8.2

Los Angeles County commented that its current internal audits of
medical assessments show that compliance is at approximately 90
percent.  We agree with that estimate of the receipt of required
medical examinations, which we refer to as routine care.  As stated
in our report, about 12 percent of young foster children did not
receive any routine health care; Los Angeles County's current
estimate of 10-percent noncompliance with its regulations regarding
medical assessments falls within our 95-percent confidence interval
cited in appendix II. 

Los Angeles County also commented that it has made a number of
changes over the last few years that were designed to meet the health
care needs of foster children.  Specifically, it discussed the HIV
risk assessment policy, the comprehensive multidisciplinary
assessment center it established at a children's emergency shelter,
and the new system of multidisciplinary assessment and medical
clinics.  We acknowledge that Los Angeles County has a policy of
evaluation for risk of exposure to HIV as an ongoing process for all
foster children.  However, in September 1994, numerous county program
officials told us that the county has no procedures to systematically
ensure that risk assessments take place; consequently, this policy
does not ensure that foster children who are at high risk for HIV
will be identified and tested.  In addition, in fall 1994, we visited
the assessment center at the children's emergency shelter and
interviewed key program officials and medical staff.  We acknowledge
that this assessment center was designed to provide a variety of
comprehensive health-related evaluations.  However, as stated in our
report, this assessment center is little used by foster children who
reside outside the emergency shelter.  Finally, as we stated in our
report, we agree that the multidisciplinary medical clinics are a
promising approach to meeting the complex health-related needs of
young foster children.  However, it was not until September 1994 that
the first of the seven planned assessment center clinics was funded
to hire staff; thus, this system is in its infancy and is
substantially untested. 


---------------------------------------------------------- Letter :8.3

We will send copies of this letter to the Secretary of Health and
Human Services and program officials in the states reviewed.  We will
also send copies to all state welfare program directors and make
copies available to others on request.  Please contact me at (202)
512-7215 if you or your staff have any questions.  Other GAO contacts
and contributors are listed in appendix III. 

Sincerely yours,

Jane L.  Ross
Director, Income Security Issues


SCOPE AND METHODOLOGY
=========================================================== Appendix I

To accomplish the objectives of our review, we obtained and analyzed
data on state foster care programs and the children in them from the
three states with the largest average monthly foster care populations
in 1991--California, New York, and Pennsylvania.  Over 50 percent of
the nation's foster children are under the jurisdiction of these
three states. 

We used a variety of approaches to meet our objectives.  We analyzed
electronic state and county foster care databases; conducted a case
file review based on generalizable random samples; interviewed HHS,
state, and county foster care officials; conducted a telephone survey
of child welfare advocacy groups and other child welfare experts;
conducted group interviews with foster parents and caseworkers;
reviewed foster care and related literature; reviewed applicable
portions of the Social Security Act and other legislation; and
reviewed foster care agency regulations and other documents.  Studies
cited in this report are listed in the bibliography. 


   STATEWIDE DATA
--------------------------------------------------------- Appendix I:1

To determine the number of foster children in different types of
placements in the states, we analyzed electronic foster care
databases for the two states where they were available, California
and New York.  State officials provided us with automated records for
all children who were in foster care at any time during calendar
years 1986 and 1991.  We could not obtain comparable electronic
records for Pennsylvania as that state has not established an
automated case record system.\14


--------------------
\14 For the three states reviewed, the 110-percent increase in young
foster children that was previously reported combines data from
electronic databases and aggregate state data.  For California and
New York, states with electronic databases, the counts are for foster
children under the age of 3 years.  For Pennsylvania, the count is
for foster children under the age of 5 years, as that state's
aggregate data did not break out children under age 3 years. 


   COUNTY CASE FILE DATA
--------------------------------------------------------- Appendix I:2

To determine the health-related services needed and received by young
foster children, their health conditions, and the types of placements
they were in, we reviewed statistically representative samples of
foster care case files for the county with the largest foster care
population in 1991 for each of the states reviewed:  Los Angeles
County, New York City,\15 and Philadelphia County.  To identify those
locations, we again used the state foster care databases for
California and New York; for Pennsylvania, we relied on information
provided by state officials.  Philadelphia County officials provided
us with an electronic database of the records for foster children in
that county in 1991. 

Before drawing the sample, we narrowed the databases to include only
foster children whose third birthday occurred no later than December
31, 1991.  This resulted in population sizes of 8,249 for Los Angeles
County, 13,171 for New York City, and 1,335 for Philadelphia County. 
Then we selected random samples from each of these locations
resulting in a total sample of 414 children.  The population sizes
and initial sample sizes are shown in table I.1. 



                         Table I.1
          
          Initial Population and Sample Sizes for
                  Children in Foster Care

                                                Percentage
                                                of initial
                             Initial     Final      sample
                Population    sample    sample        used
--------------  ----------  --------  --------  ----------
Los Angeles          8,249       137       114        83.2
 County
New York City       13,171       150       142        94.7
Philadelphia         1,335       127       104        81.9
 County
==========================================================
Total               22,755       414       360      89.8\a
----------------------------------------------------------
\a Percentage total is a weighted average showing the percentage of
the total population covered by the final samples. 

We requested all foster care case files for each child in the sample. 
A few cases were dropped from the sample because the children did not
meet the criterion of being in foster care during the review year or
were not of the appropriate age.  Other cases were dropped because
county officials could not locate the records.  Finally, we dropped
cases of children who were in foster care during our review period
for less than 30 days in Los Angeles County and New York City, and
less than 60 days in Philadelphia County.  We did this to eliminate
cases in which a child's tenure in foster care was shorter than the
time foster care agencies were allowed, by their local regulations,
to complete initial medical examinations.  This resulted in final
samples totaling 360 young foster children in our three locations. 

We examined the foster care case files for the period covering a
child's first entry into foster care until the end of the review year
or until the child was discharged from foster care, whichever
occurred earlier.  We used an automated data collection instrument to
record information from the case files.  The recorded information was
reviewed for accuracy by the individual preparing it before
finalizing each electronic record.  We also reviewed the case file
data for consistent coding among individuals; minor adjustments were
made to the coding as a result of that review. 

We analyzed the case file data using univariate and bivariate
analyses, descriptive statistical methods.  We found that for some of
the data, the results varied among the three locations; however, the
conclusions we drew from the analyses of each location were similar. 
Thus, the locations could be combined for analysis.  Finally, when
combining these data, we weighted them to adjust for disproportionate
sampling and produced aggregate estimates.  However, the results
pertain to only the three locations combined and do not necessarily
reflect populations of foster children at the state or national
level. 

For data derived from the case file review, the percentage estimates
reported in the letter and the numerical estimates reported in
appendix II are point estimates.  Because the estimates are based on
combined results from three samples, each is subject to sampling
error.  The size of the sampling error reflects the precision of the
estimate; the smaller the error, the more precise the estimate. 
Sampling errors for the estimates were calculated at the 95-percent
confidence level except where noted.  We are 95-percent confident
that the actual percentages fall within the confidence intervals
reported in appendix II.  In other words, there is a 5-percent chance
that the confidence intervals do not contain the actual population
percentages. 


--------------------
\15 New York City comprises five boroughs and is treated in the state
database as a county. 


      ANALYSIS OF PLACEMENT TYPE
      AND SERVICE DELIVERY
------------------------------------------------------- Appendix I:2.1

For the analysis comparing the subpopulations of children in kinship
and traditional foster care, we used only records of children who had
been placed exclusively in kinship care or exclusively in traditional
foster care.  Furthermore, because the sample for Philadelphia County
contained only one child who was placed exclusively in kinship care,
we eliminated that location from this analysis.  The subpopulation
sizes are shown in table I.2. 



                         Table I.2
          
                    Subpopulation Sizes

                                 Traditional       Kinship
                                 foster care          care
------------------------------  ------------  ------------
Los Angeles County                        20            33
New York City                             41            43
==========================================================
Total                                     61            76
----------------------------------------------------------
This is the second report responding to this request.  We conducted
our review for both reports between November 1992 and March 1995 in
accordance with generally accepted government auditing standards.  We
analyzed the electronic databases as provided to us by state and
county officials, and we performed limited tests of the completeness
of the case files. 


ANALYSIS RESULTS
========================================================== Appendix II

This appendix presents the numerical values for the data discussed in
the body of this report.  Where appropriate, point estimates and
confidence intervals are provided.  The appendix includes case file
review results for the review year 1991 and statewide data for
calendar years 1986 and 1991. 



                                        Table II.1
                         
                              Health Care in Three Counties


                                    Point       Point
                                estimate,   estimate,      Upper bound,      Lower bound,
Health care\a                      number     percent           percent           percent
-----------------------------  ----------  ----------  ----------------  ----------------
Services not received
-----------------------------------------------------------------------------------------
Routine care\b                      2,434        11.9              15.4               8.4
Immunizations\c                     6,885        34.3              42.7              28.0

Service received
-----------------------------------------------------------------------------------------
EPSDT                                 267         1.3               3.5               0.7
-----------------------------------------------------------------------------------------
\a We dropped cases for children who were in foster care during our
review period for less than 30 days in Los Angeles County and New
York City, and less than 60 days in Philadelphia County.  We did this
to eliminate cases in which a child's tenure in foster care was
shorter than the time foster care agencies are allowed, by their
local regulations, to complete initial medical examinations. 

\b Children are exempt from initial examination requirements if they
received an equivalent examination within 90 days before entering
foster care.  Of children who received no routine care during our
review period, three were required, because of their age at entry and
length of stay, to have an initial examination if they did not have
an examination 90 days before entering foster care.  We believe,
based on the case file data, that these children did not meet the
prior examination requirement. 

\c Includes children who received at least one immunization and
excludes children who were under 90 days of age. 

Source:  Case file review. 



                                        Table II.2
                         
                           Proportion of Children Receiving No
                           Health Services by Health Condition


                                    Point       Point
                                estimate,   estimate,      Upper bound,      Lower bound,
Health condition                   number     percent           percent           percent
-----------------------------  ----------  ----------  ----------------  ----------------
No known serious health             1,094        28.4              48.6               8.2
 problems\a
At risk for serious health            723         5.6              11.4               2.6
 problems\b
Serious health problems\c             241         2.1               8.0               0.6
-----------------------------------------------------------------------------------------
\a Includes children who had minor illnesses. 

\b Consists of prenatal drug exposure (including alcohol exposure)
and drug withdrawal or symptoms.  We considered a child to be
prenatally drug-exposed if any of the following conditions were
documented in the child's foster care records:  mother reported that
she used drugs during pregnancy, toxicology test results for mother
or infant were positive for drug use, or infant was diagnosed as
having drug-withdrawal symptoms. 

\c Consists of fetal alcohol syndrome, low birth weight, cardiac
defects or heart problems, HIV-positive or AIDS, developmentally
delayed, and other serious problems. 

Source:  Case file review. 



                                        Table II.3
                         
                          Specialized Services Received in Three
                                         Counties


                                    Point       Point
                                estimate,   estimate,      Upper bound,      Lower bound,
Services                           number     percent           percent           percent
-----------------------------  ----------  ----------  ----------------  ----------------
Medications\a                       9,607        47.0              52.9              41.2
Tests\b                             7,499        36.7              42.4              31.0
Specialized examinations\c          6,211        30.4              35.9              24.9
Hospitalizations\d                  5,025        24.6              29.8              19.4
Specialized treatments\e            4,619        22.6              27.5              17.7
Early intervention services\f         725         3.6               6.0               1.6
-----------------------------------------------------------------------------------------
\a Includes treatment for asthma, syphilis, seizures, and kidney
problems. 

\b Includes blood, laboratory, and radiology. 

\c Includes developmental, psychological, and cardiological. 

\d Includes care for HIV, pneumonia, and failure to thrive, as well
as surgery. 

\e Includes apnea monitors, infant stimulation, and speech therapy. 

\f Includes therapeutic day care and Head Start services. 

Source:  Case file review. 



                                        Table II.4
                         
                          Extent to Which Identified Needs Were
                                  Met in Three Counties


                                    Point       Point
                                estimate,   estimate,      Upper bound,      Lower bound,
Needs met                          number   percent\a           percent           percent
-----------------------------  ----------  ----------  ----------------  ----------------
Children with at least some         6,591        32.1              37.7              26.5
 needs not met
Children with all needs met\b       9,763        47.6              53.4              41.8
Children with no needs              3,924        19.1              23.1              15.2
 identified\b
-----------------------------------------------------------------------------------------
\a Point estimates do not total 100 percent because of rounding and
records lacking data on identified needs.\

\b The point estimates for the three locations varied widely in these
two categories.  The range of point estimates was 60.6 to 24.6
percent for "all needs met" and 45.6 to 4.9 percent for "no needs
identified."

Source:  Case file review. 



                         Table II.5
          
           HIV-Infected Newborns for Three States
                          in 1993

                 Ranking of selected        Estimated HIV-
                 states by estimated     infected newborns
               rates of HIV-infected        per 1,000 live
States                      newborns             births\a\
----------  ------------------------  --------------------
California                        26                  0.14
New York                           2                  1.43
Pennsylvan                        14                  0.40
 ia
----------------------------------------------------------
Note:  We calculated these rates from data supplied by CDC from its
study of blind HIV testing of newborns.  Forty-four states and the
District of Columbia participated in this study.  The six states that
did not participate were Idaho, Indiana, Nebraska, North Dakota,
South Dakota, and Vermont.  While the District of Columbia was the
only nonstate location included in this comparison of estimated
HIV-infected newborns, it ranked first. 

\a Reflects the estimates of children who are HIV-infected, excluding
those who falsely test positive at birth.  CDC's estimates of the
number of HIV-infected newborns are based on a 25-percent
transmission rate; in other words, one-quarter of the number of
newborns who test HIV-positive at birth are estimated to be
HIV-infected rather than merely carrying their HIV-infected mothers'
antibodies. 

Source:  GAO analysis. 



                                    Table II.6
                     
                      Services Received by Placement Type in
                       Los Angeles County and New York City


                                                          Differ
                                                            ence
                                                          betwee
                                                               n
                                                           point
                                   Point           Point  estima
                           Point  estima   Point  estima    tes,   Upper   Lower
                          estima     te,  estima     te,  percen  bound,  bound,
                             te,  percen     te,  percen    tage  percen  percen
Services                  number       t  number       t  points       t       t
------------------------  ------  ------  ------  ------  ------  ------  ------
Routine care\c             4,476    93.2   4,689    81.4    11.8    22.1     1.4
Immunizations\d            3,352    72.5   3,209    57.4  15.1\e    31.2    -1.1
Medications\f              2,963    61.7   1,826    31.7    30.0    46.1    13.9
Tests\g                    2,496    52.0     850    14.8    37.2    52.5    22.0
Specialized                2,002    41.7     652    11.3    30.4    44.8    15.9
 examinations\h
Hospitalizations\i         1,558    32.4   1,124    19.5  12.9\e    27.9    -2.1
Specialized treatments\j   1,300    27.0     888    15.4  11.6\e    25.6    -2.3
Early intervention           411     8.6       0     0.0     8.6    15.8     1.3
 services\k
No services received         329     6.8     925    16.0       -     0.7   -19.1
                                                           9.2\e
--------------------------------------------------------------------------------
Note:  Because the sample for Philadelphia County contained only one
child who was placed exclusively in kinship care, we eliminated that
location from this analysis. 

\a Consists of children placed exclusively in traditional foster
care. 

\b Consists of children placed exclusively in kinship care. 

\c Consists of varying amounts of medical practitioner care and/or
EPSDT examinations. 

\d Includes children who received at least one immunization and
excludes children who were under 90 days of age. 

\e Differences are statistically significant at the 90-percent
confidence level. 

\f Includes treatments for asthma, syphilis, seizures, and kidney
problems. 

\g Includes blood, laboratory, and radiology. 

\h Includes developmental, psychological, and cardiological. 

\i Includes care for HIV, pneumonia, and failure to thrive, as well
as surgery. 

\j Includes apnea monitors, infant stimulation, and speech therapy. 

\k Includes therapeutic day care and Head Start services. 

Source:  Case file review. 



                         Table II.7
          
            Increase in Kinship and Traditional
           Foster Care in California and New York
                   Between 1986 and 1991

                                                Percentage
Placements                1986        1991        increase
------------------  ----------  ----------  --------------
All children
----------------------------------------------------------
Kinship care\a          15,241      69,590           356.6
Traditional foster      64,225      80,443            25.3
 care

Young children
----------------------------------------------------------
Kinship care\a           2,941      14,072           378.5
Traditional foster      12,007      18,457            53.7
 care
----------------------------------------------------------
Note:  Counts represent children in foster care at the end of the
calendar years. 

\a Part of the increase in kinship care placements is due to a
lawsuit filed in the New York Supreme Court, Eugene F.  v.  Gross,
which sought to require New York City to follow regulations to
formally include children who are placed with relatives in the foster
care caseload and make them eligible for services. 

Source:  State electronic databases. 



                                        Table II.8
                         
                         Distribution of Children in Kinship and
                             Traditional Foster Care in Three
                                         Counties


                                    Point       Point
                                estimate,   estimate,      Upper bound,      Lower bound,
Placement type                     number     percent           percent           percent
-----------------------------  ----------  ----------  ----------------  ----------------
Kinship care                        9,976        48.8              55.0              42.7
Traditional foster care            10,773        52.7              58.9              46.6
-----------------------------------------------------------------------------------------
Note:  Point estimates represent the proportion of children who were
in each type of placement at any time during the review period. 
Since some children were in both types of placements during the
review period, point estimates total more than 100 percent. 

Source:  Case file review. 


GAO CONTACTS AND ACKNOWLEDGMENTS
========================================================= Appendix III

GAO CONTACTS

Kerry Gail Dunn, Evaluator-in-Charge, (415) 904-2000
Robert L.  MacLafferty, Assistant Director, (415) 904-2000

ACKNOWLEDGMENTS

In addition to those named above, the following individuals made
important contributions to this report:  Susan Riggio led the
fieldwork in California and Pennsylvania and coauthored the draft;
Ann Walker led the data analysis and coauthored the draft; Helen
Cregger, Sheila Murray, Tranchau Nguyen, and Cameo Zola conducted
case file reviews and interviews. 


BIBLIOGRAPHY
============================================================ Chapter 0

American Academy of Pediatrics, Committee on Early Childhood,
Adoption, and Dependent Care.  "Health Care of Children in Foster
Care." Pediatrics, Vol.  93, No.  2 (Feb.  1994), pp.  335-38. 

American Academy of Pediatrics, Committee on Practice and Ambulatory
Medicine.  "Recommendations For Preventive Pediatric Health Care."
July 1991. 

American Academy of Pediatrics, Task Force on Pediatric AIDS. 
"Guidelines for Human Immunodeficiency Virus (HIV)-Infected Children
and Their Foster Families." Pediatrics, Vol.  89, No.  4 (Apr. 
1992),
pp.  681-83. 

Besharov, Douglas J., ed.  When Drug Addicts Have
Children--Reorienting Child Welfare's Response.  Washington, D.C.: 
Child Welfare League of America, Inc., and American Enterprise
Institute, 1994. 

Center for the Future of Children and The David and Lucile Packard
Foundation.  "Drug Exposed Infants." The Future of Children, Vol.  1,
No.  1 (Spring 1991). 

Child Welfare & AIDS Project.  Child Welfare Protocols for Children
with HIV Infection:  Guidelines for Development and Evaluation. 
Berkeley, Calif.:  School of Social Welfare, University of
California, Berkeley, Jan.  1991. 

Child Welfare League of America.  Kinship Care:  A Natural Bridge. 
Washington, D.C.:  1994. 

Child Welfare League of America.  Standards for Health Care Services
for Children in Out-of-Home Care.  Washington, D.C.:  1988. 

Cohen, Felissa, R.N., Ph.D., F.A.A.N., and Wendy Nehring, R.N., Ph.D. 
"Foster Care of HIV-Positive Children in the United States." Public
Health Reports, Vol.  109, No.  1 (Jan./Feb.  1994), pp.  60-67. 

Congressional Research Service.  "Kinship" Foster Care:  An Emerging
Federal Issue.  Washington, D.C.:  Sept.  27, 1993. 

Deering's California Codes.  "Annotated Health and Safety Code of the
State of California," pp.  106-107.  San Francisco:  1990. 

English, Abigail.  "The HIV-AIDS Epidemic and the Child Welfare
System:  Protecting the Rights of Infants, Young Children, and
Adolescents." Iowa Law Review, Vol.  77, No.  4 (1992), pp.  1509-60. 

National Abandoned Infants Assistance Resource Center.  Report to the
Congress on Effective Care Methods for Responding to the Needs of
Abandoned Infants and Young Children.  Berkeley, Calif.:  Family
Welfare Research Group, School of Social Welfare, University of
California, Berkeley, undated report. 

Pinkney, Deborah.  "America's Sickest Children." American Medical
News, Jan.  10, 1994, pp.  15-19. 

Sterk, Claire.  "Cocaine and HIV Seropositivity." Lancet, Vol.  I,
No.  8593 (May 7, 1988), pp.  1052-53. 

Teare, Catherine, Abigail English, J.D., Susan Lockwood, M.S.W., and
Kirby Clark, J.D.  HIV/AIDS Policies and Guidelines:  State Child
Welfare Agencies.  Iowa City, Iowa:  University of Iowa National
Maternal and Child Health Resource Center, 1992. 

U.S.  Department of Health and Human Services.  Maternal Drug Abuse
and Drug Exposed Children:  Understanding the Problem.  Washington,
D.C.:  1992. 

U.S.  Department of Health and Human Services, Office of Inspector
General.  Using Relatives For Foster Care.  Washington, D.C.:  July
1992. 

U.S.  Department of Health and Human Services, Public Health Service,
Centers for Disease Control and Prevention.  HIV/AIDS Surveillance
Report.  Vol.  6, No.  1 (1994). 

U.S.  Department of Health and Human Services, Public Health Service,
Centers for Disease Control and Prevention, National Center for
Infectious Diseases.  Unpublished data on blind HIV testing of
newborns in 44 states and the District of Columbia.  Atlanta:  1993. 



RELATED GAO PRODUCTS
============================================================ Chapter 1

Foster Care:  Parental Drug Abuse Has Alarming Impact on Young
Children (GAO/HEHS-94-89, Apr.  4, 1994). 

Childhood Immunization:  Opportunities to Improve Immunization Rates
at Lower Cost (GAO/HRD-93-41, Mar.  24, 1993). 

Drug Abuse:  The Crack Cocaine Epidemic:  Health Consequences and
Treatment (GAO/HRD-91-55FS, Jan.  30, 1991). 

Drug-Exposed Infants:  A Generation at Risk (GAO/HRD-90-138, June 28,
1990). 

Foster Care:  Incomplete Implementation of Reforms and Unknown
Effectiveness (GAO/PEMD-89-17, Aug.  14, 1989). 

Pediatric AIDS:  Health and Social Service Needs of Infants and
Children (GAO/HRD-89-96, May 5, 1989). 

