Medicaid: Demographics of Nonenrolled Children Suggest State Outreach
Strategies (Letter Report, 03/20/98, GAO/HEHS-98-93).

Pursuant to a congressional request, GAO reported on children who are
eligible for Medicaid but are not enrolled, focusing on: (1) the
demographic and socioeconomic characteristics of children who qualify
for Medicaid, and identifying groups in which uninsured children are
concentrated and to whom outreach efforts might be expected; (2) the
reasons these children are not enrolled in Medicaid; and (3) strategies
that states and communities are using to increase employment.

GAO noted that: (1) the demographic and socioeconomic characteristics of
uninsured Medicaid-eligible children suggest that outreach strategies
could be targeted to specific groups; (2) in 1996, 3.4 million
Medicaid-eligible children--23 percent of those eligible under the
federal mandate--were uninsured; (3) the majority were children of
working poor or near poor, and their parents were often employed by
small firms and were themselves uninsured; (4) uninsured children who
are eligible for Medicaid are more likely to be in working families,
Hispanic, and either U.S.-born to foreign-born parents or foreign born;
(5) state officials, beneficiary advocates, and health care providers
whom GAO contacted cited several reasons that families do not enroll
their children in Medicaid; (6) lower income working families may not
realize that their children qualify for Medicaid, or they may think
their children do not need coverage if they are not currently sick; (7)
under welfare reform, the delinking of Medicaid and cash assistance may
cause some confusion for families, although GAO found that states were
making efforts to retain a single application and eligibility
determination process to avoid this problem; (8) in addition, many
low-income families believe that Medicaid carries the same negative
image of dependency that they attach to welfare; (9) immigrant families,
many of whom are Hispanic, face additional barriers, including language
and cultural separateness, fear of dealing with the government, and
changing eligibility rules; (10) finally, the enrollment process for
Medicaid can involve long forms and extensive documentation, which are
intended to ensure program integrity but often are a major deterrent to
enrollment; (11) recognizing these impediments, some states have
undertaken education and outreach initiatives and have tried to change
the image of the program and simplify enrollment to acquire only
necessary information; (12) these efforts include mass media campaigns
and coordination of effort with community organizations and provider
groups; (13) some states have made the enrollment process more
accessible for working families, using mail-in applications or
enrollment at sites chosen for their convenience; (14) several states
have changed the name of the program to minimize its identification with
welfare and other assistance programs; (15) many states provide
Spanish-language applications and some are working with community
groups; and (16) some states have also simplified the enrollment
procedure by shortening the enrollment form and reducing the
documentation requirements.

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

 REPORTNUM:  HEHS-98-93
     TITLE:  Medicaid: Demographics of Nonenrolled Children Suggest 
             State Outreach Strategies
      DATE:  03/20/98
   SUBJECT:  Health insurance
             Disadvantaged persons
             Children
             State/local relations
             Health care programs
             Eligibility criteria
             Population statistics
             State-administered programs
             Health care services
             Immigrants
IDENTIFIER:  AFDC
             Aid to Families with Dependent Children Program
             Medicaid Program
             Arkansas
             Massachusetts
             Georgia
             HHS Temporary Assistance for Needy Families Program
             Supplemental Security Income Program
             Children's Health Insurance Program
             CHAMPUS
             WIC
             Wisconsin
             Special Supplemental Food Program for Women, Infants, and 
             Children
             
******************************************************************
** 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.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


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


Report to the Honorable
John McCain, U.S.  Senate

March 1998

MEDICAID - DEMOGRAPHICS OF
NONENROLLED CHILDREN SUGGEST STATE
OUTREACH STRATEGIES

GAO/HEHS-98-93

Outreach to Medicaid-Eligible Children

(101574)


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

  AFDC - Aid to Families With Dependent Children
  CHAMPUS - Civilian Health and Medical Program of the Uniformed
  Services
  CHIP - Children's Health Insurance Program
  CPS - Current Population Survey
  CTS - Community Tracking Study
  HCFA - Health Care Financing Administration
  INS - Immigration and Naturalization Service
  PRUCOL - persons residing under color of law
  SIPP - Survey of Income and Program Participation
  SSI - Supplemental Security Income
  TANF - Temporary Assistance for Needy Families
  WIC - Special Supplemental Food Program for Women, Infants, and
  Children

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


B-278558

March 20, 1998

The Honorable John McCain
United States Senate

Dear Senator McCain: 

Health insurance has become increasingly important as a means for
providing children access to adequate health care, yet in 1996, 10.6
million children were uninsured.  The Congress recently demonstrated
its interest in insuring more children by committing more than $20
billion over a 5-year period to fund state expansions of children's
health insurance--either through the Medicaid program or through
insurance programs developed by states.  However, as we have reported
in the past, many uninsured children who are eligible for Medicaid
are not enrolled.\1

Concerned about this failure to enroll children in Medicaid, you
asked us to

  -- examine the demographic and socioeconomic characteristics of
     children who qualify for Medicaid, and identify groups in which
     uninsured children are concentrated and to whom outreach efforts
     might be targeted;

  -- determine the reasons these children are not enrolled in
     Medicaid; and

  -- identify strategies that states and communities are using to
     increase enrollment. 

To examine the characteristics of families of Medicaid-eligible
uninsured children, we analyzed data from the Bureau of the Census'
March 1997 Current Population Survey (CPS).  To develop a demographic
profile of uninsured Medicaid-eligible children, we linked
information on children's health insurance status with their
families' characteristics and compared them with children similar in
age and family income who are enrolled in Medicaid or who have
employment-based insurance.  In this analysis, we only considered
children who met the federal age and income standards for mandated
eligibility.  We did not include children who could be eligible
because of individual state expansions or children over age 12 who
could be eligible if their family income was low enough to receive
cash assistance in their state. 

To determine the reasons eligible children are not enrolled in
Medicaid and identify strategies that could assist in getting such
children enrolled, we interviewed state officials, beneficiary
advocates, and provider representatives in Arkansas and
Massachusetts; state officials in Georgia; and other experts.  We
also reviewed the outreach materials used by these states and by
Wisconsin.  Arkansas, Georgia, and Massachusetts were selected from
among several states that were identified by experts as having
innovative outreach programs.  Additionally, Wisconsin was identified
as being further along in its approach to welfare reform and Medicaid
outreach and eligibility determination.  State outreach programs are
designed to reach all Medicaid eligibles, and the programs we
reviewed target all eligible children, both mandated and optional. 
For more information on our scope and methodology, see appendix I. 
Our work was conducted between July 1997 and February 1998, in
accordance with generally accepted government auditing standards. 


--------------------
\1 Health Insurance for Children:  Private Insurance Coverage
Continues to Deteriorate (GAO/HEHS-96-129, June 17, 1996); Health
Insurance for Children:  Many Remain Uninsured Despite Medicaid
Expansion (GAO/HEHS-95-175, July 19, 1995). 


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

The demographic and socioeconomic characteristics of uninsured
Medicaid-eligible children suggest that outreach strategies could be
targeted to specific groups.  In 1996, 3.4 million Medicaid-eligible
children--23 percent of those eligible under the federal
mandate--were uninsured.  The majority were children of working poor
or near poor, and their parents were often employed by small firms
and were themselves uninsured.  Uninsured children who are eligible
for Medicaid are more likely to be in working families, Hispanic, and
either U.S.-born to foreign-born parents or foreign-born.  This
suggests that state outreach may effectively target working families
and qualified immigrants.  States in the West have higher numbers and
percentages of Hispanics and immigrants among their Medicaid-eligible
uninsured children, suggesting that these groups could also be
targeted.  Because nearly three-quarters of uninsured
Medicaid-eligible children live in the West and the South, states in
these regions may already face greater challenges in enrolling
eligible children than other states. 

State officials, beneficiary advocates, and health care providers
whom we contacted cited several reasons that families do not enroll
their children in Medicaid.  Lower income working families may not
realize that their children qualify for Medicaid, or they may think
their children do not need coverage if they are not currently sick. 
Under welfare reform, the delinking of Medicaid and cash assistance
may cause some confusion for families, although in a recent report,
we found that states were making efforts to retain a single
application and eligibility determination process to avoid this
problem.  In addition, many low-income families believe that Medicaid
carries the same negative image of dependency and inability to
provide for their family that they attach to welfare.  Immigrant
families, many of whom are Hispanic, face additional barriers,
including language and cultural separateness, fear of dealing with
the government, and changing eligibility rules.  Finally, the
enrollment process for Medicaid can involve long forms and extensive
documentation, which are intended to ensure program integrity but
often are a major deterrent to enrollment. 

Recognizing these impediments, some states have undertaken education
and outreach initiatives and have tried to change the image of the
program and simplify enrollment to acquire only information deemed
essential.  For example, Arkansas and Massachusetts recently
initiated major outreach efforts to publicize their expansions of
coverage for children and to reach families who are unaware of their
children's eligibility.  These efforts include mass media campaigns
and coordination of effort with community organizations and provider
groups.  Part of their campaign is to inform the public that Medicaid
is available for children in working families.  Some states have made
their application and enrollment process more accessible for working
families, using mail-in applications or enrollment at sites chosen
for their convenience.  In Georgia, for example, the state has over
140 outreach workers located at nontraditional sites, such as
supermarkets, who often work evening and weekend hours.  To minimize
its identification with welfare and other assistance programs,
several states have changed the name of their Medicaid program.  To
facilitate enrollment of Hispanics, many states provide
Spanish-language applications and some are working with community
groups.  Some states have also simplified enrollment procedures by
shortening the enrollment form and reducing the documentation
requirements.  Of states whose outreach programs we reviewed,
officials and advocates emphasized the importance of consensus and
support from a broad spectrum of political and community leaders in
gaining visibility and resources for outreach. 


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

Studies have shown that insured children are more likely than
uninsured children to get preventive and primary health care.\2

Insured children are also more likely to have a relationship with a
primary care physician and to receive required preventive services,
such as well-child checkups.  In contrast, lack of insurance can
inhibit parents from trying to get health care for their children and
can lead providers to offer less intensive services when families
seek care.  Several studies have found that low-income and uninsured
children are more likely to be hospitalized for conditions that could
have been managed with appropriate outpatient care.\3

Most insured U.S.  children under age 18 have health coverage through
their parents' employment--62 percent in 1996.  Most other children
with insurance have publicly funded coverage, usually the Medicaid
program.  Medicaid--a jointly funded federal-state entitlement
program that provides health coverage for both children and
adults--is administered through 56 separate programs, including the
50 states, the District of Columbia, Puerto Rico, and the U.S. 
territories.  Historically, children and their parents were
automatically covered if they received benefits under the Aid to
Families With Dependent Children (AFDC) program.  Children and adults
may also be eligible for Medicaid if they are disabled and have low
incomes or, at state discretion, if their medical expenses are
extremely high relative to family income. 

Before 1989, coverage expansions for pregnant women and children
based on family income and age were optional for states, although
many states had expanded coverage.  Starting in July 1989, states
were required to cover pregnant women and infants (defined as
children under 1 year of age) with family incomes at or below 75
percent of the federal poverty level.  Two subsequent federal laws
further expanded mandated eligibility for children.  By July 1991,
states were required to cover (1) infants and children up to 6 years
old with family income at or below 133 percent of the federal poverty
level and (2) children 6 years old and older born after September 30,
1983, with family income at or below 100 percent of the federal
poverty level.\4 Since 1989, states have also had the option of
covering infants with family income between 133 percent and 185
percent of the poverty level.  States may expand Medicaid eligibility
for children by phasing in coverage of children up to 19 years old
more quickly than required, by increasing eligibility income levels,
or both.  The demographic analysis in this report, however, focuses
on the group of children for whom coverage is mandated. 

The Personal Responsibility and Work Opportunity Reconciliation Act
of 1996 (P.L 104-193), also known as the Welfare Reform Act,
substantially altered AFDC and Supplemental Security Income (SSI) but
made relatively few changes to the Medicaid program itself.  The law
replaced AFDC with a block grant that allowed states to set different
income and resource (asset) eligibility standards for the new
program--Temporary Assistance for Needy Families (TANF)--than for
Medicaid.  To ensure continued health coverage for low-income
families, the law generally set Medicaid's eligibility standards at
AFDC levels in effect July 16, 1996, thereby ensuring that families
who were eligible for Medicaid before welfare reform continued to
qualify, regardless of their eligibility for states' cash assistance
programs.\5 The law tightened the criteria for children to qualify
for disability assistance through SSI, thus tightening eligibility
for Medicaid.\6 In addition, the law restricted aliens' access to
benefit programs, including SSI, and Medicaid benefits that were
conditional on receipt of SSI.  State and local governments were
given some flexibility in designing policies that governed aliens'
eligibility for TANF, Medicaid, and social services.  In a recently
released report, we studied the Welfare Reform Act and its impact on
Medicaid and found that in the states we visited, most chose to
continue to provide Medicaid coverage to previously covered groups.\7

The Balanced Budget Act of 1997 (P.L.  105-33) restored SSI
eligibility and the derivative Medicaid benefits to all aliens
receiving SSI at the time welfare reform was enacted and to all
aliens legally residing in the United States on the date of enactment
who become disabled in the future.  At the same time, states
continued to have flexibility in implementing certain benefits
policies for aliens.  Current law allows states the option of
providing Medicaid coverage to aliens who were legal permanent
residents in the country before August 23, 1996.  States also have
the option of covering legal residents who arrived after August 22,
1996, once they have resided in the United States for 5 years. 
Illegal aliens are eligible only for emergency services under
Medicaid.  (See table 1.)



                          Table 1
          
            Summary of Medicaid Eligibility for
                           Aliens

Population          Policy
------------------  --------------------------------------
Qualified aliens arriving before August 23, 1996
----------------------------------------------------------
Legal permanent     State option
residents

Asylees, refugees   Eligible for first 7 years of
                    residency; state option afterward


Qualified aliens arriving after August 22, 1996
----------------------------------------------------------
Legal permanent     Barred for first 5 years of residency;
residents           state option afterward

Asylees, refugees   Eligible for first 7 years of
                    residency; state option afterward


Unqualified aliens
----------------------------------------------------------
Illegal aliens      Emergency services only

PRUCOL aliens\a     Emergency services only
----------------------------------------------------------
\a PRUCOL (persons residing under color of law) is an umbrella term
used for aliens who are legally residing in the United States but who
do not fit in other alien categories. 

The Balanced Budget Act also made two changes that directly affect
children's coverage in the Medicaid program.  It gives states the
option of providing 12 months of continuous eligibility to children
without a redetermination of eligibility, thereby avoiding the
problem of children frequently moving on and off Medicaid as their
parents' circumstances change.  The act also allows states to extend
Medicaid coverage to children on the basis of "presumptive
eligibility" until a formal determination is made.  Under this
provision, certain qualified providers can make an initial
determination of eligibility, based on income, that an individual is
eligible.  The individual is then required to apply formally for the
program by the last day of the month following the month in which the
determination of presumptive eligibility was made. 

Finally, the Balanced Budget Act created the Children's Health
Insurance Program (CHIP), a grant program for uninsured children,
through which $20.3 billion in new federal funds will be made
available to states over the next 5 years.  CHIP has a number of
implications for Medicaid.  If a state chooses to offer coverage
through a separate program, the state must coordinate activities with
the Medicaid program to ensure that Medicaid-eligible children are
enrolled in Medicaid.\8 The Congressional Budget Office estimated
that the "outreach effect" of CHIP will result in an additional $2.4
billion in Medicaid spending over the same 5 years due to increased
enrollment of 460,000 Medicaid-eligible children each year.  States
may also use the grant funds to expand coverage under their state
Medicaid programs to reach additional low-income children, increasing
the number of children potentially eligible for Medicaid. 


--------------------
\2 Health Insurance:  Coverage Leads to Increased Health Care Access
for Children (GAO/HEHS-98-14, Nov.  24, 1997). 

\3 G.  Pappas and others, "Potentially Avoidable Hospitalizations: 
Inequities in Rates Between U.S.  Socioeconomic Groups," American
Journal of Public Health, Vol.  87, No.  5 (1997), pp.  811-22;
Andrew Bindman and others, "Preventable Hospitalizations and Access
to Health Care," Journal of the American Medical Association, Vol. 
274, No.  4 (1995), pp.  305-11; and Carmen Casanova and Barbara
Starfield, "Hospitalizations of Children and Access to Primary Care: 
A Cross-National Comparison," International Journal of Health
Services, Vol.  25, No.  2 (1995), pp.  283-94. 

\4 States have the option of dropping the resource (asset)
eligibility requirement for these expansion populations.  Forty
states have discontinued using asset tests for some or all children. 

\5 The law allows states to roll back their income eligibility levels
to those in effect May 1, 1988.  Approval from the Health Care
Financing Administration (HCFA) is required to make such a change. 

\6 These children might still qualify for Medicaid under alternative
eligibility categories, and states are required to do
redeterminations. 

\7 Medicaid:  Early Implications of Welfare Reform for Beneficiaries
and States (GAO/HEHS-98-62, Feb.  24, 1998). 

\8 States will not receive any federal payments for CHIP if they
adopt income or resource standards or methodologies for determining a
child's eligibility for Medicaid that are more restrictive than those
in effect on June 1, 1997. 


   MEDICAID-ELIGIBLE CHILDREN WHO
   ARE UNINSURED HAVE
   CHARACTERISTICS THAT CAN BE
   USED TO TARGET OUTREACH EFFORTS
------------------------------------------------------------ Letter :3

Uninsured Medicaid-eligible children differ somewhat from those
currently enrolled in Medicaid, and these differences can be used by
states to focus their outreach and enrollment efforts.\9

Overall, about 23 percent--or 3.4 million--of the 15 million children
who were eligible for Medicaid were uninsured in 1996.  Slightly over
half of the Medicaid-eligible children are insured solely by
Medicaid, while about 7 percent have both Medicaid and private
coverage.  The remainder have coverage through other public programs,
such as the Civilian Health and Medical Program of the Uniformed
Services (CHAMPUS) or the Indian Health Service.  (See fig.  1.)
Medicaid-eligible children who are uninsured have characteristics
closer to Medicaid-eligible children who are privately insured than
to those with Medicaid.\10 They are disproportionately children of
the working poor, Hispanic, and U.S.-born children of foreign-born
parents or foreign-born, and they are more likely to live in the West
and the South. 

   Figure 1:  Percentage of
   Medicaid-Eligible Children With
   Different Types of Health
   Insurance or Uninsured, 1996

   (See figure in printed
   edition.)

Note:  Uninsured children are children who are reported to have no
insurance coverage at all for the entire year.  Children reported as
having health insurance coverage may have been uninsured for part of
the year.  Children with more than one source of coverage reported
may have had duplicate coverage at the same time or may have had
different types of coverage at different times of the year.  For this
figure, more than one source of coverage is shown only for children
who have both private insurance and Medicaid coverage.  Children with
Medicare are included with the Medicaid group.  Children with other
coverage include those with insurance obtained through other public
programs. 


--------------------
\9 We define Medicaid-eligible children as those eligible for
Medicaid by federally mandated age and income criteria:  children
under 6 years old with family income at or below 133 percent of the
federal poverty level and children 6 through 12 years old with family
income at or below federal poverty level. 

\10 For a comparison of Medicaid-eligible children to all uninsured
children, see S.  Avruch and others, "The Demographic Characteristics
of Medicaid-Eligible Uninsured Children," American Journal of Public
Health, Vol.  88, No.  3 (1998), pp.  445-7. 


      UNINSURED MEDICAID-ELIGIBLE
      CHILDREN HAVE SIMILARITIES
      WITH PRIVATELY INSURED
      CHILDREN
---------------------------------------------------------- Letter :3.1

Medicaid-eligible children are more likely to be uninsured if their
parents work, if their parents are self-employed or employed by a
small firm, or if they have a two-parent family.  Children whose
parents worked at all during the year--whether full-time, part-time,
or for part of the year--are about twice as likely to be uninsured as
those whose parents were unemployed.  However, these children are
more likely to be covered by employment-based insurance.  The
explanation for this apparent paradox is that children with employed
parents are less likely to be covered by Medicaid, and
employment-based coverage does not fully compensate for low rates of
Medicaid participation.  (See table 2.)



                          Table 2
          
           Insurance Status of Medicaid-Eligible
             Children in 1996, by Parents' Work
                           Status

                              Insurance status
                  ----------------------------------------
                                                Percentage
                                                      with
                                  Percentage   employment-
Parents' work       Percentage   enrolled in         based
status               uninsured    Medicaid\a   insurance\a
----------------  ------------  ------------  ------------
Full-time for             32.0          38.5          30.6
 entire year
Less than full-           21.2          62.1          22.7
 time
Not working               13.0          81.1           8.2
----------------------------------------------------------
Note:  Table is based on 96 percent of Medicaid-eligible children
whose records matched with those of their parents.  (See app.  I.)

\a Includes children who have both Medicaid and employment-based
health insurance coverage in the same year. 

Small firms are less likely than larger firms to offer health
insurance; therefore, it is not surprising that children whose
parents are self-employed or employed by small firms are less likely
to be insured.  This could suggest selective targeting of smaller
firms in Medicaid outreach efforts, especially if it is known that
they do not offer insurance. 

Half of all uninsured Medicaid-eligible children are in two-parent
families, compared with only about 28 percent of those insured by
Medicaid.  The uninsured rate for Medicaid-eligible children is also
higher in two-parent families than in single-parent families--30
percent compared with 18 percent.  This again underscores that
successful outreach efforts need to reach beyond the single
unemployed mothers generally associated with both cash assistance
programs and Medicaid. 


      HISPANICS, U.S.-BORN
      CHILDREN OF FOREIGN-BORN
      PARENTS, AND IMMIGRANT
      CHILDREN ARE MORE LIKELY TO
      BE UNINSURED
---------------------------------------------------------- Letter :3.2

Among racial and ethnic groups, the proportion of uninsured
Medicaid-eligible children, as well as the proportion enrolled in
Medicaid, varies by racial and ethnic group.  Among uninsured
Medicaid-eligible children, Hispanics have the highest uninsured
rate, while blacks are most likely to be enrolled in Medicaid.  (See
table 3 and table II.1 in app.  II.)



                          Table 3
          
          Percentage of Medicaid-Eligible Children
            Uninsured and Percentage Enrolled in
          Medicaid in 1996, by Race and Ethnicity

                                                Percentage
                                  Percentage   enrolled in
                                   uninsured      Medicaid
------------------------------  ------------  ------------
White                                   21.1          54.1
Black                                   18.9        69.4\a
Hispanic                              29.2\b          58.8
Other, non-Hispanic                     25.6          54.9
==========================================================
Total                                   23.0          59.7
----------------------------------------------------------
\a Statistically different from percentage for whites and Hispanics. 

\b Statistically different from percentage for whites and blacks. 

In 1996, almost 9 out of every 10 uninsured Medicaid-eligible
children were U.S.-born, but many--over one-third--lived in immigrant
families.  In addition to the 11 percent who were immigrants, another
one-quarter had at least one foreign-born parent.\11 (See fig.  2.)
The large number of children in immigrant families and the high
proportion that are uninsured suggest that immigrant communities may
be promising targets for outreach.  (See table 4.) Over 70 percent of
children in immigrant families are Hispanic, suggesting that outreach
efforts be targeted to the Hispanic community as well as use
Spanish-language outreach materials and applications. 

   Figure 2:  Medicaid-Eligible
   Uninsured Children in 1996, by
   Birth Status of Child and
   Parent

   (See figure in printed
   edition.)



                          Table 4
          
          Percentage of Medicaid-Eligible Children
          Insured and Uninsured in 1996, by Birth
                 Status of Child and Parent

                    Percentage    Percentage
Characteristic         insured     uninsured  Total number
----------------  ------------  ------------  ------------
U.S.-born child           81.7          18.3    11,140,000
 with U.S.-born
 parent
U.S.-born child           70.1          29.9     2,648,000
 with foreign-
 born parent
Foreign-born              44.5          55.5       631,000
 child with
 foreign-born
 parent
----------------------------------------------------------
Note:  Table is based on 96 percent of Medicaid-eligible children
whose records matched with those of their parents.  Children were
matched to the parent with the highest workforce participation.  Less
than 1 percent of children were foreign-born children with U.S.-born
parents; these children were excluded from the table. 


--------------------
\11 We matched children to one parent--the one with the highest
workforce participation--which might lead to a slight underestimate
of children who had at least one foreign-born parent.  However, in
two-parent families, spouses generally had similar birth and
immigration status.  (See app.  I.)


      MANY UNINSURED
      MEDICAID-ELIGIBLE CHILDREN
      ARE SCHOOL-AGE OR HAVE
      SCHOOL-AGE SIBLINGS
---------------------------------------------------------- Letter :3.3

Medicaid eligibility criteria allow younger children with higher
family income to enroll.  Children under 6 years old in families with
income at or below 133 percent of the federal poverty level are
eligible for the program, according to federal mandate, as compared
with older children whose families' income must be at or below 100
percent of the federal poverty level.  As a consequence, 54 percent
of children who are Medicaid-eligible but uninsured are less than 6
years old.  Nevertheless, outreach through schools could reach some
of these younger children, since 42 percent have a school-age sibling
aged 6 to 17.  This means that it could be possible to reach about 69
percent--or 2.4 million--of uninsured Medicaid-eligible children
through schools. 


      FAMILIES OF
      MEDICAID-ELIGIBLE UNINSURED
      CHILDREN ARE LESS LIKELY TO
      ENROLL FOR FOOD STAMPS THAN
      OTHER LOW-INCOME FAMILIES
---------------------------------------------------------- Letter :3.4

Other government programs could be used to reach families of
uninsured Medicaid-eligible children, if they were using such
programs.  Use of government-subsidized services by these families
might also indicate their willingness to access certain kinds of
government-sponsored programs.  While the CPS does not have
information on use of public programs such as Head Start or the
Department of Agriculture's Special Supplemental Food Program for
Women, Infants, and Children (WIC), it does have information on
family use of the Food Stamp program. 

Compared with similar families with employment-based insurance or
Medicaid coverage for their children, children who were
Medicaid-eligible but uninsured were less likely to have been in
families that received food stamps.  Some experts have argued that
immigrant families have been less willing to apply for subsidized
benefits because of their fear of the government or language and
cultural barriers.  However, we did not find any significant
difference in use of food stamps among uninsured Medicaid-eligible
children with U.S.-born parents, foreign-born naturalized citizen
parents, and foreign-born noncitizen parents. 


      THE WEST AND THE SOUTH FACE
      DIFFERENT AND LARGER
      OUTREACH CHALLENGES
---------------------------------------------------------- Letter :3.5

The demographic makeup of the uninsured child population varies
geographically and by community, meaning that national analyses can
only suggest potential outreach targets and must be validated in the
light of local knowledge.  Nonetheless, it is clear that the West and
the South as a whole face particular challenges.  A larger proportion
of Medicaid-eligible children in these regions are
uninsured--overall, the West and the South account for 73 percent of
all uninsured Medicaid-eligible children nationwide.  (See fig.  3.)

   Figure 3:  Number of
   Medicaid-Eligible Children With
   and Without Health Insurance in
   1996, by Region

   (See figure in printed
   edition.)

Note:  Numbers in thousands. 

Several reasons may explain the differences in proportions of
uninsured Medicaid-eligible children in different regions.  Some
areas have higher proportions of workers with employment-based
insurance because of the size of local firms, type of business, or
degree of unionization.  As a result, higher proportions of workers'
dependents are insured.  Regions also differ in the number and
percentage of immigrant families and various ethnic groups.  All of
these factors can affect insurance status and how states conduct
outreach to the uninsured. 

Regions differ in the number of uninsured Medicaid-eligible children
who are Hispanic or of Hispanic descent and in the number who are
members of immigrant families.  Both Hispanic and immigrant families
are most prevalent in the West, particularly in California, where
over 60 percent of uninsured Medicaid-eligible children are Hispanic
and over 70 percent live in immigrant families.  (See tables II.2 and
II.3 in app.  II.)

Although some differences among states are due to states' demographic
characteristics, differences may also be due to states' varying
efforts to extend health insurance to those who have been unable to
receive coverage and to inform these individuals of their
eligibility.  Having a larger, more visible program is a mechanism
that may help.  Some states have expanded Medicaid eligibility
further for children than other states and may have attracted more of
their poorer Medicaid-eligible children to enroll.\12


--------------------
\12 States like Washington, Tennessee, and Minnesota developed
subsidized health insurance for low-income families, partially or
entirely funded through Medicaid.  They may also have reduced the
percentage of uninsured Medicaid-eligible children through attracting
low-income uninsured families to their program. 


   LIMITED ENROLLMENT DUE TO
   SEVERAL FACTORS
------------------------------------------------------------ Letter :4

When asked why families do not enroll their Medicaid-eligible
children in the program, state officials, beneficiary advocates,
health care providers, and other experts report a variety of
contributing factors.  Some families do not know about the program or
do not perceive a need for its benefits.  Some families, especially
those who have never enrolled in public benefit programs, may not
even be aware that they are eligible.  In addition, some parents may
associate Medicaid with welfare and dependency, and therefore have an
aversion to enrolling their children in the program.  Cultural and
language differences may limit awareness or understanding, and
immigration status may also affect a family's willingness to apply. 
Finally, the eligibility process can be difficult for working
families because of the limits on where and when enrollment can take
place, the lengthy application, and the documentation required. 
These barriers to the enrollment process can be reduced, but in
reducing the length of the application and the amount of
documentation a balance must be struck between maximizing enrollment
and minimizing program abuse. 


      FAMILIES MAY LACK KNOWLEDGE
      OF MEDICAID AND ITS
      ELIGIBILITY CRITERIA
---------------------------------------------------------- Letter :4.1

Most state officials, advocates, providers, and other experts whom we
interviewed agreed that many families are unaware of Medicaid.  Even
families who know about the program may not realize that they could
be eligible.  With the long-standing link between Medicaid and AFDC,
many families--both those who have never received welfare and those
who have--assume that if they are not receiving cash assistance, they
are not eligible for Medicaid.  Two types of families tend to be
unaware of their eligibility:  working families who assume that
Medicaid eligibility is tied to welfare eligibility and families who
were previously on welfare and believe that, because of welfare
reform, they are no longer eligible for Medicaid.  These families are
unlikely to understand that children with higher levels of family
income may be eligible for Medicaid.  Complex eligibility
rules--which can result in younger children being eligible while
their elder siblings are not--can simply add to families' confusion. 


      FAMILIES MAY NOT PERCEIVE A
      NEED FOR MEDICAID DUE TO
      GOOD HEALTH AND ALTERNATIVE
      SOURCES OF CARE
---------------------------------------------------------- Letter :4.2

Several state officials, providers, and one expert told us that some
families do not become concerned about health care access until their
children become sick and, therefore, do not enroll them in
Medicaid--especially if the children are relatively healthy.  In
addition, if families have successfully sought and received care for
their children from clinics or emergency rooms in the past without
enrolling in a health care program such as Medicaid, they are likely
to continue to seek care from these providers. 


      CULTURAL DIFFERENCES,
      LANGUAGE BARRIERS, AND
      IMMIGRATION POLICIES MAY
      KEEP SOME FAMILIES FROM
      ENROLLING IN MEDICAID
---------------------------------------------------------- Letter :4.3

State officials, advocates, and other experts told us that some
families are hesitant to enroll in Medicaid because of cultural
differences, language barriers, and their understanding of U.S. 
immigration policies.  Experts and a state official said that
cultural differences may keep immigrant families from enrolling in
Medicaid.  Language was often mentioned as a barrier.  Individuals
who cannot read the Medicaid application and informational materials
and cannot easily converse with eligibility workers by telephone or
in person are at a distinct disadvantage.  One expert told us that
the degree of acculturation has a major impact on whether an
immigrant will use public assistance of any kind.  While time spent
in the country is the main predictor of acculturation, some
individuals may not participate in mainstream society and use its
institutions even after living in the country for many years. 

According to some state officials, advocates, and experts, immigrant
families may also hesitate to enroll in Medicaid because they are
concerned that it will negatively affect their immigration status. 
Immigrants who are legal residents may be afraid that if they receive
benefits that they will be labeled a "public charge" and will have
difficulties with the Immigration and Naturalization Service (INS)
when applying for naturalization, visa renewal, or reentry into the
United States.\13 Although advocates question whether aliens
receiving benefits may be considered public charges, in some
instances actions have been taken against such individuals seeking
visa renewals.  Several advocates also told us about cases where
individuals were prevented from reentering the United States unless
they agreed to reimburse Medicaid for services paid for by the
program on their behalf--particularly in border states such as
California.\14 Publicity about such cases in the immigrant community
can deter immigrants from applying for Medicaid benefits for
themselves or their children--even in cases where the children were
born in the United States and are American citizens. 

In families where one or more adults are in the country illegally,
the reluctance to seek Medicaid benefits for a child may be even
greater.  When applying for Medicaid for children, families in some
states are asked about the immigration status of other members of the
household.  Again, advocates told us this is a deterrent to
enrollment for such families and reported that many immigrant
families, both legal and illegal, seek medical assistance through
county clinics and public hospitals because these institutions are
viewed as more sympathetic and less likely to ask questions about
immigration status. 


--------------------
\13 Aliens may be excluded from entering the United States if it is
determined that they are likely to become public charges.  If already
in the United States, they may be deported if they have become public
charges.  Neither the law nor the regulations specify whether receipt
of benefits such as Medicaid would cause someone to be deemed a
public charge. 

\14 In a December 17, 1997, letter to state Medicaid directors, HCFA
stated that when an alien has legitimately received Medicaid
benefits, the beneficiary is not indebted to the state.  The letter
also said that states do not have the authority to collect repayments
of benefits from current or former beneficiaries except in cases
where those benefits were fraudulently received or an overpayment had
occurred.  Further, the letter said that state Medicaid agencies are
not authorized to provide information about the receipt of benefits
or the dollar amount of these benefits to INS, the State Department,
or immigration judges.  In a December 17, 1997, memorandum to its
offices, INS said that it does not have the authority to require or
request that aliens repay public benefits, and lawful permanent
residents who have been outside the United States 180 days or less
should not be questioned on issues related to the likelihood that
they would be a public charge. 


      PERCEIVED NEGATIVE IMAGE
      ATTACHED TO MEDICAID MAY
      DETER ENROLLMENT
---------------------------------------------------------- Letter :4.4

State officials and other experts told us that because of its
long-standing ties with welfare and other benefit programs, many
families associate Medicaid with a family that cannot provide for
itself.  Experts report that many working poor and near poor do not
want to be labeled as welfare recipients, even if the law entitles
their children to benefits.  They often take the view that they never
have received welfare and do not want to start. 


      ENROLLMENT PROCESS CAN BE A
      BARRIER
---------------------------------------------------------- Letter :4.5

State officials, beneficiary advocates, providers, and other experts
agree that Medicaid enrollment processes and requirements have often
been barriers.  However, to ensure that all recipients of Medicaid
benefits meet income and other requirements, states have found it
necessary to develop application processes that

  -- use lengthy application forms and

  -- require extensive documentation. 

State officials, beneficiary advocates, and other experts told us
that lengthy enrollment forms and the associated documentation
requirements create a barrier for families.  Long forms are often
used when a family is applying for a combination of programs,
including Medicaid.  Numerous questions relating to income, assets,
citizenship, and family composition are used to determine eligibility
and to ensure that only those who are entitled to benefits are
enrolled in Medicaid.  In addition to length, enrollment forms often
require extensive documentation.  Families are asked to provide
paystubs, bank account statements, birth certificates, and other
documents that verify the information they provide on the forms. 
Gathering such documents can be burdensome.  For example, obtaining a
birth certificate can involve going to a different office and then
returning to the eligibility office.  Obtaining certain documents can
also require a family to pay a fee. 

A valid and reliable eligibility determination process is important
to state officials to ensure program integrity.  In addition, states
can be assessed a financial penalty by the federal government if
their error rates are too high.  In an effort to balance these needs,
most states have developed shorter forms for children who are
applying exclusively for Medicaid, primarily by dropping the asset
requirements.  Some advocates, however, are still concerned with the
length and complexity of application forms and the number of
questions they contain.  One advocate suggested that if applicants
cannot understand the form, they are not going to fill it out. 
Another advocate pointed out that some questions may be
well-intended, but they nonetheless lengthen the application.  For
example, as a way of identifying if the family may be eligible for
other benefit programs, some states' applications ask questions
related to disability.  In addition, advocates pointed out that the
documentation requirements are so stringent in some states that many
applicants are denied enrollment because they cannot produce the
documentation required.  In an earlier report, we found that such
requirements were shown to account for nearly half of all denials.\15

In addition to limits that were developed as part of a legitimate
effort to maximize the accuracy of eligibility determinations and
monitor the eligibility process, other barriers exist.  These include

  -- location of enrollment sites and enrollment hours;

  -- fluctuations in eligibility status, including the impact of
     welfare reform; and

  -- families' lack of transportation and communication problems. 

Many of the state officials and other experts with whom we spoke said
that the enrollment process used for welfare was difficult for
working families because enrollment locations are limited and open
only during typical work hours.  This makes it difficult for working
parents in families whose children may be eligible for Medicaid to
apply.  Such parents may not have the flexibility in their job to
take time off to enroll through face-to-face interviews, according to
one state official and one expert.  States are required to provide
for the receipt and initial processing of applications for pregnant
women, infants, and children at sites not used for AFDC
applications--such as federally qualified health centers and
hospitals that serve a larger share of uninsured and publicly insured
persons--but these efforts may have been limited. 

Experts also noted that the eligibility system does not accommodate
the fluctuating eligibility status of many families.  Low-income
working families may have changes in their income if they work
seasonally or change or lose jobs.  A family eligible one month may
not be eligible the next month because of an increase in family
income, but children in that family may still be covered under other
categories of eligibility.  According to experts, some states'
eligibility processes do not automatically make redeterminations to
see if children who lose their eligibility might be eligible in
another category.  If the family does not reapply, the child loses
coverage. 

Advocates have also been concerned that welfare reform may make
enrollment less likely.  Families may be confused about their
Medicaid eligibility because, prior to welfare reform, Medicaid and
cash assistance had, historically, been so closely linked.  For
example, if TANF enrollment workers focus on job search strategies
and not on benefits, families who come in may not be enrolled for
Medicaid.  In addition, some families may believe Medicaid is time
limited as is TANF. 

According to experts, advocates, and one provider, limits on the
ability to communicate and availability of transportation can be a
barrier for applicants.  In addition to difficulties for
non-English-speaking families, illiteracy may also limit a parent's
ability to enroll without substantial assistance.  Experts also
pointed out that lack of transportation to enrollment sites can be a
barrier, primarily in rural areas, but also in some urban settings. 
A family may not have a car or have limited time and money to make a
long trip to the welfare office. 


--------------------
\15 Health Care Reform:  Potential Difficulties in Determining
Eligibility for Low-Income People (GAO/HEHS-94-176, July 11, 1994). 


   SOME STATES USE INNOVATIVE
   STRATEGIES TO TARGET OUTREACH
   AND ENCOURAGE ENROLLMENT
------------------------------------------------------------ Letter :5

To enroll eligible children in Medicaid, some states are using
innovative strategies that are intended to increase knowledge and
awareness of the program and its benefits, minimize the perceived
social stigma, and simplify and streamline the eligibility process. 
Education and outreach programs are often targeted to families who
have children potentially eligible for Medicaid.  Visible support
from state leadership and partnerships with community groups are
viewed by state officials and advocates as essential to obtaining the
necessary resources to implement outreach programs.  Some states have
even renamed the Medicaid program as a way to change its image.  To
improve the enrollment process, some states have adopted strategies
to assist immigrant families or have simplified and streamlined the
eligibility process by shortening forms and accepting applications at
many new sites, as well as mail-in applications.  However, this kind
of simplification and streamlining has required state officials to
make difficult trade-offs between the need for program integrity and
higher Medicaid enrollment. 


      MULTIFACETED OUTREACH
      PROGRAMS INTENDED TO EDUCATE
      TARGETED FAMILIES
---------------------------------------------------------- Letter :5.1

The states that we contacted have developed multifaceted outreach
programs to educate families on the availability of the Medicaid
program and the importance of enrolling their children.  They
generally agreed that a successful education and outreach program
should

  -- target outreach to low-income working families with children,
     using nontraditional methods and locations, and

  -- work in collaboration with community groups, schools, providers,
     and advocates. 

These themes are broadly consistent with several findings from our
demographic analysis:  low-income working families with children have
a high uninsured rate, and most uninsured Medicaid-eligible children
are in school or have a sibling in school, which makes the schools an
available avenue for reaching children and families. 

The states that we studied have employed a variety of methods to
publicize Medicaid.  For example, Massachusetts has placed outreach
workers in health centers, hospitals, and other traditional
locations; distributed literature in schools; sent material to the
YMCA and other community groups; and worked with a supermarket chain
to place in grocery bags notices of the program.  The governor has
held several press conferences around the state to publicize the
program, and the state is working with workers in WIC clinics, who
are already trained to do income-based eligibility assessments.  The
state has also used its enrollment data to target communities that
have low levels of Medicaid enrollment and worked with local
officials to address the problem.  The state's private contractor for
managed care enrollment has also assisted with outreach through its
presentations in the community.  One advocacy group worked with the
state to develop a campaign to target high-school athletes, who are
required to have health insurance.  This campaign involved sending
posters and fold-out fliers--developed and produced with the donated
time of professionals--to athletic directors in high schools
throughout the state and establishing a pool of student athletes to
go out and talk to their peers.  In another initiative, the state
medical society is training its members' staffs to assist in
educating families about program eligibility and benefits.  Finally,
Massachusetts is making $600,000 available to help community groups
conduct outreach and educate families of uninsured Medicaid-eligible
children, with the money distributed as grants in amounts between
$10,000 and $20,000. 

In Arkansas, as part of a large media campaign that included
television and radio announcements, the state placed color inserts in
Sunday newspapers during September 1997.  These inserts provided
information on program eligibility and benefits, a toll-free number
to obtain additional information, as well as a photograph of children
with the governor endorsing the program.  The state's children's
hospital paid for the insert.  Applications are available at schools,
pharmacies, and churches, and brochures have also been placed in fast
food bags.  The state has also worked with its children's hospital to
place enrollment forms at affiliated clinics, which are located
throughout the state. 

Georgia has made a major commitment to outreach by employing over 140
eligibility workers with the specific job of getting eligible
children and families enrolled in Medicaid.  These outreach workers
are situated in numerous locations, including health departments,
clinics, and hospitals.  These workers also temporarily set up at
nontraditional sites, such as schools, community agencies, and
shopping malls.  The outreach workers are often available during
evening and weekend hours as a convenience to working families. 
Workers also make presentations regularly to community groups,
medical providers, and employers.  A flier was developed that is
targeted to employers to inform them about benefit programs for which
their employees may be eligible.  Georgia is also trying to enroll
former welfare recipients by emphasizing Medicaid enrollment as an
important part of a successful transition to work.  The state's
outreach program has also established partnerships with numerous
community groups--including local coordinating councils, local teen
pregnancy task forces, and school boards--and has used these local
partnerships to develop outreach tailored to needs and
characteristics of the communities.  The state's private contractor
for enrollment in managed care has also assisted with the outreach
program through its contacts with the community. 

In view of its recent welfare reform initiatives, Wisconsin is making
a concerted effort to ensure that Medicaid-eligible individuals
enroll in Medicaid regardless of their eligibility for the state
welfare program.  As part of this outreach effort, the state has
begun to target county eligibility workers, individual providers, and
Medicaid-eligible individuals to communicate that people may still
qualify for medical assistance apart from their eligibility for
welfare.  Additional resources have been made available for outreach,
outstationing, and training materials for staff.  To plan its
outreach efforts, the state is working with outside groups, including
the Primary Health Care Association, the state medical society,
Milwaukee County, Children's Hospital, and Marshfield Clinic. 


      FEW STATES TAILOR PROGRAMS
      TO IMMIGRANT AND ETHNIC
      COMMUNITIES
---------------------------------------------------------- Letter :5.2

We found less targeting of immigrant communities than might have been
expected from the demographic analysis, although this was in some
measure due to the characteristics of the states that we selected for
our study.  However, advocates report concern within the immigrant
communities that receiving benefits will compromise their immigration
status.  One expert told us that some states have attempted to assist
eligible immigrant families in enrolling their children by providing
enrollment information and applications in alternative languages,
particularly Spanish, and by hiring bilingual enrollment workers.  In
general, their outreach approach is similar to those tailored to
other communities but with an emphasis on particular immigrant and
ethnic cultures and languages.  Massachusetts is working with local
community groups that provide information and educate immigrants on
the availability of Medicaid.  Georgia's outreach workers give
presentations to employee groups within firms that have a large
proportion of Hispanic immigrants among their workers. 

In their outreach efforts, states face challenges with the immigrant
community because they have to take into account the recent changes
of the Welfare Reform Act and the Balanced Budget Act, which make
benefits a state option for qualified immigrants who arrived before
August 23, 1996, and bar immigrants for 5 years if they arrived after
August 22, 1996.  However, these limitations do not affect the
eligibility of native-born children in immigrant families. 


      NAME CHANGE AND OUTREACH
      INITIATIVES INTENDED TO
      PROJECT MORE POSITIVE
      PROGRAM IMAGE
---------------------------------------------------------- Letter :5.3

States have tried to change the perception that Medicaid is tied to
welfare and dependency in a variety of ways.  The most direct method
for changing the program's image is changing the program's name.  In
addition, states have advertised the program as one that is intended
for working families, while some have included policies to avoid
displacing private health insurance.  They have also adopted
alternative enrollment methods so that individuals do not have to go
to the local welfare office to enroll. 

Changing the Medicaid program's name is not new, but it has become
more widespread.\16 Massachusetts recently renamed its program
MassHealth with the intent that it would be more appealing to
beneficiaries.  MassHealth fliers describe several option plans
available (six in total), referring to them by names such as
"MassHealth Standard" and "MassHealth Basic"--names similar to
commercial health plans.  Arkansas named its Medicaid expansion
program for children ARKids 1st.  The logo for the program uses
bright colors with the "1" in 1st represented by a crayon.  Georgia
has not changed the name of Medicaid, but its outreach project is
called "Right From the Start" to project a positive message.\17

Advertisements and fliers for these programs emphasize that they are
for a broad population, not just those on welfare.  MassHealth fliers
state, "There is no reason why a child or a teen in Massachusetts
should go without health care." Massachusetts has fliers that outline
income levels for eligibility that show families with almost $2,400 a
month in income and pregnant women with income up to $3,300 a month
as eligible.  Georgia has a flier entitled, "Have you heard about
benefits for working families?," and the first program mentioned is
Medicaid for children.  Another flier targeted to families leaving
welfare to work asks the question, "Did you know you could work full
time and still receive some benefits?" (See fig.  4.)

   Figure 4:  Outreach Materials
   From Arkansas, Georgia, and
   Massachusetts

   (See figure in printed
   edition.)

To minimize the possibility of displacing private insurance, known as
"crowd out," some states have policies to address the issue.  The
Medicaid program cannot refuse enrollment to any eligible individual
based on the fact that he or she has insurance, although Medicaid is
the payer of last resort.  However, some states that have expanded
eligibility through waivers of normal program rules have been allowed
to limit eligibility if a family already has insurance.  For example,
in Arkansas, which received a waiver for its expansion, a child is
not eligible for ARKids 1st unless he or she has been uninsured for a
period of 12 months or the child lost insurance coverage during that
period through no fault of the family.  In Massachusetts (which also
has a waiver for expansion) and Georgia, officials are cognizant of
the potential dangers of crowd out.  Massachusetts, as part of
MassHealth, will subsidize the cost of insurance available to the
family. 


--------------------
\16 Some states changed their Medicaid name for pregnant women and
infants when they began outreach to these populations in the late
1980s. 

\17 States also often provide identification cards that look like
cards in commercial plans.  This is also prevalent in managed care
programs where the beneficiary gets the same managed care plan card
as non-Medicaid members. 


      STATES CHANGE ENROLLMENT
      PROCESS IN VARIOUS WAYS TO
      ATTRACT WORKING FAMILIES
---------------------------------------------------------- Letter :5.4

Some states have developed a number of strategies to make the
enrollment process easier for working families.  Several states, as
part of their outreach effort, have outstationed eligibility workers
in sites that families frequent as an alternative to enrolling at the
welfare office.  In addition, states have simplified and shortened
their enrollment applications, allowed applications by mail, dropped
asset requirements, and reduced documentation requirements.  To help
ensure continued coverage of children in families whose income
fluctuates, states can provide continuous eligibility.  Of the states
we contacted, only Arkansas has adopted continuous eligibility for a
year for children. 

Some states have adopted enrollment methods that do not require
individuals to visit a welfare office, in part to minimize Medicaid's
association with welfare and welfare families.  If families are only
seeking Medicaid enrollment for children, Massachusetts and Arkansas
allow families to ask questions and request an application by
telephone.  These two states also accept applications by mail. 
Completing applications with outreach workers at various
nontraditional sites is another way the process is made easier for
working families and those without transportation. 

Each of the states with whom we spoke had shortened and simplified
their enrollment form.  Massachusetts officials used focus groups to
find out why families did not enroll their children and how barriers
to enrollment could be removed.  Suggestions from the focus
groups--such as adding more space on the enrollment form--helped the
state design a simplified form that is easier to read.  States have
had the option of dropping the asset tests for certain populations. 
When Arkansas dropped its asset test for the ARKids 1st program, it
also dropped the related questions about assets and property,
shortening the enrollment form to four pages.  Georgia also shortened
its enrollment form and dropped the asset test. 

States are concerned with maintaining program integrity and ensuring
that benefits go only to qualified individuals.  However, 40 states
have abolished the asset test for some or all children, primarily
because the likelihood that these families have substantial assets is
low.  Table 5 shows the number of states that have made these
changes. 



                                Table 5
                
                Number of States That Have Made Changes
                       to the Enrollment Process

--------------------------------------------------------------  ------
Dropped asset test                                                40\a
Shortened eligibility form                                          30
Adopted mail-in enrollment                                          25
----------------------------------------------------------------------
\a Four states have dropped their asset test for some but not all
eligible children. 

Source:  Center on Budget and Policy Priorities. 

Few efforts have been made to address the problem of fluctuating
family eligibility status, causing children to be inappropriately
disenrolled from Medicaid.  As part of its ARKids 1st program,
Arkansas is providing 12 months of continuous eligibility to children
regardless of changes in family income, under waiver authority
granted by HCFA.  Until recently, states had to receive a waiver to
pursue such a policy.  The Balanced Budget Act, however, allows
states to adopt 12 months of continuous eligibility. 

To date, welfare reform has not significantly affected the
application process for Medicaid.  In a recent report, we found that
nine states we contacted have chosen to make few structural changes
in their Medicaid programs in the first full year of implementing
welfare reform.  For example, while the Welfare Reform Act delinked
eligibility for cash assistance and Medicaid, the states that we
contacted had generally decided not to separate Medicaid and cash
assistance program administration.  In three of the states that we
spoke with for this study, welfare applicants received a combined
form that permits families to apply for both cash assistance and
Medicaid, but families applying only for Medicaid receive a shorter
form with a subset of questions. 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

Despite the importance of and large investment in providing health
care to children in low-income families, difficulties in enrolling
them in Medicaid leave more than 3 million children vulnerable.  The
states that we reviewed recognized that uninsured Medicaid-eligible
children are generally in working two-parent families and have
targeted their outreach accordingly.  Targeting working families
raises the issue of crowd out--replacing employer-based insurance
with Medicaid--but states that we contacted have not seen this as a
major problem given the low income levels of these families.  Only
Arkansas has taken direct action to discourage employers from
dropping health insurance coverage by enforcing a 12-month waiting
period. 

We found less outreach targeted to Hispanics and immigrants, and
experts whom we interviewed said this was generally true, even in
states with large immigrant or Hispanic populations.  Immigrants,
particularly families in which the parents are not naturalized U.S. 
citizens, are likely to be a more difficult group to reach, both
because of the complexities of the law, which makes some but not all
immigrant children eligible for Medicaid, and because of the
immigrants' general wariness of government.  Some immigrant families
include children who--because they were born in this country--are
citizens and fully eligible for Medicaid. 

The states that we studied are, for the most part, using outreach and
enrollment strategies available for some time--but not necessarily
used for enrolling uninsured children.  However, other strategies
provided for by the Balanced Budget Act--such as continuous
enrollment and presumptive eligibility--have not been widely
implemented.  CHIP also has considerable potential for identifying
uninsured Medicaid-eligible children.  The law provides that any
child who applies for CHIP and is determined to be Medicaid-eligible
should be enrolled in Medicaid.  The more that states publicize CHIP,
the greater the number of uninsured Medicaid-eligible children they
are likely to identify and enroll in Medicaid--particularly if the
states' screening and enrollment process effectively identifies
Medicaid-eligible children and enrolls them in the Medicaid program. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :7

We sought comments on a draft of this report from HCFA; from state
officials in Arkansas, Georgia, Massachusetts, and Wisconsin; and
from experts on children's health insurance issues with the Southern
Institute on Children and Families and the Center on Budget and
Policy Priorities.  A number of these officials provided technical or
clarifying comments, which we incorporated as appropriate.  In
addition, HCFA noted that it had sent a letter dated January 23,
1998, to state officials to encourage them to simplify enrollment and
expand outreach to the Medicaid-eligible population. 


---------------------------------------------------------- Letter :7.1

As arranged with your office, unless you announce its contents
earlier, we plan no further distribution of this report until 30 days
after its issuance date.  At that time, we will send copies to the
Secretary of Health and Human Services, the Administrator of HCFA,
the directors of the state programs we spoke with; and interested
congressional committees.  Copies of the report will be made
available to others upon request. 

If you or your staff have any questions about the information in this
report, please call me or Phyllis Thorburn, Assistant Director, at
(202) 512-7114.  Other contributors to this report were Richard
Jensen, Sheila Avruch, and Sarah Lamb. 

Sincerely yours,

William J.  Scanlon
Director, Health Financing and
 Systems Issues


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

To examine the demographic characteristics of Medicaid-eligible
uninsured children, we analyzed the Current Population Survey (CPS),
which is used by some researchers to measure health insurance
coverage in the United States.  This technical appendix discusses the
survey, how we measured insurance coverage and estimated
Medicaid-eligible children, and how we determined parents' work
effort and immigration status.  It also discusses some concerns about
how well the CPS measures insurance coverage and compares our
estimate of the number of Medicaid-eligible uninsured children with
other analysts' estimates. 


   ABOUT THE CPS
--------------------------------------------------------- Appendix I:1

The CPS, a monthly survey conducted by the Bureau of the Census, is
the source of official government statistics on employment and
unemployment.  Although the main purpose of the survey is to collect
information on employment, an important secondary purpose is to
collect information on the demographic status of the population, such
as age, sex, race, marital status, educational attainment, and family
structure.  The March supplement of the CPS survey collects
additional data on work experience, income, noncash benefits, and
health insurance coverage of each household member at any time during
the previous year. 

The CPS sample is based on the civilian, noninstitutionalized
population of the United States.  About 48,000 households with
approximately 94,000 persons 15 years old and older and approximately
28,000 children aged 0 to 14 years old are interviewed monthly.  The
sample also includes about 450 armed forces members living in
households that include civilians and are either on or off a military
base.  For the March supplement, an additional 2,500 Hispanic
households are interviewed.  The households sampled by the CPS are
scientifically selected on the basis of area of residence to
represent the United States as a whole, individual states, and other
specified areas. 


   HOW WE DEFINED INSURANCE,
   MEDICAID ELIGIBILITY, AND
   FAMILY STATUS
--------------------------------------------------------- Appendix I:2


      HEALTH INSURANCE COVERAGE
      STATUS
------------------------------------------------------- Appendix I:2.1

Children can have multiple sources of health insurance coverage in
the same year.  The CPS asks about all sources of health insurance
coverage.  It is impossible to tell, for example, if a child is
reported as having both Medicaid and employment-based insurance,
whether the child had duplicate coverage, had Medicaid coverage first
and then employment-based coverage, or vice versa.  For this report,
children who had employment-based insurance were reported as having
such coverage even if they also had other sources of coverage. 
Likewise, children who had Medicaid coverage were reported as having
such coverage even if they had other sources of coverage.  As result,
some children were reported as having both public and private
coverage--usually Medicaid and employment-based insurance--for the
same year.  (See fig.  1.)

For this report, children who are uninsured are children for whom no
source of coverage during the entire previous year is reported.  CPS
asks specific questions about whether any members of the household
have coverage provided through an employer or union; purchased
directly; or have Medicare, Medicaid, or other public coverage. 
However, it does not directly ask whether an individual is uninsured
if no source of coverage is reported. 


      MEDICAID-ELIGIBLE CHILDREN
      WHO ARE UNINSURED
------------------------------------------------------- Appendix I:2.2

We defined Medicaid-eligible children in 1996 as children eligible by
federal mandate based on age and poverty criteria--children from
birth through 5 years old with family income at or below 133 percent
of the federal poverty level and children 6 through 12 years old with
family income at or below the poverty level. 

We used income in the immediate family rather than the household
income to calculate poverty levels.  We did this because states have
specific rules on what income can be deemed available to the child to
determine Medicaid eligibility, and it may not include income
provided to the household by people not related to the child.  In
addition, employment-based health insurance is usually only available
to immediate dependents; therefore, the income and work effort within
the nuclear family is more relevant to whether or not the child is
insured. 


      MATCHING CHILDREN WITH
      PARENTS
------------------------------------------------------- Appendix I:2.3

We matched children's records with parents' records to analyze family
characteristics.  CPS considers a family to be two or more persons
residing together and related by birth, marriage, or adoption.  The
Census Bureau develops family records for the householder (a person
in whose name the housing unit is owned, leased, or rented or, if no
such person, an adult in the household); other relatives of the
householder with their own subfamilies; and unrelated subfamilies. 
If the house is owned, leased, or rented jointly by a married couple,
the householder may be either the husband or wife.  We paired
children's records to their parents' records or, lacking a parent,
another adult relative (aged 18 through 64) in their immediate family
whom we called a parent.  After this pairing, we matched the adult
family member's record to his or her spouse's record, if any, to get
"parents" in our file.  We were not able to match all children's
records with records of parents or other relatives in their
households.  For Medicaid-eligible children, we matched 96 percent of
the children's records.  For Medicaid-eligible uninsured children, we
matched 92 percent of the children's records.  Some of our tables and
figures are based on the entire file of children's records; others
are based on the matched file and are so indicated. 


      DETERMINING PARENTS' WORK
      AND EDUCATION STATUS
------------------------------------------------------- Appendix I:2.4

Matching parents with children to analyze the association of
workforce participation and insurance for children helped us develop
a more accurate picture of uninsured and Medicaid-insured children
with working parents.  We analyzed parent work status on the basis of
information about the parent who worked the most.  (See table I.1.)
This allowed us to more accurately portray the work status of parents
in two-parent families.  Where two parents were working in the same
status--such as full-time--we matched to the first parent in that
work status. 



                         Table I.1
          
           Definition of Work Status of Parent or
                          Parents

Work status   Reported as   Definition
------------  ------------  ------------------------------
Full-time,    Full-time,    Either parent worked full-
full year     full year     time, full year.

Full-time,    Less than     Neither parent worked full-
part year     full-time,    time, full year, but at least
              full year     one worked full-time part of
                            the year.

Part-time,    Less than     Neither parent worked full-
full year     full-time,    time, but at least one parent
              full year     worked part-time for the
                            entire year.

Part-time,    Less than     Neither parent worked either
part year     full-time,    full-time or full year, but at
              full year     least one parent worked part-
                            time for part of the year.

Not working   Not working   Neither parent worked at all
                            during the entire year.
----------------------------------------------------------
We used the parent with the greater workforce participation to
determine children's birth and immigration status relative to their
parents'.  This could lead to a slight underestimate of children in
immigrant families, since in some two-parent families, spouses do not
have the same birth or citizenship status.  However, spouses
generally share similar birth and citizenship status.  We examined
birth and citizenship status of one parent compared with the other in
two-parent families and found that over 90 percent had the same birth
and citizenship status as their spouse.  Since only about half of
Medicaid-eligible children live in two-parent families to begin with,
matching to one parent would lead to over 95 percent of children
being accurately categorized, based on a match with one parent. 


   COMPARING CPS AND OTHER SURVEYS
--------------------------------------------------------- Appendix I:3

Some researchers who work with survey data to assess health insurance
status of the U.S.  population are concerned that the currently used
surveys, including CPS, may not accurately reflect health insurance
coverage in the United States.\18 CPS and the Survey of Income and
Program Participation (SIPP)--another survey that is often used to
assess health insurance coverage--report lower Medicaid coverage than
HCFA data on Medicaid enrollment.  Comparing CPS and SIPP data for
similar periods of time, some researchers have concluded that
although the CPS asks about insurance coverage for the entire
previous year, respondents are reporting coverage based on a shorter
time frame--perhaps 4 to 6 months.  Researchers at the Urban
Institute have concluded that some of the uninsured actually have
coverage, probably Medicaid coverage, and adjust their estimates of
the uninsured accordingly. 

Although health researchers are concerned that the CPS may not be
ideal for analyzing health insurance coverage, neither is any other
currently available survey.  Therefore, many researchers continue to
use it.  GAO chose to use CPS data for its analysis of children's
health insurance coverage for several reasons.  The CPS can be used
to look at trends over time, although care must be taken when making
comparisons between years because of questionnaire and methodological
changes.  It has a large sample, which gives estimates from the data
more statistical power.  It was designed so that it can be used for
some state-level estimates. 

Information from new health insurance surveys is or is becoming
available.  The National Health Interview Survey periodically asks
questions about health insurance coverage, and the Agency for Health
Care Policy and Research has released preliminary 1996 estimates of
health insurance coverage from the Medical Expenditure Panel Survey. 
The Center for Studying Health System Change has surveyed health
insurance coverage in 1996 and 1997 in its Community Tracking Study
(CTS) and is beginning to publish its data.  The Urban Institute has
also developed and fielded its own health insurance survey. 
Comparisons of these surveys with the CPS and SIPP may help
researchers more definitively agree on the number of uninsured
Americans and trends in insurance over time. 


--------------------
\18 A good summary of these issues can be found in Kimball Lewis,
Marilyn Ellwood, and John Czajkas, Children's Health Insurance:  A
Review of the Literature (Washington, D.C.:  Department of Health and
Human Services, Dec.  17, 1997). 


      COMPARING ESTIMATES OF
      MEDICAID-ELIGIBLE CHILDREN
      WHO ARE UNINSURED
------------------------------------------------------- Appendix I:3.1

Using either CPS or new CTS data, five different groups of
researchers compared estimates of uninsured Medicaid-eligible
children.  (See table I.2.) While the number of Medicaid-eligible
children and definition of Medicaid eligibility used by the
researchers differed, all came up with a similar conclusion--many
uninsured children are eligible for Medicaid.  The researchers'
estimates ranged from 24 to 45 percent. 



                                        Table I.2
                         
                           Estimates of the Number of Uninsured
                          Children Who Are Eligible for Medicaid
                               but Are Uninsured, by Source

                                                                     Uninsured children
                                                                   eligible for Medicaid
                                                                   ----------------------
                                       Age group
                                       of
                                       Medicaid-
                                       eligible                                Percentage
                           Time        uninsured   Eligibility     Number (in          of
Source         Survey      period      children    criteria         millions)   uninsured
-------------  ----------  ----------  ----------  --------------  ----------  ----------
Urban          CPS, March  1995        Less than   State-               1.6\a        24\b
Institute's    1996                    18 years    specific
TRIM2 model\a                                      poverty-
                                                   related and
                                                   AFDC-
                                                   eligible; SSI
                                                   children;
                                                   medically
                                                   needy
                                                   children;
                                                   estimated
                                                   assets

Reschovsky     CTS         late 1996-  Less than   State-                 3.1        35\c
(1997)                     early 1997  19 years    specific
                                                   poverty-
                                                   related

Thorpe         CPS, March  1995        Less than   Not given              3.3        31\c
(1997b)        1996                    19 years

GAO (1996)     CPS, March  1994        Less than   Federal-               2.9        30\b
               1995                    12 years    poverty-
                                                   related; not
                                                   state-
                                                   specific

GAO (1998)     CPS, March  1996        Less than   Federal-               3.4        33\b
               1997                    13 years    poverty-
                                                   related only;
                                                   not state-
                                                   specific

Center on      CPS, March  1994        Less than   Federal-               2.7        45\d
Budget and     1995                    11 years    poverty-
Policy                                             related; not
Priorities                                         state-
(1997)                                             specific
-----------------------------------------------------------------------------------------
\a Decreases estimate of uninsured and increases Medicaid enrollment
to adjust for differences between HCFA Medicaid enrollment data and
CPS data. 

\b Estimate of the percentage of uninsured children less than 18
years old who were Medicaid eligible. 

\c Estimate of the percentage of uninsured children less than 19
years old who were Medicaid eligible. 

\d Estimate of the percentage of uninsured children less than 11
years old who were Medicaid eligible. 

Source:  Kimball Lewis, Marilyn Ellwood, and John L.  Czajka,
Children's Health Insurance Patterns:  A Review of the Literature
(Cambridge, Mass.:  Mathematica Policy Research, Inc., 1997). 


      NONSAMPLING ERROR
------------------------------------------------------- Appendix I:3.2

Since CPS estimates come from a sample, they may differ from figures
from a complete census using the same questionnaires, instructions,
and enumerators.  A sample survey estimate has two possible types of
errors:  sampling and nonsampling.  Each of the studies mentioned
above--using either CPS or other sampling surveys--has the same
possible errors.  The accuracy of an estimate depends on both types
of error, but the full extent of the nonsampling error is unknown. 
Several sources of nonsampling errors include the following: 

  -- inability to get information about all sample cases;

  -- definitional difficulties;

  -- differences in interpretation of questions;

  -- respondents' inability or unwillingness to provide correct
     information;

  -- respondents' inability to recall information;

  -- errors made in data collection, such as recording and coding
     data;

  -- errors made in processing data;

  -- errors made in estimating values for missing data; and

  -- failure to represent all units with the sample
     (undercoverage).\19


--------------------
\19 Bureau of the Census, Current Population Survey, March 1997,
Technical Documentation (Washington, D.C.:  U.S.  Department of
Commerce, Bureau of the Census, Economics and Statistics
Administration, 1997). 


DEMOGRAPHICS OF MEDICAID-ELIGIBLE
CHILDREN WHO ARE UNINSURED
========================================================== Appendix II

Tables II.1 through II.3 provide a demographic profile of
Medicaid-eligible children in 1996. 



                                        Table II.1
                         
                            Number and Percentage of Medicaid-
                          Eligible Children Who Were Insured by
                          Medicaid or Uninsured in 1996, by Race
                                      and Ethnicity

                                Medicaid                           Uninsured
                   ----------------------------------  ----------------------------------
                         Number (in        Percentage        Number (in
                         thousands)          enrolled        thousands)        Percentage
-----------------  ----------------  ----------------  ----------------  ----------------
White                         3,175              54.1             1,238              21.1
Black                         2,883            69.4\a               784              18.9
Hispanic                      2,424              58.8             1,205            29.2\b
Other, non-                     465              54.9               217              25.6
 Hispanic
=========================================================================================
Total                         8,947              59.7             3,445              23.0
-----------------------------------------------------------------------------------------
Note:  Numbers may not add to totals due to rounding. 

\a Statistically different from rate for whites and Hispanics. 

\b Statistically different from rate for whites and blacks. 



                                        Table II.2
                         
                            Percentage of Uninsured Medicaid-
                          Eligible Children in 1996, by Race and
                                 Ethnicity and by Region

                                                                           Total
                                                                   ----------------------
                       White,      Black,                  Other,
                         non-        non-                    non-              Number (in
                     Hispanic    Hispanic    Hispanic    Hispanic  Percentage  thousands)
-----------------  ----------  ----------  ----------  ----------  ==========  ==========
Northeast                38.0        23.0        30.7         8.3         100         459
Midwest                  48.6        32.4        14.4         4.6         100         458
South                    37.1        30.6        27.4         4.9         100       1,486
West                     27.9         7.2        56.7         8.2         100       1,042
West excluding           38.8         6.3        47.7         7.2         100         481
 California
California               18.5         8.0        64.4         9.1         100         560
=========================================================================================
U.S.                     36.0        22.8        35.0         6.3         100       3,445
-----------------------------------------------------------------------------------------
Note:  Numbers may not add to totals due to rounding. 



                                        Table II.3
                         
                            Percentage of Uninsured Medicaid-
                          Eligible Children in 1996, by Parents'
                                 Birth Status and Region

                                                                         Total
                                                               --------------------------
                                      U.S.-born
                         U.S-born    child with      Foreign-
                        child and      foreign-    born child                  Number (in
                           parent   born parent    and parent       Percent    thousands)
-------------------  ------------  ------------  ------------  ============  ============
Northeast                    59.6          30.1          10.2           100           437
Midwest                      85.8           9.6           4.5           100           427
South                        71.6          19.6           8.1           100         1,394
West                         44.1          37.2          18.6           100           927
West excluding               64.0          23.6          12.3           100           434
 California
California                   26.6          49.3          24.2           100           493
=========================================================================================
U.S.                         63.8          24.8          11.0           100         3,185
-----------------------------------------------------------------------------------------
Note:  Table is based on 92 percent of uninsured Medicaid-eligible
children whose records matched with those of their parents.  Numbers
may not add to totals due to rounding. 

RELATED GAO PRODUCTS

Medicaid:  Early Implications of Welfare Reform for Beneficiaries and
States (GAO/HEHS-98-62, Feb.  24, 1998). 

Health Insurance:  Coverage Leads to Increased Health Care Access for
Children (GAO/HEHS-98-14, Nov.  24, 1997). 

Uninsured Children and Immigration, 1995 (GAO/HEHS-97-126R, May 27,
1997). 

Health Insurance for Children:  Declines in Employment-Based Coverage
Leave Millions Uninsured; State and Private Programs Offer New
Approaches (GAO/T-HEHS-97-105, Apr.  8, 1997). 

Employment-Based Health Insurance:  Costs Increase and Family
Coverage Decreases (GAO/HEHS-97-35, Feb.  24, 1997). 

Children's Health Insurance, 1995 (GAO/HEHS-97-68R, Feb.  19, 1997). 

Children's Health Insurance Programs, 1996 (GAO/HEHS-97-40R, Dec.  3,
1996). 

Private Health Insurance:  Millions Relying on Individual Market Face
Cost and Coverage Trade-Offs (GAO/HEHS-97-8, Nov.  25, 1996). 

Medicaid and Uninsured Children, 1994 (GAO/HEHS-96-174R, July 9,
1996). 

Health Insurance for Children:  Private Insurance Coverage Continues
to Deteriorate (GAO/HEHS-96-129, June 17, 1996). 

Health Insurance for Children:  State and Private Programs Create New
Strategies to Insure Children (GAO/HEHS-96-35, Jan.  18, 1996). 

Health Insurance for Children:  Many Remain Uninsured Despite
Medicaid Expansion (GAO/HEHS-95-175, July 19, 1995). 

Medicaid:  Spending Pressures Drive States Toward Program Reinvention
(GAO/HEHS-95-122, Apr.  4, 1995). 

Medicaid:  Restructuring Approaches Leave Many Questions
(GAO/HEHS-95-103, Apr.  4, 1995). 

Health Care Reform:  Potential Difficulties in Determining
Eligibility for Low-Income People (GAO/HEHS-94-176, July 11, 1994). 


*** End of document. ***