Low-Income Medicare Beneficiaries: Further Outreach and Administrative
Simplification Could Increase Enrollment (Letter Report, 04/09/99,
GAO/HEHS-99-61).

Pursuant to a congressional request, GAO provided information on the
level of enrollment in the Qualified Medicare Beneficiary (QMB) program
and the Specified Low-Income Medicare Beneficiary (SLMB) program,
focusing on: (1) the demographic and socioeconomic characteristics of:
(a) Medicare beneficiaries who enroll as a QMB or SLMB; and (b) Medicare
beneficiaries who qualify for QMB or SLMB but do not enroll; (2) reasons
why eligible beneficiaries are not enrolled; and (3) strategies to
increase enrollment.

GAO noted that: (1) although enrollment in QMB and SLMB has increased
since the programs were implemented, many potentially eligible Medicare
beneficiaries are not enrolled in these programs; (2) in 1996, about 2.2
million of an estimated 5.1 million potentially eligible Medicare
beneficiaries--about 43 percent--were not enrolled in either QMB or
SLMB; (3) in general, the characteristics of QMB and SLMB enrollees are
similar to individuals who are eligible but do not enroll, placing them
among the most vulnerable Medicare beneficiaries; (4) in addition to
having low income, these individuals tend to have health conditions
affecting their capacity to perform various activities; (5) the groups
differ in some respects, however, as beneficiaries who are eligible but
not enrolled are more likely to be 80 years of age or older or have no
health insurance coverage other than Medicare; (6) GAO's analysis also
indicates that QMB and SLMB enrollment can vary by specific demographic
characteristics; (7) for example, enrollment is relatively high among
beneficiaries who are disabled, in poor health, are members of minority
groups, are separated, or have never married; (8) conversely, enrollment
is lower for beneficiaries who are white, widowed, married, or have
Medicare coverage because of age rather than disability; (9) advocates
for low-income elderly and state officials GAO interviewed attributed
persistently low QMB and SLMB enrollment to limited program awareness
among beneficiaries and the programs' administrative complexity; (10)
potentially eligible individuals are perceived to simply be unaware of
these programs, their benefits, or their eligibility criteria; (11)
also, low enrollment in these programs is thought to result from state
cost-sharing obligations that limit states' incentives to notify and
enroll eligible individuals; (12) recently, the Health Care Financing
Administration (HCFA) and the Social Security Administration have
initiated efforts aimed at identifying strategies for increasing QMB and
SLMB enrollment; (13) HCFA has established a task force that is in the
process of identifying targets for increased enrollment and strategies
for reaching these goals; and (14) a number of states GAO contacted have
taken steps to simplify their application and enrollment processes, and
advocates and state officials who GAO interviewed suggest that expanded
administrative simplification efforts in conjunction with more creative
and targeted outreach could increase QMB and SLMB enrollment.

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

 REPORTNUM:  HEHS-99-61
     TITLE:  Low-Income Medicare Beneficiaries: Further Outreach and 
             Administrative Simplification Could Increase
             Enrollment
      DATE:  04/09/99
   SUBJECT:  Health care programs
             Health care services
             Beneficiaries
             Health statistics
             Health resources utilization
             Health insurance
             Elderly persons
             Eligibility criteria
IDENTIFIER:  Qualified Medicare Beneficiary Program
             Specified Low Income Medicare Beneficiary Program
             Medicare Qualifying Individuals Program
             Medicaid Program
             Medicare Program
             
******************************************************************
** 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 Congressional Requesters

April 1999

LOW-INCOME MEDICARE BENEFICIARIES
- FURTHER OUTREACH AND
ADMINISTRATIVE SIMPLIFICATION
COULD INCREASE ENROLLMENT

GAO/HEHS-99-61

Low-Income Medicare Beneficiaries

(101786)


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

  CPS - Current Population Survey
  ESRD - end-stage renal disease
  GPRA - Government Performance and Results Act
  HCFA - Health Care Financing Administration
  MCBS - Medicare Current Beneficiary Survey
  NGA - National Governors' Association
  OBRA - Omnibus Budget Reconciliation Act
  QI - Qualifying Individuals
  QMB - Qualified Medicare Beneficiary
  SCF - Survey of Consumer Finances
  SCHIP - State Children's Health Insurance Program
  SLMB - Specified Low-Income Medicare Beneficiary
  SSA - Social Security Administration
  SSI - Supplemental Security Income

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


B-282061

April 9, 1999

The Honorable Pete Stark
Ranking Minority Member, Subcommittee on Health
Committee on Ways and Means
House of Representatives

The Honorable Jim McDermott
House of Representatives

Medicare provides health insurance coverage to nearly 39 million
Americans who are elderly, disabled, or have end-stage renal disease
(ESRD).  However, the program's cost-sharing provisions--including
premiums, deductibles, and coinsurance--make participation in the
program difficult to afford for low-income individuals.  In 1995, the
annual cost-sharing liability for Medicare-covered services was
typically about $760 per beneficiary.  This liability represented
about 10 percent of income for a single person and about 15 percent
of income for couples at the federal poverty level.\1 While many
Medicare beneficiaries with low incomes have protection from these
costs through Medicaid--the federal-state health financing program
for low-income people--those with low incomes who do not qualify for
Medicaid face significant cost-sharing obligations. 

To assist low-income Medicare beneficiaries with potentially high
out-of-pocket costs, the Congress enacted three programs:  the
Qualified Medicare Beneficiary (QMB) program; the Specified
Low-Income Medicare Beneficiary (SLMB) program; and the Qualifying
Individuals (QI) program, whereby state Medicaid programs help bear
the beneficiary share of costs, which varies depending on the
beneficiary's income.  However, there has been continuing concern
about the level of enrollment in these programs.  Therefore, you
asked us to

  -- highlight the demographic and socioeconomic characteristics of
     (1) Medicare beneficiaries who enroll as a QMB or SLMB and (2)
     Medicare beneficiaries who qualify for QMB or SLMB but do not
     enroll,

  -- examine reasons why eligible beneficiaries are not enrolled, and

  -- identify strategies to increase enrollment. 

To perform our work, we conducted a statistical analysis of recent
surveys by the Health Care Financing Administration (HCFA), the
Census Bureau, and the Federal Reserve Board.  We also interviewed
officials at HCFA, the Social Security Administration (SSA), Medicaid
agencies in seven states, and advocates for low-income elderly.  We
conducted our work from November 1998 to March 1999 in accordance
with generally accepted government auditing standards.  (For a
detailed description of our scope and methodology, see app.  I.)


--------------------
\1 This level is based on federal guidelines prepared by the
Department of Health and Human Services.  The federal poverty level
in 1995 was $7,470 and $10,030 for couples. 


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

Although enrollment in QMB and SLMB has increased since the programs
were implemented, many potentially eligible Medicare beneficiaries
are not enrolled in these programs.  In 1996, about 2.2 million of an
estimated 5.1 million potentially eligible Medicare
beneficiaries--about 43 percent--were not enrolled in either QMB or
SLMB.  In general, the characteristics of QMB and SLMB enrollees are
similar to individuals who are eligible but do not enroll, placing
them among the most vulnerable Medicare beneficiaries.  In addition
to having low income, these individuals tend to have health
conditions affecting their capacity to perform various activities. 
The groups differ in some respects, however, as beneficiaries who are
eligible but not enrolled are more likely to be 80 years of age or
older or have no health insurance coverage other than Medicare.  Our
analysis also indicates that QMB and SLMB enrollment can vary by
specific demographic characteristics.  For example, enrollment is
relatively high among beneficiaries who are disabled, in poor health,
are members of minority groups, are separated, or have never married. 
Conversely, enrollment is lower for beneficiaries who are white,
widowed, married, or have Medicare coverage because of age rather
than disability. 

Advocates for low-income elderly and state officials we interviewed
attributed persistently low QMB and SLMB enrollment to limited
program awareness among beneficiaries and the programs'
administrative complexity.  Potentially eligible individuals are
perceived to simply be unaware of these programs, their benefits, or
their eligibility criteria.  Moreover, limited beneficiary awareness
is thought to be exacerbated by cultural and language barriers as
well as perceptions of social stigma related to enrolling in the
Medicaid-administered QMB and SLMB programs.  Enrollment can be
further hindered by a burdensome and complex application process that
can require beneficiaries to interact with more than one government
agency.  Also, low enrollment in these programs is thought to result
from state cost-sharing obligations that limit states' incentives to
notify and enroll eligible individuals. 

Recently, HCFA and SSA have initiated efforts aimed at identifying
strategies for increasing QMB and SLMB enrollment.  HCFA has
established a task force that is in the process of identifying
targets for increased enrollment and strategies for reaching these
goals.  SSA selected one state, Massachusetts, and 11 communities in
six other states to participate in a demonstration project to examine
the effects of various approaches on enrollment.  Further, a number
of states we contacted have taken steps to simplify their application
and enrollment processes, and advocates and state officials who we
interviewed suggest that expanded administrative simplification
efforts in conjunction with more creative and targeted outreach could
increase QMB and SLMB enrollment. 


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

Medicare is the nation's largest health insurance program and
provides coverage for a broad array of services.  However, many
beneficiaries purchase supplemental coverage to offset the program's
cost-sharing provisions--that is, premiums, deductibles, and
coinsurance.\2 To the extent that beneficiaries can purchase
insurance to supplement their Medicare coverage, they limit their
potential cost-sharing liability.  However, many low-income
persons--especially those with poor health--are less able to afford
such supplemental coverage. 

About 2.5 million persons who qualify for Medicare and are poor also
receive assistance from Medicaid, a joint federal-state program that
provides health care services for certain vulnerable and needy
individuals and families with low incomes and resources.\3 For those
who are eligible for full Medicaid coverage, the Medicare health care
coverage is supplemented by services that are available under their
state's Medicaid program, which may include prescription drugs and
long-term care services--generally not available under Medicare--as
well as payment of Medicare part B premiums.  Also, Medicare makes
payments for Medicare-covered services before the Medicaid program
makes any payments. 

To assist low-income Medicare beneficiaries with potentially high
out-of-pocket costs, the Congress established the QMB, SLMB, and QI
programs. 

  -- QMB, implemented in 1986,\4 is a benefit program for Medicare
     beneficiaries with incomes at or below 100 percent of the
     federal poverty level.  Under QMB, state Medicaid programs are
     responsible for these individuals' Medicare premiums,
     deductibles, and coinsurance. 

  -- SLMB, implemented in 1993,\5 requires state Medicaid programs to
     pay Medicare part B premiums (but not the deductibles or
     coinsurance) for individuals with incomes above 100 percent but
     less than 120 percent of the federal poverty level. 

  -- QI, implemented in 1998, requires state Medicaid programs to pay
     all of the Medicare part B premiums for individuals with incomes
     at least 120 percent but less than 135 percent of the federal
     poverty level, and to provide a small rebate of Medicare
     premiums for beneficiaries with incomes at least 135 percent but
     less than 175 percent of the federal poverty level.  The QI
     program is funded with $1.5 billion in federal dollars over a
     5-year period.\6 Because the funding amount is fixed, eligible
     individuals receive assistance on a first-come, first-served
     basis. 

These Medicare buy-in programs and full Medicaid have varying
eligibility criteria and benefits.  (See table 1.)



                          Table 1
          
             Medicaid Eligibility Criteria and
            Benefits for Medicare Beneficiaries
           Under Full Medicaid, QMB, SLMB, and QI

                                                Enrollment
                                                     as of
Progra  Eligibility                               December
m       criteria            Benefits                1998\a
------  ------------------  ----------------  ------------
Full    Low-income          Medicare part B      2,450,000
Medica  Medicare            premiums paid by
id      beneficiaries as    the state
        defined by each     Medicaid program
        state               and Medicaid
                            services,
                            including those
                            covered under
                            Medicare

QMB     Medicare            Medicare             2,420,000
        beneficiaries       premiums,
        whose (1) incomes   deductibles, and
        are at or below     coinsurance paid
        100 percent of the  by the state
        federal poverty     Medicaid
        level and (2)       program\c
        assets are no
        greater than twice
        the limit for
        Supplemental
        Security Income
        (SSI)\b

SLMB    Medicare            Medicare part B        290,000
        beneficiaries with  premiums paid by
        (1) incomes above   the state
        100 percent but     Medicaid
        less than 120       program\c
        percent of the
        federal poverty
        level and (2)
        assets no greater
        than twice the
        limit for SSI\b

QI      Medicare            Medicare part B         16,000
        beneficiaries who   premiums paid
        are otherwise       for income 120
        ineligible for      percent to less
        Medicaid with (1)   than 135 percent
        incomes at least    of the federal
        120 percent but     poverty level in
        less than 175       1999; a $2.23
        percent of the      premium
        federal poverty     contribution for
        level and (2)       income 135
        assets no greater   percent to less
        than twice the      than 175 percent
        limit for SSI\b     of the federal
                            poverty level
----------------------------------------------------------
\a Based on administrative data provided by HCFA. 

\b The asset limits for SSI are $2,000 for individuals and $3,000 for
couples. 

\c Individuals may also be eligible for Medicaid services. 


--------------------
\2 Part A--which covers inpatient care in a hospital or skilled
nursing facility, post-institutional home health care, and hospice
care--typically has no premiums, but deductibles for an inpatient
hospital period were $764 in 1998.  Beneficiaries pay no coinsurance
for the first 60 days of inpatient care, but they pay 25 percent of
the deductible for the 61st through 90th days, and 50 percent of the
deductible for hospitalization past the 90th day.  For part B--which
covers physician services, outpatient hospital services,
non-post-institutional home health care, and other health care
services--1998 premiums were $43.80 a month, or $526 a year.  Also,
beneficiaries must pay a coinsurance of 20 percent of allowable
expenses. 

\3 In 1996, Medicaid provided medical assistance to about 36 million
low-income individuals. 

\4 QMB was enacted as an optional benefit through the Omnibus Budget
Reconciliation Act (OBRA) of 1986.  The Medicare Catastrophic
Coverage Act of 1988 made the QMB benefit mandatory, effective
January 1, 1989. 

\5 SLMB was enacted under OBRA 1990, effective January 1, 1993. 

\6 Because the QI program did not become effective until 1998, we did
not examine enrollment in this program. 


   PROGRAM ENROLLMENT HAS
   INCREASED, BUT SOME OF THE MOST
   VULNERABLE ELIGIBLE MEDICARE
   BENEFICIARIES ARE NOT ENROLLED
------------------------------------------------------------ Letter :3

Enrollment in QMB has increased steadily since it was implemented.\7
However, nearly half of all Medicare beneficiaries who are eligible
for the QMB and SLMB programs are not enrolled.  Moreover, these
beneficiaries are some of the most vulnerable among the Medicare
population. 


--------------------
\7 Trend data on SLMB were not available. 


      QMB ENROLLMENT HAS INCREASED
---------------------------------------------------------- Letter :3.1

HCFA administrative records based on state-reported enrollment data
indicate that enrollment in QMB increased from over 760,000 in 1991
to over 2.4 million in 1998\8

(see fig.  1).  Following steady enrollment growth from 1991 to 1994,
enrollment has largely stabilized.  While enrollment appears to
increase sharply between 1994 and 1995, this increase largely
represents a change in states' reporting methods, which had
undercounted QMB enrollees.\9

   Figure 1:  Growth in QMB Part B
   Enrollment From 1991 to 1998

   (See figure in printed
   edition.)

Note:  In 1995, states changed their reporting method, resulting in
the apparent sharp increase in reported enrollment. 


--------------------
\8 Based on part B enrollment data.  Most beneficiaries have both
part A and part B Medicare coverage. 

\9 QMB enrollment was underreported prior to 1995 because some QMBs
were counted as full Medicaid recipients.  This was changed beginning
in 1995, explaining much of the apparent growth in QMB enrollment
between 1994 and 1995. 


      MANY QMB- AND SLMB-ELIGIBLE
      MEDICARE BENEFICIARIES ARE
      NOT ENROLLED
---------------------------------------------------------- Letter :3.2

While QMB enrollment has increased gradually over time, a relatively
low percentage of Medicare beneficiaries who are eligible for the QMB
and SLMB programs actually enroll.  Based on our analysis of the 1996
Medicare Current Beneficiary Survey (MCBS), an estimated 8.6 million
Medicare beneficiaries had income levels low enough to qualify them
for these programs.  Within this group, about 61 percent had assets
within the QMB and SLMB thresholds, based on data from the Survey of
Consumer Finance (SCF).  Considering both income and assets, we
estimate that about 5.1 million Medicare beneficiaries are
potentially eligible for the QMB or SLMB program, with MCBS reporting
about 2.9 million individuals enrolled in QMB or SLMB.  Therefore,
about 2.2 million--or 43 percent--of the estimated eligible
population are not enrolled.  Other analysts have examined enrollment
for QMB and enrollment for SLMB separately and found that enrollment
is higher in the QMB program but lower in the SLMB program, which
serves a population with incomes slightly higher than the QMB
population and with more limited benefits. 


      DEMOGRAPHIC PROFILES OF
      ENROLLED AND NONENROLLED
      ELIGIBLES ARE SIMILAR
---------------------------------------------------------- Letter :3.3

Based on our analysis of MCBS, the general profile of individuals who
are eligible but do not enroll in QMB and SLMB is similar to that of
program enrollees.\10 The characteristics of these two groups place
them among the most vulnerable Medicare beneficiaries.  In addition
to having a lower income than noneligible Medicare beneficiaries, QMB
and SLMB eligibles and enrollees have fewer years of education and
health conditions that limit their capacity to perform various
activities.  A relatively high percentage of both the QMB and SLMB
eligible and enrolled are female, single, living alone, or a member
of a minority group. 

While the general profiles of enrolled and nonenrolled QMB- and
SLMB-eligible individuals are similar, certain characteristics
distinguish them.  For example, QMB and SLMB enrollees are more
likely to be disabled or reside in a facility than those who are
eligible but not enrolled.  In contrast, individuals who are eligible
but not enrolled are more likely to not have health insurance
coverage other than Medicare or be 80 years of age or older. 


--------------------
\10 For our analysis, we examined the characteristics of individuals
who are eligible for QMB or SLMB based upon income alone. 


      ENROLLMENT IS HIGHEST FOR
      THOSE WHO ARE MOST
      VULNERABLE
---------------------------------------------------------- Letter :3.4

Our analysis indicates that enrollment was higher among some of the
most vulnerable beneficiaries--those in poor health; receiving
Medicare coverage because of a disability or ESRD; with difficulty
performing certain life activities; or residing in facilities, such
as a nursing home, assisted living facility, or mental health
facility.  Enrollment was also higher among individuals having 8 or
fewer years of education.  Conversely, enrollment was lower among
beneficiaries who had 13 or more years of education, were in better
health, had Medicare coverage due to age, or were living in the
community. 

QMB and SLMB enrollment was also associated with beneficiaries' race
and marital status.  For example, Asian- and African-Americans were
more likely to be enrolled than whites.  Also, beneficiaries who were
separated or never married were more likely to be enrolled than those
who were widowed or married.  (See app.  II for more detailed
information on the characteristics of QMB and SLMB enrollees,
individuals who are eligible but not enrolled, and other Medicare
beneficiaries.)


   LOW QMB AND SLMB ENROLLMENT
   ATTRIBUTED TO LIMITED PROGRAM
   AWARENESS AND ADMINISTRATIVE
   COMPLEXITY
------------------------------------------------------------ Letter :4

The state officials and advocates for low-income elderly who we
interviewed indicated that low QMB and SLMB enrollment persists
because of limited program awareness among beneficiaries and the
administrative complexity associated with the programs. 
Beneficiaries are perceived to have insufficient knowledge of the
programs, their benefits, and their eligibility criteria--a problem
exacerbated by cultural and language barriers and perceptions of
social stigma related to enrolling in Medicaid-administered programs. 
Furthermore, establishing QMB and SLMB eligibility can be a complex
process.  For potential beneficiaries, lengthy applications and
eligibility verification requirements can discourage them from
seeking enrollment.  For agencies, the division of financial and
programmatic responsibilities between the federal government and
states can provide a disincentive to assume full responsibility for
maximizing enrollment. 


      INSUFFICIENT AND INEFFECTIVE
      OUTREACH LIMITS PROGRAM
      AWARENESS
---------------------------------------------------------- Letter :4.1

Although the QMB and SLMB programs have been operable for a number of
years, most of those we interviewed reported that many potential
recipients do not enroll because they do not know the programs
existed.  Misperceptions about the programs are also thought to deter
some beneficiaries from enrolling.  For example, an individual who
meets the eligibility criteria might not apply because of a belief
that the program is intended only for poor people. Some potential
beneficiaries are thought not to apply because of their apprehensions
or misperceptions about their state's Medicaid estate recovery
practices.  These individuals may fear that, following their death,
their state will attempt to recover QMB and SLMB payments made on
their behalf through liens on their estate and jeopardize the
financial well-being of a surviving spouse or their children.  Other
potential beneficiaries think the programs are a form of welfare and
are unwilling to accept this type of assistance. 

Some states we interviewed attributed limited program awareness, in
part, to either a general lack of outreach efforts or the lack of
effective outreach.  They believe, for example, that current outreach
efforts are insufficient or ineffective in raising the level of
program awareness among beneficiaries with limited English language
skills or in allaying concerns regarding the acceptance of public
assistance.  Our analysis of MCBS data similarly suggests that
current outreach efforts may not be reaching all populations.  QMB
and SLMB enrollment is comparatively high for beneficiaries who are
ill or disabled or reside in facilities such as those that provide
long-term care.  Even without outreach, however, these individuals
are more apt to become enrolled because their health conditions
increase their number of encounters with the medical community--some
even reside in medical settings--and their caregivers have financial
incentives to ensure that they are covered.  In contrast, QMB and
SLMB enrollment is lower among beneficiaries who are aged, better
educated, in better health and with less need for medical care, or
living independently in the community--groups for whom outreach
efforts are more necessary for increasing program awareness or
addressing concerns or misperceptions. 

Through our interviews with states, we also found that most discussed
previous or ongoing outreach efforts.  Only one state reported new
outreach initiatives for increasing QMB and SLMB enrollment, and one
state reported targeting its outreach to specific groups. 


      ADMINISTRATIVE COMPLEXITY
      IMPEDES QMB AND SLMB
      ENROLLMENT
---------------------------------------------------------- Letter :4.2

Even with improved outreach, boosting enrollment in QMB and SLMB may
be undermined by the administrative complexity associated with
determining eligibility.  The application process is cumbersome and
lengthy, and other administrative processes must be coordinated among
various federal and state government agencies, given that Medicare is
administered by the federal government and Medicaid is administered
by the states. 

According to the state officials and advocates we interviewed, the
process for applying for QMB and SLMB benefits could be a key factor
limiting enrollment.  In some states, applicants are required to
complete the full Medicaid application, which can exceed 10 pages and
be difficult to read, given its small print.  In addition, applicants
may require the assistance of a state caseworker to complete the
application.  Some states require information that will allow the
verification of an applicant's reported resources--a process that can
be onerous and time-consuming to both applicants and state workers. 
Further, some states require applicants to have a face-to-face
interview at either a social service or an aging office, instead of
accepting applications over the phone, as other states do.  Requiring
face-to-face interviews likely impedes enrollment for those who are
homebound or concerned about perceived welfare stigma. 

Other administrative processes--typically those that require
coordination among state and federal agencies--can result in eligible
individuals' enrollment being delayed.  For example, state Medicaid
programs may need to coordinate with HCFA and SSA to verify
information such as enrollment in Medicare and income from Social
Security, creating the potential for administrative delays and
errors. 

Some advocates have also suggested that the financing of the QMB and
SLMB programs, with state cost-sharing responsibilities, has deterred
states from embracing the programs and created a disincentive for
states to conduct additional outreach or simplify the application
process.  A February 1998 National Governors' Association (NGA)
position on the financing arrangement supports this belief.  NGA
stated that it cannot support Medicare reform strategies, such as
increased cost-sharing obligations for the dually eligible, that
result in cost shifts to the states.\11 NGA further stated that
Medicare, as a federal program, should bear all of its costs, but if
it were to continue to make Medicaid responsible for meeting the
Medicare cost-sharing obligations of low-income beneficiaries,
"Congress should at a minimum clarify that copayments may be
reimbursed at Medicaid rather than Medicare rates.\12


--------------------
\11 National Governors' Association, Policy Positions (Washington,
D.C., Feb.  1998). 

\12 Since Medicaid reimbursement rates tend to be lower than Medicare
rates, such a change would result in cost savings to states.  Section
4714 of the Balanced Budget Act of 1997 (P.L.  105-33) clarified that
states were not required to provide payments for deductibles,
coinsurance, or copayments for the full Medicare cost-sharing amount
made under the state plan for services provided to individuals other
than Medicare beneficiaries. 


   ENHANCED OUTREACH AND
   SIMPLIFIED ENROLLMENT COULD
   INCREASE PARTICIPATION IN QMB
   AND SLMB
------------------------------------------------------------ Letter :5

The federal government has developed various strategies to boost
enrollment in QMB and SLMB.  A number of these strategies focus on
enhancing outreach to increase program awareness and simplifying the
enrollment process.  For example, as part of its Government
Performance and Results Act (GPRA) goals, HCFA has convened a task
force to develop an outreach, enrollment, and eligibility
simplification strategy for increasing enrollment of those who are
dually eligible.  SSA is conducting a pilot project intended to
increase referral of potential beneficiaries to state Medicaid
programs. 

The state officials and advocates we interviewed recommended that
outreach be improved through strategies such as increasing overall
outreach efforts; targeting groups that include large numbers of
eligible but nonenrolled individuals; and developing partnerships
with key stakeholders, such as seniors' advocates, area agencies on
aging, and other community-based organizations.  They also
recommended strategies for streamlining the application process and
providing flexibility in applying eligibility rules to make it easier
for eligible individuals to become enrolled in the programs. 


      HCFA INITIATIVES TO INCREASE
      ENROLLMENT
---------------------------------------------------------- Letter :5.1

Since the programs were developed, HCFA has made several efforts to
increase enrollment in QMB and SLMB.  Promotional efforts have
included mailing notices to prospective enrollees, distributing
pamphlets on the programs, advertising in the media, and developing a
section on QMB and SLMB in its Medicare handbook for beneficiaries. 
HCFA has also issued directives and letters to states providing
guidance on program administration and simplification.  For example,
in October 1998, HCFA wrote state Medicaid program directors
suggesting that they develop outreach and enrollment strategies
modeled on those used for the new State Children's Health Insurance
Program (SCHIP)--a strategy some advocates strongly support. 

In response to the Social Security Amendments of 1994, HCFA
established a list of newly eligible Medicare beneficiaries that
includes demographic information, such as income from Social
Security, which states can use to identify individuals potentially
eligible for QMB and SLMB benefits.\13 Currently, HCFA is seeking to
improve QMB and SLMB enrollment through one of its GPRA goals.  To
reach this goal, HCFA plans to

  -- establish targets for increased QMB and SLMB enrollment;

  -- develop an outreach, enrollment, and eligibility simplification
     strategy;

  -- identify best practices in collaboration with states; and

  -- measure progress toward meeting these goals. 

HCFA intends to recommend targets and best practices in summer 1999
and begin measuring progress toward these goals in fiscal year 2000. 


--------------------
\13 Section 154 of the Social Security Amendments of 1994 (P.L. 
103-432) directs the Secretary of Health and Human Services to
implement a method for obtaining information from newly eligible
Medicare beneficiaries that could be used to determine their QMB
eligibility and to transmit this information to the state in which
the beneficiary resides. 


      SSA EFFORTS TO INCREASE
      ENROLLMENT
---------------------------------------------------------- Letter :5.2

SSA is an important point of contact for those potentially eligible
for QMB and SLMB, not only because it is responsible for enrolling
new Medicare beneficiaries but because Social Security is a primary
source of income for many low-income beneficiaries.  However, a
majority of individuals file for Social Security benefits before age
65--when most become eligible for Medicare--and do not have ongoing
contact with SSA.\14

SSA's efforts to notify potentially eligible individuals include
sending program information in cost-of-living adjustment notices to
all Social Security recipients and providing QMB and SLMB information
and referral as part of the agency's in-person contacts and toll-free
telephone service.  SSA has also included information about the QMB
and SLMB programs in pamphlets available at Social Security offices
and distributed to interested individuals. 

SSA is currently conducting a demonstration project with selected
states to evaluate which strategies are most effective in increasing
the number of SSA referrals of potentially eligible beneficiaries to
state agencies.  The demonstration project will be conducted in
Massachusetts and 11 communities in six other states, which were
selected based on each participating state's offer to provide access
to a concentration of elderly, disabled, and low-income individuals. 
Under this project, SSA is testing four approaches.  In one approach,
SSA will use its death report process to identify potential buy-in
eligibles and refer them to the state's Medicaid office to file an
application for benefits.  In the other three approaches, SSA will
identify and send mailings to potentially eligible individuals in the
selected communities.  Respondents will be screened by SSA employees
and then referred to complete an application (1) with an SSA
employee; (2) with a state Medicaid official located in the SSA
office; or (3) with an official at the state Medicaid office,
typically at another location.  Final eligibility determinations are
still performed by the state Medicaid agency, regardless of the
approach.  The demonstration is scheduled to continue through the end
of 1999, and an evaluation of the project and findings on the
relative effectiveness of the referral methods is expected to be
released in spring 2000. 


--------------------
\14 When these individuals become eligible for Medicare, they are
automatically enrolled in part A and part B and, therefore, do not
have to contact SSA again to enroll in Medicare. 


      INCREASED AND MORE EFFECTIVE
      OUTREACH COULD INCREASE
      ENROLLMENT
---------------------------------------------------------- Letter :5.3

In addition to HCFA's and SSA's recent initiatives to increase QMB
and SLMB enrollment, the state officials and advocates we interviewed
recommended a number of strategies, some of which have been used, for
intensifying and broadening the range and scope of outreach efforts. 

  -- Target outreach to populations with particularly low enrollment: 
     individuals who are widowed, aged 65 or older, white, have 13
     years or more of education, or report good health status. 

  -- Target low-income Medicare beneficiaries with health conditions
     and high use of health care services, who are most likely to
     benefit from supplemental coverage of Medicare coinsurance and
     deductibles.  For example, Medicare benefits statements, which
     show Medicare's payments and the beneficiary share of the cost,
     could include a brief notice suggesting that low-income
     beneficiaries apply for QMB. 

  -- Use other methods and sources to provide information on QMB and
     SLMB.  For example, states could coordinate with local utility
     companies to include QMB and SLMB literature with mailings to
     subscribers.  Delaware's Division of Health and Social Services
     attempts one-on-one outreach at senior picnics, health fairs,
     and senior centers.  In Arizona, a state coalition enrolled
     volunteers to conduct door-to-door outreach.  In addition, for
     the SCHIP program, HCFA recommends that states allow application
     at a wide variety of sites, including public schools and
     school-based health clinics. 

  -- Enlist physicians and other health care professionals in
     outreach efforts, such as encouraging them to advise their
     low-income patients to apply for QMB. 

  -- Coordinate outreach with other programs providing assistance to
     low-income individuals.  For example, elderly pharmacy
     assistance programs can help identify individuals with ongoing
     prescription drug needs, who are potentially eligible for QMB
     and SLMB. 

  -- Establish partnerships with local stakeholders to increase QMB
     and SLMB enrollment.  For example, Tennessee and Arizona
     partnered with organizations such as religious organizations,
     advocacy groups, state and local agencies, voluntary health
     agencies, health professionals and providers, area agencies on
     aging, and other seniors groups to develop task forces to work
     on outreach, training, and enrollment. 

  -- Provide outreach information and applications in languages other
     than English. 

While improved outreach could improve enrollment, many of the
proposed strategies would likely require the commitment of additional
resources by states and HCFA. 


      SIMPLIFYING THE APPLICATION
      PROCESS AND ELIGIBILITY
      RULES COULD ALSO INCREASE
      ENROLLMENT
---------------------------------------------------------- Letter :5.4

State officials and advocates also suggested that additional efforts
are needed to simplify the application process and eligibility rules. 
Some approaches that they recommended include the following: 

  -- Use a shorter application form.  For example, some states have
     developed a one- or two-page application for QMB and SLMB. 

  -- Allow beneficiaries to declare the eligibility information they
     provide as true and accurate.  For example, Delaware allows a
     self-declaration that the applicant meets the asset requirements
     for enrollment in QMB and SLMB, rather than requiring
     documentation to verify assets. 

  -- Eliminate the need for applicants to come in person to Medicaid
     or other state agency offices to apply.  Some states have
     computerized portions of the eligibility determination, which
     could allow the testing of an electronic application. 
     Intermediaries such as area agencies on aging and
     community-based organizations could assist in preparing and
     transmitting applications electronically to state Medicaid
     offices.  Arizona offers help filling out applications via
     telephone.  New York is encouraging local counties to experiment
     with allowing those who are potentially eligible to enroll
     through area agencies on aging staffed with state intake
     workers.  Also, Tennessee performed a 3-month screening of
     potential beneficiaries at the U.S.  Department of Agriculture's
     commodity distribution centers.\15

  -- Relax program eligibility rules.  For example, Arizona excludes
     items such as household goods and personal effects, mineral and
     timber rights, burial and life insurance from countable
     resources.  Furthermore, most states have no asset requirements
     for SCHIP applicants. 

  -- Share automated data to improve enrollment. 

  -- Expand the use of retroactive eligibility so beneficiaries can
     be compensated for medical expenses incurred while their
     application is pending.  QMB and SLMB applications and
     eligibility determination can take 1 month or longer before
     enrollment is completed.  Potentially eligible individuals who
     have recently incurred medical expenses covered under QMB,
     therefore, may be more likely to complete the application
     process if they expect to be reimbursed for these expenses. 

For states that use a uniform application to establish eligibility
for multiple programs, developing a simpler application specifically
for QMB and SLMB may also have some drawbacks.  While a simpler
application may help improve QMB and SLMB enrollment, it would make
it more difficult to determine whether the applicant is also eligible
for other or more comprehensive programs for low-income individuals. 
For example, some states that maintain longer application forms and
require verification of assets use the information to screen the
individual for full Medicaid benefits and other programs such as
low-income housing or energy assistance.  In certain circumstances, a
streamlined QMB or SLMB application could hinder a state agency's
ability to identify applicants who would also qualify for more
comprehensive assistance or benefits. 


--------------------
\15 The Food and Nutrition Service, an agency of the U.S.  Department
of Agriculture, makes food available through various programs,
including the Emergency Food Assistance Program, the Commodity
Supplemental Food Program, and Nutrition Program for the Elderly. 


   CONCLUDING OBSERVATIONS
------------------------------------------------------------ Letter :6

As proposals to restructure and increase the long-term financial
strength of the Medicare program are considered in the Congress,
increased attention may be focused on the best approaches for
providing financial assistance to low-income Medicare beneficiaries. 
The persistence of relatively low enrollment in the QMB and SLMB
programs suggests that enhanced outreach or simplified enrollment
processes would be helpful in reaching a larger share of eligible
low-income Medicare beneficiaries.  Effective targeted outreach can
also serve as a means to optimize limited outreach resources. 
Assessment of ongoing efforts--including SSA's demonstration project,
SCHIP outreach and enrollment efforts by states, and HCFA's GPRA
efforts--could yield new strategies to increase QMB and SLMB
enrollment.  Successful approaches from these efforts could then be
widely disseminated to enhance outreach and enrollment. 


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

We obtained comments on a draft of our report from HCFA and SSA. 
HCFA generally agreed with the strategies for increasing enrollment
in the QMB and SLMB programs suggested by the advocates and state
officials we interviewed.  HCFA also indicated its commitment to
providing more effective outreach and removing administrative
barriers to enrollment and highlighted its current efforts under GPRA
to increase QMB and SLMB enrollment. 

HCFA also noted that our estimate of the population potentially
eligible for the QMB and SLMB programs is lower than their
forthcoming estimate.  Estimating the number of individuals eligible
for means-tested programs is challenging because most available
surveys have shortcomings of one kind or another.  For this reason,
we recognize that different methods can legitimately produce
different estimates of this population.  Further, as estimates of
this population are produced from surveys that are based on
statistical samples, these estimates are subject to sampling error so
that the actual level of enrollment is likely to be higher or lower
than the point estimate.  In our opinion, the differences among the
various estimates of this population narrow when these sampling
errors are taken into account. 

HCFA and SSA also noted that other researchers have found
significantly higher enrollment among QMB-eligible individuals than
SLMB-eligible individuals.  In addition, HCFA indicated that
combining these groups could mask their differences.  We acknowledge
in our report that other research has determined that the QMB program
reaches a larger portion of eligible individuals than does the SLMB
program.  We also acknowledge that demographic differences could
potentially exist in (1) the enrolled QMB and SLMB populations and
(2) the nonenrolled eligible QMB and SLMB populations.  However, our
study's objective was not to distinguish between these groups, but
rather to compare the enrolled and the nonenrolled eligible
populations for both programs.  Moreover, MCBS income data do not
permit differentiating nonenrolled QMB and SLMB eligibles, and given
the small number of SLMB enrolled and nonenrolled eligibles included
in the MCBS sample, discrete estimates about their characteristics
would not likely be reliable.  Nonetheless, this should not be a
significant limitation to our study's objective because, based on the
eligibility criteria for these programs for the time period we
examined, only about $2,000 in income separated an individual
eligible for QMB from one eligible for SLMB. 

Both HCFA and SSA suggested technical clarifications, which we
included where appropriate.  HCFA's written comments are provided as
appendix III. 


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

As agreed with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this letter until 30 days
after its issue date.  At that time, we will send copies to other
interested congressional committees and members and agency officials. 
We will also make copies available to others upon request. 

Please call me at (202) 512-7114 if you have any questions about the
information provided in this report.  The information presented in
this report was developed by N.  Rotimi Adebonojo, Senior Evaluator;
Wayne Turowski, Computer Specialist; and Mark Vinkenes, Senior Social
Science Analyst, under the direction of John Dicken, Assistant
Director. 

Sincerely yours,

Kathryn G.  Allen
Associate Director, Health Financing
 and Public Health Issues


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

We conducted an analysis of the 1996 Medicare Current Beneficiary
Survey (MCBS) on access to care to estimate the number and
characteristics of Medicare beneficiaries who enroll as a QMB or SLMB
as well as those who may qualify but do not enroll.  Given certain
limitations of MCBS, we used the March 1996 Current Population Survey
(CPS) and the 1995 Survey of Consumer Finances (SCF) to further
refine our estimates.  To examine reasons why eligible beneficiaries
do not enroll and identify strategies to increase enrollment, we
reviewed the available literature and interviewed representatives
from HCFA, which administers Medicare and Medicaid; SSA, which is
responsible for enrolling eligible individuals in Medicare; national
organizations that represent elderly and low-income persons; state
health insurance counseling agencies; and Medicaid agencies in
Arizona, California, Delaware, Michigan, Nebraska, New York, and
Tennessee.  We excluded the Qualifying Individuals program from our
review due to its recent enactment. 

We used MCBS to conduct our analysis because it (1) contains
comprehensive information on Medicare beneficiaries, including their
demographic characteristics, health status, and health care use, and
(2) relies on HCFA administrative records rather than self-reported
information for QMB and SLMB enrollment status.  This latter factor
is important because previous research suggests that QMB enrollees
and individuals who are eligible but not enrolled in the program are
not always aware of their enrollment status, which could affect the
reliability of our estimates.  Because MCBS does not contain
information on assets and only provides income information in ranges,
we also analyzed the 1995 SCF to obtain additional asset information
and the 1996 CPS March Supplement to obtain additional income
information. 

Using MCBS, we categorized Medicare beneficiaries as (1) enrolled in
QMB or SLMB, (2) eligible for QMB or SLMB but not enrolled, and (3)
ineligible for QMB or SLMB.  The first group includes any individual
enrolled in QMB or SLMB for at least 1 month.  The second group
consisted of beneficiaries with income less than or equal to $10,000,
no QMB or SLMB enrollment, and less than continuous coverage by full
Medicaid.  The third group consists of any Medicare beneficiary with
income greater than $10,000 and no QMB, SLMB, or Medicaid enrollment. 

For purposes of our analysis, we did not distinguish between QMB and
SLMB enrollees.  Likewise, individuals who were potentially eligible
for QMB were not distinguished from those who were eligible for SLMB. 
This is because the numbers of enrolled and potentially eligible SLMB
populations are relatively small and the resulting sampling errors
would have been too great to allow meaningful comparisons.  Also,
because MCBS measures income within a range rather than as a specific
amount, the survey precludes distinguishing individuals who are
potentially eligible for QMB from those who are eligible for SLMB. 

As shown in table I.1, the maximum income that an individual could
have to meet the SLMB income threshold of 120 percent of the federal
poverty level was $8,964; for a couple, this income threshold was
$12,036.  Thus, some individuals we classified as eligible for QMB or
SLMB based on the MCBS income range of less than or equal to $10,000
may have exceeded the actual income threshold for individuals. 
Similarly, some individuals we classified as not eligible for QMB or
SLMB may have met the income thresholds for couples. 



                         Table I.1
          
               QMB and SLMB Asset and Income
            Thresholds, Individuals 65 Years or
                        Older, 1995

                    Assets             Income per year
            ----------------------  ----------------------
Category    Individual      Couple  Individual      Couple
----------  ----------  ----------  ----------  ----------
QMB             $4,000      $6,000      $7,470     $10,030
SLMB             4,000       6,000       8,964      12,036
----------------------------------------------------------
To determine the extent to which the discrepancy in MCBS' income data
and the actual income requirements of the program influenced our
estimates of the eligible but nonenrolled population, we conducted an
analysis of income among Medicare beneficiaries using the 1996 CPS. 
Based on this analysis, using the SLMB income threshold of 120
percent of the federal poverty level instead of $10,000, we estimate
217,000 fewer Medicare beneficiaries could qualify for SLMB based on
income.  Therefore, our MCBS analysis, based on a $10,000 income
threshold, slightly overestimates the number of individuals
potentially eligible for QMB or SLMB. 

Because MCBS also does not include information on assets available to
Medicare beneficiaries, we analyzed the 1995 SCF.  As shown in table
I.1, in general, individuals qualifying for QMB or SLMB may not have
assets exceeding $4,000 in value ($6,000 for a couple).  While rules
for determining assets for QMB or SLMB eligibility are applied
differently by state, we generally used SSI eligibility definitions
for the purpose of establishing countable resources for QMB or SLMB
eligibility.  For example, we excluded from countable assets the
value of an individual's home and the first $1,500 in cash surrender
value of life insurance policies.  Using SCF, we estimate that
approximately 39 percent of Medicare beneficiaries with income of
$10,000 or less had countable assets above the QMB and SLMB
eligibility thresholds.  Thus, we deflated our MCBS estimate of the
number of eligible individuals based on income by 39 percent--our
estimate, based on SCF, of those who would not meet the QMB and SLMB
asset requirements. 

MEDICARE BENEFICIARY PROFILES



                           Table
          
              II.1: Percentage of QMB and SLMB
          Enrollees, Eligible but Nonenrolled QMBs
          and SLMBs, and Beneficiaries Ineligible
          for QMB or SLMB by Demographic and Other
                      Characteristics

                                                  Medicare
                                                 enrollees
                                 QMB or SLMB    ineligible
                   QMB or SLMB  eligible but    for QMB or
                      enrolled   nonenrolled          SLMB
----------------  ------------  ------------  ------------
Age
----------------------------------------------------------
Less than 65             32.2%         15.8%          6.9%
 years old
65 to 79 years            41.7          51.9          73.5
 old
80 years old or            2.1          32.2          19.6
 older

Education
----------------------------------------------------------
8 years or less           41.9          35.3          13.9
9 to 12 years             41.3          50.6          49.5
13 or more years           8.1          11.7          36.2

Other demographic characteristics
----------------------------------------------------------
Member of a               30.0          18.6           7.5
 minority group
Hispanic                  13.3          11.5           3.5
 ancestry
Female                    65.2          68.1          50.7
Single                    81.8          78.7          33.0
Live alone                44.3          48.5          24.5
Live in a                 20.7           4.6           0.9
 facility

Basis for Medicare
----------------------------------------------------------
Aged                      67.7          84.2          93.1
Disabled                  31.9          15.6           6.7
ESRD                       0.4           0.3           0.2

Insurance status
----------------------------------------------------------
Medicare only              9.3          35.4          16.6
Medicare and               2.9          45.7          79.2
 private
 insurance

Health status
----------------------------------------------------------
Fair or poor              49.4          34.6          21.2
Limits most or            23.7          20.2          11.3
 all social life

Physical difficulties
----------------------------------------------------------
Seeing                    14.1          13.0           6.5
Hearing                    9.4           8.9           6.1
Stooping or               47.6          38.6          24.9
 kneeling
Lifting 10                37.4          25.9          13.3
 pounds
Reaching over             15.4          13.8           6.9
 head
Writing                   13.2           9.3           4.8
Walking two               46.1          33.8          19.7
 blocks
----------------------------------------------------------
Source:  GAO analysis of the 1996 MCBS. 



                               Table II.2
                
                 Percentage of Individuals Potentially
                    Eligible for QMB or SLMB Who Are
                Enrolled, by Demographic Characteristics

                                                                Percen
                                                                  tage
                                                                enroll
Demographic characteristic                                          ed
--------------------------------------------------------------  ------
Age
----------------------------------------------------------------------
Less than 65 years old                                           50.6%
65 years old or older                                             28.9

Education
----------------------------------------------------------------------
8 years or less                                                   37.5
9 to 12 years                                                     29.1
13 or more years                                                  25.9

Race
----------------------------------------------------------------------
American Indian                                                   40.9
Asian/Pacific Islander                                            67.2
African American                                                  43.6
Caucasian                                                         30.2
Other                                                             38.4

Marital status
----------------------------------------------------------------------
Married                                                           30.2
Widowed                                                           27.6
Divorced                                                          37.4
Separated                                                         43.7
Never married                                                     52.9

Residence
----------------------------------------------------------------------
Community (independent)                                           29.6
Facility                                                          69.2

Basis for Medicare
----------------------------------------------------------------------
Aged                                                              28.8
Aged With ESRD                                                    74.1
Disabled                                                          50.7
Disabled With ESRD                                                58.5
ESRD                                                              39.8
----------------------------------------------------------------------
Source:  GAO analysis of the 1996 MCBS. 



                               Table II.3
                
                 Percentage of Individuals Potentially
                    Eligible for QMB or SLMB Who Are
                  Enrolled, by Health Characteristics

                                                                Percen
                                                                  tage
                                                                enroll
Health characteristic                                               ed
--------------------------------------------------------------  ------
Health status
----------------------------------------------------------------------
Excellent                                                         23.8
Very good                                                         22.2
Good                                                              32.6
Fair                                                              41.1
Poor                                                              43.4

Health limits social life
----------------------------------------------------------------------
No                                                                28.7
Some                                                              41.6
Most                                                              36.4
All                                                               38.2

Have difficulty stooping/kneeling
----------------------------------------------------------------------
No                                                                30.2
Little                                                            27.0
Some                                                              33.3
A lot                                                             33.6
Unable                                                            43.4

Have difficulty lifting 10 pounds
----------------------------------------------------------------------
No                                                                27.6
Little                                                            32.6
Some                                                              35.0
A lot                                                             35.9
Unable                                                            45.8

Have difficulty reaching over head
----------------------------------------------------------------------
No                                                                30.4
Little                                                            38.4
Some                                                              40.9
A lot                                                             32.3
Unable                                                            40.9

Have difficulty writing
----------------------------------------------------------------------
No                                                                30.2
Little                                                            34.6
Some                                                              43.5
A lot                                                             39.2
Unable                                                            49.0

Have difficulty walking two to three blocks
----------------------------------------------------------------------
No                                                                27.2
Little                                                            32.2
Some                                                              33.5
A lot                                                             38.8
Unable                                                            41.5
----------------------------------------------------------------------
Source:  GAO analysis of the 1996 MCBS. 




(See figure in printed edition.)Appendix III
COMMENTS FROM THE HEALTH CARE
FINANCING ADMINISTRATION
========================================================== Appendix II



(See figure in printed edition.)



(See figure in printed edition.)


*** End of document. ***