Veterans' Health Care: Use of VA Services by Medicare-Eligible Veterans
(Letter Report, 10/24/94, GAO/HEHS-95-13).

Medicare-eligible veterans make substantial use of Department of
Veterans Affairs (VA) services not extensively covered under Medicare.
GAO found that many of these veterans turn to VA specifically to obtain
such services, particularly prescription drugs, inpatient psychiatric
care, and long-term nursing care. Also, many Medicare-eligible veterans
who use VA health care facilities have lower incomes and less private
insurance than those who rely solely on Medicare, suggesting that
out-of-pocket costs may have influenced veterans to turn to VA for
health care. Changes to Medicare or veterans health benefits made as a
result of health care reform could significantly affect future demand
for VA health care services. Medicare changes that would add benefits,
such as outpatient prescription drugs, or reduce beneficiary cost
sharing could lower demand for VA health care services. On the other
hand, VA benefit changes, such as the elimination of restrictions on
access to outpatient services, improved access to care, and expanded
entitlement to free care, could boost demand for VA health care.
Finally, the historic reluctance of Medicare beneficiaries to enroll in
health maintenance organizations could reduce their willingness to
enroll in VA health plans as long as traditional fee-for-service care
remains available under Medicare.

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

 REPORTNUM:  HEHS-95-13
     TITLE:  Veterans' Health Care: Use of VA Services by 
             Medicare-Eligible Veterans
      DATE:  10/24/94
   SUBJECT:  Health care cost control
             Health care programs
             Medicare programs
             Long-term care
             Veterans benefits
             Veterans hospitals
             Health insurance
             Health services administration
             Health care planning
             Projections
IDENTIFIER:  Health Security Act
             Clinton Health Care Plan
             National Health Care Reform Initiative
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


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


Report to the Chairman, Subcommittee on Oversight and Investigations,
Committee on Veterans' Affairs,
House of Representatives

October 1994

VETERANS' HEALTH CARE - USE OF VA
SERVICES BY MEDICARE-ELIGIBLE
VETERANS

GAO/HEHS-95-13

VA:  Veterans' Medicare Coverage and Use


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

  HCFA - Health Care Financing Administration
  HMO - health maintenance organization
  VA - Department of Veterans Affairs

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


B-256395

October 24, 1994

The Honorable Lane Evans
Chairman, Subcommittee on Oversight
 and Investigations
Committee on Veterans' Affairs
House of Representatives

Dear Mr.  Chairman: 

Most of America's veterans now have public or private insurance
programs that provide alternatives to health care available through
the Department of Veterans Affairs (VA).  When veterans have multiple
health care options, changes in one option can have unforeseen
repercussions on the others.  Many changes were proposed for both
programs during this year's health care reform debate, including
expanding entitlement to free VA health care services and adding a
Medicare outpatient prescription drug benefit. 

Medicare-eligible veterans form a substantial portion of VA's user
group, accounting for almost one-half of the more than 2.2 million
veterans using VA health care services in 1990.\1

To get a better idea of how various health care reforms might affect
the demand for VA health care services from this user group, you
asked us to analyze factors that might affect veterans' choices to
obtain care under Medicare or through VA.  As agreed with your
office, we focused our review on the following: 

  Determining the extent to which veterans use VA to obtain services
     not extensively covered under Medicare, such as prescription
     drugs, nursing home care, and routine dental care. 

  Comparing the income and private health insurance coverage of
     Medicare-eligible veterans who chose Medicare, VA, or a
     combination of the two. 

  Examining the potential effects of proposed reforms of VA and
     Medicare benefits on Medicare-eligible veterans' demand for VA
     health care services on the basis of the above data. 


--------------------
\1 When veterans have both Medicare and VA coverage, they
overwhelmingly use Medicare.  In 1990, for example, almost 62 percent
of Medicare-eligible veterans used Medicare, fewer than 7 percent
used VA, and fewer than 8 percent used both.  The remaining 24
percent used no services under either program.  See Veterans' Health
Care:  Most Care Provided Through Non-VA Programs (GAO/HEHS-94-104BR,
Apr.  25, 1994). 


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

Medicare-eligible veterans make substantial use of VA services not
extensively covered under Medicare.  Our analysis suggests that many
Medicare-eligible veterans turn to VA specifically to obtain several
of these services, particularly prescription drugs, inpatient
psychiatric care, and long-term nursing home care.  Also,
Medicare-eligible veterans who use VA health care facilities
generally have lower incomes and less private insurance than those
who rely solely on Medicare.  This suggests that out-of-pocket costs
may have influenced these veterans' decisions to use VA for health
care services. 

Changes to Medicare or veterans health benefits made as a result of
health care reform could significantly affect future demand for VA
health care services.  Changes in Medicare to add benefits (such as
outpatient prescription drugs) or to reduce beneficiary cost sharing
could reduce demand for VA health care services.  On the other hand,
changes in VA benefits, such as the elimination of restrictions on
access to outpatient services, improved access to care, and expanded
entitlement to free care, could increase demand for VA health care. 
Finally, the historic reluctance of Medicare beneficiaries to enroll
in health maintenance organizations (HMOs) could reduce their
willingness to enroll in VA health plans as long as traditional
fee-for-service care remains available under Medicare. 


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

In 1990, VA provided health care services to over 2.2 million of the
nation's estimated 28.2 million veterans through a system of 158
medical centers, 240 outpatient clinics, 126 nursing homes, and 32
domiciliaries.\2 Of those who used VA health care services in 1990,
almost one-half were eligible for Medicare--a federal health
insurance program that covers almost all Americans aged 65 and older
and certain individuals under 65 who are disabled or suffer from
kidney failure. 

While VA offers an extensive array of services, it has eligibility
and entitlement provisions that limit the extent to which individual
veterans may be eligible for and entitled to services.  For example,
priority for receiving VA hospital and nursing home care is divided
into two categories--mandatory and discretionary.  VA must provide
cost-free hospital care and, if space and resources are available,
may provide nursing home care to veterans in the mandatory care
category.  VA may provide hospital and nursing home care to those in
the discretionary category if space and resources are available in VA
facilities. 

Included in the mandatory care category are veterans who have
service-connected disabilities,\3 are former prisoners of war, served
during the Mexican border period or World War I, or have
nonservice-connected disabilities and are unable to defray the cost
of care.\4 Veterans eligible for Medicaid, receiving VA pensions, or
having financial resources below a prescribed level are considered
unable to defray the cost of necessary care. 

VA must furnish comprehensive outpatient medical services to veterans
who have service-connected disabilities rated at 50 percent or more. 
VA may provide comprehensive outpatient care to veterans who (1) are
former prisoners of war, (2) served during World War I or the Mexican
border period, (3) are housebound or in need of aid and attendance,
or (4) are participants in VA-approved vocational rehabilitation
programs. 

VA must furnish all outpatient services needed for the treatment of
conditions related to any veteran's service-connected disability
regardless of the veteran's disability rating.  VA must also provide
hospital-related outpatient care to veterans (1) with
service-connected disabilities rated at 30 or 40 percent or (2) whose
annual incomes do not exceed VA's pension rate for veterans in need
of regular aid and attendance.\5 \6 VA may, to the extent resources
permit, furnish hospital-related outpatient care to all veterans not
otherwise entitled to outpatient care.  Additional restrictions apply
to the availability of dental care. 

Once in the VA system, however, veterans are generally offered a
broader range of services with fewer limitations and less cost
sharing than are available under other public or private health
benefits programs such as Medicare.  For example, VA offers coverage
of outpatient prescription drugs and dental care that are not
available under Medicare.  VA also has no limit on the number of days
of care that can be obtained in VA-operated hospitals and nursing
homes, while Medicare limits its coverage of general hospital care to
90 days per benefit period (with 60 nonrenewable reserve days
available during the beneficiary's lifetime) and its coverage of
inpatient psychiatric care to 90 days per benefit period with a
lifetime limit of 190 days.\7 Medicare's coverage of nursing home
care is limited to post-hospital skilled nursing care.  In addition,
VA generally requires no cost sharing for inpatient hospital care
while Medicare requires beneficiaries to pay substantial deductibles
and copayments.\8


--------------------
\2 Domiciliaries provide shelter, food, and necessary medical care on
an ambulatory self-care basis to veterans who are disabled by age or
disease but not in need of skilled nursing care or hospitalization. 

\3 A service-connected disability is one that results from an injury
or disease or other physical or mental impairment incurred or
aggravated during military service.  VA determines if veterans have
service-connected disabilities and, for those with such disabilities,
assigns ratings of from 0 to 100 percent based on the severity of the
disability.  These ratings form the basis both for determining the
amount of compensation paid to the veterans and the types of health
care services to which they are eligible and entitled. 

\4 Veterans exposed to certain toxic substances during the Vietnam
War, to ionizing radiation, or to environmental hazards during the
Persian Gulf War are also included in the mandatory care category for
treatment of conditions that may be related to such exposures. 

\5 Hospital-related care refers to those outpatient services needed
to (1) prepare for a hospital admission, (2) obviate the need for a
hospital admission, or (3) complete treatment begun during a hospital
stay. 

\6 In 1991, the income limits were either $11,409 or less for
veterans with no dependents or $13,620 or less if married or single
with one dependent, plus $1,213 for each additional dependent. 

\7 A benefit period begins with admission to a hospital and ends when
the beneficiary has been out of the hospital or any other facility
providing skilled nursing or rehabilitation services for 60
consecutive days. 

\8 In 1993, Medicare beneficiaries were required to pay an inpatient
hospital deductible amount of $676 and copayments of $169 per day for
inpatient stays of 61-90 days, $338 a day for stays of 91-150 days,
and 100 percent of costs for stays beyond 150 days.  In addition,
they had to pay a copayment of 20 percent of approved charges for
doctors' fees and other professional fees.  VA requires cost sharing
only for higher-income veterans with no service-connected
disabilities.  It requires such veterans to pay the lesser of the
cost of care or $676 plus $10 a day for the first 90 days of care and
$338 plus $10 a day for each additional 90 days. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :3

Our review consisted of four main segments.  The first segment
centered on analyzing Medicare-eligible veterans' use of a variety of
inpatient and outpatient services in fiscal year 1990 using
nationwide data files maintained by VA and by the Health Care
Financing Administration (the federal administrator of Medicare)
within the Department of Health and Human Services.\9 The second
segment dealt with the one outpatient service--prescription
drugs--not covered in the nationwide data for 1990.  Because VA does
not maintain a centralized database for prescription drugs and
because most VA medical centers do not maintain readily accessible
files that go back to 1990, we determined the number of prescriptions
filled for Medicare-eligible veterans during fiscal year 1993 and
VA's costs for acquiring the drugs from 10 of VA's 158 medical
centers.  The third segment dealt with determining the income, health
insurance, and other characteristics of Medicare-eligible veterans. 
We used VA's 1987 Survey of Veterans because it is the most recent
complete source of such data.\10 Finally, we based our analysis of
the potential effects of health reform proposals on Medicare-eligible
veterans' use of VA health care services primarily on the
Administration's proposed Health Security Act (S.  1757/H.R.  3600),
including the Mitchell (S.2357) and Gephardt proposals, and the
Dole/Packwood proposal (S.  2374) because they have provisions
relating specifically to VA. 

Our review relied extensively on computerized data from VA, the
Health Care Financing Administration, and other federal agencies that
were obtained and verified during prior reviews.  However, we
verified some of the data fields used in this review.  (App.  I
describes our methodology in greater detail.) Our work was done in
accordance with generally accepted government auditing standards
between January 1994 and August 1994. 


--------------------
\9 We used 1990 because we had developed a database on the population
of Medicare-eligible veterans through an earlier study and we also
had Medicare and VA computerized treatment records for that year. 

\10 VA, in conjunction with the Bureau of the Census, conducted the
survey to evaluate its programs and to assess the status and
well-being of veterans across the nation.  Based on survey
participants' responses to questions concerning Medicare eligibility
and use of VA and Medicare health care services, we grouped
respondents who were eligible for Medicare into dual users, VA-only
users, and Medicare only users.  We then analyzed income and
insurance coverage of the three groups.

In 1993, VA conducted an updated survey, but the data were not yet
available at the time we completed our review. 


   USE OF VA SERVICES NOT
   EXTENSIVELY COVERED UNDER
   MEDICARE IS SUBSTANTIAL
------------------------------------------------------------ Letter :4

Medicare-eligible veterans were making substantial use of VA health
care services that were not extensively covered under Medicare.  We
reviewed six such services:  outpatient prescription drugs; dental,
audiology, and optometry care; inpatient psychiatric care; and
nursing home care.  For several of the services, the data indicate
that Medicare-eligible veterans may be using VA specifically to
obtain these services. 


      MOST MEDICARE-ELIGIBLE
      VETERANS RECEIVED OUTPATIENT
      PRESCRIPTION DRUGS
---------------------------------------------------------- Letter :4.1

Medicare-eligible veterans who use VA for all or part of their
medical services receive substantial amounts of outpatient
prescription drugs.  At the 10 medical centers where we analyzed
prescription drug usage, about 72 percent of the 87,607
Medicare-eligible veterans using VA health services during fiscal
year 1993 received outpatient prescription drugs.  On average, they
received 30 prescriptions during the year at an average acquisition
cost to VA of $374 per recipient.\11 While most of the
Medicare-eligible veterans at the 10 centers received prescriptions
that cost VA less than $200 for the year, about 8 percent received
prescriptions costing VA $1,000 or more (see fig.  1).  For dual
users of VA and Medicare services, the maximum amount of
prescriptions received by one veteran was 408 prescriptions costing
VA $18,199; for VA-only users, the maximum was 420 prescriptions
costing VA $17,303. 

   Figure 1:  Medicare-Eligible VA
   Users Receiving Outpatient
   Prescriptions, by VA Costs
   (1993)

   (See figure in printed
   edition.)

Dual users may be coming to VA in part to obtain prescription
services they cannot obtain under Medicare.  Dual users had fewer VA
outpatient visits than VA-only users but obtained about the same
number of prescriptions.  Dual users averaged less than 10 outpatient
visits compared with over 13 for VA-only users; further analysis
showed similar variation across all age groups of Medicare-eligible
veterans.  This suggests that dual users obtained at least a portion
of their outpatient care through Medicare or other sources.  Similar
differences between VA-only and dual users did not, however, exist in
use of VA outpatient prescription drugs.  Although dual users were
somewhat less likely than VA-only users to receive prescription
drugs, there was little difference between VA-only users and dual
users in the average number of prescriptions received or the average
value of prescriptions received (see table 1).  This suggests that VA
coverage of outpatient prescription drugs coupled with the lack of
similar coverage under Medicare is one of the factors dual users
consider in deciding between use of their VA and Medicare benefits. 



                          Table 1
          
             Prescription Drug Use by Medicare-
             Eligible Veterans Using Outpatient
             Services at 10 Selected VA Medical
                   Centers, by Age (1993)



                         VA-             VA-  Dual     VA-
Characterist    Dual    only    Dual    only  user    only
ic             users   users   users   users     s   users
------------  ------  ------  ------  ------  ----  ------
Percent           72      78      69      77    68      77
 receiving
 prescriptio
 ns
Average           13      17       9      11     9      12
 number of
 outpatient
 visits
Average           34      35      29      29    28      27
 number of
 prescriptio
 ns
Average cost    $473    $434    $366    $349  $324    $305
 of
 prescriptio
 ns
----------------------------------------------------------
As table 1 shows, Medicare-eligible veterans under 65 years of age
also had more outpatient visits and received more prescription drugs
than older age groups.  This is consistent with Medicare eligibility
criteria.  People 65 years of age or older are generally eligible for
Medicare regardless of their health status, while people under 65
have to be disabled to qualify for Medicare.  It is not surprising,
therefore, that we found that Medicare-eligible veterans under the
age of 65 are in greater need of health services in general,
including prescription drugs. 


--------------------
\11 VA's cost is based on the cost of acquiring the ingredients for
the prescriptions and is substantially lower than the market value of
the prescriptions because (1) federal law requires pharmaceutical
companies to charge VA no more than 76 percent of the nonfederal
average manufacturer price of prescription ingredients and (2)
besides ingredient costs, retail pharmacies generally add overhead
and dispensing fees to the price that they charge consumers. 


      ONE IN FIVE
      MEDICARE-ELIGIBLE VETERANS
      VISITED A DENTAL, HEARING,
      OR VISION CLINIC
---------------------------------------------------------- Letter :4.2

About 22 percent of the approximately 1 million Medicare-eligible
veterans using VA for outpatient services in 1990 used dental,
hearing, or vision services, services not generally covered under
Medicare.  Compared with outpatient prescription drugs, the data for
these services do not show as clear a trend with regard to whether
dual users are coming to VA specifically to obtain these services or
whether they are receiving the services incidental to receipt of
other outpatient care.  Dual users were slightly less likely than
VA-only users to use dental and optometry services but slightly more
likely than VA-only users to use audiology services (see table 2). 
The frequency of visits to these clinics was about the same for
VA-only users and dual users--five clinic visits for dental services
and two clinic visits each for audiology and optometry services. 



                          Table 2
          
          Percentage of Medicare-Eligible Veterans
              Using VA Outpatient Services Who
            Received Dental, Vision, or Hearing
                      Services (1990)

                                                   VA-only
Clinic                            Dual users         users
--------------------------------  ----------  ------------
Dental                                     8            11
Hearing                                    8             7
Vision                                     8            10
At least one of the three                 20            24
 clinics
----------------------------------------------------------

      VA IS THE PRIMARY SOURCE FOR
      INPATIENT PSYCHIATRIC CARE
---------------------------------------------------------- Letter :4.3

While Medicare covers some inpatient psychiatric care,\12 VA is the
major provider of inpatient psychiatric care for Medicare-eligible
veterans.  In 1990, over 71 percent of the 41,271 Medicare-eligible
veterans using inpatient psychiatric care used only VA for inpatient
psychiatric care, our analysis of VA and Medicare treatment records
shows.  By contrast, about 24 percent used only Medicare inpatient
psychiatric care and 5 percent used both VA and Medicare. 

Medicare-eligible veterans using VA psychiatric care also tend to
have longer lengths of stay than those who obtain this care through
Medicare.  On average, users of VA facilities had 70 days of
inpatient psychiatric stay in 1990, compared with 39 days for those
who used non-VA facilities paid for under Medicare.  While the stays
by VA-only users were longer in all age categories, they became more
pronounced among the oldest veterans.  As figure 2 shows, stays for
veterans younger than 65 years old averaged 66 days for users of VA
facilities and 40 days for users of non-VA facilities paid for under
Medicare, while stays for veterans 75 years old and older averaged 93
days for VA facilities and 29 days for non-VA facilities.  The
significant difference in average length of stay between VA users and
Medicare users could be due in part to the duration limits imposed
under Medicare, particularly for veterans in the older age groups who
may have exceeded the lifetime limit of 190 days on inpatient
psychiatric care. 

   Figure 2:  Average Length of
   Stay for Inpatient Psychiatric
   Care by Medicare-Eligible
   Veterans at VA and Non-VA
   Facilities (1990)

   (See figure in printed
   edition.)


--------------------
\12 Medicare's coverage for inpatient psychiatric care is limited to
190 days in a person's lifetime. 


      VA IS AN IMPORTANT SOURCE
      FOR LONG-TERM NURSING HOME
      CARE
---------------------------------------------------------- Letter :4.4

While Medicare is the primary source of short-term skilled nursing
home care for Medicare-eligible veterans,\13 VA also plays an
important role in meeting the nursing home needs of veterans,
particularly for those whose care needs exceed the limits of Medicare
coverage.  In 1990, 68 percent of the 61,524 Medicare-eligible
veterans using nursing home care relied solely on Medicare, 31
percent relied solely on VA, and 1 percent relied on both.  However,
on average, VA users had 236 days of nursing home care in 1990,
compared with 40 days for Medicare users. 

The significant difference in the average length of stay between VA
users and Medicare users is likely due to differences in nursing home
coverage under the two programs.  Medicare covers only short-term
skilled nursing home care following a hospitalization, whereas VA
covers both short-term and long-term nursing home care and both
skilled and intermediate nursing home care.  Veterans who need care
that exceeds the limits of Medicare coverage, either because of level
of care or length of coverage limits, must rely on VA or other
programs, such as Medicaid, the primary source of public funding for
nursing home care, to meet those needs.\14

As with inpatient psychiatric care, usage patterns for VA long-term
care vary somewhat by age (see fig.  3).  The difference between VA
users and Medicare users is greatest among veterans 65 years old and
younger and 75 years old or older.  Again, this is consistent with
what one would expect if users tended to rely on VA for long-term
care.  Nursing home usage generally increases with age; those 75
years of age or older are most likely to be in nursing homes. 
Persons younger than 65 tend to use nursing homes because of severe,
debilitating diseases such as acquired immune deficiency syndrome
(AIDS) and multiple sclerosis or permanent disabilities such as
spinal cord injuries.  Persons older than 75 tend to have longer
stays in nursing homes than their counterparts ages 65 to 74 because
they have more chronic illnesses and may no longer have the informal
network of caregivers to allow them to return to their homes. 

   Figure 3:  Average Nursing Home
   Length of Stay, by Age and
   Source of Payment (1990)

   (See figure in printed
   edition.)


--------------------
\13 Medicare's nursing home coverage is limited to skilled nursing
care following a hospitalization and allows 20 days of nursing home
care during a benefit period without a copayment.  In 1993, patients
had copayments of $84.50 per day for days 21 through 100 and had to
pay all costs after 100 days. 

\14 Medicaid is a combined federal and state program of medical
assistance to certain categories of low-income persons. 


   USERS OF VA FACILITIES
   GENERALLY HAVE LOWER INCOMES
   AND LESS PRIVATE INSURANCE
------------------------------------------------------------ Letter :5

Medicare-eligible veterans who use VA services tend to have lower
incomes and less private insurance coverage than those who rely on
Medicare.  This suggests that out-of-pocket costs and the lack of
private insurance to cover them may have influenced these veterans in
their decision to use VA for health care services. 


      LOW-INCOME MEDICARE-ELIGIBLE
      VETERANS MORE LIKELY TO USE
      VA
---------------------------------------------------------- Letter :5.1

Our analysis of data from VA's 1987 Survey of Veterans shows that VA
users generally had lower incomes than Medicare-only users.  About 54
percent of the approximately 308,000 dual users and 36 percent of the
approximately 318,000 VA-only users had annual family incomes below
$10,000, compared with 30 percent of the approximately 3,675,000
Medicare-only users.  At the other end of the scale, 13 percent of
Medicare-only users had incomes higher than $40,000, while only 3
percent of dual users and 5 percent of VA-only users were at this
level (see fig.  4). 

   Figure 4:  Income Distribution
   of VA-Only, Dual, and
   Medicare-Only Users (1987)

   (See figure in printed
   edition.)

Since the time the VA survey was conducted, the situation may have
changed.  The Medicare Catastrophic Coverage Act of 1988 requires
state Medicaid programs to pay the Medicare part A and part B
premiums and all deductibles and coinsurance for most Medicare
beneficiaries with incomes below 100 percent of the federal poverty
level ($9,430 for a family of two in 1992).\15 Under the Omnibus
Budget Reconciliation Act of 1990, this requirement was made
effective January 1, 1991.  The act also required state Medicaid
programs to begin paying by 1995 the part B premiums of Medicare
beneficiaries with incomes of 120 percent or less of the federal
poverty level. 

As a result of these changes, veterans with incomes below the federal
poverty level can now obtain Medicare-covered services from
private-sector providers with no out-of-pocket expense.  Their
financial advantage in coming to VA for services may thus have
largely disappeared.  VA recently conducted a new survey of veterans
that could enable us to determine whether the availability of free
care under Medicare resulted in a decline in VA use by low-income
Medicare-eligible veterans, but data from the new survey are not yet
available. 


--------------------
\15 Part A, Hospital Insurance, helps to pay for inpatient hospital
care, posthospital care in skilled nursing facilities, posthospital
home health services, and hospice care.  Part B, Supplementary
Medical Insurance, supplements part A by helping to pay for doctors'
services, outpatient hospital services, and a number of other medical
services and supplies. 


      VA USERS LESS LIKELY TO HAVE
      INSURANCE COVERAGE
---------------------------------------------------------- Letter :5.2

Medicare-eligible users of VA services (both VA-only and dual users)
were much less likely to have private health insurance to help pay
for Medicare copayments, deductibles, and coinsurance than were
Medicare-only users.  Most Medicare-eligible veterans are retired or
no longer able to work because of disability.  For the majority of
such veterans, Medicare is the primary source of health care
coverage.  Because of the high copayments and deductibles imposed
under Medicare, many beneficiaries purchase supplemental private
health insurance policies, typically referred to as Medigap policies. 
In 1987, less than 50 percent of VA-only and dual users had private
health insurance, compared with almost 90 percent of Medicare-only
users.  The difference in private insurance coverage among the groups
was consistent across all income levels (see fig.  5). 

   Figure 5:  Private Insurance
   Enrollment Rates of VA-Only,
   Dual, and Medicare-Only Users,
   by Income Category (1987)

   (See figure in printed
   edition.)


   IMPLICATIONS OF HEALTH CARE
   REFORM PROPOSALS ON
   MEDICARE-ELIGIBLE VETERANS
------------------------------------------------------------ Letter :6

Changes to Medicare or VA health care, such as those that were
proposed under this year's health care reform debate, could
substantially affect the future demand for VA health care services. 
The specific effect on the demand for VA services would vary greatly
and depend on the changes, if any, that are made to either program as
a result of health reform. 


      PROPOSED CHANGES TO MEDICARE
      WOULD LIKELY HAVE DECREASED
      DEMAND FOR VA SERVICES
---------------------------------------------------------- Letter :6.1

Several health care reform proposals contained provisions that would
have expanded Medicare coverage or decreased Medicare out-of-pocket
costs.  For example, the administration's proposed Health Security
Act and the Mitchell (S.2357) and Gephardt proposals would have
expanded Medicare to cover outpatient prescription drugs.  In
general, such changes would tend to decrease the demand for VA
services because they would reduce the need and the accompanying
financial advantage to seek such services from VA.  High
out-of-pocket costs for a Medicare drug benefit (initially a $500
deductible and 20 percent copayment would have been imposed under the
Gephardt proposal and a deductible and 20 percent copayment under the
Mitchell proposal) would likely lessen the effect on demand for VA
outpatient drugs. 

One change to Medicare--subsidies for low-income Medicare-eligible
families under the Medicare Catastrophic Coverage Act of
1988--implemented after VA's 1987 Survey of Veterans, may also have
this effect.  As explained earlier, this change requires state
Medicaid programs to pay premiums, deductibles, and coinsurance for
most Medicare beneficiaries living below the poverty level.  In a
recent study, however, we found that many Medicare beneficiaries
eligible for such subsidies were not enrolled in the program.\16

Healthcare reform proposals generally did not contain provisions that
would have reduced Medicare benefits by restricting services or
raising deductibles or copayments.  One notable exception to this was
home health benefits.  The administration's original proposal and the
Gephardt and Mitchell proposals would have imposed a 20 percent
copayment on Medicare home health services; such services are
currently free under part A.  Such proposals would likely have
increased the demand for VA services.  This is because
Medicare-eligible veterans might turn to VA to reduce their
out-of-pocket costs or to obtain services no longer available under
Medicare. 


--------------------
\16 Medicare and Medicaid:  Many Eligible People Not Enrolled in
Qualified Medicare Beneficiary Program (GAO/HEHS-94-52, Jan.  20,
1994). 


      PROPOSED CHANGES TO VA
      COVERAGE WOULD HAVE HAD
      UNCERTAIN EFFECT ON DEMAND
      FOR VA SERVICES
---------------------------------------------------------- Letter :6.2

Some of the health reform proposals considered in this year's debate
would have substantially changed the VA health care delivery system. 
For example, the proposed Health Security Act and the Mitchell and
Gephardt proposals would have (1) transformed VA facilities into a
series of managed care plans to compete with private sector plans,
(2) changed VA eligibility requirements to allow veterans greater
access to the full range of VA services, and (3) provided financial
incentives for service-connected and low-income veterans to enroll in
VA health plans.  These changes would have essentially removed
current VA restrictions on the availability of outpatient services
covered under the standard benefits package and entitled about 3.3
million Medicare-eligible veterans with service-connected
disabilities or low incomes to cost-free care if they enrolled in VA
health plans. 

Under the Gephardt proposal, current restrictions on access to
outpatient care services not included in the standard benefit package
would also have been eliminated; core group veterans enrolling in VA
health plans would have been entitled to receive such services
(including prescription drugs and vision- and hearing-related
services) cost-free.  Coverage of dental care would not have been
changed under the Gephardt or Mitchell proposals.  Finally, the
Gephardt proposal would have created a new entitlement to nursing
home care for veterans with service-connected disabilities rated at
50 percent or higher or in nursing homes for treatment related to
their service-connected disabilities. 

The Gephardt provisions would have created strong financial
incentives for service-connected and low-income Medicare-eligible
veterans to enroll in VA health plans.  While veterans currently make
substantial use of services such as hearing and vision care not well
covered under Medicare, their access to such services is restricted
by VA's complex eligibility and entitlement provisions.  Because such
provisions would have been largely eliminated under the Gephardt
proposal, VA might have seen a significant increase in demand for
services not included in the standard benefits package.  Similarly,
creating an entitlement for cost-free nursing home care might have
increased demand for VA-supported nursing home care, particularly if
such care were provided through contracts with community nursing
homes. 

The Mitchell proposal, however, contained a provision that would have
created a strong disincentive for Medicare-eligible veterans to
enroll in VA health plans.  Veterans remaining under Medicare's
fee-for-service program or enrolling in Medicare HMOs (other than a
VA health plan) would have been ensured of receiving all medically
necessary care covered under the Medicare program.  Veterans
enrolling in VA health plans, however, would have been ensured
receipt of medically necessary care only if VA appropriations were
sufficient to cover the costs of care.  The Secretary of Veterans
Affairs would have been given the discretion to reduce the standard
benefits covered by VA health plans if sufficient funds were not
appropriated.  In effect, the financial risks of operating VA health
plans would have been shifted from the government to the health
plans' enrollees. 

The net effect of the types of changes considered in the VA system on
the demand for VA services by Medicare-eligible veterans is
uncertain.  On one hand, such changes might have enticed more
Medicare-eligible veterans to use VA services, particularly if VA
health plans included services such as dental, vision, and hearing
that are not covered under Medicare.  However, the documented
reluctance of Medicare beneficiaries to enroll in HMOs might have
made it difficult for VA to keep those who are currently dual users
of Medicare and VA, particularly if VA plans did not guarantee the
availability of Medicare benefits.  In 1993, less than 10 percent of
all Medicare beneficiaries were enrolled in HMOs.  Those who already
rely exclusively on VA would have been the most likely to remain. 

The Dole/Packwood proposal would also have made changes in the VA
health care system, basically allowing VA to function as a provider
under state health care reform programs.  While the proposal would
not have specifically reformed VA eligibility, it appears that it
might have allowed VA to provide any health care items and services
covered under state health reform programs regardless of whether the
veteran would otherwise have been eligible to receive the service
from VA.  In addition, because the proposal did not specifically
limit VA's role under state health reforms to providing services to
veterans, it is not clear whether VA facilities in states enacting
health reforms would have been able to provide services to
nonveterans, including Medicare beneficiaries.  Such changes could
have increased demand for VA health care services in states
implementing health reforms. 

In summary, significant changes in either Medicare or VA health care
benefits or cost sharing such as those considered in the recent
health reform debate could have significant effects on future demand
for VA health care services.  The reluctance of Medicare
beneficiaries to join managed care plans could, however, reduce the
effect of reforms that would convert the VA health care system into a
series of managed care plans unless the financial incentives, either
through reduced cost sharing or added benefits are sufficient to
overcome this reluctance. 

At the request of your office we did not obtain agency comments on
this report.  As agreed with your office, we are providing copies of
this report to the Ranking Minority Member of your Subcommittee, the
Chairmen and Ranking Minority Members of the Senate and House
Committees on Veterans' Affairs and the Senate and House Committees
on Appropriations, the Secretary of Veterans Affairs, and other
interested parties.  Copies also will be available to others upon
request.  Please call me at (202) 512-7101 if you or your staff have
any questions.  Major contributors to this report are listed in
appendix II. 

Sincerely yours,

David P.  Baine
Director, Federal Health
 Care Delivery Issues


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

This study consisted of four main parts.  The first part centered on
inpatient and outpatient services received by Medicare-eligible
veterans in fiscal year 1990.  It was based on nationwide data files
maintained by VA and by the Department of Health and Human Services'
Health Care Financing Administration (HCFA).  The second part dealt
with VA prescription drug usage, based on fiscal year 1993 data from
10 of VA's 158 medical centers.  The third part dealt with income and
private health insurance coverage, using data from VA's 1987 Survey
of Veterans.  Finally, we reviewed various health reform proposals to
analyze the potential effects of health reform on the demand for VA
health care services.  A brief discussion of our methodology for each
part follows. 


   ANALYSIS OF INPATIENT AND
   OUTPATIENT USAGE OF
   MEDICARE-ELIGIBLE VETERANS
--------------------------------------------------------- Appendix I:1

In an earlier study,\17 we used a combination of VA, HCFA, and other
federal records to create a database (LIVEVETS) identifying
19,119,295 of the veterans who were alive in 1990.  We decided to use
this database for this study because we already had Medicare and VA
medical data on hand for that year.  Although the LIVEVETS file does
not include every veteran (we estimate that there were about 28.2
million veterans in 1990), it includes all those veterans who are the
focus of this study--the ones who used VA medical services. 

We focused our attention on the 4,411,558 Medicare-eligible veterans
identified in the LIVEVETS file.  Based on their VA and Medicare
usage, we divided them into four population groups: 

  VA-only users (490,404 veterans),

  Medicare-only users (2,495,502),

  Dual users of VA and Medicare services (564,345), and

  Nonusers of VA and Medicare services (861,307). 

This classification provided a means to analyze and compare health
care usage rates by different groups of veterans. 

After identifying the Medicare-eligible veterans, we added selected
health care information contained in VA's automated Patient Treatment
File and Outpatient Clinic System to each record.  VA's Patient
Treatment File contains detailed records of admission, discharge,
treatment procedures, and medical specialty units used by patients
during each episode of VA inpatient care.  VA's Outpatient Clinic
System contains records of VA outpatient visits, including clinic
stops made by patients during each visit.\18 These clinic stop codes
identify the various clinics visited by patients, but not the
specific services provided by the clinics.  They do not indicate, for
example, whether a visit to the optometry clinic was to treat ocular
and vision defects or for a routine eye examination or obtaining eye
glasses. 

Using codes contained in HCFA's Medicare Automated Data Retrieval
System, we also compared selected health care services paid by
Medicare for dual users and Medicare-only users.  This system
contains detailed inpatient treatment data--admission date, total
inpatient days, procedures and diagnosis, and a transaction code that
identifies types of inpatient facilities used.  The transaction code
is particularly useful in identifying Medicare-eligible veterans who
used Medicare for inpatient psychiatric care and nursing home care. 
These data allowed us to assess whether there were differences in
usage patterns between dual users and Medicare-only users with regard
to inpatient psychiatric care and nursing home care. 


--------------------
\17 Veterans' Health Care:  Most Care Provided Through Non-VA
Programs (GAO/HEHS-94-104BR, Apr.  25, 1994)

\18 The term clinic stop is used to identify a patient encounter with
one or more providers assigned to a particular clinic during the
course of a patient's visit to a facility.  For example, laboratory,
X ray, and general medicine are shown as separate clinic stops in
VA's Outpatient Clinic System. 


   ANALYSIS OF VA PRESCRIPTION
   DRUG USAGE
--------------------------------------------------------- Appendix I:2

VA has no centralized database on prescription drugs.  According to
VA officials, each of its 158 medical centers maintains its own
pharmacy database, which is generally kept on-line for only 9 to 18
months.  After that time, the data are archived onto tapes, which are
extremely difficult to retrieve.  Because 1990 data had already been
archived, it was impractical for us to obtain prescription data for
the same 1990 study populations that we used in analyzing VA
inpatient and outpatient services.  As a result, we (1) obtained 1993
prescription drug data from a sample of medical centers and (2)
identified a new Medicare-eligible veteran population for our
analysis of VA prescription usage. 


      OBTAINING PRESCRIPTION DATA
      FROM MEDICAL CENTERS
------------------------------------------------------- Appendix I:2.1

We first worked with officials at VA headquarters' Office of Pharmacy
Services to identify medical centers where 1993 pharmacy data were
still on-line.  From a list of centers identified, we judgmentally
selected 10 centers to provide geographic and size representation as
follows: 

Albany, New York,
Bay Pines, Florida,
Columbia, South Carolina,
Hampton, Virginia,
Madison, Wisconsin,
Muskogee, Oklahoma,
Newington, Connecticut,
Phoenix, Arizona,
St.  Cloud, Minnesota, and
St.  Louis, Missouri. 

Officials at VA's Birmingham Information System Center extracted 1993
prescription data from the 10 centers we selected and provided us
data tapes that contain social security number, number of
prescriptions, and VA acquisition costs of the prescriptions for each
veteran who received prescriptions from the 10 centers during fiscal
year 1993. 


      DETERMINING
      MEDICARE-ELIGIBLE VETERAN
      POPULATION
------------------------------------------------------- Appendix I:2.2

We used VA's 1993 inpatient and outpatient data files to produce a
file containing social security numbers of all veterans who used the
10 selected medical centers for inpatient or outpatient services
during fiscal year 1993.  At our request, HCFA matched this listing
against its Medicare Standard Analytical Files to determine each
veteran's Medicare eligibility and usage status.  We matched the data
that we received from VA and HCFA to (1) identify VA users who were
Medicare-eligible and divide them into VA-only and dual users and (2)
compare the number and costs of prescription drugs received by
VA-only users and dual users. 


   ANALYSIS OF INCOME AND PRIVATE
   INSURANCE COVERAGE
--------------------------------------------------------- Appendix I:3

The 1993 and 1990 VA and Medicare data files that we analyzed did not
contain income and private health insurance data on Medicare-eligible
veterans.  For this reason, we used data from VA's 1987 Survey of
Veterans to analyze income and private insurance coverage.\19 Based
on survey participants' responses to questions concerning Medicare
eligibility and use of VA and Medicare health care services, we
grouped respondents who were eligible for Medicare into VA-only
users, Medicare-only users, dual users, and nonusers.  We then
analyzed income and insurance coverage data reported by each of these
groups of respondents. 


--------------------
\19 VA, in conjunction with the Bureau of the Census, conducted the
1987 Survey of Veterans to evaluate its programs and to assess the
status and well-being of veterans across the nation.  In 1993, VA
conducted an updated survey, but the data were not yet available at
the time we completed our review. 


   ANALYSIS OF HEALTH REFORM
   PROPOSALS
--------------------------------------------------------- Appendix I:4

We reviewed health reform proposals focusing on changes to Medicare
or VA programs to analyze their potential effects on the demand for
VA health care services.  We based our analysis primarily on the
administration's proposed Health Security Act (S.  1757/H.R.  3600)
as introduced, the Mitchell (S.  2357) and Gephardt (H.R.  3600)
proposals, and the Dole/Packwood proposal (S.  2374) because they
were the only proposals we identified that specifically addressed the
VA health care system. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II

James R.  Linz, Assistant Director, (202) 512-7116
Sophia Ku, Evaluator-in-Charge
Stan Stenersen
Evan Stoll

