VA Health Care: Assessment of VA's Fiscal Year 1998 Budget Proposal
(Stmnt. for the Rec., 05/01/97, GAO/T-HEHS-97-121).

GAO discussed the President's 1998 budget request for the Department of
Veterans Affairs (VA) health care system.

GAO noted that: (1) while VA's budget goals may be attainable, they
carry implications such as limited deficit reduction contributions and
potential risks to low-income uninsured veterans; (2) achieving
increased efficiency is not contingent on either increases in patients
served or resources; (3) VA's ongoing efforts to restructure its health
care system could yield billions of dollars in savings during the next 5
years; (4) a large part of these savings would be realized through more
efficient use of it's workforce, which will allow the existing patient
base to be served with fewer employees; (5) sufficient savings could be
generated to afford VA an opportunity to increase patients served
without new resources or increase its contribution to deficit reduction;
(6) VA can significantly decrease its reliance on appropriated resources
by using its existing authority to sell excess capacity to help other
federal agencies meet their beneficiaries' health care needs; (7) VA's
proposal to generate billions of dollars in new revenue to serve 20
percent more patients intensifies VA's direct competition with the
private sector and potentially leaves low income, uninsured veterans
vulnerable; (8) VA may be able to attain its revenue goals, only by
attracting thousands of new users who have higher incomes or public or
private insurance; (9) such new users are likely to be drawn from
private providers who may see their revenue base erode as patients shift
to VA care; (10) VA may spend unreimbursed resources on these veterans
that could reduce the availability of resources for low-income,
uninsured veterans; (11) VA faces a difficult challenge as it takes
steps to implement a new resource allocation method to improve veterans'
access to VA care and a decentralized management structure to improve
resource utilization; (12) these initiatives promise improvements in
equity and have stimulated significant changes in efficiency; and (13)
however, VA's challenge will be to adequately monitor these changes to
identify and correct unintended effects such as those that limit equity
of access.

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

 REPORTNUM:  T-HEHS-97-121
     TITLE:  VA Health Care: Assessment of VA's Fiscal Year 1998 Budget 
             Proposal
      DATE:  05/01/97
   SUBJECT:  Veterans benefits
             Veterans hospitals
             Health care programs
             Health care cost control
             Deficit reduction
             Presidential budgets
             Health services administration
IDENTIFIER:  VA Veterans Integrated Service Network
             VA Veterans Equitable Resource Allocation System
             Medicare Program
             DOD TRICARE 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.                                          *
*                                                                        *
* 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


Before the Subcommittee on VA, HUD, and Independent Agencies,
Committee on Appropriations, U.S.  Senate

For Release on Delivery
Expected at 2:00 p.m. 
Thursday, May 1, 1997

VA HEALTH CARE - ASSESSMENT OF
VA'S FISCAL YEAR 1998 BUDGET
PROPOSAL

Statement for the Record by Stephen P.  Backhus, Director
Veterans' Affairs and Military Health Care Issues
Health, Education, and Human Services Division

GAO/T-HEHS-97-121

GAO/HEHS-97-121T


(406140)


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

  DOD - Department of Veterans Affairs
  VA - Department of Veterans Affairs
  VISN - ABC
  VERA - Veterans Equitable Resource Allocation

VA HEALTH CARE:  ASSESSMENT OF
VA'S FISCAL YEAR 1998 BUDGET
PROPOSAL
============================================================ Chapter 0

Mr.  Chairman and Members of the Subcommittee: 

We are pleased to contribute this statement for the record for the
Subcommittee's deliberations on the President's 1998 budget request
for the Department of Veterans Affairs (VA) health care system.  With
a 1997 medical care appropriation of $17 billion and a declining
veteran population, VA faces increasing pressure to contain or reduce
spending as part of governmentwide efforts to achieve a balanced
budget.  Last year, we reported that VA's health care system had the
opportunity to reduce its operating costs by billions of dollars over
the next several years.\1

VA's 1998 budget proposal requests a medical care funding level of
$17.6 billion, consisting of an appropriation of almost $17 billion
and a legislative proposal to retain insurance payments and other
third-party reimbursements.\2 VA characterizes this as the first step
in a 5-year plan to reduce its per patient cost by 30 percent,
increase patients served by 20 percent, and finance 10 percent of its
expenditures using nonappropriated revenues by the year 2002.  VA
proposes to use appropriations of about $17 billion over the next 5
years and to supplement this with increases in third-party
reimbursements that are estimated to be $1.7 billion in 2002. 

Our comments focus on VA's 5-year plan, including the outlook for
attaining the stated targets and the potential effects on veterans
and others.  In addition, as requested by the Subcommittee, we also
offer our preliminary observations on VA's progress on two major
initiatives:  developing a method to more equitably allocate
resources and establishing a decentralized management structure to
more efficiently and effectively deliver services.  We plan to
provide the Subcommittee more detailed information on these two
initiatives at a later date. 

Our comments on VA's budget proposal are based on past and ongoing
work to assess operating policies, procedures, and practices of VA
hospitals and clinics.\3 We spoke with hundreds of VA officials and
examined a wide array of documents, including VA's budget submission,
annual reports, and studies done by VA's Office of Inspector General
and

others.  Our comments on VA's decentralized management and resource
allocation initiatives are based on information obtained from
discussions with officials at headquarters and seven networks as well
as a review of documents they provided. 

In summary, while VA's budget goals may be attainable, they also
carry implications such as limited deficit reduction contributions
and potential risks to low-income, uninsured veterans.  Achieving
increased efficiency is not contingent on either increases in
patients served or resources.  VA's ongoing efforts to restructure
its health care system could yield billions of dollars in savings
during the next 5 years.  A large part of these savings would be
realized through more efficient use of its workforce, which will
allow the existing patient base to be served with fewer employees. 
In fact, sufficient savings could be generated to afford VA an
opportunity to increase patients served without new resources or to
increase its contribution to deficit reduction.  Furthermore, VA can
significantly decrease its reliance on appropriated resources by
using its existing authority to sell excess capacity to help other
federal agencies meet their beneficiaries' health care needs. 

VA's proposal to generate billions of dollars in new revenue to serve
20 percent more patients intensifies VA's direct competition with the
private sector and potentially leaves low-income, uninsured veterans
vulnerable.  VA may be able to attain its revenue goals only by
attracting thousands of new users who have higher incomes or public
or private insurance.  And such new VA users are likely to be drawn
from private providers who may see their revenue base erode as
patients shift to VA care.  Moreover, VA may spend unreimbursed
resources on these veterans, which could reduce the availability of
resources for low-income, uninsured veterans. 

VA also faces a difficult challenge as it takes steps to implement a
new resource allocation method to improve veterans' access to VA care
and a decentralized management structure to improve resource
utilization.  These initiatives promise improvements in equity and
have stimulated significant changes in efficiency.  However, VA's
challenge will be to adequately monitor these changes to identify and
correct unintended effects such as those that limit equity of access. 


--------------------
\1 VA Health Care:  Opportunities for Service Delivery Efficiencies
Within Existing Resources (GAO/HEHS-96-121, July 25, 1996) and VA
Health Care:  Opportunities to Increase Efficiency and Reduce
Resource Needs (GAO/T-HEHS-96-99, Mar.  8, 1996). 

\2 This includes $123 million of administrative costs for third-party
insurance recoveries and $68 million of reimbursements for veterans
compensation and pension examinations. 

\3 A list of related GAO testimonies and reports appears at the end
of this statement. 


   BACKGROUND
---------------------------------------------------------- Chapter 0:1

VA's role in providing for the health care needs of veterans has
evolved over time.  During its first 50 years, VA predominantly
served veterans who had disabilities caused or aggravated by military
service and other low-income, uninsured veterans in need of a health
care safety net.  Over the past 10 years, VA has also served higher
income and insured veterans with nonservice-connected conditions. 
Over time, however, VA's patient base has been shifting from serving
primarily veterans with service-connected conditions to those without
service-connected conditions.  Currently, VA operates over 750
facilities, including 173 hospitals and over 400 outpatient clinics. 
These facilities serve 2.6 million of the nation's almost 26 million
veterans as well as about 300,000 nonveterans. 

In 1995, to promote greater efficiency and services to veterans, VA
created a new decentralized management structure, forming 22 Veterans
Integrated Service Networks (VISN).  These networks replaced the
previous structure's four regions and expanded local authority.  The
VISN is now the basic budgetary and decision-making unit of VA's
health care system and exercises management authority over VA
facilities in its geographic area.  This system of networks clearly
places value on efficiency and customer service, and the networks are
empowered to make a wide range of decisions regarding care delivery
options.  Under the recently enacted eligibility reform legislation
(P.L.  104-262), for example, networks can contract with a broader
range of private providers to purchase services at prices lower than
VA's costs and generate revenue by selling excess services.  In April
1997, VA implemented the Veterans Equitable Resource Allocation
(VERA) system to allocate medical care appropriations among the 22
VISNs.  VERA is intended to improve the equity of resource
distribution throughout VA's health care system. 


   EFFICIENCY SAVINGS NOT
   DEPENDENT ON INCREASED NUMBER
   OF VETERANS SERVED
---------------------------------------------------------- Chapter 0:2

Last year, we testified that VA could save billions of dollars over
the next 7 years through improved efficiency.  As noted before, the
Congress subsequently gave VA the two additional tools--eligibility
reform and expanded contracting authority--that VA said were key to
the success of its efforts to increase efficiency.  With these tools,
VA can help veterans prevent costly hospital admissions and access
lower cost services, regardless of where veterans reside.  VA's 1998
budget request, however, suggests that VA will be able to achieve 30
percent efficiency savings over the next 5 years only if it has the
additional resources to serve 20 percent more patients. 

Over the past 18 months, VA has taken aggressive steps to change the
way it operates to reduce costs and improve services to veterans. 
These initiatives are expected to save billions of dollars by
avoiding unnecessary expenditures.  Most of the initiatives involve a
resizing and more efficient use of its workforce, which accounts for
over $10 billion of VA's medical care budget.  For example, VA is
shifting patient care from inpatient to outpatient settings as well
as reducing average lengths of inpatient stays.  It is also
consolidating management and clinical services of nearby hospitals to
reduce costs.  Moreover, VA is exploring opportunities to contract
with other health care providers for services at costs lower than
VA's. 

These restructuring efforts should save billions of dollars without
attracting new users as the following examples indicate: 

  VA established a pre-admission screening process for hospitals
     that, if effectively implemented, could save $8.4 billion over
     the next 5 years. 

  VA integrated the management of two or more nearby facilities in 26
     different locations, which should result in savings of $230
     million over the next 5 years. 

  VA shifted substance abuse treatment from an inpatient to an
     outpatient setting in one service location, which is expected to
     result in savings of $10 million over the next 5 years. 

Currently, VA has teams exploring additional opportunities for
streamlining operations and reducing workforce needs.  Many of these
teams are identifying ways to use lower cost methods for delivering
services within individual facilities.  For example, many facilities
are

  reducing patient bed-days of care, including one location that
     would close seven medical wards and generate potential savings
     of almost $50 million over the next 5 years, and

  shifting inpatient surgeries to ambulatory settings, including one
     location that shifted enough workload among facilities to close
     two surgical wards and potentially save over $15 million during
     the next 5 years. 

VA also has many teams exploring ways to consolidate services at
nearby facilities.  Such actions should result in significant savings
over the next 5 years as shown by the following examples: 

  Facilities in one service area are planning to integrate eight
     pathology and laboratory medicine services into a single
     business unit with two central laboratories.  This integration
     is expected to save about $10 million over the next 5 years. 

  Facilities in another area are exploring ways to consolidate small
     purchases into one location, which is expected to save over $20
     million during the next 5 years. 

Additional savings opportunities could be available in later years
from the closing of hospitals whose workloads may be shifted to
nearby hospitals that have sufficient unused capacity to efficiently
and effectively meet veterans' needs.  For example, closing a
facility with about 300 beds could save over $100 million in overhead
costs alone during a 5-year period. 


   EFFICIENCY SAVINGS COULD
   PROVIDE OPPORTUNITIES TO SERVE
   MORE VETERANS WITHOUT
   ADDITIONAL RESOURCES
---------------------------------------------------------- Chapter 0:3

VA could expand its current patient base if its efficiency savings
exceed payroll and other cost increases.  These costs are expected to
be about $637 million in 1998 and to increase by a rate of about 4
percent a year over the next 5 years. 

The effect of VA's efficiency savings is to increase its purchasing
power each year.  For example, most of the savings are attributable
to reductions in VA's workforce, which currently numbers about
189,000 full-time equivalents.  VA may need to reduce its workforce
by about 6,800 full-time equivalents to realize an annual savings of
$637 million.  This level of reductions would decrease VA's resource
needs by comparable amounts in succeeding years.  Thus, an annual
appropriation of $17 billion could be sufficient to serve 2.9 million
patients in 2002 if efficiency savings and cost increases approximate
$637 million a year, on average.  Moreover, VA could increase its
patient base if its efficiency initiatives yield greater savings. 


   ADDING RESOURCES FURTHER
   ENHANCES VA'S OPPORTUNITY TO
   SERVE MORE VETERANS
---------------------------------------------------------- Chapter 0:4

VA's 1998 budget proposes reinvesting all efficiency savings and
using additional resources to expand its patient base.  VA expects to
add a total of $5.8 billion in new resources over the next 5 years
(from public and private insurers and others), starting with $737
million in 1998 and increasing to $1.7 billion in 2002.  VA expects
that these additional resources will allow it to increase the number
of veterans served by 587,000, which would increase its patient base
from 2.9 million to 3.5 million in 2002. 

If the targeted resource levels are attained, VA appears capable of
attracting 587,000 new users by 2002.  Recent expansion of VA's
contracting authority and veterans' eligibility for care should
facilitate creation of new access points, referred to as
community-based outpatient clinics, which along with VA's efforts to
improve accessibility of existing hospital-based clinics are likely
to attract new workload. 

For example, VA has opened or developed plans to open 86 new
community-based clinics over the last 3 years.  These clinics provide
only primary care and refer veterans to VA hospitals for more
specialized care.  Last month, we surveyed the 12 clinics that had at
least 2 years' operating experience and found that they had attracted
3,000 new veterans.  These clinics experienced the largest growth in
their initial year and smaller growth in subsequent years.  VA
estimates that the remaining 74 clinics will serve over 128,000 users
a year but has not estimated how many will be new VA users. 
Twenty-two of the new clinics estimated that between 5 and 60 percent
of the patients served will be new users, while the rest expected to
serve no new users or were unsure whether new users would be served. 

Although it plans to open many more clinics, VA told us that it is
too early to estimate how many or where they will be located.  Our
analysis suggests that VA could need between 1,200 and 1,800
additional clinics to attract 587,000 new users if each clinic
attracts between 250 and 500 new veterans.  The first 12 clinics
averaged 250 in their initial years.  These clinics also appear to
provide an affordable way for VA to attract new users. 

In addition, VA's efforts to improve veterans' access to existing
facilities should also attract new users.  These initiatives include
expanding primary care by extending operating times for
hospital-based clinics to night and weekend hours as well as ways to
reduce waiting times.  For example, one hospital-based clinic
reported enrolling 3,000 new veterans for care during the first year
after having made such accessibility improvements. 


   EXPANDING VA'S RESOURCE BASE
   POSES CHALLENGES
---------------------------------------------------------- Chapter 0:5

VA's revenue goal of $1.7 billion in 2002 includes estimated
recoveries of about $902 million from private insurance, $557 million
from Medicare, and $178 million from federal agencies and others. 
Attaining these targets may present a challenge as VA would probably
have to attract thousands of new revenue-generating veterans.  VA has
provided, however, little information on the numbers of new veterans
needed to meet revenue goals or how much of the revenue will come
from inpatient or outpatient services.  This lack of information
creates uncertainties about VA's ability to achieve its revenue
goals. 


      INCREASING RECOVERIES FROM
      PRIVATE HEALTH INSURANCE MAY
      BE DIFFICULT
-------------------------------------------------------- Chapter 0:5.1

VA currently serves insured veterans and recovers some or all of its
costs of care from insurers.  Presently, VA returns all recoveries to
the Treasury, except those needed to cover VA's billing and
collection costs.  In 1996, VA deposited $455 million into the
Treasury and used $119 million for administrative costs.  VA's
recovery of $574 million represents a decline in recoveries from
1995, despite an increase in the number of users. 

VA's ability to increase future recoveries from its current insured
patient base is uncertain for several reasons: 

  Veterans are increasingly covered by health maintenance and
     preferred provider organizations from which VA generally cannot
     recover. 

  As an increasing proportion of VA users become eligible for
     Medicare, their private health insurance becomes secondary, so
     potential recoveries drop. 

  As VA shifts from inpatient to outpatient settings, insurance
     recoveries decrease and the cost of recovery increases. 

  VA found that Medigap insurers have been paying VA too much, which
     will result in decreased future recoveries and refunds of about
     $150 million a year. 

  VA's authority to recover from private health insurance for care
     provided to service-connected veterans for non-service-related
     conditions expires September 30, 1998. 

As a result, to meet its revenue projections of $902 million from
private insurance, VA will probably have to focus its marketing
efforts on attracting veterans with fee-for-service private health
insurance.  In addition, the Congress would need to extend VA's
authorization to recover for certain services provided to
service-connected veterans. 

VA officials told us that they do not know how many veterans in their
2.9 million patient base have insurance or how many insured veterans
receive billable care.  This lack of information on key elements
affecting its projections creates considerable uncertainty about the
number of new insured users VA would need to attract in order to
generate its target revenues. 


      ATTAINING MEDICARE RECOVERY
      TARGET MAY BE DIFFICULT
-------------------------------------------------------- Chapter 0:5.2

VA proposes to collect about $557 million from Medicare in 2002 for
services provided to about 106,000 additional higher-income veterans
who are covered by Medicare.  VA currently attracts only about 1 out
of every 100 higher-income Medicare-eligible veterans--about 41,000
veterans in 1992.  It thus appears questionable whether VA will be
able to attract an additional 106,000 higher-income Medicare-eligible
veterans by the year 2002. 

VA expects to recover from Medicare, on average, about $5,300 for
each of the 106,000 additional Medicare-eligible veterans it expects
to serve in 2002, a target amount that seems achievable based on
average Medicare spending levels per patient nationwide.  However, it
may be difficult for VA to achieve this collection rate if
Medicare-eligible veterans use primarily VA services that are not
covered by Medicare, such as prescription drugs, inpatient
psychiatric care, and long-term nursing home care.  Our assessment of
Medicare-eligible veterans' use of VA services in 1994 suggests that
most of these veterans use VA, at least in part, for services not
covered by Medicare.\4


--------------------
\4 Veterans' Health Care:  Use of VA Services by Medicare-Eligible
Veterans (GAO/HEHS-95-13, Oct.  24, 1994). 


      INCREASING RECOVERIES FROM
      OTHER SOURCES APPEARS
      ATTAINABLE
-------------------------------------------------------- Chapter 0:5.3

VA proposes to collect $178 million in 2002 through sales of excess
services to federal agencies, affiliated medical schools, and others. 
This amount represents over a 300-percent increase over VA's
collections of $43 million in 1996. 

Since 1966, the Congress has expanded VA's authority on several
occasions to sell excess services in an effort to encourage VA
facilities to generate revenues in addition to those appropriated. 
Over the last 5 years, VA's sales have increased by about 37 percent,
with most sales to the Department of Defense (DOD) and affiliated
medical schools.  Last September, the Congress took another step to
expand VA's ability to generate revenue by authorizing VA to sell
excess health care services to any health care plan, insurer, or
other provider. 

VA could meet or exceed its goal of $178 million in 2002 if it
markets its excess capacity to other federal agencies.  DOD and VA
reached agreement in 1995 that VA can provide health care services to
active duty and retired members of the military and dependents
enrolled in DOD's TRICARE program.  While some VA facilities have
become TRICARE providers, most have not.  Similarly, few VA
facilities have generated revenue by serving beneficiaries of other
federal agencies, such as the Indian Health Service and the Bureau of
Prisons, even though these agencies have expressed interest in buying
VA's excess services. 


   EXPANDING VA'S RESOURCE BASE
   MAY PLACE SOME VETERANS AND
   OTHERS AT RISK
---------------------------------------------------------- Chapter 0:6

Over the last 25 years, VA has served an increasing number of
veterans without service-connected conditions, generally those
low-income veterans in need of a health care safety net.  During the
last 10 years, VA has also served higher-income and insured veterans
with its resources that were in excess of those needed to provide
care to service-connected and low-income veterans. 

Allowing VA to retain nonappropriated revenues may change VA's
perspective.  This is because the veteran population is, in effect,
likely to represent two distinct groups--non-revenue-generating
veterans and revenue-generating veterans.  Within this later group
are several potential target populations:  privately insured
veterans; Medicare-eligible veterans; higher-income veterans; and
higher-income, privately insured, or Medicare-eligible veterans. 


      NON-REVENUE-GENERATING
      VETERANS MAY BE AT RISK OF
      HAVING ACCESS LIMITED
-------------------------------------------------------- Chapter 0:6.1

VA may encounter difficulty attaining its revenue goals unless a
significant number of new users have higher incomes or insurance. 
This could create a strong incentive for VA to market services to
attract revenue-generating rather than non-revenue-generating
veterans.  This incentive could manifest itself in several ways,
including where VA decides to locate new community-based outpatient
clinics.  For example, VA recently proposed locating a
community-based clinic in a homeless shelter that VA expects could
attract 2,040 new users in need of VA's safety net and therefore not
likely to generate revenue.  By contrast, VA has also proposed
opening a new clinic in one of the country's more affluent counties. 
While the clinic is intended to improve access for current users, it
is also expected to attract patients who could ultimately generate
revenue. 


      NON-VA PROVIDERS MAY BE AT
      RISK OF LOSING WORKLOAD
-------------------------------------------------------- Chapter 0:6.2

Marketing VA services to generate revenue has the potential to draw
higher-income insured veterans from private providers who may then
see their revenue base erode, depending on the number of patients who
shift to VA care.  If VA has to aggressively attract new users who
are now receiving health care elsewhere, it will intensify the
competition between VA and other state, county, and private providers
for a larger share of a shrinking veterans' health care market. 

VA's success in attracting revenue-generating patients will be likely
to result in a shifting of health care costs from other financing
organizations to VA and to exacerbate financial hardships for those
competing health care providers that have excess capacities.  For
example, our interviews with 115 veterans using new access points
last year revealed that 70 percent had Medicare coverage, 50 percent
had private insurance, and 7 percent had Medicaid.\5 Most said they
paid for their own primary care or used insurance coverage to obtain
care at other providers before they switched to VA care. 


--------------------
\5 VA's Health Care:  Improving Veterans' Access Poses Financial and
Mission-Related Challenges (GAO/HEHS-97-7, Oct.  25, 1996). 


      VA'S PROPOSAL COULD LOWER
      CONTRIBUTION TO DEFICIT
      REDUCTION
-------------------------------------------------------- Chapter 0:6.3

VA's proposal to retain revenue generated from nonappropriated
sources would also affect VA's contribution to deficit reduction.  VA
currently returns recoveries to the Treasury, which, in effect,
reduces the government's cost of VA health care.  For example, VA
expects to return $438 million in 1997, which would reduce the amount
of government resources needed to serve VA's patient base from its
appropriated amount of $17 billion to $16.6 billion.  By contrast,
under VA's proposal it would retain insurance recoveries of $590
million in 1998, increasing the government's cost to finance VA
health care to $17.6 billion, or $1 billion more than in the previous
year. 

In addition, VA's proposal to reinvest efficiency savings and use
additional nonappropriated resources to increase the number of
patients served could affect VA's contribution to deficit reduction. 
For example, VA would need an appropriation of $17 billion a year to
serve 2.9 million users if savings equal payroll and inflation costs
between 1998 and 2002.  By contrast, VA may be able to contribute up
to $1 billion more in 2002 toward deficit reduction if annual
efficiency savings exceed cost increases by $200 million, on average,
over the 5-year period and such excess savings are returned to the
Treasury. 


   NEW ALLOCATION METHOD AND
   DECENTRALIZED MANAGEMENT SHOW
   PROMISE, BUT RISKS EXIST
---------------------------------------------------------- Chapter 0:7

VA is using a new resource allocation method and a decentralized
management structure to address two long-standing issues:  equity and
efficiency.  These initiatives are intended to improve the equity of
veterans' access to care and produce cost savings. 


      ALLOCATING RESOURCES
      EQUITABLY SEEMS ACHIEVABLE
      WITH NEW METHODOLOGY
-------------------------------------------------------- Chapter 0:7.1

VA is using the Veterans Equitable Resource Allocation (VERA) system
to allocate 88 percent of the $17 billion medical care appropriation
to the 22 networks.  This approach is a major shift away from VA's
historical process for two reasons.  First, it funds 22 networks
rather than hundreds of facilities.  Second, it allocates resources
on the basis of costs per veteran served rather than on the basis of
facilities' historical budgets.  Funding networks sends a clear
message that each facility is a part of a larger regional enterprise
charged, in part, with a mission of achieving equity of access.  VERA
recognizes that networks are the vehicles for fostering regional
change, eliminating redundancies, and facilitating cooperation among
medical facilities.  Network officials have the authority to tailor
their VERA allocations to facilities and programs, within the
parameters set by national policy and guidelines, and to integrate
services across facilities for equity and other purposes. 

The goal of VERA is to provide networks with comparable levels of
resources per veteran served.  VA implemented VERA in an attempt to
allocate patient care resources on the basis of differences in
patient needs and regional differences in the price of their care. 
To do this, VERA classifies patients into two groups--basic care and
special care--as a simple case mix adjustment.  Basic care patients
generally receive routine services that are less expensive than those
received by special care patients.  Special care patients often have
complex or chronic conditions, such as spinal cord injury or
end-stage renal disease, or require care in settings such as nursing
homes.  The VERA special care category also includes some adjustment
for age to account for the higher medical demands of older population
groups. 

VERA allocates resources to networks based on two key components: 
network workloads and national prices.  VA patient workloads are the
estimates of the number of patients--basic and special--a network may
serve.  VA also calculates workloads for research support, education
support, equipment, and nonrecurring maintenance.  To determine a
national price for each workload category, VERA divides national
resources available by the national workload for that category.  VERA
allocates funds to a network by multiplying the network's workload
numbers by their respective national prices.  In addition, VERA
adjusts for differences in regional labor costs for patient care. 

To the extent that VERA allocates comparable levels of patient care
resources for each veteran served, it provides incentives for
networks to obtain these resources by increasing workload and
decreasing costs.  Networks that increase their patient workload
relative to other networks gain resources under VERA; those whose
patient workloads decrease relative to others lose resources. 
Networks that are more efficient, that is, have patient care costs
below the national price, have more funds available for local
initiatives.  However, those with patient care costs above the
national price (that is, less efficient networks) must increase
efficiency to have such funds available.  Thus, these incentives can
result in cost savings and enhanced access for veterans.  VERA will
not be fully implemented until fiscal year 1999.  As a result, few
resources will move among networks this year.  (See fig.  1.) Five
VISNs will receive fewer dollars and 17 will receive more.\6 VERA
generally moves resources from the Northeast and Midwest, where per
veteran costs have been higher than the national average, to the
South and West where per veteran costs have been lower than the
national average.  If VA had fully implemented VERA this year, shifts
in funding among the networks would have ranged from a reduction of
14 percent to an increase of 16 percent. 

   Figure 1:  Changes Resulting
   From VERA Allocations (Fiscal
   Years 1996-97)

   (See figure in printed
   edition.)

Note:  These numbers include all six VERA expenditure categories: 
basic care, special care, research support, education support,
equipment, and nonrecurring maintenance. 

VERA, like any allocation model, has limitations.  First, VERA may
shift some resources inappropriately because it may not fully account
for justifiable differences in regional cost variations.  Although
VERA adjusts for differences in regional case mix with its basic and
special care patient categories and adjusts the allocations for
differences in regional labor costs, it assumes that all the
remaining differences are based on differences in efficiencies. 
While inefficiency is a major factor in these cost differences, other
factors may play a role.  For example, to the extent that veterans
are sicker and need more health care services in different parts of
the country, additional case mix adjustments may be necessary to
fully explain regional cost differences.  As we have said in the
past, VA needs to provide more information on why costs vary
throughout the country.\7 VA officials told us they plan to examine
this further. 

Another potential issue is that basing VERA on veteran-users may
result in underallocation of funds in areas with low usage rates.  If
these rates result from past inequities in access to services, VERA
may need to incorporate population-based data on veterans with
highest priority for receiving services rather than relying solely on
user data.\8 However, other factors, such as number of veterans with
health insurance coverage, could also affect usage rates.  Because
adequate data were not available and VA wished to implement VERA as
quickly as possible, it did not include population data in VERA.  VA
continues to examine the utility of doing so. 

VERA's incentives for lower per veteran costs and higher workload
numbers could lead to unintended consequences if not properly
monitored and corrected.  In our discussions and visits with network
and medical center officials, we found efforts under way to increase
the number of veterans served.  VA indicators for the first quarter
of fiscal year 1997 generally show increases in the number of
high-priority veterans (that is, Category A veterans) seen, and the
increases for some networks are dramatic.  We have concerns about
whether the data accurately depict changes in workload.  If the data
are reliable, we are concerned that some networks may be
inappropriately increasing their workload numbers to get more
resources under VERA.  For example, networks may be increasing
workload by increasing the number of one-visit patients.  This may be
good primary or preventive care, or it could distort VERA allocations
because only minimal services are provided to get credit for
increased workload.  In the short time since the indicators were
published, however, we have been unable to determine the accuracy of
the data and the services the new users received.  VA officials told
us that they recognize the importance of monitoring, identifying, and
correcting unintended consequences.  They said they will monitor data
used in the allocation model, including workload increases, to ensure
that they reflect changes at the network and medical center levels
that are consistent with VA-wide policy and guidance. 

Although VERA is a step toward a more equitable allocation of
resources, it does not specifically address equitable access to
services.  Networks are ultimately responsible for allocating funds
to ensure that veterans have equal access to VA services.  Each of
the networks we contacted differs in how it allocates funds.  One
funds its facilities using a flat rate for each veteran-user. 
Another uses a combination of historical funding and negotiation with
medical center management regarding new initiatives.  Still another
includes a feature in its allocation method that provides payment for
each additional veteran served.  VA officials told us they will
examine these processes to ensure that different allocation
mechanisms increase equity of access to services while addressing
other national VA goals. 


--------------------
\6 In VA's Veterans Equitable Resource Allocation System Briefing
Booklet, March 1997, VA shows that 6 networks will lose funds and 16
will gain funds in fiscal year 1997.  However, VA excludes
allocations for equipment and nonrecurring maintenance.  We included
those amounts in our calculations to show the impact of VERA more
fully.  Neither we nor VA includes funds not allocated by VERA in
these comparisons. 

\7 Veterans' Health Care:  Facilities' Resource Allocations Could Be
More Equitable (GAO/HEHS-96-48, Feb.  7, 1996) and Department of
Veterans Affairs:  Programmatic and Management Challenges Facing the
Department (GAO/T-HEHS-97-97, Mar.  18, 1997). 

\8 Category A veterans have the highest priority for receiving VA
health care services.  Included in Category A are veterans with
service-connected disabilities and those with service-connected
disabilities whose incomes fall below certain thresholds. 


      NETWORKS HAVE MADE
      SIGNIFICANT PROGRESS, BUT
      DECENTRALIZED MANAGEMENT
      POSES OVERSIGHT CHALLENGES
-------------------------------------------------------- Chapter 0:7.2

VA has taken a page from private sector organizations and empowered
the network directors by delegating broad decision-making authority
over network budgets, facility staffing, health care delivery, and
administrative functions.  This has resulted in notable
accomplishments at VA, including significant cost savings and
improvements in access. 

Decentralized decision-making at VA places a premium on effective
headquarters guidance and monitoring of VISN activities.  The
challenge is to ensure that networks have a common understanding of
VA-wide goals and legislative requirements while permitting them
flexibility in how to achieve the goals.  The challenge in monitoring
network performance is to have reliable, appropriate, and timely
indicators to ensure that problems are identified and corrected. 

VA has provided guidance to managers and staff since the beginning of
its reorganization.  For example, the Under-Secretary for Health
issued two volumes, "Vision for Change" and "Prescription for
Change," delineating the type of organization he intended VA to
become and the goals VA would strive to attain.  Network and medical
center staff told us that these publications and other
communications, such as monthly meetings between network and
headquarters managers, help develop their understanding of the
structural and operational changes being made. 

VA's new performance measurement process also plays an important
guidance role by underscoring VA-wide organizational priorities. 
These measures include key indicators such as reduced bed-days of
care and an increased percentage of surgeries performed on an
ambulatory basis.  The measures are the main components of the
network directors' performance agreements.  In networks we visited,
medical center directors' performance agreements also included these
measures.  Medical center directors we contacted told us that network
directors were exercising closer oversight of their progress in
achieving VA-wide goals than had occurred under previous
organizational structures. 

Another strategy for reducing unnecessary variation has been the use
of clinical practice guidelines.  These are intended to enhance the
quality and appropriate utilization of health care services by
reducing variations in the way a health condition--for example,
stroke--is treated.  Networks are required to adopt 12 practice
guidelines by the end of fiscal year 1997.  They can choose from
those identified by headquarters or other sources. 

Providing national guidelines but offering networks discretion on
when to follow these guidelines can create opportunities for local
innovation but problems for national oversight.  If discretion
results in variation across the system, it will be difficult for VA
to assess the impact of the guidelines.  Network flexibility may
produce tension between headquarters and networks.  For example,
officials in one network we visited told us that they preferred the
American Medical Association guidelines to the national diabetes
guidelines VA adopted. 

Headquarters, network, and medical center officials told us that
national guidance had not been sufficiently clear on whether to
notify headquarters of significant program changes at the network
level.  They told us that they had not always been clear on what
constituted "significant" changes.  In a few instances, headquarters
officials were not notified of impending network-initiated changes
such as closure of a surgical program at a medical center.  In May
and September 1996, headquarters issued guidance for networks on
prior notification and consultation with headquarters for network
actions such as restructuring clinical services--including closures
of major programs--and proposed changes to special emphasis programs
such as those for spinal cord injury and prosthetics.  VA has
additional measures planned to ensure that headquarters is involved
in significant network-initiated program changes. 

Performance measures and standards developed by headquarters are the
key components of VA's monitoring process.  Headquarters holds
network directors accountable for making progress toward VA goals by
including measures and standards of performance in the directors'
contracts.  Headquarters lengthened its list of measures for fiscal
year 1997; it now includes about two dozen indicators.  In networks
we visited, directors are monitoring medical centers on these
measures as well. 


   CONCLUDING OBSERVATIONS
---------------------------------------------------------- Chapter 0:8

VA's 1998 budget presents the Congress with a fundamental choice
about the future course of VA health care, a choice that will have an
effect on veterans, other health care providers, and efforts to
achieve a balanced federal budget.  In general, VA's proposal to
reinvest all savings and generate additional nonappropriated revenues
may intensify the direct competition between VA and other providers. 
By contrast, a decision to limit VA's retention of nonappropriated
revenues will set VA on a course to becoming a more cost-efficient
safety net for those non-revenue-
generating veterans who have no other health care options. 

Currently, there is insufficient information to understand the full
implications of VA's budget proposal.  VA states that the key
elements of its proposal--namely, a 30-percent per patient cost
reduction, a 20-percent increase in veterans served, and a 10-percent
reduction of its reliance on appropriations--are inexorably linked
but, in our view, this is not so.  It seems plausible that any number
of different scenarios could occur, depending on the magnitude of
cost savings that VA will realize through its ongoing restructuring. 

For instance, VA could operate as a health care safety net for
several years, with an appropriation of about $17 billion or less,
given VA's progress in identifying and implementing efficiency
savings.  Such efficiency savings could equal or exceed the potential
nonappropriated revenues that VA estimates it can generate over the
next 2 years if authorized to do so.  For this reason, there appears
to be time to obtain critical information from VA and others so that
VA's budget proposal may be more clearly understood and fully
debated.  In this regard, several critical issues could be addressed,
including the following: 

  Should VA reinvest all efficiency savings to expand the number of
     patients served?  If so, should VA's expansion be limited to
     certain target groups of veterans, such as service-connected,
     low-income, or uninsured veterans in need of a health care
     safety net? 

  Should VA use nonappropriated revenue sources to help finance
     increased services to higher-income and insured veterans who
     have no service-connected conditions or continue relying solely
     on appropriated resources to finance increased services for
     service-connected and low-income veterans without
     service-connected conditions? 

  Should VA reinvest savings in excess of those needed to maintain
     its current patient base in order to serve more veterans or
     should it return some or all of the excess savings as a
     contribution toward deficit reduction? 

It would be less difficult to make such choices at this time if VA
had provided a road map that clearly articulated (1) what operational
changes would be needed to move along its newly proposed competitive
course and (2) what consequences such competition would have for
veterans and others.  For example, additional information would be
helpful about how different choices may affect (1) service-connected
veterans and those in need of VA's safety net; (2) VA's existing
hospitals, clinics, and other facilities; (3) VA's workforce; and (4)
other health care providers. 

Delaying a decision on VA's legislative proposals until such critical
information is available--including a plan describing how the system
will look and operate in 2002--may result in a better legislative
decision on VA's budget proposal.  It will also afford VA and the
Congress time to better assess how VA's future resource needs may be
affected by the new decentralized management and resource allocation
initiatives. 

VA's new resource allocation process and decentralized management
structure hold promise for improved operational efficiencies and
equitable access.  Responding to VERA's incentives and VA's goals,
local managers are already producing substantial savings and
increasing the number of veterans served.  VA, however, needs to
continue examining how price and workload data are determined under
VERA to improve equity of resource allocation.  VA also needs to
carefully monitor the impact of VERA's incentives on network and
facilities performance.  This is particularly important given the
variation resulting from local managers' flexibility in the
decentralized system.  We believe that identifying and correcting
problems is essential to the success of VA's proposed 5-year plan. 


   CONTRIBUTORS
---------------------------------------------------------- Chapter 0:9

For more information about this statement, please call Paul Reynolds,
Assistant Director, at (202) 512-7109 or Bruce Layton, Assistant
Director, at (202) 512-6837. 

RELATED GAO PRODUCTS

Department of Veterans Affairs:  Programmatic and Management
Challenges Facing the Department (GAO/T-HEHS-97-97, Mar.  18, 1997). 

VA Health Care:  Improving Veterans' Access Poses Financial and
Mission-Related Challenges (GAO/HEHS-97-7, Oct.  25, 1996). 

VA Health Care:  Opportunities to Significantly Reduce Outpatient
Pharmacy Costs (GAO/HEHS-97-15, Oct.  11, 1996). 

VA Health Care:  Issues Affecting Eligibility Reform Efforts
(GAO/HEHS-96-160, Sept.  11, 1996). 

VA Health Care:  Opportunities for Service Delivery Efficiencies
Within Existing Resources (GAO/HEHS-96-121, July 25, 1996). 

VA Health Care:  Challenges for the Future (GAO/HEHS-96-172, June 27,
1996). 

VA Health Care:  Efforts to Improve Veterans' Access to Primary Care
Services (GAO/T-HEHS-96-134, Apr.  24, 1996). 

VA Health Care:  Approaches for Developing Budget-Neutral Eligibility
Reform (GAO/T-HEHS-96-107, Mar.  20, 1996). 

VA Health Care:  Opportunities to Increase Efficiency and Reduce
Resource Needs (GAO/T-HEHS-96-99, Mar.  8, 1996). 

Veterans' Health Care:  Facilities' Resource Allocations Could Be
More Equitable (GAO/HEHS-96-48, Feb.  7, 1996). 

VA Health Care:  Issues Affecting Eligibility Reform
(GAO/T-HEHS-95-213, July 19, 1995). 

VA Health Care:  Challenges and Options for the Future
(GAO/T-HEHS-95-147, May 9, 1995). 

VA Health Care:  Retargeting Needed to Better Meet Veterans' Changing
Needs (GAO/HEHS-95-39, Apr.  21, 1995). 

Veterans' Health Care:  Use of VA Services by Medicare-Eligible
Veterans (GAO/HEHS-95-13, Oct.  24, 1994). 


*** End of document. ***