VA Health Care: More Veterans Are Being Served, But Better Oversight Is
Needed (Chapter Report, 08/28/98, GAO/HEHS-98-226).

Pursuant to a legislative requirement, GAO reviewed the: (1) changes in
overall access to care, changes in access to certain specialized
services, and a comparison of changes in Veterans Integrated Service
Network (VISN) 3 (Bronx) and VISN 4 (Pittsburgh) with Department of
Veterans Affairs (VA) national data from fiscal years (FY) 1995 to 1997;
(2) extent to which VA headquarters and VISNs are working to equitably
allocate resources to facilities within VISNs; and (3) adequacy of VA's
oversight of changes in access to care.

GAO noted that: (1) overall, VISN 3, VISN 4, and VA nationally have
increased access as measured by increases in the number of veterans
served; (2) access to care, as measured by patient satisfaction, also
seems to have improved according to responses to VA surveys and
interviews GAO conducted; (3) in addition, VA has improved geographic
access to primary care by increasing the number of community-based
clinics in these two VISNs; (4) although access has increased overall,
access appears to have decreased for some specific services; (5) the two
VISNs GAO reviewed used no specific criteria for allocating their
resources to reduce historical access inequities among their facilities;
(6) VA headquarters neither provides criteria for VISNs to use to
equitably allocate resources nor reviews the allocations for equity; (7)
although VA has made progress in improving the equity of resource
allocations nationwide among the networks, it has done little to ensure
that the networks fulfill the Veterans Equitable Resource Allocation
(VERA) system's promise as they allocate resources to their facilities;
(8) although GAO prepared an overall assessment of access to care,
difficulties in working with the data cast doubt on whether VA can
perform timely and effective oversight; (9) the information GAO
developed on changes in access to care at the facility and network
levels for VISN 3 and VISN 4, as well as for VA nationally, was gathered
from many VA reports and databases--some of which had inconsistent or
incompatible information that GAO was able to resolve; (10) moreover,
medical center, VISN, and headquarters officials told GAO that such data
are not available on a routine, timely basis; (11) without such
information, it is difficult for them to say conclusively whether VA has
improved veterans' equity of access to care and whether veterans have
not been adversely affected by the many changes under way to reduce
costs and improve productivity; (12) by taking several actions, VA could
improve its oversight of changes in access to care and its resource
allocation process; and (13) these actions include improving data
collection and dissemination efforts regarding changes in access to care
and establishing criteria for VISNs to use for more equitably allocating
resources to their facilities.

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

 REPORTNUM:  HEHS-98-226
     TITLE:  VA Health Care: More Veterans Are Being Served, But Better 
             Oversight Is Needed
      DATE:  08/28/98
   SUBJECT:  Health resources utilization
             Veterans hospitals
             Veterans benefits
             Veterans
             Health care cost control
             Health care programs
             Health services administration
             Patient care services
             Federal agency reorganization
IDENTIFIER:  VA Veterans Integrated Service Network
             VA Veterans Equitable Resource Allocation System
             VA Ambulatory Care Patient Satisfaction Survey
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


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


Report to Congressional Requesters

August 1998

VA HEALTH CARE - MORE VETERANS ARE
BEING SERVED, BUT BETTER OVERSIGHT
IS NEEDED

GAO/HEHS-98-226

Access to VA Health Care

(406146)


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

  ARC - Allocation Resource Center
  FTEE - full-time employee equivalent
  HUD - Department of Housing and Urban Development
  PRRTP - Psychiatric Residential Rehabilitation Treatment Program
  PTSD - post-traumatic stress disorder
  SCI - spinal cord injury
  SMI - serious mental illness
  VA - Department of Veterans Affairs
  VAMC - Veterans Affairs medical center
  VERA - Veterans Equitable Resource Allocation
  VHA - Veterans Health Administration
  VISN - Veterans Integrated Service Network

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


B-280054

August 28, 1998

Congressional Requesters

This report presents information on changes in veterans' access to
health care in Veterans Integrated Service Network (VISN) 3
headquartered in Bronx, New York, and VISN 4 headquartered in
Pittsburgh, Pennsylvania.  It also discusses major reforms under way
in the Department of Veterans Affairs' funding and managing of health
care nationwide.  This report is required by Conference Report No. 
105-297 accompanying P.L.  105-65, Departments of Veterans Affairs
and Housing and Urban Development and Independent Agencies
Appropriations Act, 1998. 

We are sending this report to the Secretary of Veterans Affairs and
other interested parties.  We will also make copies available upon
request. 

If you have any questions about the report, please call me at (202)
512-7101 or Bruce Layton, Assistant Director, at (202) 512-6837. 
Other major contributors are listed in appendix II. 

Stephen P.  Backhus
Director, Veterans' Affairs and
 Military Health Care Issues

List of Requesters

The Honorable Ted Stevens
Chairman
The Honorable Robert C.  Byrd
Ranking Minority Member
Committee on Appropriations
United States Senate

The Honorable Bob Livingston
Chairman
The Honorable David R.  Obey
Ranking Minority Member
Committee on Appropriations
House of Representatives

The Honorable Christopher S.  (Kit) Bond
Chairman
The Honorable Barbara Mikulski
Ranking Minority Member
Subcommittee on VA, HUD,
 and Independent Agencies
Committee on Appropriations
United States Senate

The Honorable Jerry Lewis
Chairman
The Honorable Louis Stokes
Ranking Minority Member
Subcommittee on VA, HUD,
 and Independent Agencies
Committee on Appropriations
House of Representatives



EXECUTIVE SUMMARY
============================================================ Chapter 0


   PURPOSE
---------------------------------------------------------- Chapter 0:1

In the last several years, the Department of Veterans Affairs (VA)
has introduced two major initiatives to change the way it manages its
$17 billion health care system.  In fiscal year 1996, VA
decentralized the management structure of its Veterans Health
Administration (VHA) to form 22 Veterans Integrated Service Networks
(VISN) to coordinate the activities of hundreds of hospitals,
outpatient clinics, nursing homes, and other facilities.  VA expected
the geographically distinct VISNs to improve efficiency by reducing
unnecessarily duplicative services (for example, by consolidating
medical facilities and programs) and shifting services from costly
inpatient care to less costly outpatient care.  VA expected access to
improve because it could redirect resources to serve more patients. 
To accomplish these goals, VA gave each VISN substantial operational
autonomy and established performance measures to hold network and
medical center directors accountable for achieving the goals. 

In addition, in April 1997, VA began to phase in the Veterans
Equitable Resource Allocation (VERA) system to allocate resources to
the 22 VISNs.  (Implementation of VERA will be complete in about
2000.) Before VERA, each medical center received and managed its own
budget.  VA designed VERA in response to 1996 legislation that
required VA to reduce historic regional inequities in resource
allocation and improve veterans' access to care.\1 Inequities had
resulted from a dramatic shift in the veteran population from the
Northeast and Midwest to the South and West that took place without a
respective shift in resource allocation.  To allocate resources more
equitably, VA uses VERA to move funds among the networks.  Each VISN
is then responsible for allocating those resources among the
facilities in its prescribed geographic area to ensure care and
equitable access within the network and to accomplish other national
VA goals such as reducing costs. 

Concerned that some VISNs would be required to implement significant
cost-saving steps to manage within the diminished resources they
would receive under VERA and that these VISNs would reduce veterans'
access to care as a result, the Committees on Appropriations directed
GAO to examine changes in access to care in two VISNs, VISN 3
headquartered in Bronx, New York, and VISN 4 headquartered in
Pittsburgh, Pennsylvania.  VA projected that VISN 3 (Bronx) would
lose the largest proportion of resources compared with other networks
and that VISN 4 (Pittsburgh) would lose some resources, but the
change would be the smallest for any VISN.  As requested, GAO is
reporting on three issues:  (1) changes in overall access to care,
changes in access to certain specialized services, and a comparison
of changes in these networks with VA national data from fiscal years
1995 to 1997; (2) the extent to which VA headquarters and VISNs are
working to equitably allocate resources to facilities within VISNs;
and (3) the adequacy of VA's oversight of changes in access to care. 


--------------------
\1 Section 429 of P.L.  104-204. 


   BACKGROUND
---------------------------------------------------------- Chapter 0:2

VA operates one of the nation's largest health care systems,
encompassing approximately 400 service delivery locations.  In fiscal
year 1997, it provided care to about 2.7 million of the nation's 26
million veterans.  The Congress requires VA to provide services on a
priority basis to veterans with service-connected disabilities, low
incomes, or special health care needs--also referred to as Category A
veterans.  It may also provide services to other veterans as
resources allow. 

To improve the efficiency of its system and veterans' access to care,
VA is fundamentally changing its health care delivery.  Borrowing
ideas from the private sector, VA has increased its emphasis on
applying managed care practices, such as primary, outpatient, and
preventive care, and decreased its emphasis on providing inpatient
care.  VA is trying to reengineer its system, while maintaining its
core mission of efficiently managing and financing the often costly
care of special populations with chronic conditions such as spinal
cord injury (SCI) or serious mental illness. 


   RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3

Overall, VISN 3 (Bronx), VISN 4 (Pittsburgh), and VA nationally have
increased access as measured by increases in the number of veterans
served.  For example, between fiscal years 1995 and 1997, VISN 3
(Bronx) increased the number of veterans served by more than 2
percent; VISN 4 (Pittsburgh), by nearly 22 percent; and VA
nationally, by more than 5 percent.  Access to care, as measured by
patient satisfaction, also seems to have improved according to
responses to VA surveys and interviews GAO conducted.  In addition,
VA has improved geographic access to primary care by increasing the
number of community-based clinics in these two VISNs.  Although
access has increased overall, access appears to have decreased for
some specific services GAO reviewed.  For example, VISN 3 (Bronx)
served fewer patients with SCIs, and VISN 4 (Pittsburgh) served fewer
patients with post-traumatic stress disorder (PTSD). 

The two VISNs GAO reviewed used no specific criteria for allocating
their resources to reduce historical access inequities among their
facilities.  VA headquarters neither provides criteria for VISNs to
use to equitably allocate resources nor reviews the allocations for
equity.  Although VA has made progress in improving the equity of
resource allocations nationwide among the networks, it has done
little to ensure that the networks fulfill VERA's promise as they
allocate resources to their facilities. 

Although GAO prepared an overall assessment of access to care,
difficulties in working with the data cast doubt on whether VA can
perform timely and effective oversight.  The information GAO
developed on changes in access to care at the facility and network
levels for VISN 3 (Bronx) and VISN 4 (Pittsburgh), as well as for VA
nationally, was gathered from many VA reports and databases--some of
which had inconsistent or incompatible information that GAO was able
to resolve.  Moreover, medical center, VISN, and headquarters
officials told us that such data are not available on a routine,
timely basis--particularly for specific programs.  Without such
information, it is difficult for them to say conclusively whether VA
has improved veterans' equity of access to care and whether
veterans--
particularly those who had been receiving high-cost care for chronic
conditions--have not been adversely affected by the many changes
under way to reduce costs and improve productivity. 

By taking several actions, VA could improve its oversight of changes
in access to care and its resource allocation process.  These actions
include improving data collection and dissemination efforts regarding
changes in access to care and establishing criteria for VISNs to use
for more equitably allocating resources to their facilities. 


   PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4


      ACCESS TO CARE CONTINUED TO
      INCREASE UNDER VERA
-------------------------------------------------------- Chapter 0:4.1

Veterans' access to care generally continued to increase under VERA
in VISN 3 (Bronx), VISN 4 (Pittsburgh), and VA nationally as measured
by the number of veterans served (see table 1).  Increases in access
to care predate VERA but appear to have accelerated with VERA's
implementation in April 1997. 



                          Table 1
          
          Change in Patients Served, Fiscal Years
                          1995-97

                   All patients served\a
                ----------------------------
                                                Percentage
                                                 change in
                                              all patients
                  Fiscal    Fiscal    Fiscal       served,
                    year      year      year  fiscal years
Entity              1995      1996      1997       1995-97
--------------  --------  --------  --------  ------------
VISN 3 (Bronx)   148,398   148,865   151,611          +2.2
VISN 4           139,049   145,641   169,398         +21.8
 (Pittsburgh)
VA nationally   2,843,53  2,895,81  2,996,34          +5.4
                       4         9         6
----------------------------------------------------------
\a Unduplicated count of patients served each year. 

Source:  VA. 

The two VISNs and VA nationally are improving access by providing
more services on an outpatient basis; providing more health care
service locations, for example, establishing community-based and
mobile clinics; and shortening veterans' waiting times for receiving
services.  For example, VISN 3 (Bronx) opened seven new service
sites, and VISN 4 (Pittsburgh) opened four during fiscal years 1996
and 1997, improving veterans' geographic access to care.  The two
VISNs have also used outreach efforts, namely, hospital- and
community-based health fairs and screenings, to identify veterans not
previously served.  Veterans' satisfaction with access to care has
improved according to responses to VA surveys and interviews GAO
conducted. 

Although VA served more veterans in nearly all the specific services
GAO reviewed, fewer veterans received some VA services.  For example,
the number of SCI patients treated in SCI centers and clinics
declined in VISN 3 (Bronx) from 467 in fiscal year 1995 to 441 in
fiscal year 1997.  The number of PTSD patients declined in VISN 4
(Pittsburgh) from 2,173 to 2,155.  The number of patients treated
with a primary diagnosis of substance abuse declined by about 5
percent nationwide and about 3 percent in VISN 3 (Bronx). 

The two VISNs increased veterans' access to care despite reductions
in the buying power of their allocations by increasing the efficiency
of their health care delivery.  For VISN 3 (Bronx), lower VERA
allocations and VISN management decisions combined with inflation
reduced the buying power of its more than a $1 billion budget by $91
million from fiscal year 1995 to fiscal year 1997.  For VISN 4
(Pittsburgh), the buying power of its budget of more than $800
million declined $17 million in the same period.  To achieve
efficiencies, VISN 3 (Bronx) reduced its full-time employee
equivalents (FTEE) by 2,070 (about 15 percent) during this period,
for a cost reduction of about $110 million.  VISN 4 (Pittsburgh)
reduced its FTEEs by 1,485 (about 14 percent) for a cost reduction of
about $79 million.  Both VISNs also increased their productivity--for
example, by expanding clinic hours, increasing the number of
examination rooms, and improving scheduling--
resulting in each provider seeing more patients.  Officials from both
VISNs said they expect to continue increasing access to care in
fiscal year 1998 by continuing to improve efficiency and obtaining
new resources from third-party health insurance collections.  For
fiscal year 1998, VA set a collection goal of about $596 million
nationally, about $44 million for VISN 3 (Bronx), and about $36
million for VISN 4 (Pittsburgh). 


      VA HAS DONE LITTLE TO
      IMPROVE EQUITY OF RESOURCE
      ALLOCATIONS TO FACILITIES
-------------------------------------------------------- Chapter 0:4.2

As permitted under VA's decentralized management structure, the two
VISNs use different methods to allocate resources among their
facilities.  Neither VISN used criteria, however, to address
equitable allocation issues in facility allocations for fiscal years
1997 and 1998 nor did headquarters provide criteria for the VISNs to
use.  Historical inequities have existed within as well as among
VISNs, and VISN 3 (Bronx) and VISN 4 (Pittsburgh) recognized
historical inequities in their respective networks in their fiscal
year 1998 strategic plans.  Equitably allocating resources within
each VISN is important in following through on VERA's promise of a
more equitable allocation of resources for the nation's veterans. 
Without an equitable allocation of resources within their networks,
VISN 3 (Bronx) and VISN 4 (Pittsburgh) face the risk of growth in
access without progress in achieving equitable access. 


      VA OVERSIGHT OF ACCESS TO
      CARE IS INADEQUATE
-------------------------------------------------------- Chapter 0:4.3

Although GAO prepared an overall assessment of access to care,
difficulties in working with the data cast doubt on whether VA can
perform timely and effective oversight.  The information GAO
developed on changes in access to care at the facility and network
levels for VISN 3 (Bronx) and VISN 4 (Pittsburgh), as well as VA
nationally, was gathered from many VA reports and databases--some of
which had inconsistent or incompatible information.  Responsibility
for generating data and reporting results is fragmented in VA's
system; definitions for data on the number of patients served have
changed several times, reducing their comparability; and managers
lack timely and useful information on waiting times for care and
satisfaction with access.  VA managers told us they do not have
timely, comparable, and comprehensive information, particularly for
specific programs, that they need to monitor changes in access to
care--including changes in the equity of access within and among
networks--to understand what is happening locally and nationwide.  As
a result, they cannot be certain that veterans--particularly those
who have been receiving high-cost care for chronic conditions--are
not adversely affected by the many changes under way to reduce costs
and improve productivity. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5

To improve VA's oversight of changes in access to care, GAO is making
several recommendations to the Secretary of Veterans Affairs.  These
recommendations require VA to develop uniform definitions and timely
reporting of changes in access to care, develop criteria for
equitably allocating resources to facilities, and monitor
improvements to equity of access within and among VISNs. 


   AGENCY COMMENTS AND GAO'S
   EVALUATION
---------------------------------------------------------- Chapter 0:6

In commenting on a draft of this report (see app.  I), VA said it is
working to improve its information systems so that they will be more
useful to VISN and headquarters management.  VA expressed concern,
however, that GAO's recommendation to develop national criteria for
equitably allocating resources to facilities with national oversight
is contrary to VHA's reengineering philosophy, which decentralizes
authority and accountability for these allocations to the network
directors. 

GAO supports VA's intention to improve its data systems. 
Improvements are essential to allow VA managers to identify problems
and take corrective action in a more timely way to help ensure that
veterans' access to care does not deteriorate in the environment of a
transformed VA health care system.  However, GAO believes VA can
develop criteria for VISNs' use in equitably allocating resources to
their respective facilities and review VISNs' performance in
addressing these criteria without being so prescriptive that local
authority and accountability are compromised.  Having criteria does
not preclude VISNs from using different methods for allocating
resources to address local circumstances and VA's national criteria. 


BACKGROUND
============================================================ Chapter 1

The Department of Veterans Affairs (VA), the nation's largest
integrated health care system, has fundamentally changed the way it
manages and delivers health care to veterans.\2 Two major initiatives
have been central to its strategy to reduce costs and expand access. 
First, in fiscal year 1996, VA decentralized the management structure
of its Veterans Health Administration (VHA) to form 22 Veterans
Integrated Service Networks (VISN) to coordinate the activities of
hundreds of hospitals, outpatient clinics, nursing homes, and other
facilities in each area.\3 VA gave each network substantial
operational autonomy and established performance measures to hold
network and medical center directors accountable for achieving VA's
goals. 

In April 1997, VA began to phase in its second initiative, the
Veterans Equitable Resource Allocation (VERA) system for allocating
resources to the 22 VISNs.  VERA is designed to allocate comparable
resources for each veteran user among VISNs.  When VERA is fully
implemented in about 2000, VA expects it to shift resources from
VISNs in the Northeast and Midwest, such as VISN 3 (Bronx) and VISN 4
(Pittsburgh), to VISNs in the South and West, correcting historic
regional inequities in resource allocation.  This allocation method
requires that VISNs, in turn, equitably allocate the resources they
receive to their respective medical facilities. 


--------------------
\2 Vision for Change:  A Plan to Restructure the Veterans Health
Administration, VA (Washington, D.C.:  Mar.  1995); Prescription for
Change:  The Guiding Principles and Strategic Objectives Underlying
the Transformation of the Veterans Healthcare System, VA (Washington,
D.C.:  Mar.  1996); and Journey of Change, VA (Washington, D.C.: 
Apr.  1997). 

\3 VHA is the organizational unit within VA responsible for providing
medical care to eligible veterans. 


   VA HAS CHANGED ITS DELIVERY OF
   HEALTH CARE
---------------------------------------------------------- Chapter 1:1

The VA health care system, which has about 400 service delivery
locations, spent about $17 billion to provide care to approximately
2.7 million of the nation's 26 million veterans in fiscal year 1997. 
The Congress requires VA to provide services on a priority basis to
veterans with service-connected disabilities, low incomes, or special
health care needs--commonly referred to as Category A veterans.  VA
may also provide services to other veterans as resources allow. 

To improve the efficiency of its system and veterans' access to care,
VA is fundamentally changing its health care delivery.  Applying
lessons learned from the private sector's experience with managed
health care, VA has increased its emphasis on primary, outpatient,
and preventive care and decreased its emphasis on inpatient care.  VA
has a particularly challenging task because its core mission includes
caring for patients with chronic conditions such as spinal cord
injuries (SCI) or serious mental illnesses (SMI).  In addition, the
often costly and longer term care of such patients has not typically
been included on a large scale in the health care sector in
general.\4


--------------------
\4 For a discussion of related issues, see Medicaid Managed Care: 
Serving the Disabled Challenges State Programs (GAO/HEHS-96-136, July
31, 1996). 


   TWO KEY INITIATIVES HAVE
   RESHAPED MANAGEMENT OF VA
   HEALTH CARE
---------------------------------------------------------- Chapter 1:2

In the last few years, VA has undertaken two major initiatives in
reshaping its service delivery system to expand access, become a more
efficient provider of care, and improve equity of resource
allocations nationwide.  These initiatives are the decentralization
of VA's health care management structure and the creation of a new
resource allocation system. 


      VA HEALTH CARE MANAGEMENT
      STRUCTURE DECENTRALIZED
-------------------------------------------------------- Chapter 1:2.1

In fiscal year 1996, VHA shifted management authority from
headquarters to 22 newly created VISNs, each led by a director and a
staff of medical, budget, and administrative officials.  (See fig. 
1.1 for a map of the VISNs.) VISNs have been organized in part on the
basis of VHA's natural patient referral patterns and the aggregate
number of beneficiaries and facilities needed to support their care. 
VISNs have substantial operational autonomy and perform the basic
decisionmaking and budgetary duties of the VA health care system. 
Each network allocates funds and monitors the operations of the
hospitals, outpatient clinics, nursing homes, domiciliaries, and
other medical programs in its geographic area.\5

Before the creation of VISNs, each medical center received and
managed its own budget.  VISNs vary in several ways, including
geographic size--ranging from about 10,000 square miles in VISN 3
(Bronx) to 885,000 square miles in VISN 20 (Portland)--and the
services provided, reflecting, for example, historically longer
inpatient and nursing home stays in the Northeast. 

   Figure 1.1:  Veterans
   Integrated Service Networks

   (See figure in printed
   edition.)

Source:  VA. 

Under the decentralized VISN system, network and medical center
directors are held accountable for increasing efficiency and
improving access.  VISNs are responsible for improving access and
reducing costs in part by implementing efficiencies that shift
resources from costly inpatient care to less costly outpatient care. 
They can consolidate programs and facilities to eliminate duplicative
services.  Network directors are held accountable by performance
measures for systemwide and network-specific goals for increasing the
number of outpatient surgeries, reducing the use of inpatient care,
and increasing the number of high-priority veterans served who had
not previously received care from VA.  These goals are generally
similar to those identified in VA's Strategic Plan for fiscal years
1998 to 2003.\6

VA has expanded outpatient care in existing medical centers and
established more clinical settings to provide VA-provided or
-sponsored care in the community.  The Congress supported this
strategy by enacting legislation in October 1996 that eliminated
several restrictions on veterans' eligibility for VA outpatient care
and enabled VA to serve more patients on an outpatient basis.\7 This
legislation also permitted VA to contract with other health care
providers to care for veterans in community-based clinics and other
non-VA settings. 


--------------------
\5 For a discussion of VA networks, see VA Health Care:  Status of
Efforts to Improve Efficiency and Access (GAO/HEHS-98-48, Feb.  6,
1998). 

\6 The Department of Veterans Affairs Strategic Plan, Fiscal Years
1998-2003, VA (Washington, D.C.:  Sept.  30, 1997) is VA's first
strategic plan based on the requirements of the Government
Performance and Results Act of 1993.  The act requires agencies to
set goals, measure performance, and report on their accomplishments. 
The intent is for an agency to define what desired results it wishes
to achieve, identify the strategy to achieve the desired results, and
then determine how well it succeeded in reaching results-oriented
goals and achieving objectives. 

\7 P.L.  104-262. 


      SYSTEM ADOPTED TO IMPROVE
      EQUITY OF RESOURCE
      ALLOCATION
-------------------------------------------------------- Chapter 1:2.2

The Congress required VA to address inequities in the allocation of
resources nationwide.  It required VA to develop a plan for equitably
allocating resources to ".  .  .  ensure that veterans who have
similar economic status and eligibility priority and who are eligible
for medical care have similar access to such care regardless of the
region of the United States in which such veterans reside."\8 A
dramatic shift in the veteran population from the Northeast and
Midwest to the South and West without a respective shift in resource
allocations caused the inequities.  Allocations did not shift when
the veteran population did because VA allocated resources to
facilities primarily on the basis of their historical budgets, and
facilities were disproportionately located in the Northeast and
Midwest.  As a result, VA's resources were not equitably allocated
nationwide and VA could not ensure that veterans who had similar
economic status and eligibility priority had similar access to care. 

In response to the legislative requirement, VA developed VERA and
began to phase it in during fiscal year 1997.  VERA allocates
resources to the networks and provides them incentives for achieving
cost efficiencies and serving more veterans.  VERA has improved the
equity of resource allocation to networks because, compared with the
system it replaced, it provides more comparable levels of resources
to each network for each high-
priority veteran served.\9 A crucial element in VA's overall
allocation strategy is VISNs' allocations of VERA resources to their
facilities.  Each VISN is responsible for allocating resources among
its facilities to achieve equitable access within its prescribed
geographic area and to accomplish other national VA goals such as
reducing costs.  Allocations to an individual medical facility may
increase or decrease in any year regardless of whether overall
allocations to its VISN increase or decrease. 

To improve equity of resource allocation, VA uses VERA to move funds
among the networks.  Networks that increase their patient workload
compared with other networks gain resources under VERA; those whose
patient workloads decrease compared with other networks lose
resources.  More efficient networks (that is, those whose patient
care costs are below the VA national average cost) have more funds
available for local initiatives.  Less efficient networks (whose
patient care costs are above the VA national average cost), however,
must increase efficiency to have such funds available.  Because
patient costs in the Northeast and Midwest networks have generally
been above the national average, after adjustments for case mix and
labor costs, VA has projected that most VISNs in those regions will
have reduced allocations under VERA. 

VERA allocates about 89 percent of VA's medical care resources to the
VISNs at the start of a fiscal year.  For the most part, VISNs can
use this general-
purpose allocation as they deem appropriate.  VA headquarters
allocates almost all the remaining medical care appropriation to
VISNs and to facilities throughout the fiscal year for specific
purposes such as prosthetics, state veterans' homes, readjustment
counseling, and other activities.  The use of these funds is
restricted to the purpose for which they are allocated. 

The formula for the general-purpose allocation includes two key
estimates:  an estimate of the number of high-priority veterans a
network can serve for routine services (called basic care) and an
estimate of the number who could be served for more expensive,
complex care for patients with chronic conditions (called special
care).\10 The formula calculates a VISN's allocation on the basis of
the number of veterans served in each category and the average
national cost of care for a patient in each category, which is
referred to as the capitation rate.  Adjustments for regional labor
costs are also made.  The special care capitation rate is higher to
ensure that networks with disproportionately large numbers of
patients with complex or chronic conditions, such as SCI, advanced
acquired immunodeficiency syndrome, or chronic mental illness, have
adequate funds to care for patients with these more costly
conditions.\11 A facility's expenses for treating an individual
patient or a group of patients in either the basic or the special
care category, however, may exceed or be below the capitation rate
for that category. 


--------------------
\8 Section 429 of P.L.  104-204. 

\9 See VA Health Care:  Resource Allocation Has Improved, but Better
Oversight Is Needed (GAO/HEHS-97-178, Sept.  17, 1997) for a
discussion of issues related to VERA. 

\10 Each VISN also receives funding by formula for other
health-related functions, including research, education, equipment,
and nonrecurring maintenance through the use of national cost
estimates for each activity. 

\11 The VERA special care category also includes some adjustment for
age to account for expected changes in the age distribution of
veterans in a network. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:3

Concerned that some VISNs would be required to implement significant
cost-saving steps to manage within the diminished resources they
would receive under VERA and that these VISNs would reduce veterans'
access to care as a result, the Committees on Appropriations directed
us to examine changes in access to care in two VISNs:  VISN 3
headquartered in Bronx, New York, and VISN 4 headquartered in
Pittsburgh, Pennsylvania.  Additional concerns--that the quality of
care was declining in VISN 3 (Bronx) as a result of allocation
reductions--were being investigated by VA.\12 VA has projected that
VISN 3 (Bronx) would lose the largest proportion of resources
compared with other networks when VERA is fully implemented in about
the year 2000.  VA has also projected that VISN 4 (Pittsburgh) would
lose some resources, but VA projected this VISN to have the smallest
change of any VISN.  Concerns were also expressed about the method
used by VISN 4 (Pittsburgh) for allocating resources to facilities
within its area.  As requested, we are reporting on three issues: 
(1) changes in overall access to care, changes in access to certain
specialized services, and a comparison of changes in these networks
with VA national data from fiscal years 1995 to 1997; (2) the role of
VA headquarters and VISNs in determining equitable allocations to
facilities within VISNs; and (3) the adequacy of VA's oversight of
changes in access to care. 

VISN 3 (Bronx) provides VA health care in the southern Hudson River
Valley of New York, New York City, Long Island, and northern New
Jersey.  The network has six medical centers\13 composed of nine
geographically distinct facilities.  The network recently implemented
two facility integrations:  East Orange and Lyons VA medical centers
(VAMC) were integrated into the New Jersey Healthcare System in
fiscal year 1996, and the Montrose and Castle Point VAMCs were
integrated into the Hudson Valley Healthcare System in fiscal year
1997.  A unique feature of this network is the geographic proximity
of its medical centers--all are within a 60-mile radius in an area
where an estimated 1.4 million veterans lived in 1997.  VISN 3
(Bronx) covers the smallest geographic area of the 22 networks. 

VISN 4 (Pittsburgh) provides VA health care in Delaware, most of
Pennsylvania, southern New Jersey, and parts of West Virginia, Ohio,
and New York.  The network has 10 medical centers in 12 geographic
locations.\14 In October 1996, VA integrated three facilities to form
the Pittsburgh Healthcare System.  The area the network covers had
approximately 1.7 million veterans in 1997. 

To assess changes in veterans' access to services and VA's monitoring
of those changes, we reviewed data primarily from fiscal years 1995
to 1997.  We used these data to assess overall changes in access for
patients and to assess changes in access for specific patient groups: 
SCI patients; patients with mental illnesses, including SMI,
post-traumatic stress disorder (PTSD), and substance abuse; and
patients receiving surgical services.  Finally, we used the data to
assess the availability of prosthetics services.  We chose most of
these services because of concerns raised by veterans' service
organizations and others that such services' relatively high cost
could lead to pressures to decrease their availability.  We chose to
examine access to surgical services because of the major changes VA
has implemented to expand the use of outpatient surgery, while
reducing inpatient surgeries.  We collected and analyzed data on the
number of veterans served, waiting times, service locations,
financial and employee resources, and veterans' satisfaction with
access to services, among other indicators. 

We visited network offices in VISN 3 (Bronx) and VISN 4 (Pittsburgh)
and obtained documents from and interviewed network management and
staff.  In VISN 3 (Bronx), we visited three VAMCs in New York--Bronx,
Montrose, and Castle Point--and two in New Jersey--East Orange and
Lyons.  In VISN 4 (Pittsburgh), we visited the Highland Drive and
University Drive divisions of the Pittsburgh Healthcare System, and
the VAMCs in Butler, Pennsylvania, and Clarksburg, West Virginia.  In
visiting these facilities, we obtained data from and interviewed
directors, financial officers, physicians, nurses, union
representatives, and local veterans' service organization
representatives. 

We gathered information on network allocation of resources to
facilities from network offices and the facilities we visited.  We
also conducted telephone interviews with officials from the
Philadelphia, Lebanon, and Wilkes-Barre, Pennsylvania, VAMCs
concerning VISN 4's (Pittsburgh) method for allocating resources to
its facilities. 

To obtain or corroborate VA national, network, and facility data, we
also interviewed officials and reviewed documents from VHA's many
organizations.  These included the Office of Policy and Planning;
Office of Performance and Quality; Office of the Chief Financial
Officer; Office of Inspector General; Program Evaluation and Resource
Center, Palo Alto, California; Northeast Program Evaluation Center,
West Haven, Connecticut; Allocation Resource Center, Boston,
Massachusetts; National Performance Data Resource Center, Durham,
North Carolina; National Customer Feedback Center, West Roxbury,
Massachusetts; VISN Support Service Center, San Francisco,
California; and headquarters strategic health care groups on surgery,
prosthetics, and mental health.  When we identified inconsistencies
between databases, we tried to resolve them by interviewing officials
responsible for creating or maintaining the databases, updating the
databases with additional information VA provided, and requesting
special data runs with parameters that we specified.  We asked VA
officials to review the data we used in this report to ensure
accuracy. 

We performed our review in accordance with generally accepted
government auditing standards between October 1997 and August 1998. 


--------------------
\12 During our work, the Office of the Medical Inspector reported
that quality of care problems it identified in the Hudson Valley
Healthcare System were not related to changes in VISN 3's (Bronx)
allocation.  See Final Report, FDR Hospital, Montrose, N.Y., VA
Medical Center Castle Point, N.Y., VA Hudson Valley Healthcare
System, VISN 3 (6 volumes), VA Office of the Medical Inspector
(Washington, D.C.:  Dec.  1, 1997). 

\13 Medical centers in VISN 3 (Bronx) include Bronx, Brooklyn, New
York, Northport, Hudson Valley Healthcare System, and New Jersey
Healthcare System. 

\14 Medical centers in VISN 4 (Pittsburgh) include Altoona, Butler,
Coatesville, Erie, Lebanon, Philadelphia, Pittsburgh, and
Wilkes-Barre, Pa.; Wilmington, Del.; and Clarksburg, W.  Va. 


ACCESS TO CARE CONTINUES TO
INCREASE UNDER VERA
============================================================ Chapter 2

The many data sources we reviewed showed generally improved veterans'
access to care under VERA in VISN 3 (Bronx), VISN 4 (Pittsburgh), and
VA nationally.  Even as VERA began shifting some resources from
northeastern and midwestern VISNs to other areas in fiscal year 1997,
overall, these two VISNs continued to provide increased access to
care as they had before VERA's implementation.\15 Today, these two
VISNs are serving more veterans, while the distance veterans need to
travel for care in these VISNs is generally decreasing.  Veterans'
satisfaction with access to care and the amount of time veterans
report waiting for care are also improving overall.  Although access
is improving overall, it appears to have decreased for some services
in certain locations, according to our review. 

VISN and VAMC officials told us they improved access in part by
increasing their efficiency.  In doing so, they served more veterans
by using these efficiencies to offset their reduced buying power
resulting from VERA and from inflation. 


--------------------
\15 VA Health Care:  Status of Efforts to Improve Efficiency and
Access (GAO/HEHS-98-48, Feb.  6, 1998). 


   OVERALL ACCESS TO CARE IS
   IMPROVING IN VISN 3 (BRONX),
   VISN 4 (PITTSBURGH), AND VA
   NATIONALLY
---------------------------------------------------------- Chapter 2:1

VISN 3 (Bronx), VISN 4 (Pittsburgh), and VA nationally increased the
number of unique patients and those who are high priority, that is,
Category A veterans, served from fiscal year 1995 to fiscal year 1997
(see table 2.1).\16 VISN 4 (Pittsburgh) had the highest rate of
increase of any VISN.  Changes in total patients and Category A
veterans served varied significantly by VISN. 



                                         Table 2.1
                          
                          Change in Unique Patients Served, Fiscal
                                       Years 1995-97

              All patients served                  Category A veterans served
          ---------------------------              ---------------------------
                                       Percentage                               Percentage
                                        change in                                change in
                                              all                               Category A
                                         patients                                 veterans
                                          served,                                  served,
            Fiscal    Fiscal   Fiscal      fiscal    Fiscal    Fiscal   Fiscal      fiscal
              year      year     year       years      year      year     year       years
Entity        1995      1996     1997     1995-97      1995      1996     1997     1995-97
--------  --------  --------  -------  ----------  --------  --------  -------  ----------
VISN 3\a   148,398   148,865  151,611        +2.2   115,758   115,502  118,217        +2.1
Bronx       24,689    23,963   22,229       -10.0    19,636    19,387   18,806        -4.2
Brooklyn    30,878    31,033   30,033        -2.7    23,867    23,771   23,467        -1.7
Hudson      15,440    16,774   19,807       +28.3    11,442    12,549   14,478       +26.5
 Valley
 Healthc
 are
 System\
 b
New         36,805    37,417   38,442        +4.4    29,312    29,902   30,240        +3.2
 Jersey
 Healthc
 are
 System\
 c
New York    30,938    31,650   29,009        -6.2    27,154    27,677   25,430        -6.3
Northpor    22,208    22,819   23,949        +7.8    16,572    16,589   17,521        +5.7
 t
VISN 4\a   139,049   145,641  169,398       +21.8   118,240   124,817  144,018       +21.8
Altoona      7,449     8,106    8,894       +19.4     6,841     7,372    7,978       +16.6
Butler       7,146     8,869   11,333       +58.6     6,293     7,735    9,002       +43.0
Clarksbu    10,507    10,752   12,539       +19.3     9,259     9,652   11,069       +19.5
 rg
Coatesvi     8,232     9,123   12,450       +51.2     6,402     7,373   10,275       +60.5
 lle
Erie         9,359     9,729   11,401       +21.8     8,205     8,416    9,706       +18.3
Lebanon     13,655    14,230   17,025       +24.7    11,513    12,219   14,574       +26.6
Philadel    27,971    28,270   30,941       +10.6    24,256    24,451   26,942       +11.1
 phia
Pittsbur    34,220    35,469   40,154       +17.3    28,626    30,306   34,761       +21.4
 gh
 Healthc
 are
 System\
 d
Wilkes-     19,192    20,993   25,628       +33.5    17,276    18,767   22,487       +30.2
 Barre
Wilmingt    13,517    13,469   14,310        +5.9    11,638    11,714   12,242        +5.2
 on
VA        2,843,53  2,895,81  2,996,3        +5.4  2,421,47  2,451,76  2,555,5        +5.5
 nationa         4         9       46                     6         6       12
 lly
------------------------------------------------------------------------------------------
\a The total number of unique patients served by each VISN is less
than the sum of unique patients served by each of its facilities
because some patients receive care at more than one facility. 

\b The Hudson Valley Healthcare System was formed by the integration
of Montrose and Castle Point VAMCs. 

\c The New Jersey Healthcare System was formed by the integration of
East Orange and Lyons VAMCs. 

\d The Pittsburgh Healthcare System was formed by the integration of
Pittsburgh, Highland Drive; Pittsburgh, University Drive; and
Aspinwall facilities. 

Source:  VISN Support Service Center, San Francisco, Calif., and
National Performance Data Resource Center, Durham, N.C.  At our
request, these centers provided data for fiscal years 1995 to 1997
consistent with the fiscal year 1998 performance indicator.  This
indicator counts unique users for 12 months each year.  Indicators VA
used previously were not comparable. 

VISN and medical center officials told us that they wanted to
increase the number of veterans served for several reasons.  They
told us it helped them meet national VA goals for expanding access. 
Increasing this workload also increases a VISN's future allocation
because VERA generally allocates resources to each VISN on the basis
of the number of high-priority veterans (Category A) served.  In
addition, VAMC officials in VISN 4 (Pittsburgh) told us they had
another incentive to increase workload:  the network allocated some
resources to facilities for each additional veteran served in fiscal
year 1997. 

VISN 3 (Bronx), VISN 4 (Pittsburgh), and VA nationally are continuing
to increase the number of veterans they serve in fiscal year 1998. 
On the basis of veterans served through the first half of fiscal year
1998, VA projects that both the VISNs and VA will serve more Category
A veterans in fiscal year 1998 than in fiscal year 1997 (see table
2.2). 



                         Table 2.2
          
          Projected Growth in Category A Veterans
                Served, Fiscal Years 1997-98

                                   Projected
                    Category A    Category A
                      veterans   veterans to
                       served,    be served,     Projected
                   fiscal year   fiscal year    percentage
Entity                    1997          1998        change
----------------  ------------  ------------  ------------
VISN 3 (Bronx)         118,217       119,480          +1.1
VISN 4                 144,018       154,655          +7.4
 (Pittsburgh)
VA nationally        2,555,512     2,637,667          +3.2
----------------------------------------------------------
Source:  VHA 1998 2nd Quarter Network Performance Report. 

The VISNs' and VA's efforts to increase outpatient care significantly
affected the number of unique outpatients served in fiscal year 1997
(see table 2.3).  Although increases in the use of outpatient care
had been under way for more than a decade, the number of unique
patients served on an outpatient basis increased noticeably in fiscal
year 1997.



                         Table 2.3
          
           Changes in Unique Veterans Seen on an
           Outpatient Basis, Fiscal Years 1995-97

                                                Percentage
                                                   change,
                                                    fiscal
                Fiscal      Fiscal      Fiscal       years
Entity       year 1995   year 1996   year 1997     1995-97
----------  ----------  ----------  ----------  ----------
VISN 3         117,684     119,834     127,313        +8.2
 (Bronx)
VISN 4         115,268     122,631     148,323       +28.7
 (Pittsbur
 gh)
VA           2,454,936   2,502,554   2,644,722        +7.7
 nationally
----------------------------------------------------------
Source:  VA Office of Policy and Planning. 

VISN and facility officials told us that they implemented a variety
of initiatives to increase the number of patients served.  Expanding
the use of primary care teams in medical centers and community
clinics significantly contributed to serving more outpatients.\17

Medical center officials told us that to accommodate more
outpatients, they also expanded clinic hours for seeing patients,
improved scheduling, and expanded the number of examination rooms for
each provider to improve productivity. 

According to VISN and medical center officials, they could implement
these initiatives by changing their service delivery and by shifting
resources from inpatient to outpatient care.  For example, Clarksburg
VAMC officials told us they served more patients in their PTSD
program as well as shortened the time veterans spent waiting to enter
the program by changing it from an inpatient to a residential
program, reducing the number of days in the program, and changing the
staffing mix.  Both VISNs have reduced their average daily inpatient
census as has VA nationally.\18 (See table 2.4.)



                         Table 2.4
          
             Change in Inpatient Average Daily
                Census, Fiscal Years 1995-97

                                                Percentage
                                                   change,
                                                    fiscal
                Fiscal      Fiscal      Fiscal       years
Entity       year 1995   year 1996   year 1997     1995-97
----------  ----------  ----------  ----------  ----------
VISN 3           3,055       2,626       2,000       -34.5
 (Bronx)
VISN 4           2,474       2,021       1,377       -44.3
 (Pittsbur
 gh)
VA              37,003      31,666      24,047       -35.0
 nationally
----------------------------------------------------------
Source:  VA Summary of Medical Programs, Fiscal Years 1995 to 1997. 

Facilities in both VISNs also conducted active outreach efforts to
identify and serve veterans who had either never come to VA for
health care or not come in the last few years.  In VISN 4
(Pittsburgh), these efforts were supported by a full-time marketing
director for the network.  Among the outreach efforts, the Butler
VAMC held 8 health fairs in fiscal year 1996 and 19 the next year. 
These events, staffed at the medical center by physician assistants,
registered nurses, social workers, and other health professionals,
provided 15 services, including cholesterol screening, smoking
cessation and stress management, immunizations, counseling,
nutritional education, breathing tests, and more.  Furthermore, in
fiscal year 1997, Butler VAMC staff conducted 53 health screenings in
the community (twice the number conducted in fiscal year 1995),
offering some of the services provided at the health fairs. 

The VISNs and facilities we visited served more patients overall in
part by providing more services closer to veterans' homes, improving
geographic access to care.\19 VISN 3 (Bronx) established seven new
service sites in fiscal years 1996 and 1997, and VISN 4 (Pittsburgh)
provided four new services sites (see fig.  2.1).  VISN 3 (Bronx)
also improved geographic access by providing care in different areas
with mobile clinics; and both VISNs provided more specialty
outpatient care in hospitals and community clinics.  For example, the
Pittsburgh Healthcare System introduced an outpatient telemedicine
program in dermatology; the Bronx VAMC developed the capability of
providing annual physical examinations for veterans with SCI at the
medical center but in an outpatient setting; and the Clarksburg VAMC
introduced PTSD group therapy sessions at its Parsons, West Virginia,
community-based clinic. 

   Figure 2.1:  Increased Service
   Locations, VISN 3 (Bronx) and
   VISN 4 (Pittsburgh), Fiscal
   Years 1996-97

   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

Source:  VISN 3 (Bronx) and VISN 4 (Pittsburgh). 

Veterans we interviewed in VISN 3 (Bronx) and VISN 4 (Pittsburgh)
reported satisfaction with the increased availability of outpatient
care through community-based clinics and in medical centers.  They
said that the VISNs had improved their scheduling of appointments to
maximize the availability of primary care and provide more reliable
appointment times.  VA has reported similar information for the two
networks and for VA overall from responses to its Ambulatory Care
Patient Satisfaction Surveys.\20

Veterans' service organization representatives told us, however, that
difficulties remain in accessing care in some facilities.  For
example, they said that veterans who had to go to the Bronx VAMC
because the care they needed was not available in the Hudson Valley
Healthcare System had found van transportation to be a problem.  The
van made only one trip a day, and veterans sometimes had long waits
before or after being seen by physicians.  The medical centers have
taken steps to address these concerns.  For example, in May 1998, the
van service made seven round trips a day between the Hudson Valley
Healthcare System and the Bronx VAMC.  In addition, New Jersey
Healthcare System officials had to increase the number of
administrative staff at its newly opened Hackensack clinic because
the demand for services at the clinic exceeded its capacity to answer
calls and schedule appointments, making access difficult for New
Jersey veterans. 

Medical center officials and veterans' service organization
representatives told us that veterans have been delayed in getting
access to particular specialty services, notably orthopedics and
urology, in both VISNs.  Management told us that these delays were
caused by difficulties in hiring physicians for those specialties. 
The delays affected some tertiary care medical centers' patients as
well as those patients who had been referred from other medical
facilities.  To reduce delays for some orthopedic and emergency
services, the Butler VAMC has contracted with Butler Memorial
Hospital and has also referred its patients to the Pittsburgh
Healthcare System for these services.  Some veterans told us they
believe that reduced staffing has made access to nursing staff in the
inpatient setting more difficult.  Analyses that VA generated at our
request from its Inpatient Satisfaction Survey showed that veterans
in VISN 3 (Bronx) were less satisfied with access to nursing care in
fiscal year 1997 than in fiscal year 1995.\21 The VA Office of
Inspector General reported similar concerns at the Lyons VAMC.\22


--------------------
\16 VA counts unique patients using their Social Security numbers to
establish the number of unduplicated users of its health care system. 

\17 Primary care teams are intended to be a veteran's first point of
contact with the VA health care system.  They generally provide a
comprehensive range of routine services, coordinate treatment for
patients requiring specialized services, and manage the care to
ensure that appropriate services are provided and duplicative
services, such as unnecessary visits for care, are avoided.  Although
the composition of a team varies depending upon a facility's mission,
it typically includes physicians, one or more health care
professionals (for example, nurse practitioners, physician
assistants, registered and licensed practical nurses), and
administrative support staff. 

\18 See GAO/HEHS-98-48, Feb.  6, 1998, and VA Hospitals:  Issues and
Challenges for the Future (GAO/HEHS-98-32, Apr.  30, 1998) for
information on changes in VA's use of inpatient care. 

\19 See VA Community Clinics:  Networks' Efforts to Improve Veterans'
Access to Primary Care Vary (GAO/HEHS-98-116, June 15, 1998). 

\20 Performance on Customer Service Standards:  Ambulatory Care, 1995
and 1997 National Surveys, VHA National Customer Feedback Center
(West Roxbury, Mass.:  1996 and 1997). 

\21 Performance on Customer Service Standards:  Recently Discharged
Inpatients, 1995 and 1997 National Surveys, VHA National Customer
Feedback Center (West Roxbury, Mass.:  1996 and 1997). 

\22 Final Report--Inspection of Patient Care Allegations and Quality
Program Assistance Review:  Department of Veterans Affairs Medical
Center Lyons, NJ, Report Number 8HI-F03-125, VA Office of Healthcare
Inspections, Office of Inspector General (Washington, D.C.:  July 16,
1998). 


   ACCESS TO CARE IMPROVED FOR
   NEARLY ALL SELECTED SERVICES
---------------------------------------------------------- Chapter 2:2

VISN 3 (Bronx) and VISN 4 (Pittsburgh) are generally improving access
for the selected services we reviewed.  In many instances, their
improvements exceeded those of VA nationally for fiscal years 1995
through 1997.  The networks differed, however, in the extent to which
access has improved for specific services; and one or both of the
networks served fewer veterans in this period.  The services we
reviewed include surgery, mental health services (including those for
patients with SMI, PTSD, and substance abuse), treatment for patients
with SCI, and prosthetics. 

These services especially interest veterans for several reasons. 
Surgery, for example, is a key indicator of VA's success in
increasing efficiency and veterans' access to health care by
providing services in a less costly outpatient setting instead of the
higher cost inpatient setting.  The other specialized services we
reviewed interest veterans because they involve relatively high-cost
activities central to VA's mission to serve more vulnerable
populations.  Preserving access to care for these populations while
under pressure to reduce costs is an essential test of VA's efforts
to transform its health care system. 

VISN 3 (Bronx) increased access for five of the seven services we
reviewed (see table 2.5), and VISN 4 (Pittsburgh) increased access
for six of these services (see table 2.6). 



                         Table 2.5
          
            VISN 3 (Bronx) Changes in Access for
          Selected Services, Fiscal Years 1995-97

                                        Percentage change,
                                        fiscal years 1995-
                                                97
                                        ------------------
Service     Fiscal    Fiscal    Fiscal
(Indicat      year      year      year    VISN 3  National
or)           1995      1996      1997   (Bronx)        ly
--------  --------  --------  --------  --------  --------
Surgery       45.1      54.8      70.5     +56.3     +60.2
 (Percen
 tage
 perform
 ed on
 outpati
 ent
 basis
 for 11
 selecte
 d
 procedu
 res)
Mental     29,557\   31,858\   32,487\     +9.9\     +5.4\
 health\a
 (Unique
 patient
 s)
SMI\b      17,871\   18,329\   18,442\     +3.2\     +2.6\
 (Unique
 patient
 s)
PTSD\c       1,383     1,703     1,833     +32.5     +16.1
 (Unique
 patient
 s)
Substanc     8,657     8,910     8,407      -2.9      -5.2
 e
 abuse\d
 (Unique
 patient
 s)
SCI\e        467\\       477       441      -5.6      +7.0
 (Unique
 patient
 s)
Prosthet       Not    62,283    88,121   +41.5\g   +20.2\g
 ics      availabl
 (Number       e\f
 of
 orders)
----------------------------------------------------------
\a Patients with a primary diagnosis of a mental health condition who
were treated in specialized mental health programs.  These patients
include some of those treated in SMI programs and in programs for
treating PTSD and substance abuse. 

\b Includes all patients treated for SMI. 

\c Includes all patients treated for PTSD. 

\d Patients with a primary diagnosis of substance abuse who were
treated in a specialized substance abuse program. 

\e Patients treated in SCI centers or clinics. 

\f VA was unable to provide these data. 

\g Percentage change is from fiscal year 1996 to fiscal year 1997. 

Sources:  VHA Fiscal Years 1996 and 1997 Maintaining Capacity to
Provide for the Specialized Treatment and Rehabilitative Needs of
Disabled Veterans; National Mental Health Performance Monitoring
System Reports (fiscal years 1995 to 1997); Surgical Performance
Indicators from VA National Performance Data Research Center;
National Delayed Prosthetics Report; and special tabulation from
VHA's Allocation Resource Center. 



                         Table 2.6
          
           VISN 4 (Pittsburgh) Changes in Access
            for Selected Services, Fiscal Years
                          1995-97

                                        Percentage change,
                                        fiscal years 1995-
                                                97
                                        ------------------
Service     Fiscal    Fiscal    Fiscal    VISN 4
(Indicat      year      year      year  (Pittsbu  National
or)           1995      1996      1997      rgh)        ly
--------  --------  --------  --------  --------  --------
Surgery       44.6      52.8      76.0     +70.4     +60.2
 (Percen
 tage
 perform
 ed on
 outpati
 ent
 basis
 for 11
 selecte
 d
 procedu
 res)
Mental     30,716\   32,819\   34,772\    +13.2\     +5.4\
 health\a
 (Unique
 patient
 s)
SMI\b      16,757\    16,760    17,545      +4.7      +2.6
 (Unique
 patient
 s)
PTSD\c       2,173     2,103     2,155      -0.8     +16.1
 (Unique
 patient
 s)
Substanc     8,810    10,358     9,438      +7.1      -5.2
 e
 abuse\d
 (Unique
 patient
 s)
SCI\e          47\        56        61     +29.8      +7.0
 (Unique
 patient
 s)
Prosthet       Not    61,486    78,156   +27.1\g   +20.2\g
 ics      availabl
 (Number       e\f
 of
 orders)
----------------------------------------------------------
\a Patients with a primary diagnosis of a mental health condition who
were treated in specialized mental health programs.  These patients
include some of those treated in SMI programs and programs for
treating PTSD and substance abuse. 

\b Includes all patients treated for SMI. 

\c Includes all patients treated for PTSD. 

\d Patients with a primary diagnosis of substance abuse who were
treated in a specialized substance abuse program. 

\e Patients treated in SCI centers or clinics. 

\f VA was unable to provide these data. 

\g Percentage change is from fiscal year 1996 to fiscal year 1997. 

Sources:  VHA Fiscal Years 1996 and 1997 Maintaining Capacity to
Provide for the Specialized Treatment and Rehabilitative Needs of
Disabled Veterans; National Mental Health Performance Monitoring
System Reports (fiscal years 1995 to 1997); Surgical Performance
Indicators from VA National Performance Data Research Center;
National Delayed Prosthetics Report; and special tabulation from
VHA's Allocation Resource Center. 

Increasing the percentage of surgical procedures performed on an
outpatient basis has been an important VA goal since 1996. 
Performing more surgical procedures on an outpatient basis improves
access because it increases patient convenience, improves quality
because it reduces the risk of infections associated with inpatient
stays, and reduces overall costs.  VA has made increasing the
proportion of surgeries performed on an outpatient basis a critical
measure in its annual assessment of VISN performance and has selected
11 categories of procedures to track in fiscal year 1998 as
indicators of progress.\23 By mid-fiscal year 1998, VISN 3 (Bronx),
VISN 4 (Pittsburgh), and VA nationally had generally continued to
increase the proportion of these services performed on an outpatient
basis. 

In addition, both networks and VA nationally served more mental
health patients in fiscal year 1997 than in fiscal year 1995.\24 Both
VISNs increased the number of patients served mainly by increasing
the number of outpatients.  Nonetheless, VA still provides a broad
continuum of mental health services in intensive and subacute
inpatient settings, nursing homes, domiciliaries, residential
settings, and outpatient clinics located in and apart from medical
centers.  VISN 3 (Bronx) increased the number of its mental health
outpatients by 12 percent (or 3,458 unique veterans), which is more
than twice the rate of the national increase, while reducing the
number of inpatients by 14 percent (or 1,075 unique veterans), which
is approximately equivalent to the national decrease.  The New Jersey
Healthcare System and the Hudson Valley Healthcare System had the
largest increases in the number of outpatients receiving mental
health services.  By providing mental health services at newly
established community-based outpatient clinics as well as at their
medical centers, these systems served more veterans.  The largest
percentage increases in VISN 4 (Pittsburgh) were at the Coatesville
(57 percent or 1,490 patients) and Clarksburg (42 percent or 829
patients) VAMCs. 

VA has recognized the importance of providing regular follow-up
therapy upon inpatient discharge when treating patients with mental
illness.  It has established access performance measures of the
percentage of patients visiting a mental health practitioner within
30 days of discharge and within 6 months of discharge to support the
patient's transition to the home or work environment.  The proportion
of VISN 4 (Pittsburgh) patients seen within 30 days of discharge
increased from 41 percent in fiscal year 1995 to 52 percent in fiscal
year 1997; the percentage in VISN 3 (Bronx) remained constant at 50
percent--approximately equivalent to VA's overall 52 percent.  For
appointments within 6 months of discharge, VISN 4 (Pittsburgh)
increased the number from 75 to 82 percent of discharged veterans;
VISN 3 (Bronx) dropped from 76 to 73 percent; and, nationally, VA
increased from 75 to 78 percent.  The number of days between an
inpatient discharge and an outpatient visit remained the same for
VISN 3 (Bronx) mental health patients--31 days in fiscal year 1995
and fiscal year 1997--
while for VISN 4 (Pittsburgh) patients, it decreased from 40 to 34
days. 

Of the three specific mental health services we reviewed, only
changes in the number of SMI patients served were similar among VISNs
3 and 4 and VA nationally.  (See tables 2.5 and 2.6.) Both networks
and VA nationally increased the number of patients treated with an
SMI diagnosis from fiscal year 1995 to fiscal year 1997.\25 SMIs are
chronic debilitating conditions that require ongoing care.  VA is
trying to provide care to this population in less restricted
environments outside institutional settings.  We were unable to
identify VA data showing changes in the number of SMI patients served
in inpatient and outpatient settings. 

VISN 3 (Bronx) and VA nationally increased the number of PTSD
patients served; VISN 4 (Pittsburgh) treated fewer of these patients. 
VISN 3 (Bronx) served nearly a third more PTSD patients in fiscal
year 1997 than it did in fiscal year 1995; in contrast, VISN 4
(Pittsburgh) served 1 percent less of these patients.  Nationally, VA
served 16 percent more PTSD patients in fiscal year 1997 than in
fiscal year 1995. 

PTSD treatment programs have changed in recent years in the networks
and facilities we visited.  Program officials we interviewed said
that they have reduced their use of long-term inpatient treatment for
PTSD and increased the use of short-term hospital treatment with
outpatient follow-up care.  VA researchers report that such care is
equally effective but less costly.\26 For example, VA established the
Psychiatric Residential Rehabilitation Treatment Program (PRRTP) to
give medical centers another category of treatment for serving
veterans with care that is less intense than acute inpatient care but
similar to other domiciliary programs.  Clarksburg VAMC officials
told us they are converting their inpatient PTSD program to this type
of residential program.  Because many PTSD patients also have
problems associated with alcohol and other drugs, Clarksburg will
locate the PTSD program patients next to the medical center's
residential substance abuse program area to maximize and facilitate
treatment.  The Lyons VAMC has also established a PRRTP. 

Both VA nationally and VISN 3 (Bronx) treated fewer veterans with a
primary diagnosis of substance abuse in specialized substance abuse
programs.  VISN 4 (Pittsburgh), in contrast, served more such
veterans in these programs.  Nationally, the number of veterans VA
treated with a primary diagnosis of substance abuse declined by 5
percent from fiscal year 1995 to fiscal year 1997; the number VISN 3
(Bronx) treated declined by 3 percent.  VISN 4 (Pittsburgh) increased
the number of these patients treated during this period by 7 percent. 

VA's method for delivering substance abuse care has also been
changing.  VA has moved to providing treatment for alcohol and drug
dependencies on an outpatient rather than an inpatient basis to those
living in the community or in VA residential programs.  Both VISNs we
visited were making such changes.  Outpatients increased 1 percent in
VISN 3 (Bronx), 26 percent in VISN 4 (Pittsburgh), and 3 percent
nationally.  Inpatients decreased 25 percent in VISN 3 (Bronx), 63
percent in VISN 4 (Pittsburgh), and 49 percent nationally. 

VA headquarters, network, and medical center officials told us they
knew about the overall decline in the number of substance abuse
patients served but have not been able to determine the reason for
the decline.  Because substance abuse patients often have other
illnesses, they may be receiving substance abuse treatment but under
a category of care other than the specialized category of substance
abuse services, according to VA officials.  Data for such patients
might not appear in the databases we and VA used.  VA is studying its
substance abuse programs to determine the reasons for the decline. 

The number of SCI veterans who received care in SCI centers or
clinics in VA nationally and in VISN 4 (Pittsburgh) increased between
fiscal years 1995 and 1997; the number treated in VISN 3 (Bronx)
declined.  Nationally, VA served 7 percent more SCI patients (an
increase of 582 SCI veterans); VISN 4 (Pittsburgh) served 30 percent
more SCI veterans (14 additional veterans); VISN 3 (Bronx) served
about 6 percent less (a decrease of 26 veterans).  VA has 23 SCI
centers nationwide and 28 SCI outpatient support clinics that provide
less intensive care than facilities with an SCI center.  Seven of
VA's 22 VISNs do not have SCI centers and refer patients who need
such care to other VISNs.\27

VISN 3 (Bronx) integrated the provision of SCI care in the network to
coordinate and maximize use of resources at its three SCI
centers--located in the Bronx, Castle Point, and East Orange VAMCs. 
The Castle Point VAMC, part of the Hudson Valley Healthcare System,
focuses on providing long-term care for SCI patients.  Specialty
care, such as plastic surgery, orthopedic services, and comprehensive
urological care, which used to be provided at Castle Point, was
recently transferred to the Bronx VAMC.  The New Jersey Healthcare
System's SCI center at the East Orange VAMC provides short-term SCI
services, including initial screening of SCI veterans as they enter
the VA system, and respite care.  More veterans are receiving SCI
services in VISN 3 (Bronx) on an outpatient basis.  This increase has
resulted in part from an increase in the number of annual exams
performed on an outpatient basis rather than the multiday inpatient
stay that had been the previous practice.  However, the number of SCI
patients treated in VISN 3 (Bronx) has declined overall.  VISN 3
(Bronx) management, SCI product-line officials, and veterans' service
organization representatives told us they believe the decline is
partly due to patients from other VISNs, such as VISN 1 (Boston),
VISN 2 (Albany), and VISN 4 (Pittsburgh), no longer using VISN 3
(Bronx) SCI centers to the extent that they were before.  They think
that these veterans are perhaps receiving care at VAMCs closer to
their homes, but they have not directly assessed the reasons for the
decline. 

Veterans' waiting times for SCI acute-care admissions and routine
outpatient appointments improved from fiscal year 1996 to fiscal year
1997 in VISN 3 (Bronx).  The Bronx and Castle Point VAMCs met the VA
standard for acute-care admissions in fiscal year 1997.\28 Although
the East Orange VAMC shortened the average number of days SCI
patients had to wait for an acute-care admission, it still did not
meet the standard in 1997.  All three facilities met the standard for
outpatient care waiting times, however, for SCI services in both
fiscal years 1996 and 1997.  VISN 3's (Bronx) progress in meeting
these standards is similar to progress VA has made nationally.  Only
9 of VA's 23 SCI centers met the goal of immediately treating SCI
patients in need of acute specialty care in fiscal year 1996, but the
number meeting the standard in fiscal year 1997 rose to 20.  VA
facilities also improved their performance in achieving the
outpatient standard of referral within 1 week.  In fiscal year 1996,
20 facilities met the standard; in fiscal year 1997, all 23 SCI
centers met the standard.  VA attributes these improvements in part
to improved communication between SCI centers and the SCI primary
care teams at referring medical centers. 

VISN 4 (Pittsburgh) also served more SCI patients in its SCI clinics. 
VISN 4 (Pittsburgh) does not have an SCI center but has an SCI clinic
at the Pittsburgh Healthcare System.  VISN 4 (Pittsburgh) patients
needing more intensive services are referred to SCI centers in East
Orange (in VISN 3), Cleveland (in VISN 10), and Richmond (in VISN 6). 

Regarding prosthetics, both VISNs and VA nationally considerably
increased the number of prosthetics orders from fiscal year 1996 to
1997.  (See tables 2.5 and 2.6.) Prosthetic, orthotic, and sensory
aids and devices include artificial limbs and eyes, wheelchairs,
canes, ostomy appliances, artificial hips, eyeglasses, and hearing
aids.  VISN 3 (Bronx) increased its prosthetics orders by about 41
percent, more than twice the 20-percent rate for VA overall.  The
number of prosthetics orders for the first half of fiscal year 1998
in VISN 3 (Bronx) suggests that the number of orders will increase
substantially again.  VISN 4 (Pittsburgh) also exceeded the national
rate with a 27-percent increase from fiscal year 1996 to fiscal year
1997.  VA facility staff in both VISNs told us that the number of
eyeglasses and hearing aids provided has increased dramatically
because of changes in veterans' eligibility for certain services.\29

VA monitors the number of prosthetics orders delayed for
administrative reasons as a measure of access.\30 Although we
identified some problems with delayed order data (discussed in ch. 
4), VA reported progress in minimizing delays in fiscal year 1997 to
less than 1 percent of all orders, exceeding its standard of 2
percent.  Through the first half of fiscal year 1998, however, it
reported almost as many delayed orders as for all of fiscal year
1997.  VA officials told us they did not know why the delays had
increased and were working to determine the cause. 

Although VISN 3 (Bronx) had some delayed prosthetics orders in fiscal
year 1996, the network reported no delays in fiscal year 1997.  And,
in contrast to VA's national trend of increased delayed orders in the
first 6 months of fiscal year 1998, VISN 3 (Bronx) continued to
report no delayed orders.  Although exceeding VA's 2-percent
standard, VISN 4's (Pittsburgh) delayed orders increased from 0.5
percent (or 31 delayed orders a month) in fiscal year 1997 to 1.1
percent (or 71 delayed orders a month) in the second quarter of
fiscal year 1998. 

According to VA surveys, veterans in VISN 3 (Bronx) generally
reported improved access to prosthetics from fiscal year 1995 to
fiscal year 1998.  The proportion of veterans reporting they received
their prosthetic devices within 5 days of their being ordered by the
Prosthetics Office increased from 45 percent in September 1995 to 66
percent in March 1998.  Similarly, the proportion of veterans
reporting that their prosthetic devices were repaired within 5 days
of their indicating a need for repair increased from 56 percent in
September 1995 to 60 percent in March 1998.  However, the proportion
of veterans who reported getting their appointments, for example, for
a fitting of their prosthetic device, within 5 days of their initial
call for an appointment declined from 52 to 48 percent in the same
time period.  In contrast, during this same period the percentage of
VISN 4 (Pittsburgh) veterans who reported service within 5 days
increased significantly:  for device receipt, from 59 to 81 percent;
for repairs, from 61 to 78 percent; and for appointments, from 51 to
77 percent. 


--------------------
\23 To make comparisons over time, we obtained information from VA
for fiscal years 1995 to 1997 on the 11 categories of surgical and
invasive procedures used in the fiscal year 1998 indicator.  (In
previous years, different indicators were used.  For example, the
1997 indicator was based on 97 categories of procedures.) The fiscal
year 1998 indicator is based on the categories of procedures that VA
expects to be routinely performed on an outpatient basis.  VA chose
these categories because the Health Care Financing Administration's
Medicare program also expects them to be routinely performed on an
outpatient basis.  These procedures are arthroscopy, breast biopsy
(and other diagnostic procedures), bronchoscopy and biopsy of
bronchus, diagnostic cardiac catheterization, colonoscopy,
cystoscopy, eyelids and other therapeutic procedures, endoscopy
(upper), hernia repair (inguinal and femoral), laparoscopy, and lens
and cataract procedures.  The percentage of each procedure expected
to be performed on an outpatient basis varies. 

\24 These include patients with a primary diagnosis of a mental
health condition who were treated in specialized mental health
programs.  In fiscal year 1997, VA treated more than 600,000 unique
patients in these programs. 

\25 Nationally, VA increased the number of SMI patients served by
nearly 3 percent from fiscal year 1995 to more than 270,000 patients
in fiscal year 1997. 

\26 A.  Fontana and R.  Rosenheck, "Effectiveness and Cost of the
Inpatient Treatment of Post Traumatic Stress Disorder:  Comparison of
Three Models of Treatment," American Journal of Psychiatry, Vol. 
154, No.  6 (1997), pp.  758-65. 

\27 The seven VISNs without SCI centers are VISN 2 (Albany), VISN 4
(Pittsburgh), VISN 5 (Baltimore), VISN 11 (Ann Arbor), VISN 13
(Minneapolis), VISN 14 (Omaha), and VISN 19 (Denver). 

\28 VA's goals are to immediately transfer all patients needing acute
SCI specialty care to an SCI center.  The goal for waiting times for
SCI outpatient care is to provide patients with an appointment within
7 days of referral. 

\29 The Veterans' Health Care Eligibility Reform Act of 1996 (P.L. 
104-262) enabled VA to provide prosthetics to patients on an
outpatient basis and significantly expanded eligibility for
eyeglasses and hearing aids. 

\30 Data are not available on the number of unique users of
prosthetic services. 


   OVERALL INCREASES IN ACCESS
   ACHIEVED DESPITE REDUCED BUYING
   POWER OF ALLOCATED RESOURCES
---------------------------------------------------------- Chapter 2:3

VISNs 3 (Bronx) and 4 (Pittsburgh) increased access to services even
though neither VISN received allocations to offset inflation as in
the past.  The VISNs served more veterans by increasing the
efficiency of their health care delivery to offset their decreased
buying power resulting from the combined effects of VERA allocation
changes and inflation. 

Both VISNs' allocations for fiscal year 1997 gave them less buying
power than their allocations for fiscal year 1995.  The phased-in
implementation of VERA along with VISN 3 (Bronx) management decisions
in fiscal year 1997 resulted in a decline in VISN 3 (Bronx) resources
of about $21 million in year-end allocations compared with fiscal
year 1995 (see table 2.7).\31 Year-end allocations reflect the total
net impact of all VERA allocations, including both general-purpose
allocations and specific-purpose allocations made by VA
headquarters--and any reprogramming that took place during the year. 
The added impact of inflation on these resources for VISN 3 (Bronx)
resulted in an overall reduced buying power of about $91 million for
this period.\32 VISN 4 (Pittsburgh), in contrast, received an
increase in its allocation from fiscal year 1995 to fiscal year 1997
but still had a net decreased buying power of $17 million due to
inflation. 



                         Table 2.7
          
           Changes in End-of-Year VA Allocations
            and the Impact of Inflation, Fiscal
                       Years 1995-97

                   (Dollars in millions)

                                             Net  Percenta
                                        estimate        ge
                      Change            d change    change
                          in   Loss in        in        in
            Fiscal  allocati    buying    buying    buying
              year       on,     power    power,    power,
              1995    fiscal      from    fiscal    fiscal
          allocati     years  inflatio     years     years
Entity          on   1995-97       n\a   1995-97   1995-97
--------  --------  --------  --------  --------  --------
VISN 3      $1,092      -$21      -$70      -$91      -8.3
 (Bronx)
VISN 4         817       +35       -52       -17      -2.2
 (Pittsb
 urgh)
VA          16,189      +938    -1,039      -101      -0.6
 nationa
 lly
----------------------------------------------------------
Note:  We included all medical care appropriations--and for VISN 3
(Bronx) the funds it returned to headquarters in fiscal year 1997--in
our calculation of total year-end allocations.  These totals differ
from those VA publishes for the beginning of a fiscal year because at
that time VA does not know what allocations will be made for specific
purposes and program changes made throughout the year.  Thus, while
facilities have received some of their allocations for fiscal year
1998, when we prepared this report, VA had not made the final
allocations. 

\a We calculated changes on the basis of allocation data and
inflation factors provided by the Office of the Chief Financial
Officer, VHA. 

Source:  VA and our calculations. 

Officials in both VISNs told us that they compensated for the reduced
buying power of their allocations by improving efficiency.  With
13,735 full-time equivalent employees (FTEE) at the start of fiscal
year 1996, VISN 3 (Bronx) reduced staffing by 2,070 FTEEs (about 15
percent) by the end of fiscal year 1997.  This represented more than
$110 million in cost reductions that could be used for offsetting
allocation reductions and inflation.  VISN 4 (Pittsburgh) started
fiscal year 1996 with 10,850 FTEEs and reduced staffing by 1,485
(about 14 percent) through fiscal year 1997.  Thus, VISN 4
(Pittsburgh) reduced costs by about $79 million to offset the effects
of inflation.  Both VISNs also increased employee productivity.  For
example, some facilities in VISN 3 (Bronx) changed primary care
physician schedules to give the physicians more time to serve
patients; and, in fact, more patients were seen by each physician. 
Both VISNs also consolidated laundry services to serve more than one
facility from a single site. 

VISN 3 (Bronx) and VISN 4 (Pittsburgh) officials told us they expect
to continue increasing access in fiscal year 1998 by a combination of
additional efficiencies and new resources.  The new resources are
primarily from third-party health insurance collections, which
network officials told us they plan to use to offset the combined
effects of VERA allocation changes and inflation.\33 For fiscal year
1998, the first year it can retain these collections, VA set a goal
of about $596 million nationally, about $44 million for VISN 3
(Bronx), and about $36 million for VISN 4 (Pittsburgh).  For fiscal
year 1999 and beyond, the degree to which the two VISNs will be able
to offset the effects of VERA allocation changes and increased
inflation is difficult to estimate given the many factors that will
determine the resources available to them and the costs of the
services that will be needed.\34 The VISN 3 (Bronx) director
estimated the network will need additional resources in fiscal year
2000 to provide veterans appropriate access to health care, and the
VISN 4 (Pittsburgh) chief financial officer estimated it will need
additional resources in fiscal year 2001. 


--------------------
\31 VISN 3 (Bronx) returned $20 million to VA headquarters in fiscal
year 1997 because, according to VISN management, all identified needs
for that year had been met.  It was the only network to return funds
that year.  According to VA headquarters, these funds became part of
a larger specific-purpose allocation to the VISNs for information
technology.  VA allocated about $8 million of these funds to VISN 3
(Bronx) for its information technology initiatives.  VA headquarters
allocated about $4 million to VISN 4 (Pittsburgh) for its share of
this specific-purpose allocation.  VA could not provide information
on the amount that would have been allocated to VISN 3 (Bronx) for
information technology if the network had not returned $20 million. 

\32 We calculated the impact of inflation on the basis of inflation
factors that VA uses for its medical care appropriations request to
the Congress. 

\33 The Balanced Budget Act of 1997 authorized VA to retain
recoveries from third-party insurance and collections from the sale
of excess services to beneficiaries of the Department of Defense,
medical school hospitals, and other providers. 

\34 Some of these factors include how one network fares relative to
the other networks in increasing workloads and decreasing costs, the
network's success in collecting third-party payments, and the rate of
inflation.  VA is also considering changes to VERA that could affect
future resource allocations. 


VA HAS DONE LITTLE TO IMPROVE THE
EQUITY OF RESOURCE ALLOCATIONS TO
FACILITIES
============================================================ Chapter 3

Although VA nationally has made progress in improving the equity of
resource allocations among the networks, it has done little to ensure
that the VISNs allocate resources to address past inequities within
each network to ensure that veterans with similar economic status and
eligibility priority have similar access to care.  VA headquarters
has not provided criteria or guidance for improving the equity of
VISN resource allocations to facilities.  Furthermore, VA
headquarters does not review VISN allocation methods and results to
determine whether allocations within each VISN are made equitably. 
VISN 3 (Bronx) and VISN 4 (Pittsburgh) use different methods to
allocate most of the medical care resources to their facilities.  To
some extent these methods reflect differences in the resource
allocation challenges the two VISNs face. 


   VISN 3 (BRONX) METHOD REDUCED
   FACILITY ALLOCATIONS
---------------------------------------------------------- Chapter 3:1

The VISN 3 (Bronx) network leadership council met in the spring of
1996 to develop a comprehensive list of cost-saving actions that
would be needed to meet the expected allocation reductions for VERA's
implementation in fiscal year 1997.  At the time, network management
estimated that their fiscal year 1997 allocation could have been at
least $100 million (about 10 percent) less than the fiscal year 1996
allocation.  Many of the cost-
saving initiatives, including staff cuts and unit closures, that had
been identified were begun in fiscal year 1996. 

The reduction in fiscal year 1997 allocations to VISN 3 (Bronx) was
less than network management had anticipated.  To allow time for
network management to implement less costly care while improving
access, VA decided to gradually implement the allocation changes
resulting from the VERA formula by capping the amount of funds
removed from this network and others.  Nonetheless, the cost-cutting
initiatives VISN 3 (Bronx) management had begun in fiscal year 1996
continued, officials told us, because they were needed for the
current year and officials expected additional reductions in fiscal
year 1998 and beyond.  As a result, VISN 3 (Bronx) management
developed facility allocations in the context of reduced funding for
the network. 

VISN 3 (Bronx) based most of its fiscal year 1997 allocation to each
facility on resources allocated in the previous year.  The amounts to
each facility were reduced by the amount of savings in
operations--medical and support--that each medical center director
had identified in the spring of 1996.  For example, the Bronx and New
York VAMCs consolidated their laboratory operations.  Castle Point
VAMC closed surgical beds because the VISN had decided to treat
patients requiring inpatient surgery at the Bronx VAMC.  These
changes reflect in part VISN 3 (Bronx) management's approach to
building a network by integrating services among facilities.  VISN 3
(Bronx) also made additional allocations from its operating reserve
to the New Jersey Healthcare System for activating a new psychiatric
facility at the Lyons VAMC. 

Because VISN 3 (Bronx) had fewer resources, its allocation method
generally reduced allocations to each facility.  To calculate changes
in allocations, we used fiscal year 1995 as a base because VISN 3
(Bronx) and its facilities began to implement changes in fiscal year
1996 in anticipation of VERA's start in fiscal year 1997.  Five of
the network's six facilities received reduced allocations in fiscal
year 1997.  (See table 3.1.) The Brooklyn VAMC, however, received a
5.3-percent increase in fiscal year 1997 mainly because of an
increase in specific-purpose funds. 



                         Table 3.1
          
            End-of-Year Resource Allocations for
          VISN 3 (Bronx) Facilities, Fiscal Years
                          1995-97

                   (Dollars in millions)

                                        Differen
                                             ce,  Percenta
            Fiscal    Fiscal    Fiscal    fiscal        ge
Allocati      year      year      year     years  differen
ons           1995      1996      1997   1995-97        ce
--------  --------  --------  --------  --------  --------
Bronx       $143.1    $139.3    $133.5     -$9.6      -6.7
 VAMC
Brooklyn     197.3   200.7\a   207.7\a     +10.4      +5.3
 VAMC
Hudson       130.4     133.1     129.3      -1.2      -0.9
 Valley
 Healthc
 are
 System\
 b
New          267.9     264.4     252.1     -15.7      -5.9
 Jersey
 Healthc
 are
 System\
 c
New York     174.5     172.1     170.4      -4.1      -2.3
 VAMC
Northpor     143.0     137.9     132.0     -11.0      -7.7
 t VAMC
Capital       35.3      53.6      46.0     +10.6     +30.1
 account
 s\d
==========================================================
Total     $1,091.5  $1,101.0  $1,071.0    $-20.6      -1.9
 VISN 3
 (Bronx)
----------------------------------------------------------
Note:  End-of-year resource allocations include VERA general- and
specific-purpose funds and all reprogramming of resources throughout
the fiscal year. 

\a Includes funds for VISN 3 (Bronx) network office operation. 

\b Includes Montrose and Castle Point VAMCs. 

\c Includes East Orange and Lyons VAMCs. 

\d Data for capital accounts include allocations for equipment and
nonrecurring maintenance.  Comparable information across fiscal years
on funds distributed to each VAMC was not available. 

VISN 3 (Bronx) used the same basic approach for allocating its fiscal
year 1998 resources but added two new parts.\35 Each facility was
allowed to keep all funds it collected from third-party insurance and
the sale of excess services up to a facility-specific goal and 75
percent of collections above the goal.  In addition, VISN 3 (Bronx)
adjusted facilities' fiscal year 1998 allocations on the basis of
changes in their respective workloads between fiscal years 1996 and
1997.  If facilities served more patients in fiscal year 1997 than
the preceding year, the VISN increased the allocation by $2,014 for
each additional patient served.  If the facility had served fewer
patients, however, its allocation was reduced by the same amount. 

Officials at the facilities we visited told us that they generally
agree with the allocation method VISN 3 (Bronx) has used.  Facility
managers said that they like the method because it allows them to
manage a defined facility budget during the year to achieve specified
program efficiencies.  Although facility directors are expected to
manage their allocations, VISN 3 (Bronx) managers and facility
directors told us that the system is flexible enough to provide
resources for unforeseen circumstances through VISN reserves and
reprogramming. 


--------------------
\35 Final allocations for fiscal year 1998 were not available when we
prepared this report. 


   VISN 4 (PITTSBURGH) ALLOCATION
   EMPHASIZED INCENTIVES
---------------------------------------------------------- Chapter 3:2

VISN 4 (Pittsburgh), facing the prospect of a stable allocation under
VERA, did not begin planning its allocation method until fiscal year
1997.  VISN 4 (Pittsburgh) officials told us that the VISN office
prepared the allocation method and allocated resources to the
facilities with little input from the facilities.  The allocation
method was designed from the beginning with incentives for changing
how facility directors manage health care, according to these
officials.  The network's method includes incentives for improving
efficiency, operating as a network, and increasing access to care. 

VISN 4 (Pittsburgh) allocated resources to facilities in fiscal year
1997 using a multistep process.  It allocated to each facility a
fixed amount for each veteran served in fiscal year 1996--regardless
of eligibility priority--and each facility received resources based
on its number of long-term care operating beds to account for the
higher cost of such care.\36 In addition, each facility received
$1,000 for each additional veteran served, up to a facility-specific
limit.  The network expected each facility to grow at the same rate. 
In addition, the network allocated resources to "buy out" early
retiring employees.  VISN 4 (Pittsburgh) allocated other resources,
for example, from its reserve fund and investment pool, throughout
the fiscal year. 

For fiscal year 1997, VISN 4 (Pittsburgh) used its method to estimate
each facility's revenues and expenditures.  Facilities did not
receive a fixed allocation from the network as they had before. 
Instead, VISN 4 (Pittsburgh) management told us that they expected
facility directors to manage the changing revenue and expenditure
patterns throughout the year.  Nonetheless, the VISN expected to make
up any shortfalls both from VISN reserves and by redistributing
surpluses at some facilities where revenues exceeded expenditures. 
The VISN did this, for example, at the Clarksburg VAMC by allocating
an additional $2 million during fiscal year 1997 because its original
allocations fell below expenditures. 

The VISN 4 (Pittsburgh) fiscal year 1997 allocation method resulted
mainly in facility increases because the network had more resources
in fiscal year 1997 than in fiscal year 1995 (see table 3.2). 
End-of-year allocations increased in 9 of the 10 facilities during
this period.  The Erie VAMC had the highest rate of increase, which
resulted from a laundry replacement program, increased workload, a
telecommunications infrastructure project, and a telephone switch
project. 



                         Table 3.2
          
            End-of-Year Resource Allocations for
           VISN 4 (Pittsburgh) Facilities, Fiscal
                       Years 1995-97

                   (Dollars in millions)

                                        Differen
                                             ce,  Percenta
            Fiscal    Fiscal    Fiscal    fiscal        ge
Allocati      year      year      year     years  differen
ons           1995      1996      1997   1995-97        ce
--------  --------  --------  --------  --------  --------
Altoona      $31.7     $32.4     $33.3     $+1.6      +5.2
 VAMC
Butler        37.3      36.6      37.6      +0.3      +0.8
 VAMC
Clarksbu      44.3      45.1      48.2      +3.9      +8.7
 rg VAMC
Coatesvi      82.2      86.5      85.1      +2.9      +3.5
 lle
 VAMC
Erie          30.4      33.3      41.8     +11.4     +37.6
 VAMC
Lebanon       76.8      78.6      78.0      +1.2      +1.6
 VAMC
Philadel     143.7     142.3     140.7      -3.0      -2.1
 phia
 VAMC
Pittsbur     207.9   209.5\b   209.4\b      +1.5      +0.7
 gh
 Healthc
 are
 System\
 a
Wilkes-       81.6      83.6      84.5      +2.8      +3.5
 Barre
 VAMC
Wilmingt      53.0      57.3      54.2      +1.2      +2.3
 on VAMC
Capital       27.9      44.0      38.8     +10.9       +39
 account
 s\c
==========================================================
Total       $817.0    $849.4    $851.7    $+34.7      +4.3
 VISN 4
 (Pittsb
 urgh)
----------------------------------------------------------
Note:  End-of-year resource allocations include VERA general- and
specific-purpose funds and all reprogramming of resources throughout
the fiscal year. 

\a Includes data from all Pittsburgh facilities. 

\b Includes funds for VISN 4 (Pittsburgh) network office operation. 

\c Data for capital accounts include allocations for equipment and
nonrecurring maintenance.  Comparable information across fiscal years
on the funds distributed to each VAMC was not available. 

VISN 4 (Pittsburgh) changed its fiscal year 1998 allocation method in
several ways:  It (1) changed the calculation of the number of
veterans served by facility for which it received a fixed payment,
(2) reduced allocations for additional patients served, and (3) added
allocations for third-party health insurance collections.\37 In
fiscal year 1998, the VISN determined the number of facility patients
by establishing a catchment area for each facility using ZIP codes. 
A facility receives an allocation on the basis of the number of
veterans in its catchment area that VA served in a prior fiscal year. 
Because of the network's success in increasing the number of veterans
served, VISN 4 (Pittsburgh) officials told us they reduced the amount
of allocations for serving additional patients.  In addition, the
VISN allocated funds to each facility for third-party health
insurance payments each facility was expected to collect. 

The establishment of catchment areas permitted VISN 4 (Pittsburgh)
management to introduce a major change in funding facilities: 
transfer pricing.  According to VISN officials, transfer pricing is
intended to foster close working relationships among facilities in
the network, improving access.  The officials also expected it to
help lower the cost of care by introducing elements of competition. 
Under transfer pricing, the Clarksburg VAMC, for example, pays the
Pittsburgh Healthcare System for services veterans from the
Clarksburg catchment area receive in the Pittsburgh Healthcare
System.  This can happen in two ways:  Clarksburg physicians can
refer a patient from Clarksburg's catchment area to the Pittsburgh
Healthcare System for care, or a patient from the Clarksburg
catchment area can go to Pittsburgh for care without a referral.  In
either case, the Clarksburg VAMC pays the Pittsburgh facility for the
services provided at a rate equivalent to 80 percent of the Medicare
reimbursement for that service in the local area.  The Clarksburg
VAMC may also purchase care from non-VA providers, for example, for
emergency care, when it cannot access care at the Pittsburgh
Healthcare System, or if it determines that it can purchase care at a
lower cost from a non-VA provider. 

As a result of transfer pricing, the amount of resources allocated to
a facility in VISN 4 (Pittsburgh) at the beginning of a fiscal year
will probably vary significantly from the amount of its allocations
at year's end.\38 For example, the Clarksburg VAMC's allocation will
probably decline because it refers more patients to other facilities,
such as the Pittsburgh Healthcare System, which has tertiary care,
than it receives from them.  In contrast, the Pittsburgh Healthcare
System's allocation will probably increase over the year because it
serves more patients from other facilities than it refers to them. 
Thus, initial facility allocations represent resources available to
meet the needs of veterans who live in the facility's catchment area
no matter where the veterans receive services.  End-of-year
allocations represent the resources spent at a facility on the health
care it provides to veterans regardless of the catchment area in
which the veteran lives.  This means VISN 4 (Pittsburgh) facilities
have no set budgets to count on as VISN 3 (Bronx) facilities have. 
VISN 4 (Pittsburgh) officials told us they want facility managers to
learn to work with uncertain resources and financial obligations so
that they can adapt to changing health care dynamics. 

The medical center directors we spoke with from VISN 4 (Pittsburgh)
had mixed views of transfer pricing.  Directors of facilities that
refer patients to tertiary care centers generally support the
transfer pricing concept.  According to these directors, transfer
pricing provides more flexibility to hospitals that need to transfer
patients to other providers and adequately compensates hospitals
receiving patients from other areas.  They also said that transfer
pricing helps to increase veterans' access to care at tertiary
facilities by encouraging those facilities to be more customer
oriented.  Managers, physicians, nurses, and social workers at
referring facilities told us that the tertiary facilities in
Pittsburgh and Philadelphia VAMCs are now more responsive and timely
in accepting and serving veterans referred to them.  In addition,
according to directors at referring facilities, transfer pricing
provides them information on the price of providing services within
the network and outside of VA that will be useful in the future in
determining where VA can most efficiently purchase and most
conveniently provide services to veterans.  The directors of the two
tertiary care centers, however, told us that transfer pricing
requires too much effort and expense to track an essentially small
share of the health care workload and resources.  These directors
said that transfer pricing was not necessary for them to improve
relations with the primary care referring facilities. 

Both VISN managers and facility directors said that managing the
allocation process with uncertain revenues and expenditures has been
a difficult but important cultural change.  To address continuing
resource allocation issues, VISN 4 (Pittsburgh) created a Resource
Allocation Committee in 1998 composed of facility managers,
physicians, staff, and union officials. 


--------------------
\36 These allocations were made for nursing home, psychiatric,
intermediate, and domiciliary beds. 

\37 Final allocations for fiscal year 1998 were not available when we
prepared this report. 

\38 In fiscal year 1998, VISN facilities do not use transfer pricing
to pay for care in VA facilities that are located outside the VISN. 


   VA HAS NEITHER PROVIDED
   CRITERIA NOR REVIEWED VISN
   ALLOCATION METHODS FOR EQUITY
---------------------------------------------------------- Chapter 3:3

Although VERA has improved the equity of resource allocations among
networks, neither VISN 3 (Bronx) nor VISN 4 (Pittsburgh) allocates
resources to address past inequities within its network to ensure
that veterans with similar economic status and eligibility priority
have similar access to care.  Achieving equity in VISN allocations to
facilities is important because similar inequities exist within and
among VISNs.\39 VA officials told us that VISNs should allocate
resources equitably within their networks as part of VA's effort to
achieve equitable access.  However, fiscal year 1998 allocation
guidelines, which headquarters issued to VISNs in response to our
earlier recommendation, do not address equity criteria as we had
recommended.\40 Furthermore, VA headquarters officials told us that
they do not review VISN allocation methods and results to determine
if allocations within VISNs are made equitably. 

Neither VISN 3 (Bronx) nor VISN 4 (Pittsburgh) used criteria to
address equitable allocation issues in facility allocations for
fiscal years 1997 and 1998.  Management officials in both VISNs told
us they had not tried to improve the equitable allocation of
resources in their VISNs. 

The two VISNs have recognized, however, that their networks have
inequities.  For example, the VISN 4 (Pittsburgh) fiscal year 1998
strategic plan states that the proportion of eligible veterans using
services is substantially lower in the eastern part of the network
than the western part and that efforts to increase users should be
concentrated in the eastern part.\41 In addition, the percentage of
service-connected veterans who used services at each of the
facilities in VISN 4 (Pittsburgh) varied from 27 to 50 percent in
1996.  In its strategic plan, VISN 3 (Bronx) discussed lower usage
rates among veterans in the catchment area for the New York VAMC
compared with the rates in the catchment area for the Bronx VAMC,
suggesting that the former area may have less access to care.  VISN 3
(Bronx) management officials told us that they may include criteria
in the fiscal year 1999 allocation process to address equity. 

Although both VISNs' allocation methods provide incentives for
increasing the number of veterans served consistent with VERA and VA
national initiatives, neither the VISN 4 (Pittsburgh) nor the VISN 3
(Bronx) allocation method addresses access inequities identified in
their respective strategic plans.  In fact, in fiscal year 1997, the
net year-end allocations to VISN 4's (Pittsburgh) western facilities
increased more than those to the VISN's eastern facilities, where the
VISN had identified equitable access problems.  As a result, both
VISNs face the risk of growth without equity. 


--------------------
\39 Explanatory Model to Project Demand for Care at the National and
Network Level:  Analysis of Select VISNs, Abt Associates, Inc. 
(Cambridge, Mass.:  Feb.  1998). 

\40 VA Health Care:  Resource Allocation Has Improved, but Better
Oversight Is Needed (GAO/HEHS-97-178, Sept.  17, 1997). 

\41 Inequities can result from the lack of similar services for
similarly situated veterans and lack of comparable resources for
comparable workload.  See Veterans' Health Care:  Facilities'
Resource Allocations Could Be More Equitable (GAO/HEHS-96-48, Feb. 
7, 1996). 


VA OVERSIGHT IS INADEQUATE
============================================================ Chapter 4

Although we prepared an overall assessment of access to care,
difficulties in working with the data cast doubt on whether VA can
perform timely and effective oversight.  The information we developed
on changes in access to care at the facility and network levels for
VISN 3 (Bronx) and VISN 4 (Pittsburgh), as well as for VA nationally,
was gathered from many VA reports and databases--some of which had
inconsistent or incompatible information.  Moreover, medical center,
VISN, and headquarters officials told us that such data are not
available on a routine, timely basis--particularly for specific
programs.  Without such information, it is difficult for them to say
conclusively whether VA has improved veterans' equity of access to
care and whether veterans--particularly those who had been receiving
high-cost care for chronic conditions--have been adversely affected
by the many changes under way to reduce costs and improve
productivity. 


   TIMELY, COMPARABLE,
   COMPREHENSIVE DATA FOR
   MONITORING ACCESS ARE LACKING
---------------------------------------------------------- Chapter 4:1

To gather the information for this report, we extracted and
reconciled information from many VA sources, including VA's Summary
of Medical Programs, Maintaining Capacity to Provide for the
Specialized Treatment and Rehabilitative Needs of Disabled Veterans,
the VERA briefing booklets, the National Mental Health Performance
Monitoring System Reports, and reports on quarterly performance
indicators.  In addition, we used many special computer runs and data
requests from the Office of the Chief Financial Officer and its
Boston Allocation Resource Center, the Office of the Chief Network
Officer, the Office of Performance and Quality, and others. 

Most of the available data on access, according to many officials we
spoke with, are not timely enough for prudent program management. 
Information on the number of patients served in the selected programs
we reviewed, for example, is available only at 1-year intervals. 
Thus, facility, VISN, and headquarters officials told us they do not
have the information needed to assess the impact of program changes
on access for identifying and correcting problems in a timely fashion
or for measuring whether they are meeting their objectives such as
improving access to appropriate care.  Much of the data reported at
1-year intervals are not available until months after the end of the
year for which they apply. 

Although VA appears to have a great deal of data for measuring
changes in access to care, closer examination shows that different
measures are used for the same indicator, users sometimes do not
clearly understand these measures, and obtaining the same measure
over time for comparison purposes can be difficult.  To identify the
impact of change, VA's managers need data that are comparable over
time and systemwide.  VA acknowledges that its data systems need
improvement.\42

One example we identified in our discussions with managers has been a
recurring source of confusion:  counting the number of patients VA
serves.  VA reports this number in several ways--each with a separate
definition or purpose.  One way it reports this information is to
count the number of all unique patients--veterans and
nonveterans--treated.  VA does this, for example, to gauge whether it
is meeting its strategic access goal of increasing the number of
patients served from 1998 to 2002 by 4 percent each year.  A second
way, used in VA's annual Summary of Medical Programs, includes
counting veterans and nonveterans, but the totals are for inpatient
and outpatient visits rather than unique patients served.  A third
way, used by VA for determining the number of veterans served for
VERA allocations, reports Category A veterans served but also
includes some Category C veterans (lower priority veterans) and
nonveterans in its calculations and uses a different time period (3
or 5 years) to calculate patients served.  In contrast, VA's
performance indicator data include only Category A veterans and are
based on the number of unique patients served during a 1-year period. 
Moreover, VA has changed the way it calculates this fourth
indicator--twice for fiscal year 1997 and again for fiscal year 1998. 

Data on the number of patients treated in the selected programs we
reviewed have similar problems.  For example, VA SCI data indicate
that VISN 4 (Pittsburgh) treated 61 patients in its SCI clinics in
fiscal year 1997, 56 of whom received treatment at the Pittsburgh
Healthcare System SCI clinic.  Similar data were available for fiscal
year 1996, but for fiscal year 1995, VISN 4 (Pittsburgh) officials
did not have the data until VA's Allocation Resource Center (ARC)
provided the data at our request.  Staff at the Pittsburgh Healthcare
System SCI clinic told us that they treat about 100 patients a year. 
Later, the Pittsburgh Healthcare System informed us that it could not
document the number of these patients served.  Data provided by
request from ARC for all SCI patients treated, both in SCI centers
and elsewhere, showed that VISN 4 (Pittsburgh) treated 1,017 SCI
patients in fiscal year 1997, 295 of whom received treatment in the
Pittsburgh Healthcare System.  We also found similar data
inconsistencies for SCI patients treated in VISN 3 (Bronx).  VISN and
facility staff with whom we spoke did not know how VA's national data
on SCI patients are defined or how to reconcile differences between
these data and respective national and local data.  A veterans'
service organization representative told us that his organization
does not believe that the SCI data are accurate. 

VISN officials also told us that information they wanted on patient
satisfaction with access to care and reported waiting times has been
hard to obtain.  Although VA provides VISNs with an overall index of
customer satisfaction with access to care, VA does not routinely
provide them with responses to particular questions for each
facility--even though doing so would help managers identify problems
to correct.  For example, a VISN's satisfaction index may show a
composite decline in satisfaction with access to care.  From this
information, however, the VISN cannot determine by facility whether
more veterans are reporting problems getting outpatient appointments
when they want them or seeing their health care providers within 20
minutes of their scheduled appointments, one of VA's strategic goals. 
VISN 3 (Bronx) officials told us that they had difficulty getting
this information for fiscal year 1997 and they did not have it for
fiscal year 1995.  Without such information, managers cannot identify
the specific access problems they need to address.  VA's Office of
Inspector General also reported finding delays in disseminating
information from the Ambulatory Care and Inpatient Care Surveys to
managers.\43

In addition, managers we interviewed told us that indicators of
access to care for prosthetics are inadequate or not consistently
available.  One such measure is the delayed order, which is the
indicator for access to prosthetics in VA's strategic plan.  This
indicator measures an administrative action:  any prosthetic order
that the local unit did not process within 5 workdays because of
incomplete management or administrative action.  According to
headquarters officials, however, the measure may not be accurate
because of the discretion granted to facility officials in defining
the reasons for a delay.  Furthermore, the measure does not assess a
critical dimension of a veteran's access, namely, how long it takes a
veteran to receive the prescribed prosthetic device. 

VA has additional measures of prosthetics access but does not use
them as strategic indicators.  One of these is the average time
veterans wait for prosthetics appointments and receipt of prescribed
prosthetic devices, which is noted in each facility's monthly delayed
order report.  Because neither headquarters nor the VISNs routinely
monitor these data, we obtained facility-generated information.  Data
were not available for some facilities for the years requested. 
Headquarters and VISN officials told us, however, that even when
available, such data may be inaccurate because calculations at the
medical centers were not done accurately or the same number was used
every month rather than the correct number being calculated from that
month's records.  Our analysis of the available data raised concerns
similar to those raised by others.\44

VA officials told us that they think a more reliable measure comes
from a twice yearly satisfaction survey of prosthetics users asked to
report the time it takes them to get appointments, their devices, and
repairs to their devices (discussed in ch.  2).  Although VA managers
told us that they believe the survey data are generally reliable, VA
has not disseminated facility-level information for monitoring. 
However, VA's Prosthetics and Sensory Aids Service Strategic
Healthcare Group distributed VISN-level results of the March 1998
survey in June 1998, and an official in that group told us of plans
to disseminate VISN-level information every 6 months.  She said they
do not have the capability to provide facility-level information. 


--------------------
\42 VA Health Care:  VA's Efforts to Maintain Services for Veterans
With Special Disabilities (GAO/T-HEHS-98-220, July 23, 1998). 

\43 Review of Veterans Health Administration's National Customer
Feedback Center Department of Veterans Affairs Medical Center, West
Roxbury, Massachusetts, VA Office of Inspector General, Office of
Healthcare Inspections, 8HI-A28-069 (Washington, D.C.:  Feb.  4,
1998). 

\44 Thomas H.  Miller, Chairman, VA Federal Advisory Committee on
Prosthetics and Special Disabilities Programs, testimony before the
House Committee on Veterans' Affairs, Subcommittee on Health
Oversight, Washington, D.C., July 23, 1998. 


   VA DOES NOT KNOW IF ACCESS IS
   MORE EQUITABLE
---------------------------------------------------------- Chapter 4:2

In spite of VA's major effort to design and implement VERA and to
provide VISN management with the opportunity to allocate resources
more equitably within VISNs, VA does not know if it is making
progress in providing similar services to similarly situated
veterans.  VA's strategic plan does not include a goal for achieving
equitable access, and VA does not monitor the extent to which
equitable access is being achieved among or within VISNs.  Instead,
VA has focused its efforts on increasing access generally--apparently
expecting this to lead to more equitable access sometime in the
future.\45 VA officials told us they have identified no indicators to
be used for monitoring improvements in equitable access and they have
no plans to do so. 

Because VA officials are not monitoring improvements in equitable
access to care, VA does not know if changes in allocations from VERA
and other actions are equalizing access nationwide.  Without
monitoring changes in equitable access to care, VA can neither assure
stakeholders that equitable access is improving nor take corrective
actions, if needed, to improve resource allocation or other
initiatives. 


--------------------
\45 We have expressed related concerns in VA Community Clinics: 
Networks' Efforts to Improve Veterans' Access to Primary Care Vary
(GAO/HEHS-98-116, June 15, 1998). 


CONCLUSIONS, RECOMMENDATIONS, AND
AGENCY COMMENTS AND OUR EVALUATION
============================================================ Chapter 5


   CONCLUSIONS
---------------------------------------------------------- Chapter 5:1

Overall, VA has increased access to care for veterans in VISN 3
(Bronx), VISN 4 (Pittsburgh), and VA nationally, although access to
some specific services has declined.  VA has increased access mainly
by expanding outpatient services through conversion of inpatient
resources for that purpose.  This has increased the efficiency of VA
health care delivery and allowed VISN 3 (Bronx) and VISN 4
(Pittsburgh) to serve more veterans with fewer inflation-adjusted
dollars under VERA. 

Although various VA offices have a broad range of information on
access that would be useful to managers, it is often generated as
part of larger efforts for purposes other than monitoring access.  As
a result, the information is not always easily accessible or
understandable to managers for monitoring access.  Managers are
unaware of useful access information and unclear about how some of
the information is defined.  In addition, they do not know where to
go for clarification of the data when needed.  VA needs more uniform
and timely reporting of changes in access to care, including the
number and eligibility priority of patients served, waiting times for
care, and patient satisfaction for specific services at the VISN and
facility level. 

Finally, VA has not followed through on VERA's promise of more
equitably allocating resources.  VA guidance to VISNs on allocating
resources to facilities includes no criteria for VISNs to use to
achieve equitable allocation, and VA does not review VISN allocations
to assess the extent to which they improve equitable allocation of
resources.  The allocation methods of VISN 3 (Bronx) and VISN 4
(Pittsburgh) that we reviewed include no initiatives for improving
equitable allocation of resources among their facilities.  If VISNs
do not equitably allocate the resources received under VERA to their
respective facilities, historical inequities within VISNs may
continue even if VERA improves equity among VISNs. 

Acting to achieve an objective, such as implementing VERA to improve
equitable access to health care services, does not ensure meeting the
objective.  Indicators and monitoring are required to gauge whether
the action taken is having the desired effect.  Because VA has not
established measures to assess its progress in achieving equitable
access, it does not know whether it has made such progress.  It does
not know whether additional changes in resource allocation, strategic
planning, or management decisionmaking are needed to ensure more
equitable access.  Without information on changes in equitable
access, VA does not know whether the increased number of veterans it
has served has occurred at the expense of reduced access to services
for veterans who have been historically underserved. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 5:2

We recommend that the Secretary of Veterans Affairs direct the
Undersecretary for Health to

  -- develop uniform definitions and institute timely reporting of
     changes in access to care, including the number and eligibility
     priority of patients served, waiting times for care, and patient
     satisfaction for specific services at the VISN and facility
     level and

  -- develop criteria for equitably allocating resources to
     facilities and monitor any improvements in equity of access
     among and within VISNs. 


   AGENCY COMMENTS AND OUR
   EVALUATION
---------------------------------------------------------- Chapter 5:3

In an August 26, 1998, letter in response to a draft of this report
(see app.  I), VA said it is working to improve its information
systems so that they will be more useful to VISN and headquarters
management.  VA expressed concern, however, that our recommendation
to develop national criteria for equitably allocating resources to
facilities is contrary to VHA's reengineering philosophy, which
decentralizes authority and accountability for these allocations to
the network directors. 

Regarding information systems, VA's letter did not specify whether it
intends to implement our recommendation.  As our report notes,
improvements are essential to allow VA managers to identify problems
and take corrective action in a more timely way to help ensure that
veterans' access to care does not deteriorate in the environment of a
transformed VA health care system. 

We disagree that our recommendation for VA to develop criteria for
equitably allocating resources to facilities within VISNs is contrary
to VA's philosophy of decentralizing authority and accountability. 
We believe VA can develop criteria for VISNs' use in equitably
allocating resources to their respective facilities and review VISNs'
performance in addressing these criteria without being so
prescriptive that local authority and accountability are compromised. 
For example, VA has already used performance measures based on
national criteria to hold VISN directors accountable for achieving
national goals.  Having criteria does not preclude VISNs from using
different methods for allocating resources to address local
circumstances and VA's national criteria.  We still believe that if
VISNs do not improve equitable allocation of resources to their
facilities, VERA's promise of more equitable access to care will not
be achieved. 




(See figure in printed edition.)Appendix I
COMMENTS FROM THE DEPARTMENT OF
VETERANS AFFAIRS
============================================================ Chapter 5



(See figure in printed edition.)


GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
========================================================== Appendix II

GAO CONTACTS

Bruce D.  Layton, Assistant Director, (202) 512-6837
James C.  Musselwhite, Senior Social Science Analyst, (202) 512-7259

STAFF ACKNOWLEDGMENTS

The following team members made important contributions:  Donna M. 
Bulvin, Senior Evaluator; John R.  Kirstein, Evaluator; Lawrence L. 
Moore, Evaluator; and Michael J.  O'Dell, Senior Social Science
Analyst. 


RELATED GAO PRODUCTS
========================================================== Appendix II

VA Health Care:  VA's Efforts to Maintain Services for Veterans With
Special Disabilities (GAO/T-HEHS-98-220, July 23, 1998). 

Veterans' Health Care:  Challenges Facing VA's Evolving Role in
Serving Veterans (GAO/T-HEHS-98-194, June 17, 1998). 

VA Community Clinics:  Networks' Efforts to Improve Veterans' Access
to Primary Care Vary (GAO/HEHS-98-116, June 15, 1998). 

Results Act:  Observations on VA's Fiscal Year 1999 Performance Plan
(GAO/HEHS-98-181R, June 10, 1998). 

VA Hospitals:  Issues and Challenges for the Future (GAO/HEHS-98-32,
Apr.  30, 1998). 

VA Health Care:  Closing a Chicago Hospital Would Save Millions and
Enhance Access to Services (GAO/HEHS-98-64, Apr.  16, 1998). 

VA Health Care:  Status of Efforts to Improve Efficiency and Access
(GAO/HEHS-98-48, Feb.  6, 1998). 

Managing for Results:  Agencies' Annual Performance Plans Can Help
Address Strategic Planning Challenges (GAO/GGD-98-44, Jan.  30,
1998). 

VA Health Care:  Resource Allocation Has Improved, but Better
Oversight Is Needed (GAO/HEHS-97-178, Sept.  17, 1997). 

VA Health Care:  Lessons Learned From Medical Facility Integrations
(GAO/T-HEHS-97-184, July 24, 1997). 

VA Health Care:  Assessment of VA's Fiscal Year 1998 Budget Proposal
(GAO/T-HEHS-97-121, May 1, 1997). 

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

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

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


*** End of document. ***