Disabled Veterans' Care: Better Data and More Accountability Needed to
Adequately Assess Care (Letter Report, 04/21/2000, GAO/HEHS-00-57).

Pursuant to a congressional request, GAO reviewed the Department of
Veterans Affairs' (VA) compliance with the requirements to maintain
capacity and access for veterans with special disabilities, focusing on:
(1) the accuracy of the conclusions in VA's fiscal year (FY) 1998 annual
capacity report; and (2) challenges facing VA in managing its special
disability programs.

GAO noted that: (1) VA concluded in its annual report for FY 1998 that
it had maintained its capacity to treat veterans with special
disabilities; (2) however, VA's data are not sufficient to support that
conclusion because of extensive data problems, such as the use of
unreliable proxy measures to identify veterans with special
disabilities; (3) moreover, VA based its conclusion on national
statistics that indicated more special disability veterans were served
with fewer resources expended in 1998 than in 1996; (4) however, there
is considerable variability among the Veterans Integrated Service
Networks (VISN), and, in fact, some VISNs reported serving fewer
veterans; (5) in addition, VA attributes reduced expenditures and the
use of fewer resources to efficiency gains--however, because it lacks
outcome measures, VA cannot tell whether it has maintained, enhanced, or
diminished quality of care; (6) VA faces challenges in maintaining its
capacity to serve special disability populations; and (7) in particular,
the lack of a single VA headquarters unit accountable for ensuring
compliance with the capacity legislation may have caused delays in: (a)
monitoring and investigating locations where capacity appears to have
declined; and (b) fully implementing congressionally mandated
performance standards for VA employees responsible for allocating and
managing special disability program resources.

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

 REPORTNUM:  HEHS-00-57
     TITLE:  Disabled Veterans' Care: Better Data and More
	     Accountability Needed to Adequately Assess Care
      DATE:  04/21/2000
   SUBJECT:  Veterans
	     Data integrity
	     Health care services
	     Persons with disabilities
	     Accountability
	     Health care cost control
	     Rehabilitation programs
	     Veterans benefits
	     Performance measures
IDENTIFIER:  VA Veterans Integrated Service Network

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** 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.                                         **
**                                                              **
******************************************************************
GAO/HEHS-00-57

Appendix I: VA Definitions of Six Special Disabilities

22

Appendix II: Performance and Access Monitors

23

Appendix III: Comments From the Department of Veterans Affairs

34

Table 1: Change in Number of Special Disability Veterans Served and Dollars
Spent Between Fiscal Years 1996 and 1998 8

Table 2: Change in Number of Special Disability Veterans Served in Inpatient
and Outpatient Settings Between Fiscal Years 1996
and 1998 9

Table 3: Change in FTE Employees and Beds for Spinal Cord
Dysfunction and Blind Rehabilitation Between Fiscal Years
1996 and 1998 10

BROS blind rehabilitation outpatient specialist

CDR cost distribution report

DSM-IV Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition

FTE full-time-equivalent

PTSD post-traumatic stress disorder

SCI spinal cord injury

SCI&D spinal cord injury and disorder

SMI serious mental illness

TBI traumatic brain injury

VA Department of Veterans Affairs

VHA Veterans Health Administration

VISN Veterans Integrated Service Network

Health, Education, and
Human Services Division

B-283462

April 21, 2000

The Honorable Lane Evans
Ranking Democratic Member
Committee on Veterans' Affairs
House of Representatives

Dear Mr. Evans:

In 1996, the Congress expressed concern that budgetary pressures and ongoing
reorganization within the Department of Veterans Affairs (VA) health care
system could make VA's specialized programs for disabled veterans vulnerable
to inappropriate cost cutting. Section 104 of the Veterans Health Care
Eligibility Reform Act of 1996 (P.L. 104-262) requires the Secretary of VA
to (1) ensure that VA's systemwide capacity to provide specialized treatment
and rehabilitative services to veterans with spinal cord dysfunction,
blindness, amputations, or mental illness is not reduced below October 1996
levels and (2) provide veterans with reasonable access to needed specialized
care and services. VA is required to report to the House and Senate
Committees on Veterans' Affairs annually from 1997 through 2001 about its
systemwide capacity to provide this specialized care.

Although the legislation directed VA to preserve capacity and to ensure
reasonable access for veterans with special disabilities, it did not define
capacity or access or specify how each was to be measured. After
consultation with stakeholders,1 VA defined capacity as the number of
individual veterans treated within specialized inpatient units and clinics
and the dollars expended for their care. VA included number of beds and
staffing levels as additional measures of capacity for spinal cord
dysfunction and blind rehabilitation. Access was defined as timeliness in
providing services to veterans for their specialized needs. In addition, VA
planned to implement outcome measures within 2 to 3 years to evaluate
program effectiveness, regardless of resources expended, by measuring
treatment results.

This report responds to your request that we review VA's compliance with the
requirements to maintain capacity and access for veterans with special
disabilities.2 Specifically, we provide the results of our review of (1) the
accuracy of the conclusions in VA's fiscal year 1998 annual capacity report
and (2) challenges facing VA in managing its special disability programs. In
addition, we assessed whether VA has complied with section 903 of the
Veterans Programs Enhancement Act of 1998 (P.L. 105-368), which directed the
Under Secretary for Health to prescribe, by January 1, 1999, job performance
standards for employees responsible for allocating and managing special
disability program resources.

To develop this information, we met with VA officials responsible for
developing and analyzing information on the disability programs; VA
officials responsible for managing the special disability programs at the
national level; officials at Veterans Integrated Service Networks (VISN) in
Durham, N.C., Atlanta, Ga., and San Francisco, Calif.; officials at
facilities in Durham, N.C., Richmond, Va., Decatur and Augusta, Ga., and
Palo Alto and San Francisco, Calif.; and representatives from veterans'
service organizations and advisory committees with which VA is required to
consult in responding to the 1996 legislation. We also reviewed relevant VA
and advisory committee reports, policies, manuals, and publications. We
performed our work between October 1998 and January 2000 in accordance with
generally accepted government auditing standards.

VA concluded in its annual report for fiscal year 1998 that it had
maintained its capacity to treat veterans with special disabilities.
However, VA's data are not sufficient to support that conclusion because of
extensive data problems, such as the use of unreliable proxy measures to
identify veterans with special disabilities. Moreover, VA based its
conclusion on national statistics that indicated more special disability
veterans were served with fewer resources expended in 1998 than in 1996.
However, there is considerable variability among the VISNs, and, in fact,
some VISNs reported serving fewer veterans. In addition, VA attributes
reduced expenditures and the use of fewer resources to efficiency gains;
however, because it lacks outcome measures, VA cannot tell whether it has
maintained, enhanced, or diminished quality of care.

VA faces challenges in maintaining its capacity to serve special disability
populations. In particular, the lack of a single VA headquarters unit
accountable for ensuring compliance with the capacity legislation may have
caused delays in (1) monitoring and investigating locations where capacity
appears to have declined and (2) fully implementing congressionally mandated
performance standards for VA employees responsible for allocating and
managing special disability program resources. In order to ensure compliance
with the capacity legislation, we are recommending that VA designate a
single office to be accountable for fully implementing the mandate to
maintain capacity in special disability programs.

Since 1995, VA has taken significant steps to transform its health care
system from a hospital- and specialist-based system to a
prevention-oriented, community-based system with primary care as its
foundation. To accomplish this transition, VA moved from a management
structure based on 172 hospitals to one based on VISNs in 22 separate
geographic areas. These VISNs have substantial operational autonomy and are
responsible for making basic budgetary, planning, and operational decisions
to meet the health care needs of veterans living within the 22 geographic
areas. Each VISN oversees between 5 and 11 large hospital facilities, as
well as clinics and other delivery locations.

The Veterans Health Care Eligibility Reform Act of 1996 authorized new
eligibility rules for outpatient treatment that permit VA to provide medical
care in the most appropriate settings. Eligibility reform was intended to
reduce inappropriate admissions and denial of care to many veterans who were
ineligible under the old rules for outpatient treatment. In addition, VA
proposed a plan to operate within the same annual appropriation for VA
health care through 2002.3 As a result, VISN and facility managers had
strong incentives to reengineer delivery systems to offset rising health
care costs.

In this environment of shifting service delivery from inpatient to
outpatient settings, system reorganization, and no-growth budgets, section
104 of the Veterans Health Care Eligibility Reform Act directed VA to
protect services and resources committed to veterans with spinal cord
dysfunction, blindness, amputations, or mental illness. After discussions
with stakeholders, as required by the act, VA added two more special
disabilities--traumatic brain injury and post-traumatic stress disorder
(PTSD)--to this list. For the purpose of the capacity requirement, VA
limited the definition of mental illness to refer to only those veterans
with serious mental illness and created two subcategories: veterans disabled
as a result of a disorder related to substance abuse and homeless veterans
disabled as a result of serious mental illness.

House Report 104-690, which accompanied the 1996 legislation, noted that the
special disability programs constitute a vital core of VA's mission, tend to
be high-cost efforts, and are unmatched in scope and quality in the private
sector. The six special disabilities were targeted because of their close
association with service-related illnesses and the likelihood of
progressively worsening disability in the absence of specialized treatment
or rehabilitation. VA must carry out the requirements of the legislation in
consultation with the Advisory Committee on Prosthetics and Special
Disabilities Programs and the Committee on the Care of Severely Chronically
Mentally Ill Veterans,4 and VA has done so.

Although the legislation directed VA to preserve capacity and to ensure
reasonable access for veterans with special disabilities, it did not specify
how capacity and access were to be defined or measured. VA defined capacity
as the number of individual veterans treated within specialized inpatient
units and clinics and the dollars expended for the care of these veterans.
VA intends to use outcome measures, when they become available, to measure
the effectiveness of its specialized programs. At the insistence of
veterans' service organizations, number of beds and staffing levels were
included as additional measures of capacity for spinal cord dysfunction and
blind rehabilitation. VA defined access as timeliness in providing services
to veterans for their specialized needs. Although VA considered other
measures of access, the data necessary to develop these measures, primarily
the number and location of the universe of veterans with each disability,
are not generally available.

Has Been Maintained

VA's fiscal year 1998 annual capacity report concluded that VA's capacity to
treat special disability groups nationwide had been generally maintained.
This conclusion was based on national statistics that indicated that more
special disability veterans were served in 1998 than in 1996. However,
measurement results varied among VISNs, with some showing a decrease in
veterans served. In our view, the available data are insufficient to support
any conclusions because VA's workload and resource data are inaccurate. In
addition, reliable outcome measures are not available to assess whether the
quality of care provided to special disability populations has changed or is
satisfactory.

VA has asserted that capacity to treat veterans with special disabilities
has been maintained because the number of veterans treated in special
disability programs increased from fiscal year 1996 to fiscal year 1998 by 8
percent, or 28,141 individuals. However, during that same period, spending
for these programs decreased by 8 percent, or approximately $184 million.
Veterans with a serious mental illness accounted for 81 percent of the
special disability veterans served and 84 percent of the dollars spent in
the special disability categories. The number of veterans with special
disabilities served during the period increased or remained relatively
constant for all conditions except amputations. VA reported the decline in
the number of veterans with amputations as a favorable outcome of successful
efforts to prevent amputations in diabetic patients. VA expenditures
decreased for amputations, serious mental illness, and PTSD. Analysis of
expenditures for two subcategories of serious mental illness shows a
decrease of 29 percent for veterans with a disorder related to substance
abuse and an increase of 23 percent for veterans who were homeless.
Expenditures also increased for spinal cord dysfunction, traumatic brain
injury, and blind rehabilitation. (See table 1.)

Table 1: Change in Number of Special Disability Veterans Served and Dollars
Spent Between Fiscal Years 1996 and 1998

                   Veterans served            Dollars spent (in thousands)

 Disability                    Change                              Change
             FY 1996 FY 1998                FY 1996    FY 1998
                             (percentage)                        (percentage)
 Spinal cord
 dysfunction 8,598   9,252   +654 (+8)     $199,848   $202,878   +$3,030 (+2)

 Blindness   9,726   11,930  +2,204 (+23)  43,855     53,935     +10,080
                                                                 (+23)
 Traumatic
 brain injury176     189     +13 (+7)      4,439      4,906      +467 (+11)
 Amputations 4,765   4,549   -216 (-5)     5,953      5,286      -667 (-11)
 Serious
 mental      269,009 290,961 +21,952 (+8)  2,080,240  1,900,938  -179,302
 illnessa                                                        (-9)
 Substance                                                       -168,568
 abuse       107,074 106,599 -475 (-0.4)   575,902    407,334    (-29)

 Homeless    24,539  27,201  +2,662 (+11)  75,071     92,614     +17,543
                                                                 (+23)

 PTSD        39,653  43,187  +3,534 (+9)   101,882    84,112     -17,770
                                                                 (-17)

 Total       331,927 360,068 +28,141(+8)   $2,436,217 $2,252,055 -$184,162
                                                                 (-8)

aThe total for serious mental illness is more than the sum of the
subcategories listed under it because the category includes, but is not
limited to, veterans who are substance-abusing or homeless.

Source: VA's 1998 capacity report to the Congress: Maintaining Capacity to
Provide for the Specialized Treatment and Rehabilitative Needs of Disabled
Veterans (VA, June 1999).

VA's fiscal year 1998 capacity report was an improvement over its previous
reports because it included for the first time a breakout of veterans served
in both inpatient and outpatient settings. Between 1996 and 1998, the number
of veterans served in inpatient settings increased for blindness, traumatic
brain injury, the homeless subcategory of serious mental illness, and PTSD.
The shift from inpatient to outpatient care was most evident in the broad
category of serious mental illness and its subcategory of substance abuse.
The number of seriously mentally ill veterans treated in an inpatient
setting declined by approximately 19 percent, while the number of seriously
mentally ill veterans with substance abuse disorders who received inpatient
care declined by approximately 41 percent. The number of special disability
veterans treated in outpatient settings increased for all conditions with
the exception of amputations. (See table 2.)

Table 2: Change in Number of Special Disability Veterans Served in Inpatient
and Outpatient Settings Between Fiscal Years 1996 and 1998

                         Veterans served
 Disability/Setting                        Percentage change
                         FY 1996  FY 1998
 Spinal cord dysfunction
 Inpatient               5,185    5,117    -1
 Outpatient              6,599    7,576    +15
 Blindness
 Inpatient               1,607    1,976    +23
 Outpatient              9,345    11,560   +24
 Traumatic brain injurya
 Inpatient               176      189      +7
 Amputationsa
 Outpatient              4,765    4,549    -5
 Serious mental illness
 Inpatient               117,088  95,068   -19
 Outpatient              251,216  278,674  +11
 Substance abuse
 Inpatient               50,628   30,021   -41
 Outpatient              90,916   99,337   +9
 Homeless
 Inpatient               5,273    7,072    +34
 Outpatient              21,913   23,763   +8
 PTSD
 Inpatient               4,312    4,694    +9
 Outpatient              37,768   41,224   +9

aData were not reported for both inpatient and outpatient settings.

Source: VA 1998 capacity report (VA, June 1999).

For veterans disabled by spinal cord dysfunction or blindness, capacity was
also measured by staff resources--full-time-equivalent (FTE) employees--and
the number of specialized beds dedicated to veterans with these
disabilities. From 1996 to 1998, VA reported that staffing levels dropped by
12 percent (267 FTE employees) for spinal cord dysfunction and increased by
1 percent (3 FTE employees) for blind rehabilitation. Numbers of beds
declined in both areas: 15 percent (180 beds) for spinal cord dysfunction
and 7 percent (15 beds) for blind rehabilitation. (See table 3.) Veterans'
service organizations have questioned the accuracy of these numbers on the
basis of surveys they have conducted at VA facilities and have concluded
that capacity reductions have been even greater: 18 percent fewer spinal
cord care beds and 32 percent fewer spinal cord care staff resources. In
addition, the Advisory Committee on Prosthetics and Special Disabilities
Programs has questioned whether VA inappropriately included unstaffed spinal
cord and blind rehabilitation beds in the capacity report. Two of the
facilities we visited reported delays in transferring veterans with acute
care needs to a specialized spinal cord injury unit. In addition, the
national average waiting time for admission to an inpatient blind
rehabilitation program increased slightly, from 31.8 weeks in fiscal year
1996 to 33.4 weeks in fiscal year 1998. Moreover, outreach efforts by the
facility-based Visual Impairment Services Team, which provides coordinated
services to legally blind veterans, continued to increase the already
lengthy waiting lists for blind rehabilitation programs. The delays in
admission to spinal cord care beds in some areas and the lengthy waiting
times for blind rehabilitation indicate that the reduction in bed levels may
be affecting access to these services.

Table 3: Change in FTE Employees and Beds for Spinal Cord Dysfunction and
Blind Rehabilitation Between Fiscal Years 1996 and 1998

 Disability/Measure      FY 1996  FY 1998  Percentage change
 Spinal cord dysfunction
 FTE employees           2,175.8  1,909.3  -12
 Beds                    1,209    1,029    -15
 Blindness
 FTE employees           414.5    417.3    +1
 Beds                    228      213      -7

Source: VA 1998 capacity report (VA, June 1999).

While VA stated that it had maintained capacity nationally, some VISNs
appeared to be maintaining workloads and expenditures for special disability
populations, while others showed declines in veterans served and
expenditures. For example, only two VISNs maintained or increased their
workloads for all six disabilities. Another two VISNs served fewer veterans
in at least four of the six special disability groups. VA's Committee on the
Care of Severely Chronically Mentally Ill Veterans stated in its response to
the capacity report that because the measure of maintenance of clinical
effort (dollars expended) had actually decreased, VA should scrutinize those
VISNs with the largest reductions in capacity. The committee further stated
that many VISNs showed substantial increases in the numbers of veterans
treated (up to 17 percent) with relatively constant expenditures, while
other networks showed decreases in veterans treated and decreases in funds
expended. VA provided no data to show that decreased demand for services
accounted for the decrease in veterans treated.

The information management systems currently used by VA are not precise
enough to capture the information necessary to accurately calculate workload
and expenditure statistics for the special disability populations. As a
result, VA developed a complex process using eight different databases to
compile VISN and national workload and resource data for the six disabling
conditions. This process used inpatient diagnostic information, when
available, and a set of proxy measures to infer a likely condition when
diagnostic information was not available. For example, because VA's
outpatient care database does not currently include diagnostic information,5
VA identified additional patients as belonging to a special disability group
on the basis of information regarding the number of visits to specific
clinics. Thus, veterans who visited certain psychiatric/mental health
clinics at least six times were counted as disabled by serious mental
illness. Because of the lack of diagnostic information and the sometimes
inappropriate proxies used, we are not confident that the workload figures
and subsequent expenditure amounts are accurate. Stakeholders have voiced
similar concerns. For example, the cochair of the Committee on the Care of
Severely Chronically Mentally Ill Veterans stated during testimony in 1998
that "the currently available data is inadequate to comprehensively and
reliably monitor the Veterans Health Administration's efforts to maintain
capacity for these disabling conditions."

Our visits to selected field locations helped confirm the validity of our
concerns about the accuracy of VA's workload and resource data. We were
unable to validate workload and resource data contained in the capacity
report using information maintained at VA facilities because data were not
routinely available under the definitions developed for the disabling
conditions (see app. I).6 Several clinicians told us that the definitions
used in the capacity report had no clinical basis when it came to treating
patients. For example, as defined by VA's capacity report, seriously
mentally ill veterans represented about 81 percent of the universe of
veterans with special disabilities in fiscal year 1998. Yet the seriously
mentally ill category would not be tracked at the facility level because it
does not represent a meaningful grouping of patients who would receive
similar medical care.

Our site visits also found that despite 3 years of requirements to report on
capacity, management staff at VA facilities generally did not know the
definitions used by VA headquarters to identify veterans with special
disabilities or the methodology used to develop workload data for their
facilities. For example, one facility offered substance abuse treatment in a
day treatment program instead of a traditional substance abuse clinic. The
capacity report indicated that this facility experienced a 17-percent
decline in the treatment of seriously mentally ill patients with substance
abuse disorders and a 9-percent decline in expenditures for this population.
Facility officials believed that the capacity report understated workload
for seriously mentally ill patients with substance-related disorders because
VA's methodology did not include the day treatment clinic as a program
serving this special disability population.

We also identified deficiencies in the accuracy of VA's resource measures
(that is, expenditures for all programs and staffing levels and beds for
spinal cord care and blind rehabilitation). Moreover, veterans' service
organizations have reported discrepancies between the cost distribution
report (CDR), which is the data system used by VA to allocate costs, and
information reported by special disability program officials. The
information in the CDR is suspect because it relies on subjective judgments
to allocate the distribution of staff time and dollars spent in each
inpatient unit and outpatient care area. VA's Inspector General found that
service-level managers have broad discretion in selecting and applying cost
allocation techniques, leading to inconsistency, infrequent updates, and
disparate treatment of similar cost accounting issues.7 Clinical staff told
us that vacancies were at times deliberately hidden through the reallocation
of staff to special disability programs in the CDR or filled with
individuals possessing less skill and ability. Thus, these staff believe
that the CDR can be easily manipulated to create the appearance that
staffing and expenditure levels in the special disability program have been
maintained.

VA acknowledged that its data systems need improvement, and in December 1998
the Veterans Health Administration (VHA) held a Data Quality Summit to
identify issues related to the collection and use of data in VHA. Multiple
data quality issues were identified, including the completeness,
reliability, validity, and timeliness of ambulatory care data. Recognizing
problems with the data used to prepare the annual capacity report, VA used a
verification and data correction process to improve the accuracy and
reliability of data for the fiscal year 1998 report to the Congress. VA
headquarters shared preliminary data with medical centers, VISNs, and
program offices to identify problems. In addition, the most recent capacity
report reflects a closer working relationship among VA, its advisory
committees, and interested veterans' service organizations. According to VA,
this collaboration has resulted in data improvements. While the steps taken
by VA to improve its data quality are commendable, we believe that these
efforts will not bear fruit in the short term because of the myriad people
and processes at the facility, VISN, and national levels that make data
collection at VA so cumbersome.

The primary basis for VA's conclusion that it has maintained capacity is the
increased number of veterans served by the special disability programs.
However, of the five special disability programs that reported serving more
veterans, two experienced a decrease in expenditures and two a reduction in
dedicated beds. Without outcome measures, the effect of these changes on the
appropriateness and effectiveness of treatment is unclear. For example,
although the number of veterans treated for serious mental illness increased
by 8 percent from 1996 to 1998, expenditures decreased by 9 percent.
Similarly, the number of veterans treated for PTSD increased by 9 percent,
while expenditures declined by 17 percent. VA generally attributed
expenditure reductions to increases in efficiency as outpatient or
domiciliary care replaced more costly hospital inpatient treatment. Other
stakeholders review the same data and conclude that reduced expenditures
have eroded comprehensiveness and quality of care.

Facility managers we contacted were primarily concerned with maintaining
operations given the constraints of constant budgets, staff reductions, and
increasing workloads. These managers implemented various strategies to
improve efficiency while maintaining services to all veterans, including
those served in the special disability programs. These strategies included
the use of service lines,8 "hoptel" beds,9 making referrals to
community-based service providers, and shifting care to outpatient settings.
Facility officials generally believed that newly developed alternative care
settings were appropriate for special disability populations, although no
clear evidence exists to support this position.

While facility officials believed that they were meeting the demand for
special disability services, they expressed concern that additional cost
reductions might adversely affect quality of care. One facility was able to
reduce the number of inpatient psychiatric beds from over 400 to fewer than
100. Officials at this location were confident that the community
infrastructure was adequate for most veterans. VA case managers were
assigned to patients, and staff members were working with the community to
develop additional capacity as needed. In contrast, officials from another
facility stated that their community had few suitable alternatives, a
situation that led the facility's chief of staff to question the strategy of
deinstitutionalizing patients with mental illness.

Assessment of patient care outcomes for VA's special disability populations
would be a major asset in interpreting VA data and trends that showed more
veterans served with fewer resources. Although outcome measures are
difficult to develop and are not generally available in private sector
health care systems either, VA made a commitment in 1997 to develop within 2
years comprehensive and reliable measures of treatment outcome for the six
disability groups.

The fiscal year 1998 capacity report contains performance "monitors" that
are a mixture of outcome and process measures related to the care provided
to the six disability groups. The 18 performance monitors VA identified are
designed to assess quality, functional status, and patient satisfaction.
According to VA, these measures will be revised as more appropriate ones are
identified.

Data were unavailable for 7 of the 18 performance monitors identified in the
fiscal year 1998 capacity report, and information was unavailable for the 3
years from 1996 through 1998 for 15 monitors (see app. II). Some performance
monitors, such as continuity of care for previously hospitalized patients
and changes in functional status, appear to be useful indicators of quality.
However, others are more process-oriented and do not support an assessment
of possible improvements resulting from the care provided. For example, the
performance monitor for the care of veterans with serious mental illness is
a process measure of the percentage of patients who are assessed on a
one-time basis for their level of functioning, not an outcome measure of
their improvement. Furthermore, some monitors are limited to a small segment
of the population or address populations broader than the special disability
populations. For example, performance monitors for the spinal cord
dysfunction population include only those patients discharged after
inpatient treatment (about 55 percent of all spinal cord dysfunction
patients served), and measures for the serious mental illness category
include all psychiatric patients, and not just those with serious mental
illnesses.

Both advisory committees questioned the validity of VA's performance
monitors and expressed concern that insufficient progress has been made in
the development of comprehensive and valid measures of treatment outcome as
VA transitions to greater reliance on outpatient delivery systems. While
VA's development of performance monitors is a step in the right direction,
more research is needed to determine whether these measures are adequate to
assess whether the care provided to veterans in the special disability
programs is as comprehensive as, and equal in quality to, the care provided
in 1996.

Programs

Special disability services are delivered at the facility level, where VISN
and facility officials face the need to become more efficient to meet the
needs of more veterans with fewer resources. The alternative to increased
efficiency is decreased services. Accountability for maintaining capacity in
the special disability programs is currently fragmented among several
organizational units in VA, and performance standards mandated by statute
have not been fully implemented for those managing resources or allocating
them to special disability programs. VA indicated in its fiscal year 1998
capacity report that it was monitoring situations in which capacity appeared
to have declined, but VA did not respond to our repeated requests that it
identify who was responsible for this monitoring.

Organizational Units

Responsibility for implementing the mandate to maintain capacity in special
disability programs is divided among several headquarters units, including
the Office of Policy and Planning, the Chief Network Office, and the Office
of Patient Care Services. The Office of Policy and Planning is responsible
for developing the annual capacity report. This office coordinates the
development of capacity statistics and program definitions, oversees the
verification and validation process,10 and consults with internal and
external stakeholders in finalizing the capacity report. The Chief Network
Officer is the primary point of contact for the VISNs and provides
operational direction and supervision to the field through the 22 VISN
directors. The Office of Patient Care Services houses the clinically related
headquarters programs that support the delivery of patient care services in
the field. This office develops patient care policies and guidelines, acts
as program consultant to the special disability programs, and provides
advice and consultation to VISN and facility directors.

After contacting these three headquarters units, we concluded that none of
them was responsible for monitoring field locations whose capacity to serve
special disability populations appears to have declined. Each of the units
denied responsibility for monitoring and referred us to one of the other
offices as the potentially responsible unit. Despite VA's data problems,
enhanced monitoring and follow-up could be useful in mitigating the
limitations of VA's current capacity measures and performance monitors. In
addition to helping identify data reliability issues, such monitoring
efforts could bring to light legitimate concerns about the provision of
services to special disability populations in alternative care settings.

To improve accountability for maintaining capacity in the special disability
programs, the Congress, through the Veterans Programs Enhancement Act,
required VA to develop job performance standards for employees responsible
for allocating and managing resources for serving veterans with special
disabilities. The law also required that the standards include measures of
workload, allocation of resources, and quality of care indicators, and that
the standards be implemented by January 1, 1999.

As of January 2000, VA had implemented at least one quality of care
performance standard, or monitor, in each of the six special disability
programs. The 2000 VISN directors' performance agreement states that part of
the performance evaluation will be based on the results of the monitors of
capacity for special populations. The agreement includes 12 monitors related
to the disability populations. Achievement goals for VISN directors have
been established for the 2 spinal cord care monitors, but not for the
remaining 10 monitors. Without stated goals, it is unclear what would be
considered acceptable performance.

Also, the performance agreement is silent on the measurement of workload and
allocation of resources, which is required by the law. The mandatory job
performance standards cannot be considered fully implemented without
measures of workload and allocation of resources.

VA has demonstrated that measurable performance standards for key management
officials can promote change. For example, by including in the VISN
directors' performance agreements a requirement to decrease the number of
days inpatients spent in acute care in fiscal year 1998, VA reduced these
acute-bed-days from 3,430 per 1,000 veterans served in fiscal year 1994 to
1,333 per 1,000 veterans served in fiscal year 1998, a 61-percent decrease.

While questions remain about the accuracy of VA's workload and resource
data, VA has committed to work with stakeholders and veterans' service
organizations to improve the accuracy and reliability of the data. VA has
also committed itself to developing measures of quality of care in special
disability programs in order to ensure that quality is maintained or
improved. However, in view of the difficulty of developing and validating
outcome measures, it is unlikely that VA will be able to develop measures
across all special disability programs in the near future. In the meantime,
the annual capacity report can be a valuable tool for identifying specific
locations with potential problems in service delivery to special disability
populations. Enhanced monitoring of such locations could be used to augment
VA's current limited capacity measures and performance monitors.

Responsibility for managing VA's response to the capacity requirement is
dispersed among several of VA's organizational components, and none of them
has taken responsibility for investigating apparent declines in capacity or
quality of services for veterans with special disabilities. Designating a
single organization as responsible for these functions would help focus
accountability for maintaining capacity. In addition, the accountable office
could be charged with fully implementing the congressional requirement to
develop job performance standards for employees responsible for allocating
and managing resources used to serve veterans with special disabilities.

To help ensure compliance with the law, we recommend that the Secretary of
Veterans Affairs direct the Under Secretary for Health to assign lead
responsibility to a headquarters unit for

ï¿½ initiating efforts to monitor and determine the causes for apparent
declines in capacity and

ï¿½ developing job performance standards for employees who are responsible for
allocating and managing the resources used to serve veterans with special
disabilities.

In commenting on the draft report, VA generally agreed with our findings and
recommendations. VA intends to take an approach to ensuring compliance with
the law that is different from assigning lead responsibility to a
headquarters unit. VA said it would renew its commitment to using existing
coordination and issue resolution mechanisms to address compliance with the
law. We continue to believe that assigning responsibility to one office
would better ensure that capacity is accurately measured and appropriately
maintained. The coordination mechanisms have not accomplished this in the
past, and focusing one office's attention on the issue is, in our opinion,
more likely to ensure accountability in the future. VA said that a new
management structure would have to be created if it designated a single
office as responsible for ensuring compliance with the law. We believe that
a new management structure is not required and that designating an office as
accountable for compliance with the law would be sufficient. VA has tasked a
working group, the 3-year-old Performance Management Work Group, with the
development of job performance standards. This action should emphasize and
delineate responsibility for the timely completion of the job performance
standards.

VA also commented that it believes that measuring the full continuum of
care, not just the numbers of beds and FTE staff, is the most appropriate
measure of access to care. We agree with VA, but, as we have discussed in
the report, VA does not have the data or processes available to consider the
full continuum of care. In this regard, we support VA's efforts to develop
outcome measures.

VA expressed concern that the draft report placed emphasis on the
maintenance of capacity at the VISN level, noting that the law states that
capacity should be maintained at the national level. As discussed in the
report, we included information by VISN because the nationwide data hid the
variability across the VISNs. Moreover, VA stated in its 1998 capacity
report that it was monitoring situations in which capacity appeared to have
declined in particular VISNs; however, we were unable to identify who was
responsible for this monitoring. We support VA's initiative to monitor
declines in capacity at the VISN level, knowing that such an effort exceeds
the statutory requirement to maintain capacity nationally.

As agreed with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days from
the date of the report. At that time, we will send copies of this report to
the Honorable Togo D. West, Jr., Secretary of Veterans Affairs; appropriate
congressional committees; and other interested parties. We will also make
copies available to others upon request.

Please contact me on (202) 512-7101 if you or your staff have questions
about the report or need additional assistance. George Poindexter, Linda
Diggs, Marcia Mann, and William Stanco made key contributions to this
assignment.

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

VA Definitions of Six Special Disabilities

          Disability                  VA capacity report definition
                               Veterans with neurological deficit lesions
                               involving the spinal cord, including but not
                               limited to traumatic spinal cord injuries;
                               intraspinal neoplasms resulting in
 Spinal cord injury and        neurological deficit; vascular insults to
 disorders (SCI&D)             the spinal cord; cauda equina syndrome;
                               inflammatory disease of the spinal cord; and
                               diseases such as multiple sclerosis,
                               unstable traumatic lesions of the spinal
                               cord, and degenerative spine diseases are
                               considered to have SCI&D.
                               Veterans are considered disabled by
                               blindness when the best corrected central
                               visual acuity with ordinary eyeglasses or
                               contact lenses is 20/200 or less in the
 Blindness                     better eye, or when the best corrected
                               visual acuity in the better eye is better
                               than 20/200 but visual field defects exist
                               that produce a useful visual field dimension
                               of 20 degrees or less.
                               Veterans who have sustained brain trauma,
                               including from iatrogenic causes, and who
                               have motor or cognitive impairments as a
                               result of the brain injury for at least 3
                               months, are considered to have a TBI. These
                               individuals either require rehabilitative
                               services, such as acute intervention of
 Traumatic brain injury (TBI)  speech pathology and cognitive
                               rehabilitation, in the first several months
                               or are so severely impaired that their
                               rehabilitative potential is objectively so
                               low that rehabilitative services are not
                               appropriate. These patients may require
                               chronic life support measures, such as
                               mechanical ventilation.
                               Amputation applies to veterans who have a
                               full or partial amputation of a limb,
 Amputation                    including neurologic loss of a limb (except
                               from stroke) and loss of the use of a limb
                               from injuries to the brain, spinal cord, or
                               peripheral nerves.
                               Veterans who have within the past year had a
                               diagnosed mental, behavioral, or emotional
                               disorder of sufficient duration to result in
                               a disability that meets the Diagnostic and
                               Statistical Manual of Mental Disorders,
                               Fourth Edition (DSM-IV),a criteria are
                               considered to have an SMI. Disability is
 Serious mental illness (SMI)  defined as a functional impairment that
                               substantially interferes with or limits one
                               or more major life activities, such as
                               bathing, dressing, managing money, or taking
                               prescribed medication. Two subcategories
                               include veterans who have a diagnosed DSM-IV
                               substance-related disorder and homeless
                               veterans who have a disability as a result
                               of mental illness.
                               Veterans who meet the diagnostic criteria
                               for PTSD include those who

                               have been exposed to a catastrophic or
                               traumatic event involving actual or
                               threatened death or injury, or a threat to
                               the physical integrity of self or others,
                               and who have a subjective response of
                               intense fear, helplessness, or horror;

                               have intrusive recollections or
                               reexperiences of the event;
 Post-traumatic stress
 disorder (PTSD)               persistently avoid stimuli associated with
                               the trauma and have generally numbed
                               responsiveness;

                               have persistent symptoms of increased
                               arousal;

                               have symptoms that last 1 month or longer;
                               and

                               experience clinically significant distress
                               or impairment in social, occupational, or
                               other important areas of functioning.

a DSM-IV is the standard handbook for psychodiagnosis employed by clinicians
and researchers in the United States.

Performance and Access Monitors

Continued

                                               Indicator
                 Access or                        type       Portion of
  Disability    performance     Indicator      (processa     population    Method of data   Results       Comments
                 monitor                           or         covered       collection     reported
                                               outcomeb)
                                                          Only SCI&D
                                                          patients needing
                                                          urgent admission
                                                          to an SCI center                            Acute condition
                                                          are included in                             means care is
                                                          the measurement,                            required by newly
                                                          which covers all                            injured veterans
                            Admission to a                newly injured                               or by veterans
                            spinal cord injury            veterans and a                              with SCI who need
                            (SCI) center                  portion of the                  FY 96--41%  urgent care
                            within 1 day of               SCI&D populationSurvey of SCI               because of
 SCI&D         Access       request for        Process    with older      centers by the  FY 97--91%  medical
                            patients needing              injuries. VA    SCI&D Strategic             complications or
                            acute, inpatient              treats          Health Group    FY 98--100% surgical needs.
                            specialty care                approximately                               Results are
                            (average)                     400 newly                                   reported as a
                                                          injured patients                            percentage of SCI
                                                          per year. The                               centers that met
                                                          newly injured                               the requirements
                                                          make up 8% of                               established by
                                                          the SCI&D                                   the indicator.
                                                          population
                                                          treated as
                                                          inpatients.
                                                                                                      Results are
                            SCI clinic                    Covers those                    FY 96--87%  reported as a
                                                          patients seen inSurvey by the               percentage of SCI
 SCI&D         Access       appointment within Process    SCI clinics at  SCI&D Strategic FY 97--100% centers that met
                            7 days of referral
                            (average)                     facilities with Health Group                the requirements
                                                          an SCI center                   FY 98--100% established by
                                                                                                      the indicator.
                                                          Only veterans
                                                          discharged from                             Of the 23 SCI
                                                          an SCI center                               centers, 16 had
                                                          can potentially                             fewer than 30
                                                          be included in                              respondents to
                            SCI inpatients                the denominator.                            the survey;
                                                                          Patient         FY 97--55%
 SCI&D         Performance  rating VA care as  Outcome    Of the 5,117    satisfaction                according to VA,
                            "very good" to                patients                                    such small
                            "excellent"                   discharged in FYsurvey results  FY 98--55%  samples are
                                                          1998, only 550,                             regarded as
                                                          or 10%, of the                              unreliable and of
                                                          patients were                               questionable
                                                          included in the                             validity.
                                                          survey results.
                                                                                                      VA states that
                                                                                                      discharge to
                                                                                                      noninstitutional
                                                                                                      community living
                                                          Includes                                    "could be" viewed
                                                                          National
                            Discharge from SCI            veterans        Patient Care                as positive.
                                                          admitted for                    FY 97--95%  Outcome does not
 SCI&D         Performance  bed unit to        Outcome    rehabilitation  Database on                 necessarily
                            noninstitutional                              patients
                            community living              and other       discharged from FY 98--95%  relate to quality
                                                          purposes, such                              of care received
                                                          as annual exams SCI bed unit                during inpatient
                                                                                                      treatment and may
                                                                                                      be the result of
                                                                                                      socioeconomic
                                                                                                      factors.
                                                          Measures time
                                                          only for                        FY 96--31.8
                            Average waiting               patients                        weeks
                            time for admission            actually        Self-reporting              Measures
 Blindness     Access       to inpatient blind Process    admitted to     by blind        FY 97--32.2 admissions during
                            rehabilitation                program, not    rehabilitation  weeks       a 6-month period
                            program                       those on waitingcenters
                                                          list who decline                FY 98--33.4
                                                          admission                       weeks
                                                                                                      The number of
                                                                                                      veterans served
                            Number and                    There was a 36%                             is not the best
                            percentage change             increase in the                 FY 97--873  measure of
                            in the number of              number of                                   access. A
                                                          veterans seen by                FY
 Blindness     Access       veterans served by Process    a BROS,         BROS semiannual 98--1,191   numerator and a
                            blind                                         reports                     denominator are
                            rehabilitation                representing an                             needed to
                            outpatient                    increase from                   % change:   determine
                            specialists (BROS)            7.4% in 1997 to                 +36         accessibility of
                                                          10% in 1998.
                                                                                                      the program for
                                                                                                      veterans.

                                                          Patients                                    The survey
                            Inpatients of                 admitted to a VA                            response rate in
                            blind                         blind                                       1998 was 38%,
                            rehabilitation                rehabilitation  Blind                       which was an
                                                                          rehabilitation  FY 97--98%  improvement over
 Blindness     Performance  program reporting  Outcome    program, which  customer                    1997. Two of the
                            being "satisfied"             accounts for 17%
                            or "completely                of the blinded  satisfaction    FY 98--98%  nine blind
                            satisfied" with VA            veteran         survey                      rehabilitation
                            care                          population                                  centers had less
                                                          seeking care                                than 30 responses
                                                                                                      to the survey.
                                                          Impossible to                   FY 96--3
                                                          determine                       days
                            Average waiting               without                                     There is no
 TBI           Access       time for admission Process    numerators and  Survey of TBI   FY 97--2    indication of the
                            to a designated               denominators,   centers         days        size of the
                            TBI bed                       which were not                              survey sample.
                                                          provided in the                 FY 98--2
                                                          report                          days
                                                          Impossible to                   FY 96--6
                            Average waiting               determine                       days
                            time for                      without                                     There is no
 TBI           Access       first-time TBI     Process    numerators and  Survey of TBI   FY 97--5    indication of the
                            outpatient                    denominators,   centers         days        size of the
                            appointment                   which were not                              survey sample.
                                                          provided in the                 FY 98--7
                                                          report                          days
                            Discharge of
                            "first-admission"                                                         Only percentages
                            TBI patients from                                                         were provided,
                                                          Measures only   VA Functional               with a footnote
 TBI           Performance  TBI network        Outcome    first-admission Status Outcome  FY 98---68% stating that some
                            medical
                            rehabilitation                TBI patients    Database                    were based on a
                            beds to the                                                               very small number
                            community                                                                 of discharges.
                                                          Measures much
                                                          more than
                                                          patient
                            Delayed prosthetic            population with National        FY 96--1.3%
 Amputation    Access       orders             Process    amputations and delayed
                            (cumulative)                  was deleted fromprosthetic      FY 97--0.7%
                                                          the list of     order report
                                                          access measures
                                                          by VA for FY
                                                          1998
                                                                                                      The number of
                                                                                                      inpatient
                                                                                                      rehabilitation
                                                                                                      units has
                            Discharge of lower            Unable to                                   decreased and, as
                            extremity amputees            determine what  VA Functional   FY 96--82%  a result, the way
                                                                                                      care is delivered
 Amputation    Performance  from inpatient     Outcome    percentage of   Status and      FY 97--77%  to veterans
                            rehabilitation                amputees the    Outcome
                            units to community            data provided byDatabase                    undergoing
                            setting                       VA represent                    FY 98--77%  amputations has
                                                                                                      also changed,
                                                                                                      making
                                                                                                      comparisons by
                                                                                                      fiscal year
                                                                                                      difficult.
                                                                                                      Indicator
                                                                                                      included in
                            Patients at risk                              External peer               Veterans
                            for foot                                      review                      Integrated
                                                          Sample of
 Amputation    Performance  amputation who     Process    diabetic        program--data   Data not    Service Network
                            were referred to a                            collected by    available.  (VISN) directors'
                            foot care                     patients        outside                     performance
                            specialist                                    contractor.                 measures under
                                                                                                      clinical practice
                                                                                                      guidelines
                            Patient
                            satisfaction with             Survey to be
 Amputation    Performance  VA-issued lower    Outcome    administered to National survey Data not
                            extremity                     lower extremity of patients     available.
                            prosthetic limb               amputees
                                                          Indicator
                                                          includes only
                                                          those veterans
                                                          discharged from                             The indicator for
                                                          a general                                   veterans with a
                                                          psychiatry unit                             psychiatric
                            Veterans seen in              (approximately                  FY 96--52%  diagnosis who
                            any psychiatric               33% of the      Patient                     were seen within
 SMI           Access       outpatient clinic  Process    veterans treatedTreatment File  FY 97--53%  30 days of
                            within 30 days                for SMI); unableand outpatient              discharge is part
                            after discharge               to determine if files           FY 98--58%  of the FY 2000
                                                          this includes                               VISN directors'
                                                          all discharges                              performance
                                                          from psychiatry                             measures.
                                                          or just patients
                                                          determined to
                                                          have an SMI.
                                                          Indicator
                                                          includes only
                                                          those veterans
                                                          discharged from
                                                          a general
                                                          psychiatry unit                 FY 96--34
                                                                                          days
                            Average time to               (approximately  Patient
                                                          33% of the
 SMI           Access       first outpatient   Process    veterans treatedTreatment File  FY 97--32
                            visit following                               and outpatient  days
                            discharge                     for SMI); unablefiles
                                                          to determine if
                                                          this includes                   FY 98--31
                                                          all discharges                  days
                                                          from psychiatry
                                                          or just patients
                                                          determined to
                                                          have an SMI.
                            Patients with at
                            least one Global                                                          Would need
                            Assessment of                 Includes all                                comparison of
                            Functioning score             patients seen in                            functional scores
                            if seen in any                a psychiatric                               to measure
 SMI           Performance  Veterans Health    Process    inpatient or    Mental health   Data not    outcome of
                            Administration                outpatient      package         available.  treatment, which
                            mental health                 setting, not                                VA intends to
                            inpatient or                  just those with                             accomplish in FY
                            outpatient setting            SMI                                         1999.
                            in FY 1998
                                                          The majority of
                                                          substance abuse
                                                          care is now
                                                          delivered in an
                                                          outpatient
                            Veterans seen in              setting. This
                            any substance                 measure includesPatient         FY 96--38%
 SMI--substance             abuse outpatient              only the 28% of Treatment File
 abuse         Access       clinic within 30   Process    veterans treatedand outpatient  FY 97--41%
                            days after                    for substance   files
                            discharge                     abuse in an                     FY 98--41%
                                                          inpatient
                                                          setting and
                                                          includes more
                                                          than SMI
                                                          patients.
                                                          The majority of
                                                          substance abuse
                                                          care is now
                                                          delivered in an
                                                          outpatient                      FY 96--27
                            Average time to               setting. This                   days
                            first outpatient              measure includesPatient
 SMI--substance             visit for veterans            only the 28% of Treatment File  FY 97--28
 abuse         Access       discharged from    Process    veterans treatedand outpatient  days
                            substance abuse               for substance   files
                            programs                      abuse in an                     FY 98--27
                                                          inpatient                       days
                                                          setting and
                                                          includes more
                                                          than SMI
                                                          patients.
                                                          Covers veterans
                            Patients                      receiving
                            (percentage) that             inpatient or
                            demonstrate                   outpatient care
 SMI--substance             improvement in the            and would       Mental health   Data not
 abuse         Performance  drug and alcohol   Outcome    potentially     package         available.
                            use scores in the             include more
                            Addiction Severity            than SMI
                            Index                         patients in the
                                                          sample
                                                                                                      Because of the
                                                                                                      closure of
                            VA had not                                                                substance abuse
                            developed an                                                              and general
                            access indicator              Not determined  Not determined              psychiatry beds,
                                                          at the time the at the time the             VA did not
 SMI--homeless Access       for the homeless   Process    FY 1998 capacityFY 1998         Data not    include access
                            subcategory of SMI                                            available.
                            at the time the FY            report was      capacity report             data for these
                            1998 report was               published       was published               patients and may
                            published.                                                                not include them
                                                                                                      in the future
                                                                                                      because of small
                                                                                                      sample sizes.
                                                          May include
                                                          veterans without
                            Veterans who                  SMI when looking
                            acquired living               at homeless
                            arrangements at               population
                            discharge from a              discharged from
                            Domiciliary Care              a Domiciliary                   FY 96--51%
                            for Homeless                  Care for        Northeast                   Sample size was
 SMI--homeless Performance  Veterans program   Outcome    Homeless        Program         FY 97--52%  not provided in
                            or Health Care for            Veterans programEvaluation                  report.
                            Homeless Veteran              or Health Care  Center files    FY 98--52%
                            community-based               for Homeless
                            contract                      Veteran
                            residential care              community-based
                            program                       contract
                                                          residential care
                                                          program
                                                          May include
                                                          veterans without
                            Veterans who                  SMI if looking
                            obtained                      at homeless
                            employment at                 population
                            discharge from a              discharged from                 FY 96--49%
                            Domiciliary Care              a Domiciliary   Northeast                   Sample size was
 SMI--homeless Performance  for Homeless       Outcome    Care for        Program         FY 97--52%  not provided in
                            Veterans program              Homeless        Evaluation                  report.
                            or community-based            Veterans programCenter files    FY 98--54%
                            contract                      or
                            residential care              community-based
                            program                       contract
                                                          residential care
                                                          program
                            Veterans with                 May include
                            mental illness                veterans without
                            (including                    SMI if looking
                            substance abuse)              at homeless
                            who had a                     population
                            follow-up mental              discharged from 1999 data to be
                            health outpatient             a contract      collected and
                                                                          summarized by
 SMI--homeless Performance  visit within 30    Process    Domiciliary Carethe Northeast   Data not
                            days of discharge             for Homeless                    available.
                            from a contract               Veterans programProgram
                            Domiciliary Care              or              Evaluation
                            for Homeless                  community-based Center.
                            Veterans program              residential care
                            or community-based            program
                            residential care
                            program
                                                          Covers PTSD
                            Veterans with a               population who
                            primary diagnosis             received                        FY 96--64%
                            of PTSD seen in               inpatient care, Patient
 PTSD          Access       any psychiatric    Process    which accounts  Treatment File  FY 97--65%
                            outpatient clinic             for             and outpatient
                            within 30 days                approximately   files           FY 98--68%
                            after discharge               11% of the
                                                          population
                                                          Covers PTSD
                                                                                          FY 96--30
                            Average time to               population who                  days
                            first outpatient              received        Patient
                                                          inpatient care,
 PTSD          Access       visit for veterans Process    which accounts  Treatment File  FY 97--28
                            discharged with a                             and outpatient  days
                            primary PTSD                  for             files
                            diagnosis                     approximately                   FY 98--26
                                                          11% of the
                                                          population                      days
                            Veterans
                            (percentage)
                            treated for PTSD                                                          VA is using
                            in specialized                Covers PTSD                                 functional scores
                            PTSD programs with            population                                  to determine the
 PTSD          Performance  at least one       Process    treated in both Mental health   Data not    number of
                            Global Assessment             inpatient and   package         available.  veterans disabled
                            of Functioning                outpatient                                  by PTSD for FY
                            score                         programs                                    1998.

                            Change in PTSD
                            symptoms on the                                               FY
                            short form of the             Covers PTSD     The Northeast   96--39.05
                            Mississippi Scale             population      Program
                                                                          Evaluation      FY
 PTSD          Performance  from admission to  Outcome    admitted to a   Center is using 97--37.19
                            follow-up 4 months            specialized
                            after discharge               intensive PTSD  a
                            (national average             program         self-reporting  % change:
                            for adjusted mean                             survey.
                            scores)                                                       -1.86
                            Change in alcohol
                            abuse symptoms as
                            measured by the                                               FY
                            Alcohol Abuse                                 The Northeast   96--0.169
                            Composite of the              Covers PTSD     Program
                            Addiction Severity            population      Evaluation      FY
 PTSD          Performance  Index from         Outcome    admitted to a   Center is using 97--0.137
                            admission to                  specialized     a
                            follow-up 4 months            intensive PTSD  self-reporting  % change:
                            after discharge               program         survey.
                            (national average                                             -0.032
                            for adjusted mean
                            scores)
                            Change in drug
                            abuse symptoms as
                            measured by the                                               FY
                            Drug Abuse                                    The Northeast   96--0.071
                            Composite of the              Covers PTSD     Program
                            Addiction Severity            population      Evaluation      FY
 PTSD          Performance  Index from         Outcome    admitted to a   Center is using 97--0.059
                            admission to                  specialized     a
                            follow-up 4 months            intensive PTSD  self-reporting  % change:
                            after discharge               program         survey.
                            (national average                                             -0.012
                            for adjusted mean
                            scores)
                            Change in
                            occupational
                            functioning as
                            measured by the
                            number of days the            Covers PTSD     The Northeast
                            veteran has been              population      Program
                                                                          Evaluation
 PTSD          Performance  employed during    Outcome    admitted to a   Center is using Data not
                            the past 30 days              specialized                     available.
                            at admission and              intensive PTSD  a
                            follow-up 4 months            program         self-reporting
                            after discharge                               survey.
                            (national average
                            for adjusted mean
                            scores)

                            Veterans                                                      FY
                            successfully       Process    Covers PTSD     The Northeast   96--62.7%
                            contacted for      (step to   population      Program
 PTSD          Performance  outcome assessment collect    admitted to a   Evaluation      FY
                            after discharge    data for   specialized     Center's        97--66.6%
                            from an intensive  an outcome intensive PTSD  response rate
                            PTSD program       measure)   program         for survey      % change:
                                                                                          +3.9

aA measure that focuses on a process that leads to a certain outcome and
that, when executed well, will increase the probability of achieving a
desired outcome.

bA measure that indicates the result of the performance or nonperformance of
a function(s) or process(es).

Comments From the Department of Veterans Affairs

(406174)

Table 1: Change in Number of Special Disability Veterans Served and Dollars
Spent Between Fiscal Years 1996 and 1998 8

Table 2: Change in Number of Special Disability Veterans Served in Inpatient
and Outpatient Settings Between Fiscal Years 1996
and 1998 9

Table 3: Change in FTE Employees and Beds for Spinal Cord
Dysfunction and Blind Rehabilitation Between Fiscal Years
1996 and 1998 10
  

1. Stakeholders included members of the Advisory Committee on Prosthetics
and Special Disabilities Programs and the Committee on the Care of Severely
Chronically Mentally Ill Veterans.

2. We provided preliminary information on the results of our review in a
briefing to your staff on May 24, 1999.

3. Nevertheless, VA's fiscal year 2000 budget appropriation for health care
was increased by $1.7 billion over the fiscal year 1999 budget
appropriation.

4. The members of the Advisory Committee on Prosthetics and Special
Disabilities Programs are from veterans' service organizations,
universities, and private sector health care providers. In accordance with
the Veterans Health Care Eligibility Reform Act, members of the Committee on
the Care of Severely Chronically Mentally Ill Veterans must be employees of
the Veterans Health Administration with expertise in the care of the
chronically mentally ill and must be appointed by VA's Under Secretary for
Health.

5. VA has begun implementation of a Decision Support System that will
capture diagnostic information for outpatient clinic visits and will provide
cost accounting information.

6. Other groups, such as the minority staff of the Senate Committee on
Veterans' Affairs, were also unable to verify workload and resource data
because of problems in obtaining reliable or comparable data across
facilities. See Minority Staff Review of VA Programs for Veterans With
Special Needs , prepared for Senator John D. Rockefeller IV, July 27, 1999.

7. VA Office of Inspector General, Evaluation of Medical Center Investment
in Ambulatory Care Infrastructure , Report Number 9AY-A19-078 (Washington,
D.C.: VA, Mar. 31, 1999).

8. A service-line model is a health care organizational model based upon
providing a comprehensive set of clinical and administrative services to
meet the needs of a particular segment of the market (for example, veterans
with mental illness or spinal cord dysfunction). Budgetary, personnel, and
reporting authorities vary in the different service-line models.

9. Hoptels are temporary lodging, usually within facilities, that provide a
cost-effective alternative to inpatient admissions.

10. VA refers to this as the error detection and correction process in its
fiscal year 1998 capacity report.
*** End of document. ***