VA Health Care: Experiences in Denver and Charleston Offer
Lessons for Future Partnerships with Medical Affiliates
(28-APR-06, GAO-06-472).
The Department of Veterans Affairs (VA) maintains affiliations
with medical schools, including the Medical University of South
Carolina (MUSC) and the University of Colorado at Denver and
Health Services Center and University of Colorado Hospital (UCH),
to obtain enhanced medical care for veterans. As part of their
plans for new medical campuses, both UCH and MUSC proposed
jointly constructing and operating new medical facilities with VA
in Denver and Charleston, respectively. This report discusses (1)
how VA evaluated the joint venture proposals for Denver and
Charleston and the status of these proposals, (2) the challenges
these proposals pose for VA, and (3) the lessons VA can learn
from its experiences in Charleston and Denver for future
partnerships.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-06-472
ACCNO: A52725
TITLE: VA Health Care: Experiences in Denver and Charleston
Offer Lessons for Future Partnerships with Medical Affiliates
DATE: 04/28/2006
SUBJECT: Evaluation criteria
Health care facilities
Hospital planning
Joint ventures
Lessons learned
Medical schools
Strategic planning
Veterans hospitals
Veterans' medical care
Charleston (SC)
Denver (CO)
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO 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. **
** **
******************************************************************
GAO-06-472
* Report to the Chairman, Committee on Veterans' Affairs, House of
Representatives
* April 2006
* VA HEALTH CARE
* Experiences in Denver and Charleston Offer Lessons for Future
Partnerships with Medical Affiliates
* Contents
* Results in Brief
* Background
* VA Lacks Departmental Criteria to Evaluate Joint Venture
Proposals
* VA Does Not Have Criteria at the Departmental Level to
Evaluate Joint Venture Proposals Consistently
* VA Decided to Construct Stand-alone Facility in Denver, but
Negotiations about Location Continue
* Lack of Leadership Buy-in and Miscommunication
Prolonged Negotiations and Created Atmosphere of
Mistrust
* VA and MUSC Have Identified Multiple Options for Sharing
Resources and Space, but No Decision Has Been Made
* Limited Collaboration and Communication Stalled
Negotiations in Charleston for About 3 Years
* Joint Venture Proposals in Charleston and Denver Pose Multiple
Challenges
* VA's Current Sharing Arrangements with Medical Affiliates
and DOD Could Be Instructive As VA Considers Current and
Future Joint Ventures
* VA Could Learn Several Lessons from Its Experiences with Denver
and Charleston Joint Venture Proposals
* Lack of Criteria at Departmental Level Results in
Inconsistent Evaluations
* Lack of Communications Strategy Leads to Misinformation and
Confusion
* Lack of Leadership Support Can Hinder Negotiations
* Lack of Extensive Collaboration Hampers Negotiations
* Conclusions
* Recommendations for Executive Action
* Agency Comments
* Scope and Methodology
* Description of Planning Models Identified by Steering Group
* Related GAO Products
Report to the Chairman, Committee on Veterans' Affairs, House of
Representatives
April 2006
VA HEALTH CARE
Experiences in Denver and Charleston Offer Lessons for Future Partnerships
with Medical Affiliates
Contents
Tables
Figures
April 28, 2006Letter
The Honorable Steve Buyer Chairman Committee on Veterans' Affairs House of
Representatives
Dear Mr. Chairman:
For decades, the Department of Veterans Affairs (VA) has maintained
partnerships, or affiliations, with university medical schools to obtain
medical services for veterans and provide training and education to
medical residents. These affiliations help VA fulfill its mission of
providing health care to the nation's veterans. Today, VA has affiliations
with 107 medical schools, including the Medical University of South
Carolina (MUSC) in Charleston, South Carolina, and the University of
Colorado's School of Medicine-through the University of Colorado at Denver
and Health Sciences Center and the University of Colorado Hospital
(UCH)-in the Denver, Colorado, area.1 For example, many MUSC physicians
serve as residents at VA's medical facility in Charleston, the Ralph H.
Johnson VA Medical Center, and UCH physicians do the same at VA's Eastern
Colorado Health System in Denver. As part of their plans for new medical
campuses, UCH and MUSC proposed jointly constructing and operating new
medical facilities with VA in Denver and Charleston, respectively.
Although VA has a long history of partnering with its medical affiliates
for training and education as well as purchasing medical services from its
medical affiliates, this type of joint venture would represent a departure
from VA's typical relationship with its medical affiliates.
In addition to partnering with university medical schools, VA manages a
diverse inventory of real property to provide health care to veterans.
However, many of VA's facilities were built more than 50 years ago and are
no longer well suited to providing accessible, high-quality,
cost-effective health care in the 21st century. To address its aging
infrastructure, VA initiated the Capital Asset Realignment for Enhanced
Services (CARES) process-the most comprehensive, long-range assessment of
its health care system's capital asset requirements ever undertaken by VA.
In February 2004, the CARES Commission-an independent body charged with
assessing VA's capital assets-issued its recommendations regarding the
realignment and modernization of VA's capital assets necessary to meet the
demand for veterans' health care services through 2022. In Denver, the
commission recommended VA build a replacement medical center with the
Department of Defense (DOD) on the former Fitzsimons Army Base in Aurora,
Colorado, which is just outside of Denver. The commission also noted
widespread agreement among stakeholders that the new VA facility should be
located near a new UCH facility, which was being constructed at the
Fitzsimons site. In Charleston, the commission did not recommend replacing
the current medical facility. However, the commission recommended that,
among other things, VA promptly evaluate MUSC's proposal to jointly
construct and operate a new medical center with VA in Charleston, noting
that such a joint venture arrangement could serve as a possible framework
for partnerships in the future. In responding to the commission's
recommendations in May 2004, the Secretary stated that VA would build a
replacement VA medical center on the Fitzsimons site with "some shared
facilities with the UCH" and continue to "consider options for sharing
opportunities with MUSC."2
This report discusses (1) how VA evaluated the joint venture proposals
involving its facilities in Charleston and Denver and the status of these
proposals; (2) the challenges that VA faces in sharing facilities and
services with its medical university affiliates in Charleston and Denver;
and (3) the lessons VA can learn, if any, from its experiences in
Charleston and Denver if it moves forward with other partnerships. To
address these objectives, we analyzed agency documents and interviewed
officials at VA, MUSC, UCH, and the University of Colorado at Denver and
Health Sciences Center. We also met with local stakeholders, such as
officials from the Fitzsimons Redevelopment Authority (FRA) in Aurora, the
mayors of Charleston and Aurora, and representatives from the VA employee
union in each location to obtain their perspectives on the joint venture
proposals and to obtain information on local capital asset planning and
its impact. We also discussed the CARES review process and CARES
Commission recommendations with VA officials. We assessed the reliability
of the information obtained from VA, MUSC, and UCH. We concluded that the
information was sufficiently reliable for our purposes. We also reviewed
GAO's body of work on VA's management of its capital assets and on leading
practices for realigning federal agency infrastructure, collaboration
among organizations, and organizational transformations.3 In September
2005, we also testified before the Subcommittee on Health, Committee on
Veterans' Affairs, House of Representatives, on the joint venture proposal
involving VA's facility in Charleston and MUSC.4 Although we examined the
joint venture proposals for VA's Denver and Charleston facilities and the
associated studies and planning documents, we did not evaluate the merits
of the proposals. We conducted our work from June 2005 through February
2006 in accordance with generally accepted government auditing standards.
Results in Brief
Because it lacks criteria at the departmental level, VA evaluated
proposals for joint ventures with local medical affiliates in Denver and
Charleston using criteria developed specifically for each location, and
while VA opted to build a stand-alone facility in Denver, it is still in
the process of considering a joint venture in Charleston. In both
locations, multiple iterations of the joint venture proposals have been
considered, and negotiations between VA and its medical affiliates have
stretched over a number of years, delaying decisions regarding the
facilities in Denver and Charleston. For example, in Denver, VA officials
at the facility and network level5 and UCH officials in 1999 began
informally discussing the possibility of constructing and operating a
joint facility on the former Fitzsimons army base in Aurora, Colorado,
which was closed as part of DOD's base realignment and closure process.
Negotiations over different aspects and revisions of the joint venture
proposal continued until late 2004, at which time VA decided against a
joint facility with UCH. Similarly, negotiations over the joint venture
proposed between VA and MUSC in Charleston began in 2002, and, to date, no
decisions have been made. However, a steering group composed of VA and
MUSC officials issued a report in December 2005 that outlined options for
a joint venture ranging from sharing of medical services, which could
occur with VA maintaining its existing building, to a large, new building
with space for both VA and MUSC. Complex arrangements such as the joint
venture proposals in Denver and Charleston require extensive negotiations
between the potential partners. However, negotiations in both locations
were hampered by limited communication and collaboration and lack of VA
leadership support, among other things. For example, in Charleston, for a
2-year period after the joint venture was proposed, there was little
communication between VA and MUSC, which caused negotiations to stall
during this period. In Denver, top VA leadership was not fully supportive
of exploring a joint venture with UCH, resulting in delays in negotiations
and misunderstandings between VA and UCH.
Joint ventures proposed by VA's medical university affiliates in Denver
and Charleston present several challenges to VA. These challenges include
addressing institutional changes for VA and institutional differences
between VA and its medical affiliates, identifying legal issues and
seeking legislative remedies, and balancing funding priorities. For
example, capital expenditures for a joint venture would have to be
considered in the context of other VA capital priorities, and VA would
have to ensure that a joint venture would allow VA to fulfill its other
departmental missions, such as supporting national, state, and local
emergency management. In addition, a joint venture would also require VA
to depart from its traditional health care model of providing medical
services in house and adopt one that includes sharing capital assets with
an affiliate. Although addressing these issues will be difficult, the
VA-MUSC steering group has begun to work through some of these challenges.
Its experiences in Denver and Charleston offer several lessons for VA as
it considers other similar opportunities. One of the most important
lessons to emerge from VA's experience with the joint venture proposals in
Denver and Charleston is the need to develop criteria at the departmental
level to evaluate the merits of joint venture proposals on a consistent
basis. A set of criteria for evaluating decisions regarding
infrastructure, including joint ventures, would enhance the transparency
of these decisions and help ensure that the decisions were made in a
manner that was fair to joint venture partners and other stakeholders,
such as veterans and employees. The lack of departmental-level criteria
forced VA to evaluate the proposals without a consistent framework that
would allow VA to determine and assess the effects of each proposal on
medical care cost and quality within the context of its overall mission.
Another important lesson of VA's experience in Denver and Charleston is
the need for a communications strategy for communicating with its medical
affiliate and stakeholders. Such a strategy could help facilitate
negotiations with the medical affiliate as well as help address concerns
voiced by veterans and employees, such as the impact of the joint venture
on patient care. A communication strategy would help ensure that these
groups receive a message that is consistent in tone and content. The lack
of such a strategy contributed to breakdowns in communications in both
Denver and Charleston during key points of the negotiations and hindered
progress. For example, in Charleston, there was limited communication
between VA and MUSC for about 2 years; as a result, negotiations stalled.
Other lessons that VA could take away from its experiences in Denver and
Charleston include that negotiations are facilitated by VA leadership
support for exploring the possibility of joint ventures and extensive
collaboration between the potential joint venture partners.
To better position VA to consider future joint venture proposals with
medical affiliates, we are recommending that the Secretary develop
criteria at the departmental level for evaluating joint venture proposals
on a consistent basis and a strategy for communicating with the
department's affiliates and stakeholders about joint venture proposals. VA
reviewed a draft of this report and agreed with the report's conclusions
and recommendations.
Background
VA manages a large health system for veterans, providing health care
services to over 5 million beneficiaries. The cost of these services in
fiscal year 2005 was over $30 billion. According to VA, its health care
system now includes 157 medical centers, 862 ambulatory care and
community-based outpatient clinics, and 134 nursing homes. VA health care
facilities provide a broad spectrum of medical, surgical, and
rehabilitative care. The management of VA's facilities is decentralized to
21 regional networks referred to as Veterans Integrated Service Networks
(networks). The Charleston facility is part of Network 7, or the Southeast
Network, and the Denver facility is located in Network 19, or the Rocky
Mountain Network.6
To meet its mission of serving the needs of the nation's veterans, VA
partners with medical universities and DOD. In 1946, VA established a
program to enter into partnerships with medical universities, now referred
to as academic affiliations, to provide high quality health care to
America's veterans and to train new health professionals. Today, VA
maintains affiliations with 107 of the nation's 126 medical schools. In
addition to academic affiliation agreements, VA purchases clinical
services from medical schools for the treatment of veterans.7 Similarly,
in 1982, the VA and DOD Health Resources Sharing and Emergency Operations
Act (Sharing Act)8 was enacted to provide for more efficient use of
medical resources through greater interagency sharing and coordination.
For example, the VA Medical Center in Louisville and the Ireland Army
Community Hospital in Fort Knox, Kentucky, have engaged in sharing
activities to provide services to beneficiaries that include primary care,
acute care pharmacy, ambulatory, blood bank, intensive care, pathology and
laboratory, audiology, podiatry, urology, internal medicine, and
ophthalmology. Given the importance of these partnerships to VA's ability
to meet its mission, VA's 2003-2008 strategic plan includes goals for
sustaining partnerships with medical universities and sharing resources
with DOD.
VA's Denver and Charleston medical facilities have long-standing
affiliations with local medical universities. VA's facility in Denver is
affiliated with the University of Colorado's School of Medicine-through
the University of Colorado at Denver and Health Sciences Center and
UCH-and VA's facility in Charleston is affiliated with MUSC. These
affiliations provide both VA facilities with the majority of VA's medical
residents who rotate through all VA clinical service areas. Both VA
facilities also purchase a significant amount of medical services from
their affiliates. Specifically, the Denver facility annually obtains $9
million worth of services from UCH, and the Charleston VA facility buys
$13 million in services annually from MUSC. The medical services purchased
are in such areas as gastroenterology, infectious disease, internal
medicine, neurosurgery, anesthesia, pulmonary, and cardiovascular
perfusion. In addition to these services, VA also has a medical research
partnership with MUSC for a mutually supported biomedical research
facility, the Thurmond Biomedical Research Center. Table 1 provides more
detailed information about the VA facilities in Charleston and Denver and
their medical affiliates.
Table 1: Information on VA's Facilities in Charleston and Denver and
MUSC's and UCH's Facilities
Charleston, Denver,
South Carolina Colorado
VA's Ralph H. MUSC VA's UCH
Johnson Eastern
Medical Center Colorado
Health
System
Year built: Main 1966 1955 1951 1921
hospital
Square footage 352,000 1,530,125a 670,000 1,200,000b
Beds 126 709 128 420
Inpatient 4,510 28,591 4,750 19,622
admissions
Outpatient visits 370,917 551,914c 481,769 546,248
Annual operating $160 $559 $267 $490
budget (in
millions)
Source: GAO presentation of VA, MUSC, and UCH data.
aThe MUSC medical center includes the Institute of Psychiatry (62,299
square feet), Children's Hospital (347,697), North Tower and Main Hospital
(545,201), Rutledge Tower (383,752), and Charleston Memorial Hospital
(191,176).
bBy 2008, UCH will have approximately 1.8 million gross square footage of
fully operational facilities, with about 1.5 million gross square footage
dedicated to clinical and patient care services. This number does not
reflect gross square footage at UCH's current facility in Denver.
cThis number does not include same-day surgeries (6,802) or emergency
visits (35,375).
VA Lacks Departmental Criteria to Evaluate Joint Venture Proposals
VA evaluated the joint venture proposals for its facilities in Denver and
Charleston on an ad hoc basis because it lacks criteria at the
departmental level to evaluate such proposals consistently. VA has decided
against a joint facility in Denver, but it is still in the process of
considering such a facility in Charleston. In both locations, multiple
iterations of the joint venture proposals have been considered, and
negotiations between VA and its medical affiliates have stretched over a
number of years. Negotiations in both locations were hampered by limited
communication and collaboration, a lack of top VA leadership support for
the proposals, and no single VA point of contact for the medical
affiliates.
VA Does Not Have Criteria at the Departmental Level to Evaluate Joint
Venture Proposals Consistently
VA does not have criteria at the departmental level that could be used to
evaluate joint venture proposals on a consistent basis. Consequently, VA
officials identified factors for considering the specific joint venture
proposals in Charleston and Denver. Some of the identified factors were
consistent between the two locations, and others were site-specific, but
it is not clear how any of the factors weighed in VA's consideration of
the proposals.
In studies and correspondence regarding the joint venture proposal in
Denver, VA officials identified several factors that they believed to be
important in considering the joint venture proposal. In particular, in
correspondence between VA and UCH in 2002, and again in 2004, the
Secretary of VA identified four major considerations-(1) maintaining VA's
identity; (2) maintaining VA's governance; (3) balancing and evaluating
priorities within VA's capital asset program, including the CARES process;
and (4) securing funding. In 2002, a VA taskforce composed of
headquarters, network, and facility officials examined the potential for a
VA-UCH joint facility and identified additional factors critical to the
decision-making process. These factors included maintaining VA's
commitment to providing health care to meet veterans' unique needs and
research programs, VA's aging infrastructure, and the gap between health
care demand and capacity and funding. Another consideration that arose
through the course of negotiations was VA's space requirements for a new
facility and the associated acreage of land needed and available on the
Fitzsimons campus. VA did not indicate how the factors identified in the
studies or correspondence weighed in its decision making regarding the
joint venture proposal.
Similarly, the VA-MUSC steering group identified a set of criteria to help
identify and analyze the joint venture proposal for Charleston. As shown
in table 2, these criteria include enhanced quality and service and
financial viability. The steering group's report did not indicate how or
why these criteria were chosen, provide an explanation of the individual
criterion, or indicate the relative importance of the criteria. While the
steering group used the criteria in identifying and evaluating options for
further consideration, it is not clear how, if at all, these criteria will
be used by VA leadership in making a final decision on the joint venture
proposal. In meetings with VA officials about the joint venture proposal
in Charleston, officials identified other considerations that could
influence the decision-making process, including the condition of the
existing VA facility and the need to balance investment priorities across
the region and nation.
Table 2: Criteria Identified by the VA-MUSC Steering Group
Criterion
o Enhance quality and service
o Improve access
o Financial viability
o Optimal legal authorities
o Enhance efficient infrastructure sharing
o Maximize land utilization and development
o Collaborative governance structure
o Maintain unique VA identity
o Become a regional center of excellence
o Enhance Department of Defense services
o Produce serendipitous win-wins
o Serve as a national model for collaborations
Source: Collaborative Opportunities Steering Group, Final Report, December
2005.
VA Decided to Construct Stand-alone Facility in Denver, but Negotiations
about Location Continue
VA has decided against a fully-integrated facility with UCH in Denver.
Negotiations between VA and the University of Colorado at Denver and
Health Services Center and UCH stretched over a number of years, and a
number of different options were considered. The lack of leadership buy-in
and miscommunication about VA's intentions regarding the future Denver
facility prolonged negotiations and created an atmosphere of mistrust
between the parties. Figure 1 provides a time line of key events in the
negotiations between VA and UCH.
Figure 1: Time Line of Key Events in the Negotiations between VA and UCH
In 1995, the University of Colorado decided to relocate its Health
Sciences Center campus, including its affiliated UCH, from downtown Denver
to the former Fitzsimons Army Medical Base located in nearby Aurora,
Colorado, which was closed as part of DOD's base realignment and closure
process. UCH determined that its facility in downtown Denver lacked the
space to accommodate its patient population and that there was little room
for expansion. The availability of land at the Fitzsimons site offered an
opportunity for UCH to move and expand the size of its campus. When
Fitzsimons closed, DOD turned a portion of the 577 acres the base occupied
over to the U.S. Department of Education so that it could convey land to
public educational institutions.9 The University of Colorado applied for
and received 227 acres from the U.S. Department of Education, and the
University leased about 55 acres to UCH for its new inpatient and
outpatient pavilions. The majority of the land at Fitzsimons-about 332
acres-was purchased by FRA for $1.85 million.10 FRA plans to develop a
biomedical research park on this land. The remaining land at Fitzsimons is
owned by the City of Aurora, a private entity, and a nonprofit
organization.
In late 1999, VA officials at the facility and network level and UCH
officials began to informally discuss the possibility of relocating VA's
Denver medical center to the Fitzsimons campus. UCH and VA officials were
concerned that UCH's move to Fitzsimons, about 6 miles from its downtown
Denver location, would strain their affiliation because of the amount of
time it would take doctors to travel between the facilities. The UCH
president also suggested that colocating the UCH and VA medical center at
Fitzsimons could achieve cost efficiencies through integrating inpatient
activities, such as medical and surgical specialty labs, and sharing some
patient treatment. In considering a possible joint venture, facility and
network VA officials worked with UCH officials to examine options for
moving VA's medical center to the Fitzsimons campus as well as sharing
services and facilities with UCH. In particular, these officials jointly
funded a study to determine the feasibility and cost of different options,
including constructing free-standing facilities with limited sharing to
jointly constructing and operating a new fully-integrated facility at
Fitzsimons. The study, completed in 2001, concluded that a fully
integrated, or joint, facility was the most cost-effective option. A
second study commissioned by VA's Network 19 in 2002 also analyzed a range
of options, including a joint VA-UCH facility; but this study did not
recommend which option to pursue. These studies were shared with VA's
central office, veteran service organizations, and the Congress, and
became the basis of the joint venture proposal and negotiations.
The Secretary of VA established a task force to examine the joint venture
proposal to integrate the Denver medical center and UCH on the Fitzsimons
campus in July 2002. The task force was composed of VA officials at the
departmental, network, and facility levels. In September 2002, the task
force issued a draft report, which examined seven alternatives-ranging
from maintaining the status quo to constructing a fully integrated
facility with UCH. The task force's report presented advantages and
disadvantages of each alternative. It did not recommend which alternative
to pursue.
In September 2002, the president of UCH sent a letter to the Secretary of
VA asking that VA make a decision within 1 year regarding moving the VA
facility to the Fitzsimons campus. In October 2002, the Secretary
responded that VA could not commit to a joint UCH-VA hospital within that
time frame. The Secretary indicated that a number of important questions
remained unanswered, including how the joint hospital would be governed.
Furthermore, he noted that the proposal to relocate the Denver medical
center to Fitzsimons had to be evaluated in the context of the CARES
Commission report, which was not scheduled to be completed until the
following year. The Secretary's response effectively ended discussions
about constructing and operating a fully-integrated facility with UCH.
In January 2003, VA began developing a proposal for a joint VA-DOD
facility on the Fitzsimons campus. Specifically, the proposed joint
federal facility would house VA and DOD, and the two entities would share
some medical services and equipment. The joint VA-DOD facility, which was
referred to as the federal tower, would be built on UCH-leased land at
Fitzsimons and would be connected to UCH's inpatient pavilion by a 2-story
clinical facility. (See fig. 2.) The clinical facility would house
operating rooms, imaging, and pathology laboratories, among other things,
that would be shared by VA, UCH, and DOD. With this concept in hand, VA,
UCH, and DOD began discussions about the availability of land adjacent to
the UCH inpatient pavilion for the federal tower. In August 2004, the UCH
president estimated that 18 acres of land was available adjacent to the
UCH facility for the federal tower. However, soon thereafter, a survey of
the land indicated that approximately 12 acres were available for the
federal tower once easements and setbacks were taken into account.
Figure 2: Federal Tower Proposal at Fitzsimons
In December 2004, in a letter to UCH, the Secretary stated that the
approximate 12 acres would be insufficient to meet VA's space requirements
for a new medical center. Specifically, the Secretary stated that
predesign planning for the new facility revealed that VA needed
approximately 1.46 million square feet to meet the specialized needs of
veterans and DOD patients. To accommodate these space requirements, VA's
architectural firm outlined three design options-ranging from a 6-story VA
hospital on 38 acres to a 8- to 10-story VA hospital on 20 acres. Based on
this analysis and other considerations, the Secretary concluded that VA
needed about 38 acres on the Fitzsimons campus for the joint VA-DOD
facility. This decision ended negotiations over building the federal tower
on UCH-leased land and connecting it to UCH's inpatient pavilion with a
clinical facility. UCH subsequently decided to use the land adjacent to
the inpatient pavilion for other purposes.
After land negotiations with UCH ended, VA officials began looking for a
new location on the Fitzsimons site for a stand-alone VA medical center.
The conference report accompanying VA's appropriation act for fiscal year
2004 directed VA to continue efforts to "co-locate the Denver VA medical
center with ... [UCH] at the Fitzsimons campus."11 While there is no
statutory requirement to locate the VA medical center at Fitzsimons, VA
considers this language in the conference report to express the will of
Congress and, as a result, has gone forward with efforts to purchase
property from FRA for such a purpose. In July 2005, VA signed a memorandum
of understanding with FRA to set forth the conditions under which VA and
FRA will proceed with discussions that may lead to the purchase and
conveyance of about 40 acres located on the southeast corner of the
Fitzsimons campus. (See fig. 3.) According to FRA officials, this piece of
land is currently owned by FRA and three other entities. According to a VA
official, in February 2006, VA offered FRA $16.50 per square foot for the
FRA-owned portion of the land. (VA is in the process of surveying the land
to determine the total square footage.) The VA official responsible for
the land negotiations at Fitzsimons told us that VA's offer is valid for 6
months, and that VA expects to finalize the purchase of the FRA-owned
portion of the land by the end of this fiscal year. VA is currently
negotiating with the other three land-holding entities about the purchase
of their land. According to the VA official, he does not foresee any "show
stoppers" in the negotiations with these three entities, and therefore VA
expects to reach agreement with these entities in the coming months.
Figure 3: Location of Land That VA Is Currently Pursuing at Fitzsimons in
Relation to UCH facilities
Lack of Leadership Buy-in and Miscommunication Prolonged Negotiations and
Created Atmosphere of Mistrust
Negotiations between VA and UCH on the different joint venture proposals
were hampered by a lack of VA leadership buy-in and miscommunication. For
instance, although VA officials at the facility and network levels worked
with UCH officials in developing the joint facility proposal, the current
network director told us that the Secretary was never fully supportive of
this concept. Rather, according to the network director, the Secretary
envisioned a stand-alone facility adjacent to the UCH complex. When VA
decided to pursue a stand-alone facility, UCH officials said they felt as
though they had been misled by VA officials, including the Secretary,
about VA's interest in a joint facility. Further, in a correspondence from
the UCH president to VA in 2004, the UCH president noted that a
freestanding VA medical center on the Fitzsimons campus was never
discussed. UCH officials also told us that at no time did UCH ever
consider a freestanding facility for VA on its new campus because there
would be limited opportunities for sharing capital and operating costs. In
addition, there was miscommunication about the amount of land available
for a federal tower and VA's space requirements. Specifically, UCH
officials told VA officials that there were about 18 acres available for
the federal tower; however, the survey revealed that only a little more
than 12 acres were available. In addition, in December 2004, the Secretary
of VA informed UCH that VA needed 1.46 million square feet for its new
facility. According to UCH officials, these space requirements ran counter
to estimates that were discussed with VA facility and network level
officials and, according to the UCH president in 2004, would result in a
facility that was about 50 percent larger than the existing VA medical
center in Denver.12 These events contributed to an atmosphere of mistrust
between VA and UCH.
VA and MUSC Have Identified Multiple Options for Sharing Resources and
Space, but No Decision Has Been Made
VA has not made a decision regarding a joint venture with MUSC.
Negotiations between VA and MUSC have stretched over a number of years and
have been hampered by limited collaboration and communication among the
parties. VA's Under Secretary for Health and the president of MUSC are
currently considering the results of a recent report that identifies and
analyzes options for sharing facilities and space in Charleston. Figure 4
provides a time line of key events in the negotiations between VA and UCH.
Figure 4: Time Line of Key Events in the Negotiations between VA and MUSC
In November 2002, the president of MUSC sent a proposal to the Secretary
of VA about partnering with MUSC in the construction and operation of a
new medical center in phase II of MUSC's construction project. Under
MUSC's proposal, VA would vacate its current facility and move to a new
facility located on MUSC property. MUSC also indicated that sharing
medical services would be a component of the joint venture. Although VA
and MUSC currently share some services, the joint venture proposal,
according to MUSC officials, would have increased the level of sharing of
medical services and equipment, thereby creating cost savings for both VA
and MUSC.
To meet the needs of a growing and aging patient population, MUSC has
undertaken a multiphase construction project to replace its aging medical
campus. Construction on the first phase began in October 2004. Phase I
includes the development of a 4-story diagnostic and treatment building
and a 7-story patient hospitality tower, providing an additional 641,000
square feet in clinical and support space. Phase I also includes the
construction of an atrium connecting the two buildings, a parking
structure, and a central energy plant. Initial plans for phases II through
V include diagnostic and treatment space and patient bed towers. According
to MUSC officials, as of September 2005, there are approximately 24 months
remaining for the planning of phase II. As shown in figure 5, phases IV
and V would be built on VA property. In particular, phase V would be built
on the site of VA's existing medical center.
Figure 5: MUSC's Construction Plan, 2002
Note: The circle highlights some of VA's existing property.
In response to MUSC's proposal, VA formed an internal workgroup composed
of officials primarily from VA's Network 7 to evaluate MUSC's proposal.
The workgroup analyzed the feasibility and cost-effectiveness of the
proposal and issued a report in March 2003, which outlined three other
options available to VA: replacing the Charleston facility at its present
location, replacing the Charleston facility on land presently occupied by
the Naval Hospital in Charleston, or renovating the Charleston facility.
The workgroup concluded that it would be more cost-effective to renovate
the current Charleston facility than to replace it with a new facility.
This conclusion was based, in part, on the cost estimates for constructing
a new medical center. In April 2003, the Secretary of VA sent a response
to the president of MUSC, which stated that if VA agreed to the joint
venture, it would prefer to place the new facility in phase III-which is
north of phase I-to provide better street access for veterans. (See fig. 6
for MUSC's proposal and VA's counterproposal.) In addition, the Secretary
indicated that MUSC would need to provide a financial incentive for VA to
participate in the joint venture. Specifically, MUSC would need to make up
the difference between the estimated life-cycle costs of renovating the
Charleston facility and building a new medical center-which VA estimated
to be about $85 million-through negotiations or other means. The Secretary
stated that if these conditions could not be met, VA would prefer to
remain in its current facility.
Figure 6: MUSC's Proposal (2002) and VA's Counter Proposal (2003)
Note: The circle highlights some of VA's existing property.
The MUSC president responded to VA's counterproposal in an April 2003
letter to the Secretary of VA. In the letter, the MUSC president stated
that MUSC was proceeding with phase I of the project and that the joint
venture concept could be pursued during later phases of construction. The
letter did not specifically address VA's proposal to locate the new
facility in phase III, or the suggestion that MUSC would need to provide
some type of financial incentive for VA to participate in the joint
venture. To move forward with phase I, the MUSC president stated that MUSC
would like to focus on executing an enhanced-use lease (EUL) for Doughty
Street. Although MUSC owns most of the property that will be used for
phases I through III, Doughty Street is owned by VA and serves as an
access road to the Charleston facility and parking lots. The planned
facility for phase I would encompass Doughty Street.13 (See fig. 7.)
Therefore, MUSC could not proceed with phase I-as originally planned-until
MUSC secured the rights to Doughty Street. To help its medical affiliate
move forward with construction, VA executed an EUL agreement with MUSC in
May 2004 for use of the street.14 According to the terms of the EUL, MUSC
will pay VA $342,000 for initial use of the street and $171,000 for each
of the following 8 years.
Figure 7: Construction of Phase I of MUSC's Project, July 2005
Note: The photograph shows the initial construction for phase I of MUSC's
project. Doughty Street will be encompassed by MUSC's new facility.
To facilitate negotiations on the joint venture proposal, a congressional
delegation visited Charleston to meet with VA and MUSC officials to
discuss the joint venture proposal on August 1, 2005. After this visit, VA
and MUSC agreed to jointly examine key issues associated with the joint
venture proposal. Specifically, VA and MUSC established the Collaborative
Opportunities Steering Group (steering group). The steering group is
composed of five members from VA, five members from MUSC, and a
representative from DOD, which is also a stakeholder in the facility
health care market.15 The steering group chartered four workgroups:
o The governance workgroup examined ways of establishing organizational
authority within a joint venture between VA and MUSC, including shared
medical services.
o The clinical service integration workgroup identified medical services
provided by VA and MUSC and opportunities to integrate or share these
services.
o The legal workgroup reviewed federal and state authorities (or
identified the lack thereof) and legal issues relating to a joint venture
with shared medical services.
o The finance workgroup provided cost estimates and analyses relating to
a joint venture with shared medical services.
The steering group and workgroups were intended to help VA and MUSC
determine if the joint venture proposal would be mutually beneficial. On
December 7, 2005, the steering group issued its final report to the Under
Secretary for Health and to the president of MUSC. According to the
report, the steering group concluded that the most advantageous options
were those that provide a revenue stream for VA and provide MUSC access to
new space without capital financing. Therefore, the group explored
construction models that incorporated benefits to both organizations that
included taking advantage of VA's access to capital financing and access
to MUSC revenue streams. As shown in table 3, the report identifies six
planning models, ranging from constructing a new medical facility with
space for VA and MUSC, to sharing that could occur with VA maintaining its
existing facility. Four of the models-A, A-1, A-2, and B-include varying
levels of shared space between VA and MUSC. These four models also call
for VA to overbuild the facility-that is, build it bigger than VA
needs-and lease the excess space to MUSC, thus providing VA with a revenue
stream to offset some of the cost of construction. The amount of excess
space built and leased by VA varies among the four models. Any option
pursued that involves VA building a new medical facility over-capacity for
the purpose of leasing the underutilized space requires close scrutiny,
since real estate leasing agreements are currently not part of its
mission. In addition, such options would also require specific
congressional authorization and appropriation since the costs of any of
the planning models identified would exceed $7 million, the threshold for
such action.
Table 3: Description of Planning Models Identified by Steering Group
Model Description
Model A Construct a new, oversized VA medical center to replace all VA
services. Excess capacity is leased to MUSC.
Model A-1 Construct a new, oversized VA medical center to replace all VA
services. Excess capacity is leased to MUSC. MUSC would
construct an adjacent tower.
Model A-2 Construct a new, oversized VA medical center to replace all VA
services, with administrative and clinical services located in
separate buildings. Excess capacity is leased to MUSC.
Model B Construct a new, slightly oversized VA medical center to replace
all VA services. Excess capacity is leased to MUSC.
Model C Construct a new VA medical center, with no excess space
available for leasing. Additional sharing between VA and MUSC
consists of shared high tech equipment and contracts for
services.
Model D VA remains in its current facility, with renovations as
appropriate. Additional sharing between VA and MUSC consists of
shared high tech equipment and contracts for services.
Source: GAO summary of information in the Collaborative Opportunities
Steering Group, Final Report, December 2005.
The steering group's December 2005 report does not recommend an option
that VA should pursue. Rather, the report outlines the perceived
advantages and disadvantages, as well as the costs, of each option.16 (See
app. I for the advantages, disadvantages, and costs of the different
models.) However, the report does note that two options were rejected by
steering group members. In particular, the finance workgroup rejected
Model A-2-which included an oversized new VA medical center and separate
buildings for administrative and clinical services-because, among other
things, the construction of a separate building to house administrative
services was not cost-effective. Additionally, MUSC deemed Model B-which
included a replacement VA medical center with moderate excess space to
lease to MUSC-not to be a viable option because it did not meet its total
bed replacement needs. Although the report identifies options that provide
a revenue stream for VA, the report notes that there is not sufficient
revenue or cost avoidance in any of the models for VA to achieve a 30-year
payback on the construction investment. According to VA officials, the
next step is for the Capital Asset Board of the Veterans Health
Administration to make a recommendation regarding the options contained in
the report. VA expects the Capital Asset Board's recommendations by the
end of April 2006.
Limited Collaboration and Communication Stalled Negotiations in Charleston
for About 3 Years
Prior to the summer of 2005, limited collaboration and communication
generally characterized the negotiations between MUSC and VA over the
joint venture proposal. In particular, before August 2005, VA and MUSC had
not exchanged critical information that would help facilitate
negotiations. For instance, MUSC did not clearly articulate to VA how
replacing the Charleston facility, rather than renovating it, would
improve the quality of health care services for veterans or benefit VA.
MUSC officials had generally stated that sharing services and equipment
would create efficiencies and avoid duplication, which would lead to cost
savings. However, MUSC had not provided any analyses to support such
claims. Similarly, as required by law,17 VA studied the feasibility of
coordinating its health care services with MUSC, pending construction of
MUSC's new medical center. This study was completed in June 2004. However,
VA officials did not include MUSC officials in the development of the
study, nor did they share a copy of the completed study with MUSC. VA also
updated its cost analysis of the potential joint venture in the spring of
2005, but again, VA did not share the results with MUSC. Because MUSC was
not included in the development of these analyses, there was no agreement
between VA and MUSC on key input for the analyses, such as the specific
price MUSC would charge VA for, or the nature of, the medical services
that would be provided. As a result of the limited collaboration and
communication, negotiations stalled-prior to August 2005, the last formal
correspondence between VA and MUSC leadership on the joint venture
occurred in April 2003.
Joint Venture Proposals in Charleston and Denver Pose Multiple Challenges
The joint venture proposals under consideration in Charleston and
previously proposed in Denver raise a number of challenges for VA and its
medical affiliates. These challenges-which were identified by VA, MUSC, or
UCH officials as well as previous studies prepared for or by VA, MUSC, or
UCH-include addressing institutional changes for VA and institutional
differences between VA and its medical affiliates, identifying legal
issues and seeking legislative remedies, and balancing funding priorities.
Although addressing these issues will be difficult, it is not
insurmountable, as evidenced by the VA-MUSC steering group's efforts to
address some of these challenges, as well as by VA's past partnerships
with some medical affiliates and DOD.
o Addressing institutional changes and differences: The joint ventures
proposed in Charleston and Denver pose a series of institutional changes
for VA and reveal a number of institutional differences between VA and its
medical affiliate that would need to be reconciled. Specifically, as an
in-house health care service provider with other departmental priorities,
by jointly constructing and operating a hospital with a nonfederal health
system, VA would deviate from its current health care model. Although VA
purchases significant amounts of medical services from its medical
affiliates, the relationship between VA and its affiliates has centered on
providing enhanced care for veterans as well as training medical school
residents and conducting medical research. According to VA, altering this
historical relationship to include jointly constructing and operating
facilities would introduce legal, administrative, and management
complexities that might require additional authorities. In addition,
according to VA and some stakeholders, a joint facility could diminish
VA's identity by deviating from a VA medical facility that treats only
veterans to one with a mixed-patient population served by providers from
different health systems. Hence, if maintaining VA's identity is important
to VA leadership, steps would need to be taken to protect VA's identity in
a joint facility.
Adding to the challenge of expanding affiliation relationships to include
joint ventures involving major capital are inherent differences between VA
and its medical affiliates-from their missions to their funding processes.
For example, in addition to its mission of providing care for our nation's
veterans, VA is also responsible for supporting national, state, and local
emergency management and serving as backup to DOD during war and other
national emergencies. In addition, funding decisions for both VA and MUSC
must go through several layers of review. VA's major capital investments
(over $7 million) must be evaluated at multiple levels within VA and
approved by the Office of Management and Budget and by Congress,18 while
such investments by MUSC must be approved by its board, and if requiring
state funds, by the state legislature. These differences would need to be
considered in any joint venture between VA and a medical affiliate.
o Identifying legal issues and seeking legislative remedies: Joint venture
proposals raise many complex legal issues. The specific legal issues
raised depend on the type of joint venture proposed, but many involve real
estate, construction, contracting, and employment. In Charleston, the
legal workgroup identified VA's and MUSC's legal authorities, or lack
thereof, on numerous issues relating to each option considered. The legal
workgroup concluded that VA has the legal authority to pursue any of the
six planning models identified but that specific considerations would
arise for each model. For example, legislative authorization and
appropriation are required for any major VA construction project over $7
million. In addition, while VA is authorized, under its EUL authority, to
lease underutilized real property for up to 75 years, the authorization
does not provide for building a new medical facility over-capacity for the
purpose of leasing the underutilized space.19
o Developing appropriate governance plans: A venture involving a jointly
operated facility would require the parties to agree to a plan for
governing it. Any governance plan would have to maintain VA's direct
authority over and accountability for the care of VA patients. In
addition, if shared medical services were a component of a joint venture
between the VA and an affiliate, the entities would need a mechanism to
ensure that the interests of the patients served by both are protected
today and in the future. For instance, VA might decide to purchase
operating room services from its affiliate. If the sharing agreement were
dissolved afterwards, it would be difficult for VA to resume the
independent provision of these services. Therefore, a clear plan for
governance would ensure that VA and its affiliate could continue to serve
their patients' health care needs as well as or better than before. To
address possible governance issues in Charleston, the steering group
recommended instituting a joint governance council that would include a
nonaffiliated third party to oversee the sharing relationship in areas
other than research and educational activities. The joint governance
council's decisions would be advisory in nature-and not legally binding-in
order not to undermine the current authority of VA or MUSC.
o Balancing funding priorities: VA leadership must weigh joint venture
opportunities against VA's capital assets and health care service needs
throughout the nation when making funding decisions and recommendations.
VA operates a nationwide health care system for veterans, including 157
medical centers and over 800 clinics. According to VA, its capital
requirements are significant given the amount of real property it owns and
uses and the age and condition of most of its facilities. Further, in
2004, the Secretary of VA estimated that implementing CARES will require
additional investments of approximately $1 billion per year for at least
the next 5 years, with substantial infrastructure investments then
continuing indefinitely. Balancing these competing capital requirements is
made more difficult by the fiscal challenges facing the federal
government. Given the size of the government's projected deficit, VA, like
other federal agencies, could face constrained budgets in the future,
making funding of even high priority capital requirements challenging.
Additional challenges are likely to be identified as VA continues to
explore the proposed joint venture with MUSC or other possible joint
ventures in the future. In particular, should VA decide to pursue a joint
venture with MUSC or other medical affiliates in the future, it would
likely face additional challenges during the implementation phase. For
example, due to the inherent differences in the purposes for which VA's
and MUSC's information management systems were designed, the systems would
not be compatible. According to MUSC officials, VA's and MUSC's
computerized patient record systems are different, and their billing
systems are incompatible. Therefore, at least initially, the systems would
not be integrated, and parallel systems would need to be implemented-which
could result in added costs in terms of staff time and raise the potential
for errors.
VA's Current Sharing Arrangements with Medical Affiliates and DOD Could Be
Instructive As VA Considers Current and Future Joint Ventures
Partnerships with other health providers are not new to VA. For instance,
the Mike O'Callaghan Federal Hospital, an integrated federal hospital
jointly constructed by the VA and Air Force in Las Vegas, Nevada,
currently serves as a model of joint operation and shared medical
services.20 However, joint ventures of this magnitude with DOD are
limited. Further, VA has not entered into a joint venture with a medical
affiliate of the magnitude proposed in Charleston or Denver. However,
there are instances of significant capital ventures between VA and its
affiliates involving high-priced medical equipment. For example, VA's
Western New York Healthcare System in Buffalo, New York, houses a Positron
Emission Tomography (PET) scanner that was purchased by its affiliate. In
exchange, VA purchases scans from its affiliate for veterans and provides
operational and administrative staff to support the equipment.
These past capital ventures are on a smaller scale than the joint ventures
proposed in Charleston and Denver, but they could be somewhat instructive
as VA considers current and future joint venture proposals and attempts to
address the associated challenges. For example, in these past capital
ventures, VA had to ensure that veterans received the appropriate access
to equipment and services, and VA accomplished this through the terms and
conditions outlined in the contract. In addition, VA had to address
governance, legal, and information management challenges in establishing
these capital-sharing arrangements. The difficulty of addressing such
challenges, however, likely increases as the complexity and magnitude of
the proposed joint venture grows.
VA Could Learn Several Lessons from Its Experiences with Denver and
Charleston Joint Venture Proposals
Because VA may explore the possibility of entering into partnerships with
other medical affiliates in the future, the lessons learned from VA's
experiences in Charleston and Denver could be instructive. It is possible
that more opportunities for similar joint ventures or sharing arrangements
will present themselves in the coming years. In particular, our analysis
of VA data on its major medical facilities indicates that 43 percent of
these facilities, like the medical center in Denver, consist of buildings
with an average age of over 50 years, although some have undergone
extensive renovations over the years. Given the age of these facilities,
many of them may need to be replaced or extensively renovated in the
future. Additionally, disasters, such as Hurricane Katrina, could force
unplanned renovations or replacements. As VA moves forward in making
necessary renovations or replacements throughout the country, there could
be opportunities for joint ventures with its medical affiliates. VA will
have to determine if these opportunities are in the best interest of the
federal government and our nation's veterans.
The lessons that emerged from our work in Charleston and Denver reflect
how the absence of practices that we have emphasized in previous reports
can hamper effective consideration of potential joint ventures. These
reports examine leading practices for realigning federal agency
infrastructure, collaboration among organizations, and organizational
transformations.21 The lessons include establishing criteria to evaluate
the joint venture proposal, obtaining leadership buy-in and support for
the joint venture, ensuring extensive collaboration among stakeholders,
and developing a strategy for effective and ongoing communications.
Lack of Criteria at Departmental Level Results in Inconsistent Evaluations
One of the most important lessons from VA's experiences in Denver and
Charleston is that the absence of criteria at the departmental level to
evaluate joint venture proposals can result in inconsistent evaluations,
misunderstandings, and delays. The joint venture proposals for VA's
medical centers in Denver and Charleston presented VA with a new
opportunity-that is, the proposals involved joint construction and service
sharing on a scale beyond anything VA had experienced in partnering with
its medical affiliates in the past. VA did not have criteria at the
departmental level for evaluating and negotiating joint venture proposals,
which led to inconsistent evaluations of the Denver and Charleston
proposals. For instance, in Denver, VA facility and network officials
worked collaboratively with UCH officials on the joint venture proposals,
including jointly funding a study to assess the feasibility and cost of
various options. In contrast, VA facility and network officials did not
include MUSC officials in the development of the study that examined the
feasibility of coordinating VA's health care services with MUSC, nor did
they share a copy of the completed study with MUSC. This contributed to
the negotiations between VA and MUSC stalling for over 2 years. VA
officials in Denver also told us that the lack of departmental criteria
hampered negotiations, and noted that on the basis of their experience a
common tool or process is needed to assess joint venture proposals so that
they can be evaluated consistently.
As we have emphasized in previous work on realigning federal
infrastructure, a set of criteria for evaluating decisions regarding
infrastructure enhances the transparency of these decisions and helps
ensure that the decisions are made in a manner that is fair to all
stakeholders and that is efficient and effective.22 Although we recognize
that every joint venture is likely to be different, criteria would
establish a framework for evaluating future joint venture proposals. In
addition to identifying the factors VA would consider in evaluating
proposals and indicating how these factors would be measured, the criteria
would help ensure that proposals are evaluated consistently-regardless of
location or officials involved. The criteria would also serve to
communicate VA's expectations for joint ventures. That is, they would
identify what VA is looking for in potential joint ventures, such as
improved medical care for veterans and reduced operating costs. By
documenting and sharing these criteria with potential partners, VA would
help ensure that its positions are understood from the outset and thus
eliminate possible misunderstandings. The VA-MUSC steering group's efforts
to identify criteria to evaluate the Charleston proposal and the studies
conducted in Denver could serve as starting points for the development of
criteria.
Lack of Communications Strategy Leads to Misinformation and Confusion
VA's experiences in Denver and Charleston highlighted the fact that the
absence of sustained communication with potential joint venture partners
and stakeholders as well as within VA can be detrimental to negotiations.
Breakdowns in communication occurred in both locations during key points
of the negotiations and hindered progress. For example, in Charleston,
there was limited communication between VA and MUSC for about 2 years; as
a result, negotiations stalled. In addition, in both locations, a primary
point of contact-either a single individual or a group-was not identified
to represent VA's position in negotiations with the medical universities.
Rather, various VA officials at the facility, network, and departmental
levels often maintained separate contacts with UCH and MUSC officials. As
a result, according to MUSC and UCH officials, they received mixed signals
as to VA's intentions regarding the proposals. Similarly, MUSC and UCH
also contacted and communicated with different VA officials at the
facility, network, and departmental levels, which also led to confusion.
In our previous work on organizational transformations, we have noted that
creating an effective, ongoing communication strategy is essential to
implementing significant organizational changes like the joint ventures
proposed in Charleston and Denver.23 Such a strategy should entail
communicating information early and often to help build an understanding
of the purpose of the planned change and build trust among VA and its
medical affiliates as well as stakeholders, such as employees and
veterans, who could have concerns over such issues as the impact of a
joint venture on patient care. The strategy should also encourage
communication by facilitating a two-way honest exchange with, and allow
for feedback from, stakeholders. A communications strategy can also help
ensure that these groups receive a message that is consistent in tone and
content. Sharing a consistent message with stakeholders helps reduce the
perception that others are getting the "real" story when, in fact, all are
receiving the same information. The strategy should also make it clear
that it is essential to have a primary point of contact with the necessary
authority to negotiate effectively with partners, make timely decisions,
and move quickly to implement top leadership's decisions regarding the
joint venture. Good communication is central to forming the effective
internal and external partnerships that are vital to the success of
transforming endeavors such as joint ventures. In Charleston, the steering
group has taken steps to improve communication by establishing a plan for
VA and MUSC to share information about the potential joint venture with
stakeholders such as employees and veterans groups.
Lack of Leadership Support Can Hinder Negotiations
Another lesson that emerged from VA's experience with the joint venture
proposals for Denver and Charleston is that leadership buy-in and support
are critical. The proposed joint venture in Denver did not come to
fruition largely because VA leadership never fully supported the concept.
In particular, when the joint venture was first proposed, UCH and VA
officials at the network and facility levels worked extensively together
on the proposal. Top level VA management, however, was not involved in
these efforts. Moreover, in response to UCH's request for a 1-year time
frame for a decision regarding a joint facility, in October 2002, the VA
Secretary wrote that VA "cannot now commit to a joint University-VA
hospital within the one-year timetable you propose. However, I feel
strongly that we should not preclude a freestanding VA medical center at
Fitzsimons in the future." According to UCH officials, this decision was
unexpected given the fact they had worked closely with VA facility
officials on a possible joint venture. Certainly it is the VA Secretary's
prerogative to extend or withhold support for different proposals, and the
Secretary must determine whether the proposals are in the best interest of
veterans. However, VA's experiences in Denver and Charleston indicate that
without such support negotiations for joint ventures will be hampered.
Our previous work on organizational transformation indicates that support
from top leadership is indispensable for fundamental change, such as a
joint venture.24 Top leadership's clear and personal involvement in the
transformation represents stability for both the organization's employees
and its external partners. Top leadership must set the direction, pace,
and tone for the transformation. Likewise, when a transformation requires
extensive collaboration with another organization, as would be the case
with a joint venture, committed leadership at all levels is needed to
overcome the many barriers to working across organizational boundaries. If
VA decides to pursue a joint venture with MUSC in Charleston, or other
similar projects with medical affiliates or other partners, success will
hinge on the level of support the project receives from top VA management.
Lack of Extensive Collaboration Hampers Negotiations
Another lesson that emerged from the experiences in Denver and Charleston
is that a lack of, or limited, collaboration hampers negotiations. For
example, in Charleston, VA and MUSC did not initially exchange or share
critical information, such as the feasibility study, which contributed to
the negotiations stalling from about 2003 to 2005. In addition, until the
VA-MUSC steering group was formed in Charleston, there was limited
collaboration between VA and its stakeholders. This heightened the
stakeholders' anxiety about the proposed joint venture and led to the
spread of misinformation about the proposed joint venture. In Denver,
although VA officials from the facility and network level and UCH
officials met frequently after UCH proposed the joint venture, VA
officials with the necessary decision-making authority were not involved
in these initial discussions. Consequently, when the Secretary of VA
decided against a joint venture in Denver, UCH officials felt misled,
which resulted in an atmosphere of mistrust between the entities.
Our previous work on collaboration between organizations suggests several
practices that VA might benefit from as it continues to consider a joint
venture in Charleston as well as other such opportunities that may occur
in the future.25 These practices include ensuring the involvement of key
stakeholders, defining and articulating a common outcome, establishing
mutually reinforcing or joint strategies, identifying and addressing needs
by leveraging resources, and agreeing on roles and responsibilities. The
VA-MUSC steering group illustrates how some of these practices can be
implemented. For example, the steering group was led by senior VA and MUSC
officials and consisted of VA and MUSC staff who have knowledge in key
areas (e.g., finance). In addition, the communications plan the VA-MUSC
steering group established includes a presentation to use when
communicating with stakeholders about the joint venture proposal.
Conclusions
To address future health care needs of veterans, VA's challenge is to
explore new ways to fulfill its mission of providing veterans with quality
health care. The prospect of jointly constructing and operating medical
facilities with medical affiliates presents an opportunity for VA to
consider the feasibility of expanding its relationships with university
medical school affiliates to include the sharing of medical services in an
integrated hospital. This is just one of several ways VA could provide
care to veterans. It is up to VA, working with its stakeholders, and
Congress to determine if expanding VA's relationship with medical
affiliates to include joint ventures-of the scale proposed in Denver and
Charleston-is in the best interest of the federal government and the
nation's veterans, as well as how such joint ventures fit within the
context of the CARES framework.
VA will be in a better position to consider future joint ventures if it
learns from its experiences with the joint venture proposals in Denver and
Charleston. Among these lessons is the importance of leadership support
and extensive collaboration. In addition, VA's experiences in Denver and
Charleston indicate that having a set of criteria at the departmental
level would provide a clear basis for making decisions on joint venture
proposals. Although each proposal will likely be somewhat unique, and
should be evaluated on its own merits and circumstances, criteria provide
a framework for future evaluations and negotiations. A set of criteria at
the departmental level helps ensure that proposals are evaluated in a
consistent fashion across the country as well as communicates VA's
expectations for joint ventures. Another important lesson is that a
strategy for communicating with its medical affiliates and stakeholders,
including veterans and employees, can help VA avoid the problems that
hampered progress in negotiations over the Denver and Charleston joint
venture proposals. A communications strategy helps build understanding and
trust between VA and its medical affiliates and stakeholders as well as
helps ensure that these groups receive a message that is consistent in
tone and content. Establishing a set of evaluation criteria and a
communications strategy are tangible steps VA could take to better
position itself in considering future joint venture proposals.
Recommendations for Executive Action
To ensure that there is a clear basis for evaluating future joint venture
proposals as well as to help ensure early and frequent communication
between VA and its medical affiliates and stakeholders during
negotiations, we recommend that the Secretary of VA take the following two
actions:
o Identify criteria at the departmental level for evaluating joint venture
proposals. In order to foster an atmosphere of collaboration, VA should
share these criteria with potential joint venture partners.
o Develop a communications strategy for use in negotiating joint venture
proposals.
Agency Comments
We provided a draft of this report to VA for its review and comment. On
April 10, 2006, VA's audit liaison provided VA's comments on the draft
report via e-mail. VA agreed with the report's conclusions and
recommendations. We also provided UCH and MUSC officials portions of the
draft report that related to their joint venture proposals. UCH and MUSC
officials provided technical clarifications to these portions of the draft
report, which we incorporated where appropriate.
Scope and Methodology
To address our objectives, we analyzed VA, UCH, and MUSC planning
documents, presentations, and studies related to the joint venture
proposals as well as correspondence between VA and these medical
affiliates regarding the proposals. We also examined the recommendations
of the CARES Commission and the Secretary's CARES Decision report, VA's
5-year capital plan (2005-2010), and federal statutes and accompanying
reports. In addition, we interviewed officials from VA, DOD, MUSC, and UCH
to obtain information on the history and status of the joint venture
proposals as well as the challenges associated with implementing such
proposals. We also interviewed local stakeholders, including officials
from the Fitzsimons Redevelopment Authority in Aurora, Colorado, the
mayors of Charleston and Aurora, and representatives from the VA
employees' unions in each location to obtain their perspectives and to
obtain information on local capital asset planning and its impact. We also
toured VA and MUSC facilities in Charleston and VA and UCH facilities in
the Denver area. Finally, we synthesized information obtained from VA,
MUSC, and UCH officials and reviewed our past work on organizational
transformation and collaboration among organizations to identify lessons
learned from VA's experiences with joint venture proposals in Charleston
and Denver.26 Although we examined the joint venture proposals for VA's
Denver and Charleston facilities and the associated studies and planning
documents, we did not evaluate the merits of the proposals. We assessed
the reliability of the information obtained from VA, MUSC, and UCH. We
concluded that the information was sufficiently reliable for our purposes.
We are sending copies of this report to congressional committees with
responsibilities for veteran issues; the Secretary of Veterans Affairs;
and the Director, Office of Management and Budget. We also will make
copies available to others upon request. In addition, this report will be
available at no charge on GAO's Web site at h ttp://www.gao.gov.
If you or your staff have any questions on matters discussed in this
report, please contact me on (202) 512-2834 or at g oldsteinm@gao.gov.
Contact points for our Offices of Congressional Relations and Public
Affairs may be found on the last page of this report. Key contributors to
this report include Chris Bonham, Nikki Clowers, Daniel Hoy, Jennifer Kim,
Edward Laughlin, Susan Michal-Smith, James Musselwhite, Jr., and Michael
Tropauer.
Sincerely,
Mark L. Goldstein Director, Physical Infrastructure Issues
Appendix I
Description of Planning Models Identified by Steering Group
Model A: Construct a
new, oversized VA
medical center to
replace all VA services.
Excess capacity is
leased to MUSC.
Advantage Disadvantage Construction Life cycle
cost cost
Provides VA with $6.7 Requires the $545.6 million $4.3 billion
million annually to largest + activation
enhance care construction
investment by VA
Provides VA with extra MUSC construction
bed capacity to meet must meet VA
disaster needs security
requirements,
raising
construction costs
Model A-1: Construct a
new, oversized VA
medical center to
replace all VA services.
Excess capacity is
leased to MUSC. MUSC
would construct an
adjacent tower.
Advantage Disadvantage Construction Life cycle
cost cost
Reduces VA's Requires large $368 million + $4.2 billion
construction investment investment by VA activation
for construction
Provides MUSC with a VA security and
lower construction cost safety
specifications may
need to be met if
buildings are
connected
Provides VA with $4.2 Separate
million annually to construction
enhance care reduces the
coordination and
compatibility of
space
Provides VA with extra Eliminates some VA
bed capacity to meet lease revenue and
disaster needs negatively affects
payback
Model A-2: Construct a
new, oversized VA
medical center to
replace all VA services,
with administrative and
clinical services
located in separate
buildings. Excess
capacity is leased to
MUSC.
Advantage Disadvantage Construction Life cycle
cost cost
Reduces VA's Volume of Not calculated Not
construction investment administrative calculated
space is small and
construction costs
are similar to
Model A
Builds adjacency of VA Some operational
and MUSC inefficiencies
arise through lack
of adjacency
Provides VA with MUSC construction
additional annual must meet VA
revenue to enhance care security
requirements,
raising
construction costs
Model B: Construct a
new, slightly oversized
VA medical center to
replace all VA services.
Excess capacity is
leased to MUSC.
Advantage Disadvantage Construction Life cycle
cost cost
Somewhat less initial VA MUSC does not $444.4 million $4.2 billion
investment required consider this model + activation
viable since it
must still
construct beds
elsewhere
Provides VA with $4.9 Requires second
million annually to largest investment
enhance care by VA for
construction
Provides VA with extra MUSC construction
bed capacity to meet must meet VA
disaster needs security
requirements,
raising
construction costs
Model C: Construct a new
VA medical center, with
no excess space
available for leasing.
Additional sharing
between VA and MUSC
consists of shared high
tech equipment and
contracts for services.
Advantage Disadvantage Construction Life cycle
cost cost
Provides VA with Results in $317.2 million+ $4.1 billion
state-of-the-art patient demolition of a activation
care and administrative facility currently
space, including private considered to be
and semiprivate patient sound
rooms and efficient
energy systems
Potentially improves Requires a sizable
adjacency of VA and MUSC construction
investment by VA
Provides VA with Eliminates revenue
opportunities to address sources that would
specialty care at a offset construction
network level costs through space
and service sharing
Potential for VA and Duplicates
MUSC to reduce the size construction of
of new facilities by some services that
planning to share some could potentially
capacity through be shared
contracts
Maintains greater VA and
MUSC autonomy
Avoids further
investment into VA's
aging infrastructure
Provides VA with $4.3
million annually to
enhance care
Model D: VA remains in
its current facility,
with renovations as
appropriate. Additional
sharing between VA and
MUSC consists of shared
high tech equipment and
contracts for services.
Advantage Disadvantage Construction Life cycle
cost cost
No up-front construction Existing facility $27.1 million $3.9 billion
costs for VA is dated and will
require continued
investment
Continues use of a sound Replacement may be
facility necessary at some
future date
Opportunities for Duplicates
increased sharing have construction of
been identified that do services that could
not necessarily require potentially be
construction shared
Provides VA with Charleston area
opportunities to address veterans may have
specialty care at a to travel to other
network level VA medical centers
for some care
Maintains current access Requires that the
facility be
renovated to
withstand disasters
Maintains more VA and Less bed capacity
MUSC autonomy for potential
disaster and DOD
needs
Existing facility
may not withstand
strong hurricane or
flood
Source: Collaborative Opportunities Steering Group, Final Report, December
2005.
Related GAO Products
Results-Oriented Government: Practices That Can Help Enhance and Sustain
Collaboration among Federal Agencies. GAO-06-15 . Washington, D.C.:
October 21, 2005.
VA Health Care: Preliminary Information on the Joint Venture Proposal for
VA's Charleston Facility. G AO-05-1041T. Washington, D.C.: September
26, 2005.
VA Health Care: Key Challenges to Aligning Capital Assets and Enhancing
Veterans' Care. GAO-05-429 . Washington, D.C.: August 5, 2005.
Federal Real Property: Further Actions Needed to Address Long-standing and
Complex Problems. GAO-05-848T . Washington, D.C.: June 22, 2005.
U.S. Postal Service: The Service's Strategy for Realigning Its Mail
Processing Infrastructure Lacks Clarity, Criteria, and Accountability.
GAO-05-261 . Washington, D.C.: April 8, 2005.
VA Health Care: Important Steps Taken to Enhance Veterans' Care by
Aligning Inpatient Services with Projected Needs. GAO-05-160 . Washington,
D.C.: March 2, 2005.
High-Risk Series: An Update. GAO-05-207 . Washington, D.C.: January 2005.
Budget Issues: Agency Implementation of Capital Planning Principles Is
Mixed. GAO-04-138. Washington, D.C.: January 16, 2004.
Federal Real Property: Vacant and Underutilized Properties at GSA, VA, and
USPS. GAO-03-747. Washington, D.C.: August 19, 2003.
VA Health Care: Framework for Analyzing Capital Asset Realignment for
Enhanced Services Decisions. GAO-03-1103R. Washington, D.C.: August 18,
2003.
Results-Oriented Cultures: Implementation Steps to Assist Mergers and
Organizational Transformations. GAO-03-669 . Washington, D.C.: July 2,
2003.
VA Health Care: Improved Planning Needed for Management of Excess Real
Property. GAO-03-326. Washington, D.C.: January 29, 2003.
Major Management Challenges and Program Risks: Department of Veterans
Affairs. GAO-03-110. Washington, D.C.: January 2003.
High-Risk Series: Federal Real Property. GAO-03-122. Washington, D.C.:
January 2003.
VA Health Care: VA Is Struggling to Address Asset Realignment Challenges.
GAO/T-HEHS-00-88 . Washington, D.C.: April 5, 2000.
VA Health Care: Improvements Needed in Capital Asset Planning and
Budgeting. GAO/HEHS-99-145 . Washington, D.C.: August 13, 1999.
VA Health Care: Challenges Facing VA in Developing an Asset Realignment
Process. GAO/T-HEHS-99-173 . Washington, D.C.: July 22, 1999.
VA Health Care: Capital Asset Planning and Budgeting Need Improvement.
GAO/T-HEHS-99-83 . Washington, D.C.: March 10, 1999.
(543137)
www.gao.gov/cgi-bin/getrpt? GAO-06-472 .
To view the full product, including the scope and methodology,
click on the link above.
For more information, contact Mark L. Goldstein at (202) 512-2834 or
goldsteinm@gao.gov.
Highlights of GAO-06-472 , a report to the Chairman, Committee on
Veterans' Affairs, House of Representatives
April 2006
VA HEALTH CARE
Experiences in Denver and Charleston Offer Lessons for Future Partnerships
with Medical Affiliates
The Department of Veterans Affairs (VA) maintains affiliations with
medical schools, including the Medical University of South Carolina (MUSC)
and the University of Colorado at Denver and Health Services Center and
University of Colorado Hospital (UCH), to obtain enhanced medical care for
veterans. As part of their plans for new medical campuses, both UCH and
MUSC proposed jointly constructing and operating new medical facilities
with VA in Denver and Charleston, respectively.
This report discusses (1) how VA evaluated the joint venture proposals for
Denver and Charleston and the status of these proposals, (2) the
challenges these proposals pose for VA, and (3) the lessons VA can learn
from its experiences in Charleston and Denver for future partnerships.
What GAO Recommends
GAO recommends that VA establish criteria for evaluating joint venture
proposals and a strategy for communicating with stakeholders when
negotiating such proposals. VA agreed with GAO's conclusions and
recommendations.
VA evaluated the joint venture proposals for its medical facilities in
Denver and Charleston using criteria developed specifically for each
location, and while VA opted to build a stand-alone facility in Denver, it
is still considering a joint venture in Charleston. Because the proposals
involved joint construction and service sharing on a scale beyond anything
VA had experienced with its medical affiliates in the past, VA did not
have criteria at the departmental level to evaluate the proposals on a
consistent basis in both locations. In both locations, negotiations
between VA and its medical affiliates stretched over a number of years, in
part because they were hampered by limited collaboration and
communication, among other things. While VA decided against a joint
venture in Denver, it has made no decision on Charleston. A VA-MUSC
steering group, formed last summer to study the joint venture proposal in
Charleston, issued a report in December 2005 that outlined the advantages
and disadvantages of different options.
The joint ventures proposed in Denver and Charleston present a number of
challenges to VA, including addressing institutional differences between
VA and its medical affiliates, identifying legal issues and seeking
legislative remedies, and balancing funding priorities. For example,
capital expenditures for a joint venture would have to be considered in
the context of other VA capital priorities. Although addressing these
issues will be difficult, the VA-MUSC steering group's efforts could
provide insight into how to tackle them.
VA's experiences with joint venture proposals in Denver and Charleston
offer several lessons for VA as it considers similar opportunities in the
future. One of the most important lessons is that having criteria at the
departmental level to evaluate joint venture proposals helps to improve
the transparency of decisions concerning joint ventures and VA's ability
to ensure that the decisions are made in a consistent manner across the
country. Another key lesson is that having a strategy for communicating
with stakeholders, such as employees and veterans, helps VA build
understanding and trust among stakeholders. The following table identifies
these and other lessons from VA's experiences in Denver and Charleston.
Lessons Learned from VA's Experiences in Denver and Charleston
o Criteria at the departmental level help provide clarity and
consistency in evaluation approach.
o A communications strategy helps avoid misinformation and
confusion.
o Leadership support facilitates negotiations.
o Extensive collaboration assists negotiations.
*** End of document. ***