[House Hearing, 116 Congress]
[From the U.S. Government Publishing Office]
VISTA TRANSITION: ASSESSING THE FUTURE OF AN ELECTRONIC HEALTH RECORDS
PIONEER
=======================================================================
HEARING
before the
SUBCOMMITTEE ON TECHNOLOGY MODERNIZATION
of the
COMMITTEE ON VETERANS' AFFAIRS
U.S. HOUSE OF REPRESENTATIVES
ONE HUNDRED SIXTEENTH CONGRESS
FIRST SESSION
__________
THURSDAY, JULY 25, 2019
__________
Serial No. 116-28
__________
Printed for the use of the Committee on Veterans' Affairs
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Available via the World Wide Web: http://www.govinfo.gov
______
U.S. GOVERNMENT PUBLISHING OFFICE
40-857 WASHINGTON : 2021
COMMITTEE ON VETERANS' AFFAIRS
MARK TAKANO, California, Chairman
JULIA BROWNLEY, California DAVID P. ROE, Tenessee, Ranking
KATHLEEN M. RICE, New York Member
CONOR LAMB, Pennsylvania, Vice- GUS M. BILIRAKIS, Florida
Chairman AUMUA AMATA COLEMAN RADEWAGEN,
MIKE LEVIN, California American Samoa
MAX ROSE, New York MIKE BOST, Illinois
CHRIS PAPPAS, New Hampshire NEAL P. DUNN, Florida
ELAINE G. LURIA, Virginia JACK BERGMAN, Michigan
SUSIE LEE, Nevada JIM BANKS, Indiana
JOE CUNNINGHAM, South Carolina ANDY BARR, Kentucky
GILBERT RAY CISNEROS, JR., DANIEL MEUSER, Pennsylvania
California STEVE WATKINS, Kansas
COLLIN C. PETERSON, Minnesota CHIP ROY, Texas
GREGORIO KILILI CAMACHO SABLAN, W. GREGORY STEUBE, Florida
Northern Mariana Islands
COLIN Z. ALLRED, Texas
LAUREN UNDERWOOD, Illinois
ANTHONY BRINDISI, New York
Ray Kelley, Democratic Staff Director
Jon Towers, Republican Staff Director
SUBCOMMITTEE ON TECHNOLOGY MODERNIZATION
SUSIE LEE, Nevada, Chairwoman
JULIA BROWNLEY, California JIM BANKS, Indiana, Ranking Member
CONOR LAMB, Pennsylvania STEVE WATKINS, Kansas
JOE CUNNINGHAM, South Carolina CHIP ROY, Texas
Pursuant to clause 2(e)(4) of Rule XI of the Rules of the House, public
hearing records of the Committee on Veterans' Affairs are also
published in electronic form. The printed hearing record remains the
official version. Because electronic submissions are used to prepare
both printed and electronic versions of the hearing record, the process
of converting between various electronic formats may introduce
unintentional errors or omissions. Such occurrences are inherent in the
current publication process and should diminish as the process is
further refined.
C O N T E N T S
----------
Thursday, July 25, 2019
Page
Vista Transition: Assessing The Future Of An Electronic Health
Records Pioneer................................................ 1
OPENING STATEMENTS
Honorable Susie Lee, Chairwoman.................................. 1
Honorable Jim Banks, Ranking Member.............................. 2
WITNESSES
Dr. Paul Tibbits, Executive Director, Office of Technical
Integration Office of Information and Technology, Department of
Veterans Affairs............................................... 4
Prepared Statement........................................... 21
Accompanied by:
Mr. Charles C. Hume, Acting Assistant Deputy Under Secretary
for Health for Office of Health Informatics, Veterans
Health Administration, Department of Veterans Affairs
Dr. Thomas O'Toole Senior Medical Advisor, Office of the
Assistant Deputy Undersecretary for Health for Clinical
Operations, Veterans Health Administration, Department of
Veterans Affairs
Mr. John Short, Chief Technology and Integration Officer,
Office of Electronic Health Record Modernization,
Department of Veterans Affairs
Ms. Carol Harris, Director, Information Technology Acquisition
Management, Government Accountability Office................... 5
Prepared Statement........................................... 28
VISTA TRANSITION: ASSESSING THE FUTURE OF AN ELECTRONIC HEALTH RECORDS
PIONEER
----------
Thursday, July 25, 2019
Committee on Veterans' Affairs,
U. S. House of Representatives,
Washington, D.C.
The Subcommittee met, pursuant to notice, at 10:15 a.m., in
Room 210, House Visitors Center, Hon. Susie Lee [Chairwoman of
the Subcommittee] presiding.
Present: Representatives Lee, Lamb, Cunningham, Banks, and
Watkins.
OPENING STATEMENT OF SUSIE LEE, CHAIRWOMAN
Ms. Lee. Good morning. Thank you all for being here. This
hearing will now come to order.
During the '70s, a dedicated group of programmers and
clinicians began a health care transformation as they built
what would become the Veterans Health Information Systems and
Technology Architecture, or what we know as VistA. It was the
beginning of an age of personal computer and these IT pioneers
saw the potential for bringing computing power to the health
care space. The Department of Veterans Affairs was an early
innovator and adopter of the electronic medical record, and
established itself as a leader in health care IT.
Today, we have clinicians and researchers across VA using
IT tools and powerful health data to improve care and find
medical breakthroughs. However, the VA is at a technology
crossroads and what began as a guerilla IT project has sprawled
into a massive, decentralized system in an archaic coding
language, and within the VA, there are at least 130 versions or
instances of VistA across 1500 sites. No version is the same
and the system connects to various applications and devices
through interfaces.
VistA serves many offices, programs, staff, and veterans,
but it has surpassed its technology life span.
VA has struggled to modernize VistA and past attempts to
replace it or update it have not been successful, and now the
VA is pursuing an approach with the acquisition of a commercial
electronic health records system. However, the transition from
one system to another is not a simple matter of just flipping
the switch; it is a painstaking process that you all are aware
of and that involves technical challenges, as well as policy
changes. There are many stakeholders who want to understand the
impacts of the transition and how their equities in VistA will
be affected.
VA has told the Subcommittee that there is a plan in draft
to address both the technical and policy side of the transition
from VistA to Cerner's electronic health record, but that plan
is not expected to be completed until the fall of 2019. This
plan will require the concurrence of the Office of Information
and Technology, the Veterans Health Administration, and the
Office of Electronic Health Record Modernization.
There are many unknowns in this transition. It is important
that the VA's strategy be well timed to identify those unknowns
and to mitigate potential disruptions to the health care and
research. The fact that this plan is still being formulated is
concerning. Further, as the Government Accountability Office
will discuss today, the VA does not yet have a reliable
accounting of all the costs associated with VistA management,
and there is still ongoing work to understand all of the
instances of VistA and to define them. We also need the VA to
arrive at a transparent and accountable decision as to what
VistA management will mean going forward, so that there are not
gaps in care, that valuable research is not disrupted, and that
expectations are established and met.
VistA cannot remain a static system over the 10 years that
EHRM implementation will take. And, additionally, at least 40
percent of VistA will not be in Cerner, and this Subcommittee
would like more information how VA will manage those
functionalities and potentially modernize them in the future.
We think there are opportunities for VA to be forward-
thinking in the transition and to harness the innovative
approach that drove the creation of VistA. The pilot to move
instances of VistA to the cloud has potential, but we need more
information to understand its feasibility from a cost and
impact perspective. At minimum, we need to maintain the legacy
system until it has been fully replaced or modernized, but if
there are potential efficiencies and health care innovations to
be gained, we should identify them and also consider those
opportunities.
I thank all of the witnesses for being here today and look
forward to your testimony. And I now would like to recognize my
colleague Ranking Member Banks for 5 minutes to deliver his
opening remarks.
Mr. Banks?
OPENING STATEMENT OF JIM BANKS, RANKING MEMBER
Mr. Banks. Thank you, Madam Chair.
It is no longer possible to talk about VistA without
discussing Cerner and vice versa. Although the goal of VA's
electronic health record modernization is to replace VistA and
CPRS, these legacy systems will exist alongside Cerner for at
least the next 9 years; that means they have to interoperate.
This mixed environment will be extremely challenging, in which
some medical centers will still use VistA while others use the
Cerner EHR.
Up until now, this Subcommittee has focused on the total
cost of ownership of VistA versus the total cost of
implementing and operating Cerner. I still believe that is an
important question and one we have yet to receive a
satisfactory answer to, but the complexity of the mixed
environment is the biggest difficulty confronting VA.
Some key questions are, how will the Cerner data flow back
into VistA? How will scheduling information be integrated
across the two environments? Will referrals be transmitted
uniformly in both systems? And how will different data be
aggregated for reporting an analysis?
We are still in the middle of the beginning of the EHRM
overall, but VA is nearing the end of its plan design and
configuration process; in other words, the rubber is hitting
the road.
With the MISSION Act implementation deadline behind us, the
Veterans Health Administration and the Office of Information
Technology appear to be reallocating personnel and executive
attention to EHRM, and that is very good news. VA just
completed the sixth of eight National Workflow Council
meetings. New technical obstacles are being identified,
especially with the data migration into Cerner and
interoperability in this mixed environment.
At the outset of EHRM, the team made ambitious promises to
migrate substantially more patient data into Cerner than DoD
determined was feasible in MHS GENESIS. That optimistic plan
seems to have run into technical difficulties. This is not a
foregone conclusion and there may be good reasons why; I hope
to get explanations for that this morning.
Relatedly, Cerner's Healthy Intent Population Health
Software seems to have morphed from a vehicle for feeding data
into the Millennium EHR to another repository of patient data
that clinicians may have to access alongside community. Without
a doubt, snags like this are inevitable in a project of this
magnitude. The timeline is getting tight, but the important
thing is that constraints are acknowledged and any tradeoffs
that must be made to resolve them are presented transparently.
On the other hand, everyone in VA always expected that
created the system interfaces between VistA and Cerner would be
a tall order. There are 73 different groups of interfaces
ranging in size and difficulty.
I am glad to see OIT assign more personnel, including some
of their very best people, to this effort. I want to know how
this work is being organized and whether it is being approached
in a manner that will reduce rather than add complexity in the
mixed environment. I am skeptical, though, that all the
technical constraints are known and there aren't more
intractable difficulties waiting to be discovered.
As we pass through September and the end of the plan design
and configuration process for EHRM, VA may be presented with a
choice, a choice to take the system live more quickly with
initial, some would say limited sets of capabilities, or
proceed more gradually with a complete set of capabilities. I
expect that decision to be made in VHA based on input from the
affected medical centers and I will support the decision
wholeheartedly if I believe it is made for the right decisions.
So with that, Madam Chair, I yield back.
Ms. Lee. Thank you, Mr. Banks.
I would now like to introduce the witnesses we have before
the Subcommittee today. Dr. Paul Tibbits is the Executive
Director of the Office of Technical Integration within the
Office of Information and Technology at the Department of
Veterans Affairs. Dr. Tibbits is accompanied by Charles Hume,
Assistant Deputy Under Secretary for Health for the Office of
Health Informatics, and Dr. Thomas O'Toole, who is the Senior
Medical Advisor both within the Veterans Health Administration,
as well as John Short, Chief Technology and Integration Officer
in the Office of Electronic Health Record Modernization.
I would also like to introduce Carol Harris, who is the
Director of Information Technology Acquisition Management at
the Government Accountability Office.
We will now hear the prepared statements from our panel
Members. Your written statements in fact will be included in
the hearing record. And, without objection, Dr. Tibbits, you
are recognized for 5 minutes.
STATEMENT OF PAUL TIBBITS
Dr. Tibbits. Good morning, Chairwoman Lee, Ranking Member
Banks, and Members of the Subcommittee. Thank you for the
opportunity to testify today about the Department of Veterans
Affairs IT modernization efforts, including the electronic
health record modernization and VistA, also the program you
mentioned earlier.
The Office of Information and Technology pioneered VistA to
support the clinical, administrative, and financial operations
of the Veterans Health Administration. Since its creation,
VistA has evolved into an enterprise planning tool, used by
multiple VA administrations. Today, VistA supports over 150
applications and the operations of more than 1500 VA clinics
and VA medical centers. There are 130 unique instances of VistA
nationwide that share core functionality, but are customized to
each VAMC's needs and populations.
VistA has served VA and veterans for over 40 years, but it
does not possess the modern capabilities that medical providers
and veterans deserve. VistA's required critical upgrades alone
could cost several billion dollars over the years and
maintenance costs are higher. It is not interoperable with the
Department of Defense, which keeps the health information of
servicemembers and future veterans; instead, VA staff must use
separate viewers to see the DoD data.
In May of 2018, VA awarded Cerner a contract to replace
VistA with Cerner Millennium, a commercial, off-the-shelf
solution currently deployed by the Department of Defense. VA is
working with Cerner to achieve initial operating capability and
deploy Cerner Millennium beginning in the spring of 2020 in the
Pacific Northwest.
As the nationwide Cerner rollout progresses, VA will
decommission VistA instances as necessary. However, during the
transition period, VA must maintain VistA to ensure current
patient record accessibility and continued delivery of quality
care.
The cost of sustainment. GAO's report projects VA will
spend $426 million to sustain VistA in fiscal year 2019. VA is
currently developing a methodology to update the cost data and
thereby define VistA, a recommendation in the GAO report.
We expect VistA to run without service degradation until
all VAMCs are running in the new electronic health record
solution. Sustainment costs during the transition include
development for new capability and interfaces, congressional
mandates, maintenance, and other costs.
The estimated minimum costs for VistA during the 10-year
transition period is $4.89 billion, excluding any new required
development.
Our long-term strategy. VA is leveraging more efficient
means of sustainment, including OI&T's shift to a development
and operations approach that develops, enhances, maintains, and
rolls out better products more quickly. VAMCs will be required
to run the nationally-released gold version of VistA, creating
a common set of software routines where possible.
OI&T follows VA's guidance on needed patches and upgrades
to VistA. These will continue as normal throughout the rollout
of Cerner.
The newly-formed Office of Technical Integration
facilitates communication and planning between program offices
that are implementing the systems to replace VistA. OI&T is
currently piloting a program to migrate all 130 instances of
VistA to the cloud.
In conclusion, until the new electronic health record
solution is implemented across the VA enterprise, VistA remains
VA's authoritative source of veteran data. Sustaining VistA for
the duration of the electronic health modernization ensures
that VA continues to provide uninterrupted care and services.
Madam Chair, Ranking Member, Members of the Subcommittee,
thank you for the opportunity to appear before you today to
discuss OI&T's progress towards VistA transition. I look
forward to continuing to work with the Subcommittee to address
our greatest priorities.
This concludes my testimony and I look forward to answering
your questions.
[The prepared statement of Paul Tibbits appears in the
Appendix]
Ms. Lee. Thank you.
Now Ms. Harris?
STATEMENT OF CAROL HARRIS
Ms. Harris. Thank you. Chair Lee, Ranking Member Banks, and
Members of the Subcommittee, thank you for inviting us to
testify today on VA's health information system, referred to as
VistA. As requested, I will briefly summarize the findings from
our report on this very mission-critical system.
VA provides health care services to roughly 9 million
veterans and their families and relies on VistA to do so;
however, the system is over 30 years old, is costly to
maintain, and does not fully support exchanging health data
with DoD and private health care providers. As such, VA has
work underway to replace the system with a commercial one;
however, the Department plans to continue using VistA during
its decade-long transition to the new system. This morning, I
would like to highlight three key points from our report.
First, VA lacks a comprehensive definition of VistA, but
additional work is planned that could address the gaps. To
maintain internal control activities over an IT system and its
related infrastructure, organizations should be able to define
the physical and performance characteristics of the system, as
well as the environment in which it operates.
VA maintains multiple documents and a database that
describes parts of VistA; it has also conducted multiple
analyses to better understand customization of the system
components at various medical facilities, yet the existing
information in aggregate does not provide a thorough
understanding of the local customizations reflected in about
130 versions of VistA that support health care delivery at more
than 1500 sites. According to program officials, the
decentralization of VistA's development is a reason why they
have not been able to fully define it.
Cerner's contract to provide the new electronic health
record system calls for the company to conduct comprehensive
assessments to identify site-specific requirements where its
system is to be deployed. Three site assessments have been
completed thus far and additional ones are planned. If these
assessments provide a complete understanding of the 130 VistA
versions, the Department should be able to define VistA and be
better positioned to transition to the new system.
Now my second point. VA believes VistA has cost $2.3
billion between 2015 and 2017, but this figure is neither
reliable nor comprehensive. VA can only reliably account for 1
billion of the $2.3 billion total. The source data for the
remaining $1.3 billion, which largely accounted for VistA's
infrastructure, related software, and personnel costs were not
well documented. As a result, VA's subject matter experts were
unclear on how to account for VistA versus non-VistA costs.
Furthermore, the Department omitted costs related to additional
hosting and data standardization and testing from the total
spend.
Given these issues, the Department is not in a position to
accurately report annual costs to develop and sustain VistA. As
such, VA lacks reliable information needed to make critical
management decisions for sustaining the many versions of VistA
over the next 10 years until Cerner is fully deployed.
My third point. VA has initiated a number of activities to
transition from VistA to the Cerner system. Among other things,
VA has taken steps to establish and staff a program office, as
well as form a governance structure. The Department's actions
in these critical areas are ongoing. Furthermore, additional
actions are in progress to address our recommendations from
September 2018 to clearly define the role and responsibilities
of the Joint DoD and VA Interagency Program Office.
As the Department continues to work toward acquiring a new
electronic health record, it will be important for VA to fully
implement the recommendation we made in our report for
improving the reporting of VistA costs. Doing so is essential
to helping ensure that decisions related to the current system
are informed by reliable cost information.
That concludes my statement and I look forward to
addressing your questions.
[The prepared statement of Carol Harris appears in the
Appendix]
Ms. Lee. Thank you. I would now like to recognize myself
for 5 minutes to ask questions and I will first start with Ms.
Harris.
In your report, you just stated that the VA identified $2.3
billion in VistA costs between 2015 and 2017, yet only--VA
couldn't demonstrate reliability on $1.3 billion of that
alleged VistA expenses. Can you explain the significance of
what that lack of reliability means?
Ms. Harris. Sure. Chair Lee, more than half of VA's
reported $2.3 billion costs couldn't be verified based on the
source data that we looked at in our review and this is of
concern, because without reliable information VA will not be in
a position to make critical management decisions about the
system and the system will be sustained for the next 10 years.
So that is the major problem.
Ms. Lee. In your opinion, based on your past work with VA,
do you expect the actual VistA-related costs to be more or less
than the $2.3 billion?
Ms. Harris. It will likely be more than the $2.3 billion,
because VA has omitted key costs from that 2.3 initial estimate
that they provided to us, things like additional hosting as one
example.
And just as an example of that with the additional hosting,
last June the VA told us the cost for this particular line item
was about $238 million per year. Shortly thereafter, they told
us that the cost was actually $950 million, and in the end,
they reported zero dollars per year. And so when we talked to
VA's subject matter experts, they agreed that the $950 million
was off base, but the fact that that additional hosting line
item was not included in the $2.3 billion estimate suggests
that the number is higher.
Ms. Lee. Okay. Thank you.
And the GAO, you made a recommendation in your report. And,
Dr. Tibbits, I would like to ask, will the VA concur with that
recommendation and how do you plan to address this cost-
reliability issue?
Dr. Tibbits. Great, Chairwoman Lee. Yeah, absolutely, our
concurrence is on the way in. I, in fact, saw the signed-out
version a few days ago. So, yes, we intend to fully concur with
the report and the recommendations.
I guess I should introduce here the notion of the
Technology Business Management framework, TBM. TBM is the
framework that we are using with to properly categorize and
classify information technology costs, we are working very
closely with OMB to implement that framework. Our fiscal year
2021, in September of this year, will be submitted in
accordance with that TBM framework.
As you might well imagine, a certain maturation will go on.
The first implementation of that might require additional
refinements and enhancements later on for sure, but we intend
to fully comply with that TBM standard and, in so doing,
address the GAO findings and recommendations.
Ms. Lee. Thank you for that. I am happy to hear that, but I
want to know, what has prevented the VA from implementing this
cost methodology in the past?
Dr. Tibbits. Well, let me separate my answer into two
parts. First of all, this cost methodology that I just
mentioned is relatively new as a commercial standard. It began
around 2012 and I don't remember exactly when between 2012 and
now, but somewhere in there OMB decided to make it a Federal
standard. I don't know exactly when that happened, though, but
I would say the TBM standard itself is relatively new. That is
one part of my answer. The other part is, fiscal discipline
with respect to information technology has been evolving over
time; we are very interested in improving it all the time. We
have been on a trajectory to try to improve it over time,
hence, we fully agree with the GAO recommendations.
Some of the methodology we have used, for example, on the
personnel cost that the GAO representative mentioned, we have
not up to now seen the need, I guess I would say, to classify
personnel costs by system. So we have personnel costs and we
have system costs, but mapping personnel costs to system costs
is not something we have done up to now. So we will in the
future, obviously, consistent with this TBM framework, but that
is a matter of those mappings and things that just were not
considered necessarily high priority at the time. I can't tell
you further why that was, it is just about I have exhausted my
knowledge on the subject.
Ms. Lee. All right, thank you. And I am out of my time and
I now recognize Congressman Banks.
Mr. Banks. Thank you, Madam Chair.
Dr. Tibbits, in your testimony you seem to have adopted
figures that GAO says is unreliable: four hundred and twenty
six million dollars to sustain VistA for 2019 and $4.89 billion
over the next 10 years, which is roughly ten times the 2019
number. Do you stand by the VistA cost information that VA gave
to GAO?
Dr. Tibbits. Yes, it is the using the--for the parts that
GAO is referring to that are unsubstantiated, we had to use
some form of estimation methodology; we did that, and it is the
best we can do at the time up to now. That will certainly
improve over the future as we move further into implementation
of this TBM framework, but--
Mr. Banks. Ms. Harris--
Dr. Tibbits [continued]. --those are the best numbers we
have at the time, yes.
Mr. Banks [continued]. --do you have a response to that or
anything to add to that?
Ms. Harris. The number that was reported, the $2.3 billion
number, was never intended to be projectable, because it is
not, and the $2.3 billion number is not reliable; only 1
billion of that figure was found to be reliable. So the
projections that Dr. Tibbits stated does not come from the GAO
report.
Mr. Banks. Okay, interesting.
Dr. Tibbits, the purpose of figuring out how much VistA
costs is to compare it to Cerner, but I don't see VA making
much effort to argue that EHRM is going to save money, all
things considered. Is there ever going to be a business case
demonstrating savings even over the very long term or is that
just unrealistic?
Dr. Tibbits. Well, obviously, with--first of all, with
respect to the TBM framework, again, certain Cerner costs will
be incorporated into that TBM framework. So, from a
transparency perspective, it will be included in all of our IT
reporting. That said, the major motivation for going to Cerner,
as I think all of you are aware of from the determination and
findings, is to strengthen information interoperability with
the Department of Defense.
So, yes, what the cost will turn out to be is very
important. We certainly will make a great effort to make that
very clear to whoever needs to know what that is, for our own
internal management purposes as well. But, as I say, it is the
well-being of the servicemember and veteran that is our
principal motivation for going to Cerner, not necessarily an
economic argument.
Mr. Banks. Okay, let's move on.
Mr. Short, has it been decided whether to keep VA's My
Healthy Vet patient portal and integrate it with Cerner, or
adopt the Cerner patient portal and integrate it with VistA, in
the mixed environment?
Mr. Short. Sir, at initial IOC go-live we will be rolling
out the Cerner patient portal the same as DoD rolled out. While
we are doing that, there is the initial enhancements going on
in the patient portal for all the requirements that the
Connected Care My Healthy Vet team has laid out with our
program office.
In addition to that, Dr. Kroupa, myself, and the Office of
Connected Care are doing a review currently and we will over
the next couple months on what is the final answer to your
question, and that is, will we integrate into My Healthy Vet or
will we take all that functionality and put it in the
commercial platform to make sure that it is a seamless,
integrated view for the veteran.
Either way that goes, it will be integrated into the VA.gov
portal, so all the veterans can go to one place, have one
experience to access their health care.
Mr. Banks. Okay. So, Dr. O'Toole, is VHA confident that the
Cerner patient portal can integrate with VistA in all respects
and meet your needs?
Dr. O'Toole. The driving force for all of this is to
ensure, particularly during the IOC Block 1 and Block 2
implementations, is that the local facilities and local
facility leadership and front-line providers are going to be
comfortable with the interface with Cerner and that front-line
clinicians will feel confident that patient safety will not be
compromised, and that is really our driving force.
To date and through the workshop processes and through the
local workshop efforts, all indications are that the patient
safety and patient care will not be compromised and will be
done efficiently, but this is something that we are monitoring
closely, and this is something that clearly is of highest
priority moving forward.
Mr. Banks. Okay. So, Mr. Short, I would be remiss if I
didn't ask you about the firm, it still has not been
established. When is this supposed to happen? And, given the
continued delay, how has the timeline for it to evolve into its
various stages of operating capability changed?
Mr. Short. Sir, I can tell you that there is continual
meetings on a weekly basis with DoD and VA. There may be a week
or two here and there because of schedules that they did not
meet, but routinely they meet on a regular basis and they are
continuing to make progress. I know that some of the dates and
announcements haven't come that the Hill has requested. I will
have to take that question for the record; I don't have any new
dates.
Mr. Banks. My time has expired.
Ms. Lee. Thank you.
I now recognize Mr. Lamb for 5 minutes.
Mr. Lamb. Dr. Tibbits, I think it is a little hard for
veterans in particular to understand how we are going to spend
$5 billion over 10 years on a legacy system that we are trying
to replace when the cost of the new system is $10 billion. I
mean, essentially, we are spending half of what we are doing on
the new system to just maintain the old one and that may not
even represent all the costs.
So can you explain to me how I can explain to veterans in
my community, what are the drivers of that cost to maintain and
upgrade and sustain VistA over the next decade? What is making
us spend the bulk of that $5 billion?
Dr. Tibbits. Yes, certainly. Thank you for your question.
Well, first of all, I think everyone understands and we
have had this, I think, out there for broad-based
understanding, the complexity of VistA itself due to its age
and that complexity drives costs. So understanding the
interconnections--understanding and deal with the
interconnections inside of VistA, understanding what to put a
new capability into VistA, understanding break-fix work in
VistA is complicated. However, to your point, that high
maintenance cost, if you will, is part of the concern of what
led us to the conclusion on top of the information
interoperability, which was our primary reason, but getting out
of the complexity and costs driven by that complexity is part
of the reason why we wanted to move out of VistA.
That said, it is a 10-year period. We have to account for
time to learn lessons as we go through this implementation
process and at the same time, as you recognize, we have to
continue to deliver quality care to veterans at the sites that
have not yet received Cerner--
Mr. Lamb. But I guess are there any particular tasks or
contractors that drive that $5 billion cost more than others? I
mean, it is one thing to say it is complexity, yes, I
understand that, but how does it end up being $5 billion? It
just seems like so much.
Dr. Tibbits. Well, that is not dramatically different in
any way than our past experience and I would say, no, there is
not any particular one contractor. The answer to the veterans'
question that you asked me is to maintain quality service for
those veterans at the places that haven't received Cerner, that
is the bottom-line answer to a veteran. But, no, it is not one
particular contractor; it is the overall complexity.
We have a network, a mosaic of contractors that are
supporting VistA, keeping it up and running, and we--I guess I
should hasten to add here, however, our migration to the cloud
for VistA, we are anticipating cost savings from that migration
to the cloud, which the first instance we have now successfully
completed. So we believe that the remaining will be an equally
efficient and effective migration. That will serve to keep the
ongoing maintenance costs under control, I guess I can say.
Mr. Lamb. Okay. Ms. Harris, I know this is an issue you
have stayed with for a long time, the EHR implementation and
everything, was this foreseen 5, 10, 15 years ago, whenever?
Did we understand in the past what we were spending on VistA
and was that used as an argument that maybe we should have
started this whole replacement earlier? Can you give me a
little bit of the history on that?
Ms. Harris. Well, with regards to VistA, I mean, even at
this time right now, VA is unable to draw a circle around it
and that is something that has persisted over the past 10--
since the inception of VistA, because of the decentralized
nature of how VistA was developed. And as a result of that
decentralization, which began in the '80s, VA is not in a
position to be able to at least effectively draw that circle
and that perimeter around what is and isn't VistA, and, as a
result, they aren't able to accurately report the annual
development and sustainment costs.
So, because of that lack of, I guess, management in the
beginning where there was a disciplined approach to
understanding and documenting the physical and performance
characteristics of the system, that is why they are in the
position that they are in at this time. And the inability to be
able to draw that perimeter is why they don't have accurate
costs and why at this time they don't have an accurate basis
for an ROI as to, you know, for moving to the Cerner system.
Mr. Lamb. Thank you.
I am out of time; I will yield back.
Ms. Lee. Thank you. I will now ask a few more questions.
I wanted to follow up with Ms. Harris. This TBM methodology
that Dr. Tibbits discussed, do you believe that this approach
will be sufficient?
Ms. Harris. I do not believe so. Until VA can fully define
VistA, they will not be in a position to be able to accurately
report the costs. I think the two go hand in hand and the
definition of VistA is foundational. So, whether they use TBM
or another type of methodology, the core issue remains that the
definition of VistA is not fully defined and that is the
problem.
Ms. Lee. And can you be a little more specific when you
base it--you know, it sounds like just defining the nature of
the beast is the real issue here--just improving that accuracy,
what do you foresee needs to be done?
Ms. Harris. Understanding the 130 versions of VistA, the
performance characteristics, as well as the environment in
which those instances of VistA are operating. So having those
clearly detailed and defined, adequately defined, is critical,
so that is what I mean.
Ms. Lee. All right, thank you.
Dr. Tibbits, you have a plan to transition into the Cerner,
but also continuing to support VistA. Why is--you are making
this plan for the transition, but after you have already begun
the implementation, why is that?
Dr. Tibbits. Well, I would say right now what we are
doing--and I will ask John Short to elaborate in a moment, but
actually the thinking and planning for that transition began
long ago when the determination and findings was written and
the Department decided to go in this direction for a lot of
reasons, which I will skip over right now, but the principal
one being information interoperability for the benefit of
servicemembers and veterans.
So the planning itself began, what IO--the proximity to the
initial operating capability, as we get closer and closer to
that, interact more and more with Cerner itself, with the
health care professionals in VHA, we are learning more as we go
long. We have demonstrations, you have heard already about the
workshops, we have had six of them already. So those are
intensive interactions with respect to understanding clinical
workflows and all those things, data migration, et cetera, all
that is going to go into the actual documented plan.
So there is a lot of learning that has had to happen in
order to actually put a pen to paper on a document called a
plan, a pivot, we call it a pivot plan, but the process of
thinking and gathering the information to do that has been
going on for several years already during this entire ramp-up
leading to the award decision, the award, and now the
interactions with Cerner.
Ms. Lee. Yeah, I guess, you know, my concern is looking at
all of the costs. Like we have a cost estimate that Cerner is
going to cost $10 billion, you have a cost estimate of $4.8
billion to maintain VistA. We don't have any confidence, A, in
what VistA actually entails, so I don't think we any confidence
in that $4.8 billion. But then, more importantly, that makes me
have less confidence in the $10 billion estimate for Cerner as
well, and at what point do we sit down and really lay out
exactly what this is going to cost? I mean, money does not grow
on trees. So at what point do we lay out exactly what the costs
are?
Dr. Tibbits. So I think, as Ms. Harris pointed out earlier,
part of this learning process, you mentioned I think the on-
site surveys that are done in anticipation of the Cerner
rollout, they are called current-state reviews. So, at those
current-state reviews happen, certainly in a very definitive
way we will understand everything about VistA interfaces and
everything else at that site in anticipation of Cerner being
implemented at that site.
So, as the waves roll forward, we will become more and more
definitive about the cost estimates that we have to live with
now. So that process is ongoing. As the GAO pointed out, we
have conducted that process already at the first three sites.
We are very confident that that process is going to yield very
complete information based on the actual experience we have
with it to date.
Ms. Lee. Thank you. I am out of my time and I will now
recognize Ranking Member Banks.
Mr. Banks. Thank you, Madam Chair.
Mr. Short, please give us an update on the data migration.
What data in terms of types and magnitude do you currently plan
to migrate into Cerner?
Mr. Short. Sir, the data we have identified was identified
by the Chief Medical Officer and her clinical staff working
with VHA. All the clinically relevant data, which includes 73
billion records--let me explain what a record is: it is an
encounter, a lab report, a vital sign, each one of those is an
individual record in VistA. So, initially, the initial load
from VA to Cerner is 77 billion of those records. The oldest
one is back from the early '80s, a lab report, and we can give
you more details on that for the record, if you would like. Of
those, in terms of 21 different clinical domains that were
identified by VHA and CMO office, those records moved from VA
to the Kansas City data center, into a data repository, in
preparation for loading into the Cerner Healthy Intent
platform. So, over the next 30 days, it is intended to move
that data into the Healthy Intent platform.
When we go live at a site, the current plan is for March
2020, the initial set of data domains that would be available
would be ten of those 21 inside the EHR itself, but all 21
clinical domains will be available to the clinicians and other
caregivers, MVBA, as needed, in the Healthy Intent viewer. So
they will have the long record, all records available from DoD
and VA that are in Healthy Intent, they will be able to see all
those in the Healthy Intent viewer, and the initial clinically
relevant records that they have prioritized for go-live will be
in the EHR.
Mr. Banks. So will all of this patient data be accessible
in Millennium or will a user need to look in another system
such as Healthy Intent?
Mr. Short. The CMO office in VHA determines some of the
domains they don't want in Millennium. The initial set of data
that will go in Millennium is ten data domains, within 5 to 8
months after we go live, we will add additional data domains.
So at that point 18 of those 21, the most clinically relevant
ones that they want in Millennium, will be in there.
Additionally, they have identified to have 3 years of
records as the baseline that they want in there. For different
purposes and reasons, that is the baseline they determined, and
they briefed to the Under Secretary of Health's office.
And so that way all those records will be in Millennium to
trigger clinical decision support and other information.
However, if they need to pull in additional data further back,
they can do that, or they can just view it in Healthy Intent.
Mr. Banks. Okay. How many of the VistA-to-Cerner interfaces
have been completed now and when is the deadline to complete
all of them? My understanding is that this deadline has come
sometime before the go-live deadline.
Mr. Short. Sir, there is 73 go-live minimum interfaces,
system interfaces required; of those, there are a number of the
interfaces that were already completed that we are reusing from
DoD and a number of them from a commercial. So 12 of those
system interfaces were already developed for DoD, so except for
the testing in the end for VA use from the user level, those
are complete.
And then, additionally, there is 25 interfaces that are
commercial system interfaces that they are going to be able to
reuse. And so, except for the testing and then validation by
the user, those are already complete because they are reusing
those.
Mr. Banks. Okay. What is the deadline to determine which
VistA modules get replaced by which Cerner's software package
or other companies' software, and which VistA modules have you
yet to determine a plan for?
Mr. Short. So all the clinical VistA modules with the
exception of prosthetics will be replaced by the Cerner
platform between the initial go-live and the IOC period. At the
initial go-live, the different modules that will either be
integrated versus replaced is being determined over the next 2
weeks. Dr. Kroupa, CMO for OEHRM, is meeting with Spokane and
Puget Sound functional staff and facility directors to go over
the 313 Cerner capabilities and validating which ones they will
have at go-live. And at that point we will know whether it
would be two or five modules of VistA we will still integrate
with, but by the end of the IOC exit it will be either one or
no VistA models clinically relevant that we will use.
Mr. Banks. All right. Dr. O'Toole, really quick, what is
VHA's expectation for the Cerner data from the early sites
coming back into VistA at the later sites? In other words, how
seamless should the view of patient data be for VistA users in
the mixed environment?
Dr. O'Toole. The expectation is that it is possible that
one will need to use multiple systems in the context of a
clinical encounter, whether it is looking at past chest X-rays
to determine, you know, how things looked previously, or other
clinical examples of that sort. The challenge for us, though,
is to ensure that it can be done efficiently, whether it is
going to the joint legacy view or other mechanisms, or being
able to look at the Cerner interface. This is what the Spokane
and Seattle IOC visits are going to be looking at within this
context of specific clinical scenarios and clinical needs to be
able to determine if it could be done efficiently and timely.
And, if it can and it is sanctioned and agreed to by local
leadership and front-line providers, then it will be
proceeding, but the expectation is that there will be clinical
scenarios where both interfaces are going to be needed.
Mr. Banks. All right. My time has expired.
Ms. Lee. Thank you. I would now like to recognize Mr.
Watkins for 5 minutes.
Mr. Watkins. Thank you, Madam Chairwoman.
Ms. Harris, your testimony indicates the VA could not give
you accurate numbers as to the costs to maintain VistA, because
there is not an adequate methodology to determine the costs
belonging--what costs belong to VistA. What kind of methodology
does the VA need and how is it going to be developed?
Ms. Harris. Mr. Watkins, thank you for the question. So the
finding that we had was that VA lacks a documented methodology
for accounting for what is and isn't VistA. We don't have any
recommendations related to the type of methodology that is
necessary, but what is most important is that, whatever process
that they choose, that it is documented and vetted throughout
the organization.
Mr. Watkins. Okay, thanks.
Dr. Tibbits, where are you in the process of developing
this methodology?
Dr. Tibbits. Yes. As I said earlier, we completely agree
with the GAO report and the representative's current remarks.
I did mention earlier TBM and, as indicated in the prior
discussion, TBM is only part--the Technology Business Manager
framework was only part of the answer; the definitional
boundary of CHS is clearly an important part of the answer as
well. The two of those combined together is what is going to
wind up with being our methodology.
I would say, in our response to GAO, we have indicated that
I think at the next update, I believe that is 120 days from
now, we will have a final answer as to what that methodology
will be.
Mr. Watkins. Thank you. Dr. Tibbits, your testimony
references a pilot program to move VistA data to the cloud.
Apparently, this has already been successfully accomplished at
one location. What is the scope of this pilot program? How much
VistA data are you considering eventually moving to the cloud?
Dr. Tibbits. So, let me be clear, it is not just VistA
data. We are moving VistA in its entirety, so the ultimate
scope of whatever instances of VistA remain operational as the
Cerner platform rolls out. So, as things stand today, the scope
would be 130 instances, but by the time we get VistA actually
moving and Cerner rolled out, it is probably going to be a
smaller number than that. The initial wave we are envisioning
right now is 70, seven zero, 70. Because of their current
location, the DoD facility, which is closing, we have to make
sure we get those initial 70 moved first, because there is a
date certain by which that facility will close.
Mr. Watkins. And how long and how much will it cost to move
all 130?
Dr. Tibbits. I will have to get back to you on the exact
cost figures. And we do have a schedule, again, driven by the
DISA data center closure. I just happen not to remember that
date right now, I will be happy to get that back to you, but
the schedule for that first 70 is absolutely fixed because of
that first closure date by DISA.
Mr. Watkins. So I have got to yield my time.
Ms. Lee. Mr. Hume, we have heard from the VA on multiple
occasions that Cerner's Millennium will only replace 60 percent
of VistA's capabilities or functionality, and then that the EHR
may have to link back to VistA to fulfill the other 40 percent.
Can you address what functions make up this other 40 percent?
Mr. Hume. Yes, ma'am. The bulk of those other
functionalities are being replaced by other modernization
systems, the financial management modernization system and the
supply chain modernization with the Defense Medical Logistics
Standards support system. I will defer to Mr. Short for the
details, but there is a small percentage of capabilities beyond
that are not being replaced by one of those three modernization
systems and we are in the process of identifying the solution
to that. It may be an interface to VistA for some time, a
replacement by a commercial product; we have yet to work that
out.
Mr. Short, do you want to comment?
Mr. Short. Yes, thank you.
Ma'am, initially at IOC go-live, five to seven of the VistA
clinical modules will be interfaced to, but the IOC exit the
plan is to only have a dependency on one VistA modules being
prosthetics and the solution for that, Cerner is developing
additional clinical content and some IP development to make
sure that all the nuances of prosthetics that VA has could be
added to their platform, which will be beneficial to anyone
else using that platform as well.
The other portions of VistA, the other 40 percent, a large
portion of that are base core functionalities of VistA, it has
nothing to do with any functionality at all. Like an XML
parser, you know, like to be able to split out data, that is
something that only if you need to use a system is that
capability necessary, like an operating system is only
important for an application. So those things go away when the
application functionality goes away.
The other items Mr. Hume mentioned are business systems,
accounting, acquisitions tracking, not medical-related, but
tied into health care.
Ms. Lee. So just thinking about the costs. So you have
these other capabilities, you have plans to modernize or
replace those capabilities, where is that cost coming from? Is
that included in the $10 billion that we have planned for
Cerner, is that outside of it? Is that part of the $4.89
billion projected for VistA? Where are those costs coming from?
Dr. Tibbits. Well, Madam Chair, if I understand your
question correctly, with respect to the major efforts that
address the 40 percent, FMBT, Financial Management Business
Transformation, that is our ERP replacement, and DMLS, which is
our supply chain modernization, they have their own cost
boundaries and cost definitions. So that would not be part of
the VistA boundary--
Ms. Lee. So it is in addition?
Dr. Tibbits [continued]. --no--or the Cerner boundary, no.
Those are all--
Ms. Lee. But it is not included in your $5 billion--
Dr. Tibbits. Correct.
Ms. Lee [continued]. --to maintain VistA. So this is we
have another cost on top of that to take care of this 40
percent?
Dr. Tibbits. Right. Those are programs of record and have
been in our budget submission now for a few years, the ERP
replacement, FMBT, and DMLS, yes, those are separate programs
already included in our budget submissions.
Ms. Lee. Okay. So just a question, then will Cerner be
responsible for addressing any of this 40 percent, or is this
all being taken care of?
Dr. Tibbits. No, the 40 percent are the other systems.
Ms. Lee. The other stuff?
Dr. Tibbits. So that is FMBT, Financial Management Business
Transformation, DMLS, and then the remaining things that John
Short just talked about, which might actually no longer be
needed at all, some technical things, XML parser and whatnot.
So, no, the Cerner is the 60 percent part of the question.
Ms. Lee. Okay, all right. So at go-live, how is the VA
going to address these capabilities in Cerner that are not
going to meet clinical needs such as prosthetics and where
there is no alternative product?
Dr. Tibbits. So I am going to ask perhaps Chuck Hume to
comment on that in a minute. The prosthetics community, of
course, is working very intensively with us. I have personally
sat in on many of those meetings. I think the short-term
approach, if I can say that, is to maintain a prosthetic system
and build an interface over to that prosthetic system, until
such time as that functionality is adequately developed and
represented in the Cerner product itself.
So, as I think all of you are well aware of, prosthetics is
a very well developed, very sophisticated capability at the VA,
not something that Cerner necessarily encounters to that extent
in their commercial practice, and so it is not surprising to us
that they have to beef up that capability. But, in the
meantime, I believe our short-term answer is to maintain our
prosthetics system and interface that as necessary.
Ms. Lee. Thank you.
I now recognize Ranking Member Banks.
Mr. Banks. Thank you, Madam Chair.
Dr. O'Toole, I want to make sure that I understand the data
migration answer that we discussed a little bit ago. Are you
saying that the VHA physicians don't want all patient data to
be in Millennium?
Dr. O'Toole. No, sir, I am not saying that. I think the
issue is some--as we roll out and, obviously, with the
staggered rollout across sites, and for veterans who may be
migrating across systems, there may be instances where data may
not initially be available on the Cerner platform, but it is
available on the legacy platform, particularly longitudinal
data going back. And from a clinical perspective and seeing a
patient where having that longitudinal history is going to be
necessary to provide their care, it is going to be important to
be able to have access to both the legacy systems, as well as
the current systems of care. So it is not an issue of
preference, it is a matter or issue of practicality and good
care.
Mr. Banks. Okay. Mr. Hume and Mr. Short, how many other
technology projects in VA have dependencies with EHRM? And can
you list them, if you can, and tell me who is responsible for
each set of dependencies?
Mr. Hume. Well, the predominant systems would be those we
talked about, the financial management modernization and the
supply chain modernization, each of those programs. The
immediate relationship is with the supply chain modernization,
the Defense Medical Logistics Support System, that system is to
roll out to the sites that are modernizing to Cerner 4 months
in advance of that, so that we can make sure that those
interfaces are functioning.
We are fortunate that we are adopting the Defense Medical
Logistics Support System, which they have already interfaced
with Cerner as part of their rollout under MHS GENESIS.
Mr. Short, do you want to add anything?
Mr. Short. The two programs Mr. Hume mentioned, Terry
Riffel and Harry Oland are the two people, the first FMBT and
second one the DMLS, that are the SES executives over those
programs. So both those programs have a dependency on some of
our functionality and OHEM has a dependency on theirs. OHEM
also has a dependency on the joint legacy viewer during the
transition period, because there is some functionality that for
some work-arounds until all capabilities are released and
tested and validated that they will need to use the joint
legacy viewer at the transitional sites.
There are some ancillary systems that we have some
dependencies on, and we can take that for the record and
document that for you.
Mr. Banks. Okay. Mr. Short, I read the Secretary a letter
last month about patient matching. As you know, it is key to
quality and interoperability. I appreciate the thorough
response, but I would like you to explain one of the
statements. It says, quote, ``A single EHR solution between VA
and DoD will guarantee 100 percent patient matching within the
new EHR solution for servicemembers and veterans,'' end quote.
Does that pertain to VA and DoD or VA and the MISSION Act
providers?
Mr. Short. Sir, I am not sure if it pertains to the MISSION
providers, I would believe it pertains to the first, DoD and
VA. I can get back to you for the record on the second
question.
To answer part of that question, the Joint Patient Identity
Management Service that we developed with DoD and we have
tested out, what we have used to make sure that we have
maintainability, we have a single EHR with an overlapping
customer base, as you can imagine. DoD and VA and beneficiaries
and veterans, servicemembers can go back and forth, Active duty
members are seen at VA hospitals at times, et cetera, you could
have a mismatch if you had different identity systems saying,
no, this is John Short or that is John Short. And so by having
one system with everything worked out in the background
maintains we do that.
But for the record, on the other part of your question, I
will take that back.
Mr. Banks. Okay. Last question. What is VA's--Mr. Short,
for you as well--what is VA's goal for patient matching with
the MISSION Act providers in Cerner and how are you going to
achieve it?
Mr. Short. Our goal is to have complete patient matching to
ensure that everything is completely safe, accurate for every
patient, that the veterans that deserve care get the care, and
get the right care and the right prescriptions. So, for the
record, I can take it back on our plans; I don't have that with
me today.
Mr. Banks. Thank you very much. I yield back.
Ms. Lee. Thank you.
Dr. Tibbits, there are many entities outside of VA using
VistA that have agreements, like OSEHRA and World VistA. This
Committee has heard from several of these groups with concerns
about the future of their access to VistA code and possible
future innovations. And I wanted to ask you, how is the VA
leveraging outside experience through these groups to further
the instances of VistA?
Dr. Tibbits. Well, as we mentioned earlier, first of all,
maintaining VistA over the 10-year roll-out period of Cerner is
very important to us, critical to veteran care. So we are going
to continue to focus on doing that. I can say that in the past
from the open-source community we have certainly obtained very
valuable contribution to FileMan, which is the underlying
database in VistA. How that relation--so there have been
additions and actually that FileMan upgrade was a substantial
one, not some minor tweaks, from the open-source community--how
that will play out in the future, I am not sure I know enough
to exactly tell you that yet, other than we will continue to
maintain VistA for the roll-out period; number two, we will
continue to make available whatever the VistA code is at that
point in time to those communities, we have no reason to stop
any of that.
Since there is a 10-year roll-out period and since the
roll-out process is geographic, not functional, the additional
functionality and patching will have to continue for the
majority of that 10 years until the last site gets turned off.
So, with respect to those outside entities that are using
VistA, they certainly have plenty of time to prepare for what
might eventually happen 10 years from now, it is not going to
be a surprise to them in any way.
Ms. Lee. Do you have an agreement; do you have any
licensing agreements with those groups and is there like a stop
date at 10 years?
Dr. Tibbits. Licensing, I think I am going to--we would be
best advised to take that for the record. Licensing is very
complicated when you get into Apache II licenses and commons
and all that sort of stuff. So OSEHRA is quite expert at
license management. I think we should take that for the record
and get back to you on the license questions.
Ms. Lee. Okay. Thank you.
Dr. O'Toole, while the VA is using the electronic--the dual
records, what clinical impacts are expected and tolerated, and
which ones would be unacceptable?
Dr. O'Toole. Thank you. It is an extremely important issue
and challenge for us. I think the expectation is that there
will be workflows that require dual system use for different
clinic scenarios. The challenge point and the things that we
are going to be looking for are, one, clearly, how will that
impact in terms of efficiency of patient care and the amount of
time that it takes to care for a patient within those clinical
settings. The expectation with the initial IOC roll-out sites
is that clinical time needs to be extended for each clinical
visit to ensure that adequate time is made available. We are in
the process of expanding the traveling nursing corps at our IOC
sites to enhance the staff capabilities there in order to
ensure that.
The biggest challenges and the biggest risks to us, I
think, are really related to complex clinical scenarios where
patients may be migrating across multiple settings or where
longitudinal care is critical to clinical decision making, and
that is something that we are in the process of looking at very
closely within the context of the IOC capabilities to ensure
that those workflow processes are identified in advance, that
clinicians up front know what to expect and what the workflow
processes will be, but it is something we will be monitoring
and watching very closely through this process.
Ms. Lee. Thank you. Just one last question.
Ms. Harris, obviously, there are a lot of uncertainties in
the potential solutions that we are hearing today and, from a
management perspective, do you have concerns and is the VA
taking on risk that it may not be aware of, in your opinion?
Ms. Harris. Well, we have ongoing work for the Subcommittee
related to the transition plans and activities that are
underway. I think that having effective plans is a very
critical thing and having plans that are at the right level of
detail is certainly very critical.
I think that one of the things that we have some questions
about at this time relates to the clinical workflows and when
that will be completed and the level of granularity of those
workflows in time for the IOC deployment. The timing of those
two activities is something that we have some questions on and
whether the VA will be in a position to be able to complete
those workflows in time for the deployments at those IOC sites,
that is something that we have some questions about at this
time.
Ms. Lee. Thank you.
Well, this now concludes the Subcommittee hearing. I wanted
to thank all of the witnesses for being here today, thank Ms.
Harris for your report. We are heartened that the VA will take
the recommendation of the GAO and has begun implementing the
methodology, and we look forward to having transparent updates
as we go along.
From my point of view, you know, continuing, Mr. Short, a
lack of plan on joint governance continues to be a problem with
the rollout of this program and our lack of having knowledge of
what the plans are, when we can expect to see a joint
governance really continues to concern us. And it is really,
mostly for me about the risk of the rollout in this contract. I
mean, this was a fixed-price contract, VA implemented it with
indefinite deliverable, indefinite quality, which really would
have shifted a lot of the risk onto the contractor, but with
lack of knowledge of really what the extent of VistA is, to me,
shifts a lot of that risk back onto the VA.
And when we start to talk about the cost, you know, the
billions and billions of dollars of cost of this project, I
just have concern and I hope that we can continue to have some
transparency as we roll out. And when we get to specific
decision points, to be able to stand up and make the proper
decision based on the status of where we are at the time would
be my hope as we move forward, especially given the track
record that we have had in trying to update VistA multiple
times in the past. And ultimately, you know, improved health
care for our veterans is really the focus that we all and I
know, Dr. O'Toole, we are all focused on, and obviously the
interoperability being the number one objective in this
rollout.
And so as we move forward, again, we thank you all for
being here and continue to want to have that transparency, so
we can make sure that ultimately, we are delivering the best
care possible to veterans in our country. And thank you all for
being here.
And I would like to thank the witnesses. I hope that we
will work together with this Subcommittee as we continue this
oversight.
All Members will have 5 legislative days to revise and
extend their remarks and include extraneous material. And this
hearing is now adjourned.
Thank you.
[Whereupon, at 11:23 a.m., the Subcommittee was adjourned.]
A P P E N D I X
----------
Prepared Statement of Paul Tibbits
INTRODUCTION
Good morning Chairwoman Lee, Ranking Member Banks, and
distinguished Members of the Subcommittee. Thank you for the
opportunity to testify today about the Department of Veterans Affairs'
(VA) IT modernization efforts, including the Electronic Health Record
Modernization (EHRM) initiative and the Veterans Health Information
Systems and Technology Architecture (VistA): the system at the center
of that effort.
I am accompanied today by Charles C. Hume, Assistant Deputy Under
Secretary for Health Informatics, Veterans Health Administration; John
Short, Chief Technology and Integration Officer, Office of Electronic
Health Record Modernization; and Dr. Thomas O'Toole, Senior Medical
Advisor, Office of the Assistant Deputy Undersecretary for Health for
Clinical Operations, Veterans Health Administration.
OVERVIEW
VA is committed to providing exceptional care, services, and a
seamless, unified experience to our Veterans. The Office of Information
and Technology (OIT) collaborates with various VA offices to achieve
this mission through the delivery of state-of-the-art technology,
including a modernized Electronic Health Record (EHR).
VA was an early pioneer of the EHR. We developed VistA to support
the clinical, administrative, and financial operations of the Veterans
Health Administration (VHA). Today, VistA and its integrated systems
provide an integrated EHR for Veteran care and services. It supports
over 150 applications, including the operations of more than 1,500 VA
facilities, from small outpatient clinics to large VA Medical Centers
(VAMC). There are 130 unique instances of VistA nationwide at four
Regional data centers, apart from Manila which has an onsite instance.
Each of the 130 VistA instances share a standard core of functionality
but are customized to each VAMC's needs and patient population. VistA
is also enhanced by many third-party commercial off-the-shelf (COTS)
products which further customizes the environment. One instance of
VistA, at Valley Coastal Bend, was successfully migrated to the cloud
on June 22, 2019, which is the future direction for VistA instance
maintenance until they are subsumed by Cerner Millennium.
Like any IT system, VistA requires updates and maintenance to keep
it functioning at a high level. Critical upgrades to the system could
be extremely costly over the years, and maintenance costs are even
higher. Often, it becomes more expensive to maintain a legacy system
than to replace it.
VistA has served VA and Veterans well, but after nearly 40 years in
operation, we are also aware of its limitations. It does not possess
the modern capabilities, analytics, and functionalities that medical
providers and Veterans expect and deserve. It is not interoperable with
other Federal records systems, including those at the Department of
Defense (DoD) which contain the health information of Servicemembers
who will eventually enter our system as Veterans. Instead, VA staff
must use a separate viewer to see DoD's data and yet another system to
provide allergen and medication alerts to VistA.
To modernize VA's legacy EHR systems and achieve interoperability
with DoD and community care providers, VA decided to transition to a
new EHR solution. In May 2018, VA awarded Cerner a contract to replace
VistA with a COTS solution, Cerner Millennium, which is also currently
being deployed by DoD.
VA is working with Cerner to achieve Initial Operating Capability
(IOC) in the Pacific Northwest, where DoD has already deployed the MHS
GENESIS system, which is at its core, Cerner Millennium. Beginning in
Spring 2020, VA will deploy its new EHR solution in that region.
Through the IOC period, VA will maximize efficiencies by building upon
lessons learned from DoD. VA will then deploy its new EHR solution
across the VA enterprise. During implementation of the new EHR
solution, VA will need to maintain VistA systems for a period of time.
This ensures that current patient records remain accessible and that
there will be no interruption in the delivery of quality care.
Keep in mind the Pacific Northwest region is only a small fraction
of the VistA ecosystem. Instances occur across the country and it's
even more important during the pre-deployment reviews that VA
identifies the unique differences to effectively reach IOC on schedule.
OIT has completed infrastructure readiness assessments for the IOC
sites. More importantly, VistA is not only an EHR system; it is a
complex system more like an Enterprise Resource Planning (ERP) with a
variety of capabilities and functionalities, including financial,
administrative, and supply chain management functions. It supports not
only VHA but may be used by the Veterans Benefits Administration (VBA)
and the National Cemetery Administration (NCA).
FACILITIES USE OF VISTA DURING EHRM
For the aforementioned reason, VA can only fully retire VistA when
every capability and functionality used by a facility is replaced by a
modernized replacement system, whether by Cerner or additional systems.
During the transition to VA's new EHR solution, VA facilities will
continue to use their instance of VistA. VA is undertaking several
concurrent modernization projects such as the following:
Defense Medical Logistics Standard Support (DMLSS), a
system that will manage all VHA supply chain functionality except for
pharmacy, patient specific prosthetics, and possibly IT equipment; and
Financial Management Business Transformation (FMBT),
which will replace VA's current Financial Management System.
COSTS OF SUSTAINMENT
For the purposes of ensuring uninterrupted health care delivery, VA
will continue to use VistA until all legacy systems are replaced by the
new solution. It currently costs VA $426 million to sustain VistA
through Fiscal Year (FY) 2019 based on the GAO-19-125 report. VA is
developing projected sustainment costs over the course of VA's new EHR
solution implementation.
Currently, there is no VistA sustainment cost reduction directly
tied to the new EHR solution rollout. VistA is expected to run without
service degradation until all VAMCs have been migrated to the new EHR
solution, at which time the redundant VistA modules will be
decommissioned. VistA modules that are not replaced by the new solution
will be maintained until replacement capabilities are developed. The
cost to maintain VistA will increase as we must include development for
new capabilities and interfaces, Congressional mandates, cloud costs,
hiring and retention of VistA support resources, and maintenance. The
estimated minimum cost for VistA during this 10-year transition period
is $4.89 billion, not including any required development. VA is
currently developing a methodology to update the cost data and thereby
define VistA, which was also a recommendation by GAO in a recent draft
report.
LONG-TERM STRATEGY FOR SUSTAINMENT
VA is constantly looking for more efficient ways to sustain VistA
throughout the course of the EHRM effort. The following are some of the
key strategies:
Development Operations Approach - OIT is shifting to a
DevOps approach focused on collaboration, innovation, Agile principles,
and automation-so that it can develop, enhance, maintain, and roll out
better products at a faster pace than using the existing separate
development and operations processes.
VistA Standardization - VAMCs will be required to run the
nationally released ``Gold'' version of VistA. A waiver process will
allow for critical modifications. In addition to having a common set of
software routines for each VistA instance, there are some additional
normalization activities that includes the work on terminology
extensions to account for local differences and others that will need
to be addressed to ensure complete standardization of as much of the
VistA database/file system as possible. VA's goal is for all VistA
instances to be standardized.
Merging Resources - OIT is merging VistA teams and
resources for maximum efficiency throughout OIT.
Maintain excellent customer support - Responding to
patient safety issues; hiring and retention of VistA support resources;
maintaining security and compliancy (scans and remediation, 508, ATO,
etc.); refreshing hardware (life-cycle upgrade, hardware, cloud etc.);
maintaining software versions/upgrades; decommissioning of VistA
products as appropriate.
Office of Technical Integration (OTI) - OTI facilitates
communication and planning between OIT and various program offices that
are implementing the systems that will replace VistA. OTI will track
and report progress from these program offices, facilitate real-time
conflict resolution, and manage risks between programs.
VA Enterprise Cloud (VAEC) - OIT is currently piloting a
program to migrate all 130 instances of VistA to the VAEC. Last month,
OIT successfully migrated the first VistA instance to the cloud-a
historic milestone and strong first step toward full cloud migration
for VistA. Over the next year alone, VA will migrate 70 more instances
of VistA from the St. Louis Defense Information Systems Agency (DISA)
data center into the cloud.
CONCLUSION
As VistA functionality is replaced by a COTS solution and other
systems, VA can decommission VistA products as appropriate. Until the
new EHR solution is implemented across the VA enterprise, VistA remains
VA's authoritative source of Veteran data. Sustaining VistA for the
duration of our EHRM effort ensures that Veterans continue to receive
uninterrupted care and services while VA looks to the future and
improves the Veteran experience.
Madam Chair, Ranking Member, and Members of the Subcommittee, thank
you for the opportunity to appear before you today to discuss OIT's
progress toward VistA transition. I look forward to continuing to work
with this Subcommittee to address our greatest priorities. This
concludes my testimony, and I look forward to answering your questions.
1. Acknowledgement of GAO Report
2. Definition of VistA
a. Definition of VistA
i. Electronic Health Record
ii. Interoperability
iii. Other Functionalities
b. Definition of Instances of VistA
c.Explanation of Variation in Instances
d.Plans to Further Define VistA
3. Assessment of Costs of VistA Sustainment
a.Note on GAO Report Assessment
b.Methodology
c.Comprehensive Total Cost Assessment
d.Limitations
4.Need for Sustainment
a.Facilities Use of VistA During EHRM
5.Long-Term Strategy for Sustainment
a.Consolidation of Teams and Resources
b.OTI
c.Cloud Migration
6.Activities to Prepare VistA Transition
a.Establishment of Program Office
i.Governance Structure
b.Role of OIT
c.Assessment of Initial Sites
d.Initial Operating Capability
e.System Implementation
1. Acknowledgement of GAO Report
The Department of Veterans Affairs (VA) Office of Information and
Technology (OIT) acknowledges the Government Accountability Office's
(GAO) report released in July 2019, titled ``ELECTRONIC HEALTH RECORDS:
VA Needs to Identify and Report System Costs'' regarding the costs and
requirements of sustainment of the Veterans Health Information Systems
and Technology Architecture (VistA) system during VA's transition to
Cerner Millennium and other systems intended to replace VistA
functionality.
Under the section titled ``Recommendation for Executive Action,''
GAO recommended that the Assistant Secretary for Information and
Technology and Chief Information Officer work with the Under Secretary
for Health to develop and implement a methodology for reliably
identifying and reporting the total costs of VistA sustainment. The
report states that this methodology should include steps to define
VistA and include planned sustainment activities. OIT acknowledges this
recommendation and is currently developing such a methodology and
continues to conduct current, ongoing, and planned sustainment
activities. OIT presents this written testimony to provide further
information regarding current and ongoing efforts related to VistA
sustainment and the Electronic Health Record Modernization (EHRM)
effort.
2a. Definition of VistA
VistA is VA's comprehensive information system for Veteran care and
services. It supports a complex set of clinical, administrative, and
financial operations for the Veterans Health Administration (VHA).
VistA is an architecture that includes servers, personal
workstations, and a variety of applications within the supporting
infrastructure including data centers, storage, and messaging
technologies. It provides a wide variety of functionalities and
therefore may also support functions outside of VHA.
VistA supports over 150 applications and the operations of more
than 1,500 VA facilities. Applications focus on clinically-relevant
record keeping that improves patient care by improving clinical and
administrative decision-making. Facilities range from small clinics
that provide solely outpatient care to large medical centers with
significant inpatient populations and their associated specialties.
VistA is deployed across VHA at more than 1,500 sites of care,
including Veterans Affairs Medical Centers (VAMC), Community Based
Outpatient Clinics (CBOC) and Community Living Centers (CLC), as well
as at nearly 300 VA Vet Centers. VistA was designed and often developed
and implemented jointly by VHA clinicians and IT personnel at VHA
facilities. It has been in use since 1983, nearly 40 years.
2ai. Definition of VistA: Electronic Health Record
VistA is VHA's full-featured Health Information System and
electronic health record (EHR). It contains an EHR for each patient and
supports the clinical, administrative, and financial functions of VAMCs
and VA facilities across the country. VistA interfaces with
applications through messaging protocols and reporting mechanisms.
2aii. Interoperability
As an EHR, VistA sends and exchanges stored health data with other
VA systems, other Federal agencies (e.g., Department of Defense),
health information exchange networks, community care providers, and
more than 100 commercial off-the-shelf (COTS) products. VistA is not
currently interoperable with the Department of Defense (DoD), so VA
users instead use the DoD/VA Joint Legacy Viewer (JLV), a Web-based
graphical user interface. Additionally, VA and DoD share allergens and
medication data with each other in a system called the Health Data
Repository (HDR) which feeds data to other systems that can alert VA
clinicians while using VistA.
2aiii. Other functionalities
As the GAO report notes, VistA provides functionality beyond
traditional EHRs. It exchanges information with many other applications
and interfaces. It provides a variety of other functionalities
including asset management, financial transaction management, a billing
system, and supply chain management. These functions primarily support
VHA facilities, but instances of VistA may also be used by local
Veterans Benefits Administration (VBA) and National Cemetery
Administration (NCA) facilities and cemeteries. For example, a local
cemetery may use VistA for its supply chain management needs.
2b. Definition of Instances of VistA
There are 130 instances of VistA across the VA enterprise. An
instance of VistA is an occurrence of the system that serves a VAMC and
its associated clinics, and other potential VA facilities within a
defined geographical region. Generally, there is one instance of VistA
per health care system or VAMC and associated clinics. However, over
the years, some VAMCs have been consolidated onto the same VistA
instance, so there is not exactly a 1:1 ratio of instance and site.
Each instance also consists of the hardware and software used to
provide VistA capabilities for a health care system.
2c. Explanation of Variation in Instances of VistA
Each instance of VistA may have slight modifications and variations
that serve requirements unique to that geographical region. However,
the code between instances has been made nearly identical through work
over the last 6 years through the VistA Evolution Program.
Implementation of the new EHR solution will help consolidate and
standardize VistA instances. OIT is working to avoid any changes to
VistA which could needlessly alter VistA's configuration prior to full
implementation of the new EHR solution would complicate and delay
implementation efforts.
2d. Plans to Further Define VistA
VA is currently developing a methodology to refine the definition
of VistA.
3a. Note on GAO Report Assessment
The GAO report examined cost data provided by OIT and VHA
associated with the development and sustainment of VistA for FYs 2015,
2016, and 2017 only.
3b. Methodology
VA is currently developing a methodology to update the sustainment
cost data.
3c. Comprehensive Total Cost Assessment
Cost data has been updated. It currently costs VA $426 million to
sustain VistA through FY 2019. VA is developing projected sustainment
costs over the course of the new EHR solution implementation. VA's
estimated minimum cost for VistA during this 10-year transition period
is at least $4.89 billion, not including newly required development. VA
is currently developing a methodology to update the cost data and
redefine VistA.
3d. Limitations
During the transition from VistA to the new EHR solution, the two
systems will need to be operated in parallel. In addition, VistA is
expected to run without service degradation until all VAMCs have been
migrated to Cerner, at which time the redundant VistA modules will be
decommissioned. For these reasons, there is currently no VistA
sustainment cost reduction directly tied to the EHRM effort.
4. Need for Sustainment
Further, VistA modules whose functionality is not replaced by
Cerner Millennium will need to be maintained until replacement
solutions are developed and deployed. For example, Cerner Millennium
does not replace some financial management and supply chain management
functions provided by VistA. Other programs, such as Financial
Management Business Transformation (FMBT) and Defense Medical Logistics
Standard Support (DMLSS) will replace those functionalities. VistA
cannot be decommissioned until all current functionality is replaced by
a modernized replacement system, whether Cerner Millennium or
otherwise. However, these other programs are expected to be developed
and implemented on a shorter timeline (i.e., less than 10 years). Since
implementation of the new EHR solution is currently projected to take
10 years, the EHRM effort is the ultimate driver of VistA sustainment
and full transition.
4a. Facilities Use of VistA During EHRM
Facilities will continue to use their instance of VistA until other
concurrent modernization projects have replaced all functionalities of
that VistA instance. Only then can the facility fully transition from
VistA to the new EHR solution and other replacement solutions.
5. Long-Term Strategy for Sustainment
Despite the need to maintain Vista over the course of the EHRM
effort and the development and implementation of additional modernized
replacement systems, there are current and ongoing efforts to reduce
some costs of sustainment and make transition efforts more efficient.
For example:
5a. Consolidation of Teams and Resources
OIT is consolidating teams and resources between Transition,
Release and Support (TRS) and Enterprise Program Management Division
for maximum efficiency.
5b. OTI
The newly established Office of Technical Integration (OTI)
facilitates communication and planning between OIT and various program
offices that are implementing the systems that will replace VistA. OTI
will track and report progress from these program offices, facilitate
real-time conflict resolution, and manage risks between programs.
5c. Cloud Migration
OIT is currently piloting a program to migrate all 130 instances of
VistA to the VA Enterprise Cloud (VAEC). Last month, OIT successfully
migrated the first instance of VistA to the cloud. This is a
significant achievement which will support VA's ``Cloud First'' policy
and modernization initiatives as established by the Federal Chief
Information Officer.
Over the next year alone, VA will migrate 70 more instances of
VistA from the St. Louis Defense Information Systems Agency (DISA) data
center into the cloud. Hosting VistA in the cloud is more cost-
effective than hosting in physical data centers. It allows OIT to make
updates more quickly, saving labor hours. It also improves system and
application speed and performance and is more scalable, making it more
valuable to OIT's business partners.
6. Activities to Plan for Transition
VA is working closely with DoD during this major business
transformation. DoD and VA have appointed co-chairs for all efforts. VA
is collaborating with Cerner to understand the technical support
requirements to connect to the Cerner Millennium Cloud Data Center and
to develop the processes necessary to accommodate emerging
technologies. VA is also working with its community care partners,
focusing on interoperability and bidirectional information exchange.
To allow for seamless interoperability between Cerner and VistA
over the course of implementation, JLV will be enhanced to include a
Cerner viewer. This will allow sites that have not yet transitioned to
access new electronic data repositories and to create stand-alone
technical solutions to share data with the new EHR solution. In
addition, the requisite interfaces with VistA and the new product
capabilities and related workflows will be fully tested before
transitioning to the operational environment.
During this time, Cerner HealtheIntent will become the
authoritative data store for Veteran health care information, since it
is populated with all Veteran information and since information from
VistA sites will be written into HealtheIntent real-time through VDIF,
the middleware.
6a. Establishment of Program Office
To establish a leadership accountable for planning and executing
the EHRM effort and addressing difficulties to ensure program success,
VA established the Office of Electronic Health Record Modernization
(OEHRM) in June 2018. OEHRM's initial Program Management Plan guides
management and defines program policies and processes.
6a. Governance Structure
OEHRM is comprised of three management structures. The Chief
Medical Office oversees strategy and planning; communication efforts
for business process changes; and user testing, training, and
deployment. The Technology and Integration Office provides technical
leadership, management, and oversight and supports interoperability
with DoD. Lastly, the Program Management Office provides program
support through adherence to cost, schedule, and performance
objectives. OEHRM has a governance structure that is intended to allow
leadership to address technical and functional issues as well as joint
management issues that may arise between VA and DoD during the process
of their respective EHR implementation efforts. The structure consists
of a Steering Committee; a Governance Integration Board, which oversees
a Technical Governance Board and Functional Governance Board; and the
Electronic Health Record Councils.
OIT is also working closely with DoD on the organizational
development of the Federal Electronic Health Record Modernization
(FEHRM) Program Office. The FEHRM Program Office will serve as the re-
chartered DoD/VA Interagency Program Office (IPO). In short, OIT is
working collaboratively with VHA, OEHRM, IPO/FEHRM, and their
associated partners to achieve successful implementation, leverage
lessons learned and best practices, leverage common infrastructure,
innovate to improve business processes, and facilitate effective
adjudication of issues.
16b. Role of OIT
OIT plays several roles in this business transformation process.
According to established baseline standards for initial operating
capability (IOC), OIT is responsible for upgrades to the IT
infrastructure and local area network infrastructure. These
infrastructure upgrades are critical to success of the deployment of
the new EHR solution.
OIT is also involved in the following areas:
Coordination, planning, and budgeting: OIT works
collaboratively with OEHRM, based on requirements submitted in VA IT
Process Request (VIPR), to provide planning, budgeting, project
management, infrastructure assessments, and other support to EHRM.
Fielding: In support of VHA and the IOC/VAMC sites, OIT
focuses on the infrastructure line of effort to ensure that all aspects
of the network architecture will support accessing the new EHR solution
and associated systems, within VHA-defined service levels response
times.
Access Management: OIT coordinates closely with Office of
Operations, Security, and Preparedness in development and
implementation of access control (PIV cards), and with OEHRM and DoD
for secure access to Cerner Millennium resources in the DoD Medical
Community of Interest (MedCOI) environment.
Cybersecurity: OIT is in close coordination with DoD
regarding shared network security standards and reciprocity between DoD
and VA systems.
End user experience with system performance: OIT
organizes activities among multiple stakeholders to manage service
provision and system access.
OIT is working closely with VHA and OEHRM to plan an accelerated
implementation of the Cerner Standalone Scheduling module; design
system interfaces between legacy applications and the Cerner Millennium
suite; adjudicate requests for legacy VistA upgrades against pending
Cerner Millennium functionality; and design service desk interface
tools and business rules to improve collaboration with end user
reported issues.
6c. Assessment of Sites
VA is currently conducting site assessments at IOC sites to refine
requirements and prepare for implementation. VA has identified three
primary IOC sites for assessment: VA Puget Sound Health Care System,
American Lake Division; VA Puget Sound Health Care System, Seattle
Division; and Mann-Grandstaff VA Medical Center. During assessment at
these sites, VA determined that some infrastructure and workstations
would need to be updated to achieve compatibility with Cerner
Millennium. In addition, sites that offer such VA services as
telehealth and behavioral and mental health services would need
additional attention from Cerner to meet business and system
requirements. These site assessments are intended to produce lessons
learned and ease deployment and implementation efforts at the rest of
the sites Nation-wide.
In the Pacific Northwest, there are the following:
5 VA Health Care Systems;
6 VA Medical Centers (VAMC);
9 Outpatient clinics;
17 Vet Centers; and
34 Community-based outpatient clinics (CBOC).
6d. Initial Operating Capability
VA is working with Cerner to implement the new EHR solution at
three IOC sites in the Pacific Northwest. As DoD has already deployed
to this region, VA selected the Pacific Northwest to maximize
efficiencies through DoD's lessons learned. This strategy also allows
VA to leverage DoD's data hosting environment and adopt enhanced
cybersecurity protocols to facilitate interoperability.
For FY 2019, OIT is accessing OEHRM infrastructure funding to
support IOC with network switch upgrades, bandwidth upgrades, Wi-Fi
implementation and upgrades, new endpoint devices, surge implementation
support, and monitoring tools and licenses.
6e. System Implementation
After implementation at the IOC sites in the Pacific Northwest, VA
will deploy the new EHR solution across the enterprise. As previously
discussed, VA will maintain and support VistA until full Cerner
implementation. This ensures that current patient records are
accessible and that there will be no interruption in the delivery of
quality health care to our Nation's Veterans.
Prepared Statement of Carol C. Harris
ELECTRONIC HEALTH RECORDS
VA Needs to Identify and Report Existing System Costs
Chair Lee, Ranking Member Banks, and Members of the Subcommittee:
Thank you for the opportunity to participate in today's hearing
regarding the Department of Veterans Affairs' (VA) health information
system-the Veterans Health Information Systems and Technology
Architecture (VistA)-which has been essential to the department's
ability to deliver health care to veterans. This technically complex
system has been in operation for more than 30 years, is costly to
maintain, and does not fully support exchanging health data with the
Department of Defense (DoD) and private health care providers.
VA has initiated a major program to replace the VistA electronic
health record (EHR) with a commercial-off-the-shelf (COTS) product. The
department plans to start deploying its new EHR system in March 2020.
However, VA sites are to continue using VistA until they receive the
new system during a phased transition over the next 10 years.
We recently reviewed key aspects of VistA in response to a request
from the House Committee on Veterans' Affairs. We examined, among other
things, the extent to which VA has defined VistA and the department's
annual costs to develop and sustain the system.
At your request, my testimony for this hearing summarizes the
findings discussed in our report on VistA, which is being released
today.\1\ More detailed information on our objectives, scope, and
methodology for that work can be found in the issued report.
---------------------------------------------------------------------------
\1\ GAO, Electronic Health Records: VA Needs to Identify and Report
System Costs, GAO 19 125 (Washington, D.C.: July 25, 2019).
---------------------------------------------------------------------------
We conducted the work on which this statement is based in
accordance with generally accepted government auditing standards. Those
standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe that
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.
Background
VA's mission is to promote the health, welfare, and dignity of all
veterans in recognition of their service to the Nation by ensuring that
they receive medical care, benefits, social support, and lasting
memorials. In carrying out this mission, the department manages one of
the largest health care delivery systems in the United States that
provides enrolled veterans with a full range of services. These
services may include primary care; mental health care; and outpatient,
inpatient, and residential treatment. The Veterans Health
Administration (VHA), one of the department's three major components,
is responsible for overseeing the provision of health care at all VA
medical facilities.
Information technology (IT) is widely used and critically important
to supporting the department in delivering health care to veterans. As
such, VA operates and maintains an IT infrastructure that is intended
to provide the backbone necessary to meet the day-to-day operational
needs of its medical centers and other critical systems supporting the
department's mission. The infrastructure is to provide for data
storage, transmission, and communications requirements necessary to
ensure the delivery of reliable, available, and responsive support to
all VA staff offices and administration customers, as well as veterans.
The Office of Information and Technology (OIT) is responsible for
managing the majority of VA's IT-related functions. The office provides
strategy and technical direction, guidance, and policy related to how
IT resources are to be acquired and managed for the department.
VistA's Role at VA
VA provides health care services to approximately 9 million
veterans and their families and relies on its health information
system-VistA-to do so. VistA has been essential to the department's
ability to deliver health care to veterans. It was developed based on
the collaboration between staff in the VA medical facilities and VHA IT
personnel. Specifically, clinicians and IT personnel at the various VA
medical facilities collaborated to define the system's requirements
and, in certain cases, carried out its development and implementation.
As a result of these efforts, the system has been in operation since
the early 1980s.\2\
---------------------------------------------------------------------------
\2\ VistA began operation in 1983 as the Decentralized Hospital
Computer Program. In 1996, the name of the system was changed to the
Veterans Health Information Systems and Technology Architecture,
referred to as VistA.
---------------------------------------------------------------------------
VistA supports a complex set of clinical and administrative
capabilities. It is comprised of an architecture that ties together
servers and personal computer workstations with various applications
within VA facilities and the supporting infrastructure, such as data
centers, storage, and messaging technologies. The core system and
database code are programmed in the MUMPS programming language.\3\
Among other things, VistA contains an EHR for each patient and supports
clinics and medical centers.
---------------------------------------------------------------------------
\3\ The Massachusetts General Hospital Utility Multi-Programming
System, now referred to as M, or MUMPS.
---------------------------------------------------------------------------
In addition, the system provides functionality beyond the EHR and
exchanges information with many other applications and interfaces. For
example, the system also provides the functionality of a time and
attendance program, asset management system, library, and billing
system, among other things.
Users interact with VistA through a number of interfaces that
connect stored health data. These interfaces enable the system to
communicate (send or exchange data) with other VA systems, as well as
with other Federal agencies (e.g., DoD), health information exchange
networks, and COTS products. According to OIT officials, applications
either interface with VistA directly through a messaging protocol\4\ or
extract data from the system via a reporting mechanism.
---------------------------------------------------------------------------
\4\ VistA uses, for example, application programming interfaces,
remote procedure calls, and Health Level 7 messaging to communicate
with COTS software, selected IT systems of other Federal agencies, and
health information exchange networks.
---------------------------------------------------------------------------
The Computerized Patient Record System is a graphical user
interface to VistA that runs on workstations, laptops, and tablets and
enables the department to support clinical workflows. Specifically, the
Computerized Patient Record System enables the department to create and
update an individual EHR for each VA patient. Among other things,
clinicians can order lab tests, medications, diets, radiology tests,
and procedures; record a patient's allergies or adverse reactions to
medications; request and track consults; enter progress notes,
diagnoses, and treatments for each encounter; and enter discharge
summaries.
According to VHA officials, there are also more than 100 COTS
products that interface with VistA. In addition to these commercial
products, medical equipment or devices at local facilities may also
require interfaces to the system, and these vary on a site-by-site
basis.
VA Has about 130 Different Versions of VistA
Over the last several decades, VistA has evolved into a technically
complex system that supports health care delivery at more than 1,500
locations,\5\ including VA Medical Centers, outpatient clinics,
community living centers, and VA vet centers. Customization of the
system by local facilities has resulted in about 130 clinical versions
of VistA-referred to as instances.\6\
---------------------------------------------------------------------------
\5\ The VHA Business Function Framework (Version 2.11, May 2016) is
the department's architectural model that describes the core functions
related to delivering health care services and supporting the needs of
veterans, health care providers, and resource partners.
\6\ A customization might include modifications required to address
state and local laws regarding health care, such as those related to
the inputs, outputs, and data required to produce a death certificate.
A clinical VistA instance includes the EHR. There are a limited number
of VistA instances that do not support clinical functions.
---------------------------------------------------------------------------
According to the department, no two VistA instances are identical.
Further, each instance is comprised of over 27,400 routines (executable
modules of code), which are logically grouped into products or modules.
VistA products or modules can also be comprised of one or more software
applications that support health care functions, such as providing care
coordination and mental health services. The department reported that
there are approximately 140 to 200 products or modules that comprise
the system.\7\
---------------------------------------------------------------------------
\7\ Within VistA, nationally released and supported software are
referred to by VA as Class I software. In addition, instances may also
be comprised of Class II (regionally deployed and supported) and Class
III (locally deployed and supported) software.
---------------------------------------------------------------------------
The 130 clinical instances of VistA are operated from four regional
VA data centers.\8\ Users interact with the system through the
Computerized Patient Record System. Aggregated clinical data from every
instance of the system are located on servers hosted at VA's National
Data Center.\9\
---------------------------------------------------------------------------
\8\ According to VA officials, there are about 39 additional
instances of VistA that are older and nonoperational but contain
records and must be maintained or have their data migrated for
maintenance.
\9\ The National Data Center is located in Austin, Texas.
---------------------------------------------------------------------------
Over time, VA has identified the need for enhancements and
modifications to VistA in order to ensure that the system keeps up with
current technology and health care delivery. However, according to the
department, the system has become difficult and costly to maintain.
This is a result of, for example, being programmed in MUMPS, a language
for which there is a dwindling supply of qualified software developers.
It is also due to years of decentralized customization of the system by
staff members who were permitted to develop and implement applications
at the local level.
OIT and VHA Share Responsibilities for VistA
OIT and VHA serve as the technical and functional leaders,
respectively, for the department's health care delivery and, together,
they have worked to develop and maintain VistA for decades.
Specifically, OIT is responsible for managing the majority of VA's IT-
related functions. The office provides strategy and technical
direction, guidance, and policy related to how IT resources are to be
acquired and managed for the department.
According to the department, OIT's mission is to collaborate with
its business partners (such as VHA) and provide a seamless, unified
veteran experience through the delivery of state-of-the-art technology.
The Assistant Secretary for Information and Technology/Chief
Information Officer (CIO) serves as the head of OIT and is responsible
for providing leadership for the department's IT activities.
The CIO also advises the Secretary regarding the execution of VA's
IT systems appropriation, consistent with the Federal Information
Technology Acquisition Reform Act.\10\ For fiscal year 2019, the
department has been appropriated $4.1 billion for IT. According to VA's
budget documentation, about $1.2 billion of this amount is intended to
support IT staffing and associated costs for approximately 8,100 full-
time employees.
---------------------------------------------------------------------------
\10\ Provisions in IT acquisition reform legislation (commonly
referred to as the Federal Information Technology Acquisition Reform
Act, or FITARA) require covered executive branch agencies, including
VA, to ensure that the CIO has a significant role in the decisionmaking
process for IT budgeting, and in the management, governance, and
oversight processes related to IT. See Carl Levin and Howard P. `Buck'
McKeon National Defense Authorization Act for Fiscal Year 2015 , Pub.
L. No. 113-291, div. A, title VIII, subtitle D, 128 Stat. 3292, 3438-
3450 (Dec. 19, 2014).
---------------------------------------------------------------------------
VHA provides information and expertise to OIT to support the
department's health-related information systems. For example, VHA
officials help identify clinical and business needs used to inform IT
requirements development.\11\ The Under Secretary for Health is the
head of VHA and is supported by the Principal Deputy Under Secretary
for Health, four Deputy Under Secretaries for Health, and nine
Assistant Deputy Under Secretaries for Health.
---------------------------------------------------------------------------
\11\ VHA is responsible for the Medical Support and Compliance
budget, which includes ``necessary expenses in the administration of
the medical, hospital, nursing home, domiciliary, construction, supply,
and research activities, as authorized by law.''.
---------------------------------------------------------------------------
VA Has Begun to Acquire a New EHR System
After nearly 2 decades of pursuing multiple efforts to modernize
VistA, in June 2017, the former VA Secretary announced that the
department planned to acquire the same EHR system that DoD is
acquiring-Cerner Millennium.\12\ According to the department, it has
chosen to acquire this product because Cerner Millennium should allow
VA's and DoD's patient data to reside in one system, thus, potentially
reducing or eliminating the need for manual and electronic exchange and
reconciliation of data between two separate systems.
---------------------------------------------------------------------------
\12\ In July 2015, DoD awarded a $4.3 billion contract for a
commercial EHR system developed by Cerner-Cerner Millennium-to be known
as MHS GENESIS. The transition to the new system began in February 2017
in the Pacific Northwest region of the United States and is expected to
be completed in 2022. The former Secretary of VA signed a
``Determination and Findings,'' to justify use of the public interest
exception to the requirement for full and open competition, and
authorized VA to issue a solicitation directly to Cerner. A
``Determination and Findings'' means a special form of written approval
by an authorized official that is required by statute or regulation as
a prerequisite to taking certain contract actions. The
``determination'' is a conclusion or decision supported by the
``findings.'' The findings are statements of fact or rationale
essential to support the determination and must cover each requirement
of the statute or regulation. FAR, 48 C.F.R. Sec. 1.701.
---------------------------------------------------------------------------
Accordingly, the department awarded an indefinite delivery,
indefinite quantity contract to Cerner Corporation in May 2018 for a
maximum amount of $10 billion over 10 years. Cerner is to replace the
130 instances of VistA with a standard COTS system to be implemented
across VA. This new system is to support a broad range of health care
functions including acute care, clinical decision support, dental care,
and emergency medicine. When implemented, the new system will be
expected to become the authoritative source of clinical data to support
improved health, patient safety, and quality of care provided by VA.
The Electronic Health Record Modernization (EHRM) program is
responsible for managing the Cerner contract implementation. For fiscal
year 2019, the program was appropriated about $1.1 billion for planning
and managing the transition from VistA to Cerner.\13\
---------------------------------------------------------------------------
\13\ The EHRM appropriation is in addition to the $4.1 billion
appropriated for IT in 2019.
---------------------------------------------------------------------------
Further, the department has estimated that an additional $6.1
billion in funding, above the Cerner contract amount, will be needed to
fund additional project management support supplied by outside
contractors, government labor costs, and infrastructure improvements
over the 10-year contract period.
VA plans to deploy the new EHR system at three initial operating
capability sites within 18 months of October 1, 2018,\14\ with a phased
implementation of the remaining sites over the next decade. Each VA
medical facility is expected to continue using VistA until the new
system has been deployed. The three initial deployment sites, located
in the Pacific Northwest, are the Mann-Grandstaff, American Lake, and
Seattle VA Medical Centers and related clinical facilities that operate
the same instances of VistA. These are the first locations where the
system is expected to ``go live.''
---------------------------------------------------------------------------
\14\ Initial operating capability is the contract milestone in
which the system is intended to meet minimum operational capabilities.
---------------------------------------------------------------------------
The task order to deploy the Cerner system at the three initial
sites provides a detailed description of the steps Cerner needs to take
in order to reach initial operating capability at the Mann-Grandstaff
site in March 2020, and at the Seattle and American Lake sites in April
2020. According to the schedule, the initial operating capability sites
are expected to be operational by July 2020.
VA Has Undertaken Efforts to Define VistA, but Additional Work Remains
In order to maintain internal control activities over an IT system
and its related infrastructure, organizations should be able to define
physical and performance characteristics of the system, including
descriptions of the components and the interfaces.\15\ Further,
consistent with GAO's Cost Estimating and Assessment Guide, a
comprehensive system definition should identify customization and the
environment in which the system operates.\16\ While defining a complex
IT system can be challenging, having an adequate understanding of its
characteristics will better position the organization to
comprehensively project and account for costs over the life of a system
or program as well as identify specific technical and program risks.
Definition of VistA remains important because VA plans to continue
using the system during the department's decade-long transition to the
Cerner system.
---------------------------------------------------------------------------
\15\ GAO, Standards for Internal Control in the Federal Government,
GAO 14 704G (Washington, D.C.: September 2014).
\16\ GAO, GAO Cost Estimating and Assessment Guide: Best Practices
for Developing and Managing Capital Program Costs, GAO 09 3SP
(Washington, D.C.: March 2009).
---------------------------------------------------------------------------
VA maintains multiple documents and a database that describe parts
of VistA, including various components and interfaces. However, despite
these existing sources, OIT officials acknowledged that there is no
comprehensive definition of the VistA system. Consequently, VA has
completed a number of efforts to better define VistA and understand the
environment in which it operates and additional work is planned in the
future.
Specifically, VA has documented descriptions of the system,
including the components that comprise it. These descriptions are
documented in multiple sources: the VA Monograph, VA Systems Inventory,
and VA Document Library.
The VA Monograph is a document maintained by OIT that
provides an overview of VistA and non-VistA applications used by
VHA.\17\ According to VHA officials, the VA Monograph is the primary
document that describes the components of the system. The Monograph
describes VistA in terms of modules. For modules identified, including
VistA modules, information such as the associated business functions,
VA Systems Inventory identification number, and a link to the VA
Document Library for additional technical information are provided.
---------------------------------------------------------------------------
\17\ VA, VA Monograph (Washington, D.C.: February 2019).
---------------------------------------------------------------------------
The VA Systems Inventory is a database maintained by OIT
that identifies current IT systems at the department, including systems
and interfaces related to VistA.\18\ For systems identified, the
database includes information such as the system name, the system
status (i.e., active, in development, or inactive), and related system
interfaces.
---------------------------------------------------------------------------
\18\ VA, VA Directive 6404: Department of Veterans Affairs VA
Systems Inventory (VASI) (Washington, D.C., Feb. 23, 2016). According
to VA Directive 6404, the VA Systems Inventory is the authoritative
data source for VA's IT systems. OIT is responsible for the development
and sustainment of the inventory.
---------------------------------------------------------------------------
The VA Document Library is an online resource for
accessing documentation (i.e., user guides and installation manuals) on
the department's nationally released software applications, including
VistA.\19\
---------------------------------------------------------------------------
\19\ The VA Document Library includes links to documentation on VA
software organized into the following categories: Clinical,
Infrastructure, Financial-Administrative, HealtheVet, and Benefits.
VA has taken additional steps to further define the system. For
example, EHRM program officials recognized the need to further
understand the customization of VistA components at the various medical
facilities and have conducted analyses to do so. These analyses
---------------------------------------------------------------------------
include:
Variance analysis: As part of its VistA Evolution
program,\20\ which has focused on standardizing a core set of VistA
functionality, the department implemented a process to compare the
instances of VistA installed at sites to the Enterprise Standard
version.\21\ The results of this analysis allowed the department to
assess the criticality of each variance, which is expected to help with
VA's transition to the Cerner system.
---------------------------------------------------------------------------
\20\ In December 2013, VA initiated VistA Evolution, a joint
program between OIT and VHA that focused on implementing a collection
of projects to improve the efficiency and quality of veterans' health
care. Specifically, it focused on modernizing the VistA system,
increasing the department's data exchange and interoperability with DoD
and private sector health care partners, and reducing the time it takes
to deploy new health information management capabilities.
\21\ The Enterprise Standard version of VistA represents the
compilation of different historical releases of VistA where patches
have been installed.
---------------------------------------------------------------------------
Module analysis: EHRM program subject matter experts
undertook an analysis that involved reviewing and assessing
capabilities provided by VistA modules. This analysis enabled
department officials to determine whether the capability provided by a
VistA module could be provided by the Cerner system, or whether another
COTS solution would be required to support this function going forward.
Visual mapping: EHRM program officials also directed an
analysis that involved developing a notional visual mapping of VA's
health care applications, components, and supporting systems within the
health delivery environment. The results of this analysis provided a
description of the current state of one instance of VistA and the VA
health environment, which is intended to inform the department of
possible opportunities for business process and IT improvements as it
proceeds with the Cerner acquisition.
Nevertheless, even with these analyses, VA has not yet fully
defined VistA, including, for example, identifying performance
characteristics of the system and describing the environment in which
it operates. The department's three sources that describe VistA and the
additional analyses undertaken do not provide insight into site
specific customizations of the system. For example, the VA Monograph
does not include information on module customization at local
facilities. In addition, according to OIT officials, the systems
inventory does not reflect differences among the 130 different
instances of VistA and does not take into consideration regional and
local customizations of related components. Further, the visual mapping
analysis noted that there was not full insight of the intertwined
structure of data and applications or the various local customizations
of VistA.
EHRM program officials stated that they have not been able to fully
define VistA and understand all local customizations due to the
decentralization of the development of the system and its evolution
over more than 30 years. They explained that VistA's complexity is
partly due to the various instances of the system, compounded by local
customizations, which have resulted in differences in VistA instances
operating at various facilities.
According to EHRM program documentation, Cerner's contract calls
for the company to conduct comprehensive assessments to capture the
current state of technical and clinical operations at specific
facilities, as well as identify site-specific requirements where the
Cerner system is planned to be deployed. As of June 2019, Cerner had
completed site assessments for the three initial operating capability
sites in the Pacific Northwest and had planned additional assessments
at future deployment sites. The initial site assessments included,
among other things, an assessment of the unique VistA instances and the
environment in which the system operates. The continuation of planned
site assessments should provide a thorough understanding of the 130
VistA versions, help the department better define VistA, and position
it for transitioning from VistA to Cerner's COTS solution.
VA Identified Total VistA Costs of about $2.3 Billion between 2015 and
2017, but Could Not Sufficiently Demonstrate the Reliability of All
Data and Omitted Other Costs
When using public funds, an agency must employ effective management
practices in order to let legislators, management, and the public know
the costs of programs and whether they are achieving their goals. To
make those evaluations for a program or for a system as large and
complex as VistA, a complete understanding of the system and reliable
cost information is required.\22\ By following a methodology and
utilizing reliable data, an agency can ensure that all costs are fully
accounted for, which in turn, better informs management decisions,
establishes a cost baseline, and enhances understanding of a system's
performance and return on investment.\23\
---------------------------------------------------------------------------
\22\ In the case of VistA, costs reflect the complexity of the
system itself and the environment in which it operates, beyond a single
program.
\23\ GAO's Cost Estimating and Assessment Guide describes a
methodology for compiling an exhaustive and structured accounting of
all resources and all costs required to develop and sustain a
particular program or, in this case, a system. Specifically, the
methodology describes the importance of documenting which costs are
included and how they are calculated in detail, step by step, to
provide enough information so that someone unfamiliar with the program
or system could easily recreate or update cost calculations. Further,
the methodology should include all assumptions and explanations for why
particular data sets are chosen and why these choices are reasonable to
allow for the assessment of the total accounting and the reliability of
the cost data.
---------------------------------------------------------------------------
Fundamental characteristics of reliable costs are that they should
be accurate (unbiased, not overly conservative or optimistic), well-
documented (supportable with source data, clearly detailed
calculations, and explanations for choosing a particular calculation
method), credible (identifying any uncertainty or biases surrounding
data or related assumptions), and comprehensive (costs are neither
omitted nor double counted). Identification of VistA's costs remains
important because VA plans to continue using the system during the
department's transition to the Cerner system over the next decade.
VA identified costs for VistA and its related activities adding up
to approximately $913.7 million, $664.3 million, and $711.1 million in
fiscal years 2015, 2016, and 2017, respectively-for a total of about
$2.3 billion over the 3 years.\24\ However, the department could not
sufficiently demonstrate the reliability of certain costs that were
identified. In addition, VA identified other categories of VistA-
related costs, but omitted these costs from the total.
---------------------------------------------------------------------------
\24\ We previously testified in June 2018 that preliminary costs
reported by VA for VistA and related activities included approximately
$1.1 billion, $899 million, and $946 million in fiscal years 2015,
2016, and 2017, respectively, for a total of about $3.0 billion over 3
years to support the system (see GAO, VA IT Modernization: Preparations
for Transitioning to a New Electronic Health Record System Are Ongoing,
GAO 18 636T (Washington, D.C.: Jun. 26, 2018)). Since that time,
updates were made in OIT's budget tracking tool and EHRM program
officials revised the approach to estimating certain types of costs.
VA Did Not Sufficiently Demonstrate the Reliability of Data for All
---------------------------------------------------------------------------
VistA Costs
Of the $2.3 billion total costs for VistA, VA demonstrated that
only approximately $1 billion of these costs were reliable.
Specifically, OIT officials identified VistA-related costs within seven
categories. The officials were able to sufficiently explain why these
categories were included in the development and sustainment costs for
VistA and how they were documented by the department; the officials
also presented detailed source data for our examination. As a result of
our review, we determined that the cost data for these seven categories
were accurate, well-documented, credible, and comprehensive and, thus,
sufficiently reliable.\25\
---------------------------------------------------------------------------
\25\ OIT program costs excluded pay and administrative costs, which
are not tracked within OIT by program.
---------------------------------------------------------------------------
Table 1 provides a summary of the program costs identified for
VistA by OIT and VHA for fiscal years 2015 through 2017 that we
determined to be reliable.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
As shown in the table, VA identified costs for the following seven
categories for fiscal years 2015 through 2017:
VistA Evolution - The VistA Evolution program costs were
associated with VistA strategy, system design, product development, and
program management. These costs totaled approximately $549.6 million.
Interoperability - The Interoperability program focused
on sharing electronic health data between VA and non-VA facilities,
including private sector providers and DoD.\26\ For example,
interoperability costs were associated with architecture, strategy, the
Interagency Program Office, product development, and program
management.\27\ These VistA-related costs totaled approximately $140.2
million.
---------------------------------------------------------------------------
\26\ The Interoperability program was previously reported under the
Electronic Health Record Interoperability program.
\27\ Provisions included in the National Defense Authorization Act
for Fiscal Year 2008 required VA and DoD to, among other things,
jointly develop and implement fully interoperable EHR systems or
capabilities and establish an Interagency Program Office to be a single
point of accountability for their efforts. According to the act, the
office was given the function of implementing, by September 30, 2009,
EHR systems or capabilities that would allow for full interoperability
of personal health care information between the departments. Pub. L.
No. 110-181, Sec. 1635, 122 Stat. 3, 460-463 (2008).
---------------------------------------------------------------------------
Virtual Lifetime Electronic Record (VLER) Health - This
program focused on streamlining the transition of electronic medical
information between VA and DoD.\28\ These VistA-related costs were
associated with product development and program management and totaled
approximately $81.2 million.
---------------------------------------------------------------------------
\28\ VLER Health initially started in 2009. According to VA, this
program is now referred to as the Veterans Health Information Exchange.
---------------------------------------------------------------------------
Contracts - Contract costs for VistA Evolution included
VHA's obligations associated with workload management, change
management, clinical requirements, and clinical interoperability. These
VistA-related costs totaled approximately $202.8 million.
Intergovernmental personnel acts - Intergovernmental
personnel acts are agreements for the temporary assignment of personnel
between the federal, state, and local governments; colleges and
universities; Indian tribal governments; federally funded research and
development centers; and other eligible organizations. These costs
accounted for VHA's need to use outside experts from approved entities
for limited periods of time to work on VistA Evolution assignments. The
total VistA-related costs were approximately $2.4 million.
Memorandums of understanding - According to VHA,
memorandums of understanding are agreements used by the administration
to obtain the services of personnel between VA entities for VistA-
related activities. These agreements accounted for approximately $2.3
million.
Pay - Costs in this category included salaries for VHA
staff who worked on VistA-related projects as well as travel, training,
and supply costs associated with employment. These costs totaled
approximately $34.1 million.
However, VA was not able to sufficiently demonstrate the
reliability of approximately $1.3 billion in costs related to VistA.
Specifically, OIT officials identified the additional legacy VistA
costs that generally fell into three categories:
Legacy VistA: Infrastructure, hosting, and system
sustainment - Legacy VistA costs are generally related to the
maintenance of fully operational items, such as VistA Imaging and
Fileman-two key components related to VistA's operation.\29\ The costs
also included obligations for costs related to hosting health data in
both VA and non-VA facilities.\30\ The OIT officials and subject matter
experts estimated these total costs to be approximately $343 million
during fiscal years 2015 through 2017.
---------------------------------------------------------------------------
\29\ According to the VistA 4 Product Roadmap, VistA Imaging is the
clinical imaging interface designed and developed by VHA to incorporate
image and document data, and attach said data to the veteran's EHR. It
also provides specific applications used for Telehealth. File Manager
(referred to as FileMan) serves as the data base management system for
VistA, providing both structure for the data in VistA's database and
the interface to VistA's data.
\30\ Co-location is when an instance of VistA is hosted in a data
center with other systems and includes costs, for example, of leasing
space and related utilities.
---------------------------------------------------------------------------
However, we were not able to determine the reliability of these
costs because, for example, source data were not well documented;
changes in the cost information provided to us during our review
indicated that the cost data may not be credible; and subject matter
experts were unclear about how to separate VistA costs from non-VistA
costs.
Related software - Related software costs are associated
with the software supporting, or closely integrated with, VistA that
were identified by EHRM officials, yet not tracked directly for one of
the VistA-related programs. Both OIT and VHA identified software
licensing costs as VistA-related obligations. The EHRM program reported
these costs to be approximately $389 million in total during fiscal
years 2015 through 2017.
However, we were not able to determine the reliability of the costs
in this category for a variety of reasons, including that source data
were not well documented. In addition, VA officials were not clear
regarding how the total amounts in each category should be divided
between OIT and VHA. Given this confusion, we were not able to
determine if the costs were fully accurate or credible.
OIT personnel (pay and administrative) - According to
EHRM officials, OIT does not track labor costs by program. Instead, the
department provided estimations of the amount of salaries paid to OIT
government staff working on activities such as VistA Evolution, program
management, and overall support of VistA and related applications. OIT
personnel costs were estimated by the EHRM program office to be
approximately $544 million total during fiscal years 2015 through 2017.
However, we were not able to determine the reliability of costs in
this category because assumptions made for estimating the personnel and
salary costs were not well documented and could not be verified.
VA Omitted Certain Costs from the Total Cost of VistA
In addition, VA omitted certain VistA costs from the total costs
identified for fiscal years 2015, 2016, and 2017. Specifically, VA
omitted the following costs:
Additional hosting - OIT officials stated that additional
costs related to hosting health data by an outside vendor, as well as
hosting backup VistA instances at each of the medical center sites,
should also be included in the total costs for VistA; however, VA
omitted these costs from the total for fiscal years 2015 through 2017.
Specifically, according to the officials, calculating costs for these
hosting activities requires subject matter experts to identify
equipment, space, utilities, and maintenance costs for resources
allocated specifically for VistA. However, the department has not yet
developed a methodology to calculate the costs. The officials said they
were working on identifying a reliable approach for calculating these
costs in the future.
Data standardization and testing - OIT officials stated
that additional costs related to work on clinical terminology mapping
and functional testing were not included in the total costs for VistA
for fiscal years 2015 through 2017. This work related to mapping
existing clinical data to national standards and making updates to
VistA or the Joint Legacy Viewer and included mapping data and building
test scripts and reports.\31\ OIT officials noted that this work had
been critical to the VistA Evolution program, but they did not provide
actual cost data in this category.
---------------------------------------------------------------------------
\31\ The Joint Legacy Viewer is a web-based graphical user
interface, first released in 2013, that was developed jointly by VA and
DoD. This tool provides a near real-time, integrated, and chronological
view of EHR information contained in VistA and existing DoD systems, as
well as data from some third-party providers. The Joint Legacy Viewer
allows VA clinicians to view a read-only display of patient data from
DoD as well as from a number of other medical providers.
---------------------------------------------------------------------------
The lack of sufficiently reliable and comprehensive costs indicates
that the department is not positioned to accurately report the annual
costs to develop and sustain VistA. This is due in part to VA not
following a well-documented methodology that describes how the
department determined the total costs for the system. In lieu of a
methodology, OIT officials said that leadership and staff from the
program took efforts to identify and track the cost components and
contracts associated with the system. However, they noted that costs
associated with VistA were not all clearly labeled as VistA in an IT
system and it was necessary to estimate other costs. The officials were
also unable to verify how VistA-related costs were separated from other
department costs in all areas and subject matter experts were not
consistently familiar with the estimation methods employed and how
VistA was defined for the purposes of calculating costs. Further, VA
officials noted that they were still working on the best approach to
identifying and calculating omitted costs.
Without documenting the methodology for what costs are to be
included and how they were identified and calculated, VA's total does
not accurately reflect the development and sustainment costs for VistA.
As a result, the department, legislators, and the public do not have
the comprehensive, reliable information needed to understand how much
it actually cost to develop and maintain the system. Further, VA does
not have the reliable information needed to make critical management
decisions for sustaining the many versions of VistA over the next 10
years until the Cerner system is fully deployed.
Implementation of GAO's Recommendation Could Help Ensure VA Reliably
Reports VistA Costs
In our report, we are making a recommendation for VA to improve its
reporting of VistA's costs. Specifically, we are recommending that the
department develop and implement a methodology for reliably identifying
and reporting the total costs of VistA. The methodology should include
steps to identify the definition of VistA and what is to be included in
its sustainment activities, as well as ensure that comprehensive costs
are corroborated by reliable data. In written comments on a draft of
the report, the department agreed with the recommendation and stated
that it will provide the actions it plans to take to address this
recommendation within 180 days.
In conclusion, although VA is not likely to be positioned to retire
VistA for at least another 10 years, the department lacks the
comprehensive and reliable cost information needed to make critical
management decisions for sustaining the system. As the department
continues to work toward acquiring a new electronic health record, it
will be important for VA to take actions to address our recommendation
for improving the reporting of VistA costs. Doing so is essential to
helping ensure that decisions related to the current system are
informed by reliable cost information and that there is an accurate
basis for reporting on the return on its investment for replacing
VistA.
Chair Lee, Ranking Member Banks, and Members of the Subcommittee,
this completes my prepared statement. I would be pleased to respond to
any questions that you may have.
GAO Contact and Staff Acknowledgments
If you or your staffs have any questions about this testimony,
please contact Carol C. Harris, Director, Information Technology
Management Issues, at (202) 512-4456 or [email protected]. Contact
points for our Offices of Congressional Relations and Public Affairs
may be found on the last page of this testimony statement. GAO staff
who made key contributions to this testimony are Mark Bird (Assistant
Director), Rebecca Eyler, Jacqueline Mai, Monica Perez-Nelson, Scott
Pettis, Jennifer Stavros-Turner (Analyst in Charge), and Charles
Youman.
GAO's Mission
The Government Accountability Office, the audit, evaluation, and
investigative arm of Congress, exists to support Congress in meeting
its constitutional responsibilities and to help improve the performance
and accountability of the Federal government for the American people.
GAO examines the use of public funds; evaluates Federal programs and
policies; and provides analyses, recommendations, and other assistance
to help Congress make informed oversight, policy, and funding
decisions. GAO's commitment to good government is reflected in its core
values of accountability, integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony
The fastest and easiest way to obtain copies of GAO documents at no
cost is through GAO's website (https://www.gao.gov). Each weekday
afternoon, GAO posts on its website newly released reports, testimony,
and correspondence. To have GAO e mail you a list of newly posted
products, go to https://www.gao.gov and select ``E-mail Updates.''
Order by Phone
The price of each GAO publication reflects GAO's actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black
and white. Pricing and ordering information is posted on GAO's website,
https://www.gao.gov/ordering.htm.
Place orders by calling (202) 512-6000, toll free (866) 801-7077,
or TDD (202) 512-2537.
Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional
information.
Connect with GAO
Connect with GAO on Facebook, Flickr, Twitter, and YouTube.
Subscribe to our RSS Feeds or E-mail Updates. Listen to our
Podcasts.
Visit GAO on the web at https://www.gao.gov.
To Report Fraud, Waste, and Abuse in Federal Programs
Contact FraudNet:
Website: https://www.gao.gov/fraudnet/fraudnet.htm
Automated answering system: (800) 424-5454 or (202) 512-7700
Congressional Relations
Orice Williams Brown, Managing Director, [email protected], (202)
512-4400, U.S. Government Accountability Office, 441 G Street NW, Room
7125, Washington, DC 20548
Public Affairs
Chuck Young, Managing Director, [email protected], (202) 512-4800,
U.S. Government Accountability Office, 441 G Street NW, Room 7149,
Washington, DC 20548
Strategic Planning and External Liaison
James-Christian Blockwood, Managing Director, [email protected], (202)
512-4707, U.S. Government Accountability Office, 441 G Street NW, Room
7814, Washington, DC 20548