[House Hearing, 116 Congress]
[From the U.S. Government Publishing Office]
THE FUTURE OF VA SCHEDULING: IMPLEMENTING A COMMERCIAL OFF THE SHELF
SCHEDULING SOLUTION AT THE DEPARTMENT OF VETERANS AFFAIRS
=======================================================================
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, SEPTEMBER 26, 2019
__________
Serial No. 116-37
__________
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
40-993 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, September 26, 2019
Page
The Future Of VA Scheduling: Implementing A Commercial Off The
Shelf Scheduling Solution At The Department Of Veterans Affairs 1
OPENING STATEMENTS
Honorable Susie Lee, Chairwoman.................................. 1
Honorable Jim Banks, Ranking Member.............................. 3
WITNESSES
Mr. John H. Windom, Executive Director, Office of Electronic
Health Record Modernization, U.S. Department of Veterans
Affairs........................................................ 5
Prepared Statement........................................... 25
Accompanied by:
Mr. Dominic Cussatt, Principal Deputy Assistant Secretary,
Office of Information and Technology, U.S. Department of
Veterans Affairs
Dr. Michael Davies, Senior Advisor to the Assistant, Deputy
Under Secretary for Health Access, Veterans Health
Administration, U.S. Department of Veterans Affairs
Dr. Laura Kroupa, Chief Medical Officer, Office of Electronic
Health Record Modernization, U.S. Department of Veterans
Affairs
Mr. John Short, Chief Technology and Integration Officer,
Office of Electronic Health Record Modernization, U.S.
Department of Veterans Affairs
Mr. Larry Reinkemeyer, Assistant Inspector General for Audits and
Evaluations, Office of the Inspector General, U.S. Department
of Veterans Affairs............................................ 6
Prepared Statement........................................... 27
THE FUTURE OF VA SCHEDULING: IMPLEMENTING A COMMERCIAL OFF THE SHELF
SCHEDULING SOLUTION AT THE DEPARTMENT OF VETERANS AFFAIRS
----------
September 26, 2019
Committee on Veterans' Affairs,
U. S. House of Representatives,
Washington, D.C.
The Subcommittee met, pursuant to notice, at 10:02 a.m., in
Room 210, House Visitors Center, Hon. Susie Lee [Chairwoman of
the Subcommittee] presiding.
Present: Representatives Lee, Lamb, Cunningham, Banks,
Watkins, and Roy.
OPENING STATEMENT OF SUSIE LEE, CHAIRWOMAN
Ms. Lee. Good morning. The hearing will come to order.
First of all, I would like to thank Ranking Member Banks,
as well as all of our witnesses, for being here today and
taking the time to present to us. Thank you.
We have spent a lot of time in this Congress talking about
the Electronic Health Record Modernization and VA's Cerner
Millennium platform. While the success of that transition
obviously is critically important, it is by no means the only
major technology modernization project underway at the
Department of Veterans Affairs.
For almost 20 years, VA has attempted to update its patient
appointment scheduling system. And I had to get a
representation of all the iterations that have happened in the
past 20 years, which is behind me. The old system, the VistA
legacy system is written in archaic programming language, its
user interface is confusing and cumbersome, and complicated
workflows require schedulers to memorize hundreds of rules and
apply them appropriately when scheduling an appointment;
obviously, in need of an update.
And the VA tried to update this by developing a graphical
user interface and automating some of the scheduling rules.
Unfortunately, it was not much more than a fresh coat of paint.
The project dubbed VistA Scheduling Enhancement, or VSE, did
not address data quality issues, did not bring the VA much
closer to the current state of health care information systems.
Also, as the Office of Inspector General found in its
August report on VSE, there were problems with the management
of the program, including leadership turnover and inadequate
requirements development. These seemed to be the same issues
that recur again at every VA modernization program.
I will add further that the history of the VSE indicates
that VA did not know what it wanted from the scheduling system
before it signed the contract and began developed, and I am
concerned that some of that uncertainty continues.
In 2015, the VA contracted for $624 million to implement
another commercial off-the-shelf scheduling system, the Medical
Appointment Scheduling System, or MASS, which was then
downgraded to a pilot.
In April of last year, the VA actually completed that pilot
program in Columbus, Ohio and by all reports, including the
VA's own assessment, the pilot was a huge success. VA found
increased scheduling system efficiencies, improved timely
access to services for veterans, increased productivity, and a
substantial decrease in overtime. However, just 8 months after
that implementation, the VA opted to not expand that successful
pilot any further.
Now the VA plans to implement Cerner's scheduling solution,
but on a separate and faster deployment schedule than the rest
of the electronic health record. I support efforts to improve
scheduling, increase transparency into wait times, and ensure
that all veterans spend less time waiting for appointments.
As we discussed at our Full Committee hearing in July, wait
times continue to be complicated and an unfortunate situation
at the VA that must be addressed.
The VA originally notified Congress of its intent to
implement the Cerner Scheduling Solution in December of 2018.
At that time, VA told Congress that the deployment would be
completed in 2023. VA also told Congress that this would be a
better, more cost-effective solution. However, it has been
difficult to get information to support those assertions.
Last week, representatives from the Office of Electronic
Health Record Modernization briefed staff that the current plan
is now completion in 2025, 2 years longer than was originally
notified to Congress. OEHRM contends that this project will not
add additional costs to the $16 billion EHRM project, but will
require moving funds forward from later in the project.
Basic contracting procedures require that VA should compare
available solutions and decide what works best for its needs.
That is basic-level market research and requirements
development.
I have several serious reservations with this mid-flight
change of approach. First, common business and management
practice would dictate that VA would conduct an apples-to-
apples comparison of off-the-shelf solutions, especially given
the investment of resources and apparent success of the MASS
pilot program.
Second, I wonder why none of this would have been done
before proceeding with the Cerner Scheduling Solution
implementation. It just seems a little backwards to me.
VA said it plans to begin the Cerner implementation in
Columbus. So just months after the successful completion of one
scheduling solution, VA is going to scrap it implement another.
I expect our VA witnesses here today to explain how that is not
wasteful. Further, there is a lot of change being planned,
obviously, in VA health care and change fatigue is a real
concern. So I plan on hearing how the VA is planning to
mitigate that.
I am also concerned that the VA made assertions to Congress
before it had any actual analysis of user needs, of costs, or
of benefits. Despite being 6 months out from beginning system
implementation, the plan seems to be in rough shape. This
include costs which the VA has said won't be finalized until
November.
It is my hope that we can end this hearing with more
information and reassurance that VA has its plan on track. We
need specifics on cost, schedule and change management plan,
and the infrastructure investments needed for this ambitious
project.
I thank all of the witnesses for taking time to be here and
for your work on this incredibly important project, and I look
forward to hearing your testimony.
I would now like to recognize my colleague, Ranking Member
Banks, for 5 minutes to deliver any opening remarks.
Mr. Banks?
OPENING STATEMENT OF JIM BANKS, RANKING MEMBER
Mr. Banks. Thank you, Madam Chair.
The appointment scheduling system is the VA IT system most
badly in need of an overhaul, more so than the financial
system, the claims-paying system, or even the EHR. That was
true 5 years ago when the secret wait list scandal in Phoenix
broke and, despite some incremental improvements to VistA, it
is still true today.
VA spent much of 2018 deciding which scheduling system to
select. The decision in December to go with Cerner did not come
as much of a surprise. After all, VA already committed to buy
the Cerner scheduling package along with the EHR. The idea to
put the scheduling implementation on a separate faster track
seems justified. However, my colleagues and I have spent all
year for the details and analysis, but that information is only
now starting to emerge. I need to see a lot more before I can
put my confidence in this plan.
First and foremost, I was disappointed to see the 3-year
scheduling implementation stretch out to 5 years before it even
begins, and I am not so sure that this still qualifies as,
quote, ``accelerated.''
Secondly, I disagree with the decision to select Columbus,
Ohio, where the EPIC scheduling system was installed as a
pilot, as the first site for the Cerner scheduling system. This
EPIC pilot has by all accounts been very successful. It is true
that Columbus is a convenient site to install Cerner scheduling
because the technical groundwork has already been laid, but by
pulling the plug on EPIC, VA is forfeiting the opportunity,
which may become valuable in the future, to encourage greater
interoperability between Cerner and EPIC.
Thirdly, I cannot help but notice the disparity between the
analysis VA performed by the Columbus EPIC pilot and the
minimal analysis of what the Cerner scheduling implementation
will entail. Cerner and EPIC are both commercial software, but
there seems to be an assumption that everything known about
EPIC is also true about Cerner.
Accelerating the Cerner scheduling implementation still
seems to be the right thing to do. That being said, as
information continues to come in, I am prepared to reevaluate,
and I hope VA will do the same. It is very important that the
political inertia never be allowed to take precedence over the
facts on the ground.
Relatedly, I would like to address a few other topics
pertaining to EHR modernization. The formal partnership between
OIT and OEHRM to build the data interfaces that was so
encouraging at the time of our last hearing seems to have
crumbled. This is a large and critical task and I don't believe
any entity can handle it alone.
Today is an unfortunate anniversary. The Secretaries of DoD
and VA signed their joint commitment statement exactly 1 year
ago today, but the FEHRM still has not been established. The
interim Director and Deputy Director have been working together
amicably for several months, but without an organization
beneath them their effectiveness is limited.
I think we must be realistic about what is achievable at
this point. We need to turn the page on the idea the firm is
standing up the EHRs. The critical decision-making window is
closing. The firm can still be helpful as a governance body to
sustain the EHRs, but we have to calibrate expectations
accordingly.
Before I yield, I want to also revisit the issue of EHRM
cost estimates. Last year, I expressed frustration with the
explanations VA was providing as to the assumptions underlying
the cost estimate. Mr. Lamb and I sent a letter requesting the
raw numbers. What we received in response was the same top-line
estimates and, every time we scrutinized an assumption, it
would suddenly change. OEHRM seems to have reacted to the
Subcommittee's oversight by zealously defending its funding.
The cost estimate is still very much a black box and it is
clear that the actual spending rate is trending farther below
the original projection. While that seems positive at first
glance, OEHRM seems to have achieved it by shifting more costs
onto other organizations within VA, and I intend to monitor
that very closely.
With that, Madam Chair, I yield back.
Ms. Lee. Thank you. I would now like to introduce the
witnesses we have before the Subcommittee today.
Mr. John Windom is the Executive Director of the Office of
Electronic Health Record Modernization at the Department of
Veterans Affairs. Mr. Windom is accompanied by Dominic Cussatt,
Principal Deputy Assistant Secretary, Office of Information and
Technology; Dr. Michael Davies, Senior Advisor to the Assistant
Deputy Under Secretary for Health Access at the Veterans Health
Administration; Dr. Laura Kroupa, the Chief Medical Officer for
the Office of Electronic Health Record Modernization; and John
Short, Chief Technology and Integration Officer for the Office
of Electronic Health Record Modernization.
We also have Mr. Larry Reinkemeyer, Assistant Inspector
General for the Audits and Evaluations from the VA's Office of
Inspector General.
Thank you all for being here.
We will now hear the prepared statements from our panel
members. Your written statements in full will be included in
the hearing record without objection.
Mr. Windom, you are recognized for 5 minutes.
STATEMENT OF JOHN WINDOM
Mr. Windom. Thank you, Madam Chair Lee.
Good morning, Madam Chair Lee, Ranking Member Banks, and
distinguished members of the Subcommittee. First, I want to
thank you and the members of the Subcommittee for your
unwavering support of the Department of Veterans Affairs'
Electronic Health Record Modernization effort. Without your
steadfast support, VA would not be able to deliver this
critical capability in support of our veterans. Thank you for
the opportunity to testify in support of the VA's initiative to
modernize clinical scheduling by accelerating the
implementation of the Cerner Scheduling Solution.
I am accompanied today, as you just mentioned, by Dominic
Cussatt, Principal Deputy Assistant Secretary for the Office of
Information and Technology; Dr. Laura Kroupa, Chief Medical
Officer for the Office of Electronic Health Record
Modernization; Mr. John Short, Chief Technology and Integration
Officer for the Office of Electronic Health Record
Modernization; Dr. Michael Davies, Senior Advisor to the
Assistant Deputy Under Secretary for Health Access, Veterans
Health Administration.
VA currently manages clinical scheduling using the Veterans
Health Information Systems and Technology Architecture, also
known as VistA. According to a VA study, VistA scheduling does
not provide VA with the requisite functionality, usability, and
overarching business benefits. The outdated user interface and
cumbersome manual processes create inefficiencies and prevent
schedulers from viewing the medical provider's complete picture
of available appointments.
As a result, in 2018, VA piloted the Medical Appointment
Scheduling System, also known as MASS, a commercial resource-
based scheduling solution in Columbus, Ohio, to replace the
clinic-based VistA scheduling system. The pilot demonstrated
that resource-based solutions, improve timely access for
veterans, increase provider productivity, and enhance schedule
efficiencies.
Further, the resource-based solution increased visibility
of available appointments, allow providers a comprehensive view
of their entire day, and enable staff to efficiently manage
resources needed for appointments. Because a resource-based
scheduling solution supports delivering better health care for
our veterans, VA will implement CSS to bring these benefits to
all veterans. VA's EHR Modernization contract contains the
licenses to implement CSS across the enterprise to fulfill
interoperability objectives. With congressional consent, VA
will ultimately seek to pull forward funding from OEHRM's life
cycle cost estimate to achieve this initiative. Like MASS, CSS
is a resource-based scheduling solution and will be implemented
in a number of VA facilities in advance of the full EHR
Modernization capabilities.
The Chalmers P. Wylie Ambulatory Care Center in Columbus,
Ohio will serve as the pilot for CSS and will go live in April
of 2020. The Louis Stokes VA Medical Center in Cleveland, Ohio
will serve as the next and larger pilot site for CSS. VA will
leverage the architecture and lessons learned from the MASS
solution by collaborating with key holders' stake from MASS
implementation to ensure these lessons learned are incorporated
into VA's new scheduling plan. VA has established a dedicated
pillar or division within OEHRM to provide oversight of the CSS
integration, deployment, and change management activities.
Further, the pillar will collaborate with partners such as VHA,
OIT, and Veterans Benefits Administration to successfully
implement the CSS solution.
Accelerating CSS implementation will enable VA to provide a
resource-based scheduling solution across the enterprise
sooner, and also replace VistA's Scheduling Enhancement, VSE,
which is the current temporary bridge for scheduling needs.
In August 2019, VA's Office of Inspector General assessed
VA's management of VSE and recommended broadly the VA improve
project management oversight. VA concurred with OIG's
recommendations and is implementing a new process to
independently ensure that IT projects deliver the intended
outcomes.
As demonstrated by our efforts, it is clear VA is committed
to providing the best care to our Nation's veterans. Through
the CSS initiative, VA will provide a state-of-the-market
scheduling solution that upholds the Department's commitment to
improve care and access to staff and on behalf of our veterans.
Madam Chair, this concludes my opening statement. I am
happy to answer any questions that you and the members of this
Subcommittee may have. Thank you once again.
[The prepared statement of Paul Tibbits appears in the
Appendix]
Ms. Lee. Thank you, Mr. Windom.
Mr. Reinkemeyer, you are now recognized for 5 minutes.
STATEMENT OF LARRY REINKEMEYER
Mr. Reinkemeyer. Thank you.
Chairwoman Lee, Ranking Member Banks, and members of the
Subcommittee, thank you for the opportunity to discuss the
Office of Inspector General's oversight of the Department of
Veterans Affairs medical scheduling enhancement efforts.
Our August 2019 report on VA's implementation of the VistA
Scheduling Enhancement Project examined whether the Office of
Information and Technology and the Veterans Health
Administration effectively managed its implementation of the
VSE project.
Although VSE is a relatively small program, originally
planned for a little over $4 million, VSE was intended to
provide essential near-term enhancements. VSE represented a
short-term fix by updating the graphical user interface. VSE
did not change any of the functionality of the VistA scheduling
system, only the look of the screens. Essentially, the screens
would now resemble the calendar screens you might see in your
own Outlook calendar. This seemingly small change was expected
to significantly reduce the time it took schedulers to schedule
appointments.
Since the 1980s, VHA has relied on the VistA system to make
and track patient medical appointments. The technology
underlying this legacy scheduling system used by VA medical
facilities became cumbersome, outdated, and unable to handle
the complexities and volume of VHA scheduling requirements as
they developed and expanded over time. The scheduling system
was also not designed to integrate mobile, Web, and telehealth
scheduling.
As described in more detail in my statement for the record,
VA began to update its VistA legacy program in fiscal year 2000
with the launch of the replacement scheduling application, a
commercial off-the-shelf software program, and ultimately
settling on the VSE as a short-term solution. VSE was a near-
term fix while MASS was to be developed almost concurrently and
would represent the long-term solution. Then MASS was put on
hold to place more emphasis on VSE and, a few months after
that, MASS was restarted and essentially deployed to one
location.
Now VA intends to fast-track the scheduling component of
the new electronic health record, and it stopped any
significant work on VSE and MASS.
VSE's planned completion date of November 2015 extended
well into 2018, a delay of almost 3 years, and the cost
increased from a little over $4 million to almost $7 million.
Our audit identified three key findings. First, VSE
requirements were not properly defined to meet user needs.
Because VSE was intended to be a quick and simple short-term
solution, VA did not put enough effort into developing and
validating the requirements. Users were left out of the
requirements discussion and, as a result, a number of
functionality and usability issues surfaced. For example,
schedulers needed to toggle back and forth between VistA, VSE,
and the patient records system, which effectively negated any
time savings.
Second, insufficient testing during the development phase
led to unidentified deficiencies and, once deficiencies were
identified, contractors failed to address them.
From June 2015 through July 2016, a number of problems were
identified at the initial deployment site in Asheville, North
Carolina, most significant of which was system slowness.
Additional deficiencies included limitations on canceling and
scheduling appointments. Many of these functional deficiencies
could be traced back to inadequate requirements determination.
Third, staff turnover in key management positions delayed
the development and implementation of VSE due to the loss of
project and program knowledge.
We recommended that the VA Assistant Secretary for
Information and Technology, who is also the Chief Information
Officer, enforce required project management processes to
ensure project planning requirements are adequately defined and
supported before starting information technology projects. VA
concurred with the recommendation and requested closure, as it
has implemented a new program management review process through
a policy memorandum signed on July 15th, 2019; we have not
closed that recommendation. We will monitor the Department's
progress and we will follow up on the implementation of the
policy memorandum to ensure it addresses the intent of the
recommendation.
Madam Chairwoman and Ranking Member Banks, this concludes
my statement, and I would be pleased to answer any questions
you or other members may have.
[The prepared statement of Paul Tibbits appears in the
Appendix]
Ms. Lee. Thank you, Mr. Reinkemeyer. I will now recognize
myself for 5 minutes.
When VA first notified Congress of its intention to deploy
the CSS separate from EHRM, the timeline for completion was to
be at the end of 2023, now we are hearing the timeline is the
end of 2025.
Last year, shortly after the successful MASS pilot in Ohio,
an executive from Leidos told the Washington Post, quote, ``We
communicated to the VA that we are able to do a national
deployment in 24 months and we can do it for less than $350
million.
So, based on that information, three issues rise to the top
for me. First, as Ranking Member Banks pointed out in his
opening statements, the very definition of interoperability
means that two systems, especially two systems that top-tier
EHR vendors, would be able to work together. Many health care
systems that have done this have this very same arrangement.
Secondly, Cerner's solution would not completely be
deployed until 4 years after the MASS system would have also
been in place across the finally. And then, finally, the costs
that were projected by Leidos to be $350 million.
Mr. Windom, can you--you know, I just--can you explain the
logic behind making this decision?
Mr. Windom. Ma'am, when we awarded the Electronic Health
Record Modernization contract, it was in--supported deploying
the full suite. Interoperabilities are more readily achieved
objectives by being on the same common solution.
There is a cost to integrating platforms. If you are trying
to implement, in this case the EPIC solution with the Cerner
solution, that is an integration cost and a time-consuming
consideration that had not even been evaluated. We paid for the
Cerner licenses as part of the EHR contract, it made business
sense to not duplicate that payment by installing another
system.
I fortunately led the DoD effort in the acquisition of the
Cerner Millennium solution under a competitive environment. The
scheduling systems are similar, they are both resource-based
solutions, and so sticking with a platform where software
updates would be, if you will, facilitated by that common
solution, it just made sense.
We are--the 3-year verse 5-year deployment track, the 3-
year deployment track was a number drawn without all of the
comprehensive research that is necessary to develop an
integrated schedule. Five years is more appropriate, especially
appropriate for doing it right in a risk-mitigated fashion and
support of the least amount of disruption for our veterans.
I can't speak to the Leidos declaration. I can tell you it
was $700 million ceiling contract, so I am not sure why it
would be articulated that they could be doing it for half the
price. So, I can't speak to that. All I can tell you, ma'am, is
that this is hard, and we are mitigating risk in the best
interest of our veterans, and we think we have a prospective
timeline that does that and delivers the requisite
capabilities.
Ms. Lee. A couple things. First of all, probably my biggest
concern is, I understand that having one system makes
interoperability much easier, but the bottom line is, we are on
the MISSION Act, we are going to be requiring community
providers to talk with the VA. So is your plan that everyone is
going to have to have a Cerner operating system? I mean,
wouldn't you would want interoperability with other vendors?
Mr. Windom. Yeah, I think when you consider, ma'am--the
answer is yes, we agree. When you consider the fact that we
have 130 disparate solutions within the framework of VA today,
integrating--or incorporating another integration challenge.
One of the toughest parts of our job is that we have to
integrate into an existing solution and minimize the disruption
that takes place to our care provisions, and so this is a
measure that minimized that disruption.
We can evolve into the interoperability platform that you
just recommended, and I think that is likely forthcoming, but I
guess I want to make sure it is clear as well is that the
existing contract expires in June of 2020. So we do not have
the opportunity to stay on the present scheduling platform
deployed to the Columbus pilot, hence our desire to get there
first, replace that expiring contract with a new contract to
take us into the future.
Ms. Lee. Thank you. My time is up. I will recognize Ranking
Member Banks for 5 minutes.
Mr. Banks. Thank you, Madam Chair.
Dr. Davies, I supported the decision to put the scheduling
project on a faster track than the rest of the EHR, I supported
the decision to wait until Cerner goes live at Spokane before
starting the scheduling project, but all that was based on
finishing in 2023. VA needed a modern scheduling system years
ago; how can we say 2025 is an acceleration?
Dr. Davies. As Mr. Windom said, Mr. Banks, the date of 2023
was reached in December when there was little visibility into
what the actual project would look like. And now that we have
greater visibility into what Cerner's scheduling implementation
is going to take and we have an opportunity to reflect back on
what happened in Columbus through the lessons learned, we
realize that to do this right, to do the organizational change
management, to get the benefits for the patients of shorter
waiting times, for the providers with doing more work, more
completed appointments, and the schedulers for being able to
incorporate this into the work, we need to take the time to do
it right, start with the pilots in Columbus and Cleveland, and
then use that experience to go as fast as we can through the
rest of the country as it makes sense.
Mr. Banks. How much is that acceleration going to cost?
Dr. Davies. I would defer that question to Mr. Windom.
Mr. Banks. Dr. Davies, how much of the cost is expenses
being pulled forward and how much would be the new expenses?
Dr. Davies. Sir, I would ask Mr. Windom to weigh in on
that.
Mr. Banks. Mr. Reinkemeyer, in your written--do you want me
to defer?
All right. Mr. Windom, do you want to answer that?
Mr. Windom. Sir, we think there are three cost pools.
Again, the infrastructure cost, which is the piece that we are
still evaluating. We want to make sure--we think that the delta
between the EHR, first of all, the primary cost of being
brought forward from out years, if Congress concurs.
The infrastructure piece is that tough piece to calculate,
because what we don't want to do is overly invest in
infrastructure to deliver scheduling and then when the
remainder of the best of suite comes around, some new
innovative approach renders that infrastructure upgrades
obsolete.
So we want to update the infrastructure in the right way
and so we are seeking to finalize the capture of those costs,
sir. So we should have--within a matter of weeks, we should
have actually detailed cost estimates that we would enter into
negotiations with Cerner on.
Mr. Banks. Okay, we look forward to that.
Mr. Reinkemeyer, in your written testimony you state that
the VA expected the VistA Scheduling Enhancement would save
schedulers time, but, quote, ``it failed to deliver on that
promise.'' Can you elaborate on how VSE was supposed to do this
and how you determined that it failed?
Mr. Reinkemeyer. Not via--I cannot elaborate in a technical
sense, but I can tell you that what we found was that during
the requirements development process where the Department was
trying to establish what they needed from VSE, they really
excluded the users. And as part of that--or, you know,
essentially, without getting the user's involvement, the
requirements were not developed adequately, they were not the
requirements, the right requirements, and then when they went
to test it at the initial operating capability site in
Asheville, North Carolina, that was some of the problems that
they discovered. That, you know, having to toggle back and
forth between the VSE system and the computerized patient
records system and VistA was just taking too long.
Mr. Banks. Dr. Davies, do you agree with that?
Dr. Davies. I think it is important to remember that the
users were thrilled at the time that we actually had VSE
funded, that we actually had funding to improve scheduling. The
users had one week to put in contract requirements that would
kick this off, so there was no time during that time in history
for those requirements to be fully elaborated, nor to bring in
the users. So we had to depend on the analysis of alternatives
that had been done 3 years earlier to come up with the broad
statements that we knew were the direction that needed to be--
to go in, and then use the contract time to more fully
elaborate the requirements.
Mr. Reinkemeyer is right, I mean, the users were not
brought in at that time; they were brought in later, so it
became something that was improved over time.
Mr. Banks. Thank you. I yield back.
Ms. Lee. Thank you. I now recognize Mr. Lamb for 5 minutes.
Mr. Lamb. Thank you, Madam Chairwoman.
I am not sure who knows this, so anyone can feel free to
speak up, but the original contract for MASS was $624 million;
is that a correct number?
Okay, I am getting nods, just for the record. That's good.
How much of that $624 million did we spend?
Dr. Davies. Well, I will speak up and say I believe we
spent $17 million on the pilot in Columbus, and then I believe
we also spent another $2.5 million initially on the contract to
do planning for national roll-out. There may be other costs
that I am not aware of, but those are the two numbers that I am
aware of.
Mr. Lamb. Okay. Anyone--
Mr. Windom. Congressman Lamb, I would ask that we take that
for a look-up. That is an OIT project and we would have to
reach into the OIT portfolio to pull out what was expanded
against the ceiling of that existing contract.
Mr. Lamb. Okay. Yeah, I mean, that is an important number
for us to have when we are looking at all the money that is
being spent on something that we are not actually going to use.
Now, in your comments, I think it was Dr. Davies who said
that there were lessons learned from MASS in Columbus and that
that is actually why you want to debut the Cerner solution
there, because you think that there are--what I took from what
you are saying is you think there are aspects of what was
learned in MASS that could be incorporated into Cerner, do I
have that right?
Mr. Windom. Yeah, I would offer, sir, that the driving
element for replacing Columbus is the expiring contract that
supports the present EHR system. That expires in June of 2020;
hence we are trying to get there and replace it before that
contract expires and there is no software support after that.
So it is our intent to replace out a system and then--and,
oh, by the way, as it provides a system that has already
transitioned to a commercial-based, resource-driven scheduling
system, and we feel like that is a way to facilitate better
understanding for deployment across the enterprise.
Mr. Lamb. Yeah, that part I understand, but you chose
Columbus as a pilot site--maybe not you personally, but VA
chose Columbus as a pilot site originally knowing about the
length of that contract. So you chose Columbus for whatever
advantages it presented, inked a $624 million contract, and
then set aside the thing that we built with that $20 million.
So that drives a little bit of the skepticism that you may
be perceiving. I understand it if you have decided that, since
we are going with Cerner, we might as well have a Cerner
scheduling platform, but we are right to be concerned about
sort of the sunk costs of this other idea called MASS, right?
Mr. Windom. I agree, sir, with the sunk-cost element. I
guess I would state that the MASS contract was awarded in
August of 2015, the D&F that established that VA would move to
the same common solution that DoD was on and Cerner Millennium
was awarded in May of 2017. So the commitment to that pilot
site had taken place nearly 2 years in advance of the
contemplation of a D&F.
Mr. Lamb. Right. So that is why I am saying I understand
where you want to go with this now. I think we are still quite
puzzled at the timelines and how long it is going to take and
what the advantages are.
But I guess with my remaining minute and a half I would
like to know, what protocol are in place right now to make sure
that we maximize the lessons learned in Columbus for the Cerner
solution? In other words, how are you going to be sure that
Cerner builds you something there that builds on what seems to
be the strength of MASS?
I mean, I am no expert, but what I can read here makes it
seem like the costs went down, the wait times went down, less
overtime, that kind of thing. So what specifically is going to
take place?
Dr. Davies. We have three things that--first of all, there
are documents that have our written lessons learned that have
been shared among the teams. Secondly, I think we have learned
that it is critical to have the software implemented right. So
we have teams doing site assessments, talking to the people who
did the MASS contract in Columbus, the old contractor, and
connecting them with the new contractor to take advantage of
the work that has already been done.
Secondly, the training of the schedulers. We are
incorporating VA schedulers into the training going forward,
because we learned how important it is to implement the new
system in the VA environment and it takes those VA trainers who
understand the work to be able to do the training.
And, thirdly, we are going to sit down with every single
provider and make sure they understand what their old schedule
looked like and what their new schedule is going to look like
in the old system. Keep in mind that in the old system, we
really didn't have a schedule, we had sort of a data-collection
system where each provider had multiple schedules. In Columbus,
there was an average of six profiles or grids or schedules for
every day that someone came in to work.
We are now translating that into one sort of Outlook-
Calendar based schedule where you can see your work from
beginning to end, and the providers need to be involved in how
their time is deployed and what they are doing.
Mr. Lamb. Thank you. And I am out of time, so I will just
cut you off, but I just want you to know that that last point
is the one I think a lot of us care about the most. We talk to
the providers when we are back in our districts or when we are
visiting these hospitals, and they absolutely have to be
included to the maximum extent possible. I know you all and
Cerner have tried to do that, so please keep your focus on
that.
Thank you.
Ms. Lee. Thank you.
I now recognize Mr. Watkins for 5 minutes.
Mr. Watkins. Thank you, Madam Chair.
Mr. Windom, VA clearly thinks Cerner's scheduling system is
good product, we do too, but the Department has cycled through
three different systems in the last several years. What
assurance can you give us that this will be the last VA
scheduling project?
Mr. Windom. Sir, our goal, our overall goal obviously
revolves around interoperability and delivering a state-of-the-
market solution to our users, our clinicians primarily. And so
what we are getting in this commercial-based product known as
Cerner Millennium is an evolving product that is leveraging the
best of the commercial environment, the best of VA innovation,
and will evolve to the needs of the Department, I think,
realtime.
I know the Columbus site deployment of a different solution
inspires concern, but, again, when you look at the timelines of
how those decisions were made, we would be remiss if we didn't
go there first. The infrastructure needs are minimum, such that
we can inject a very similar functional resource planning tool.
And so I think what you will see is a--again, when we set
these project timelines as well of 5 years, please understand,
we are looking to optimize efficiencies. What we want to give
you is a realistic timeline and then impress you with our
deployment efforts and actually deliver sooner.
So, sir, I think delivering capabilities in this fashion
where these areas don't have to wait on the full implementation
are going to benefit our veterans as a whole.
Mr. Watkins. Thank you.
Dr. Kroupa, I understand the VA has decided not to purchase
the dental electronic health record from the Henry Schein
Company. Your original plan was to implement this system
alongside Cerner. Can you explain the thinking here?
Dr. Kroupa. Certainly. Thank you.
So when our dental council evaluated the Henry Schein
product, they had concerns that it would not meet all of their
needs. And we consulted with DoD who was using that product,
and the decision was made that we needed more time to develop
the functionality that they needed. And so our best course of
action for our initial operating capability was to have an
interface to the current dental record manager that we're using
now and then take some time to develop out a better product for
us to use.
Mr. Watkins. Thank you, Doctor.
Mr. Cussatt, VA schedulers will continue using the VistA
scheduling enhancement system until Cerner is fully
implemented. But the VistA scheduling enhancement still has 23
outstanding defects that impact its usability.
What is your plan to fix the defects?
Mr. Cussatt. Thank you, Mr. Watkins, for the question.
Our plan is to keep VSE in sustainment. We see it as in
sustainment and we will fix issues as they occur. So we do
intend to fix any issues that pop up during the remaining life
cycle of VSE.
Mr. Watkins. Mr. Reinkemeyer, do you have anything to add
to that?
Mr. Reinkemeyer. No.
Mr. Watkins. I yield the balance of my time.
Ms. Lee. Thank you.
I now recognize Mr. Roy for 5 minutes.
[Pause]
Ms. Lee. Okay. We will reset.
[Pause]
Mr. Roy. I apologize for that. I thought we were going to
the other side of the aisle, so I was distracted for a moment.
I appreciate you all being here. I appreciate the
difficulty of the task that we are discussing. My background,
for what it is worth, I have a masters in management
information systems, dealt with system implementations in
previous lives, previous jobs. I was also in government as the
first assistant attorney general of Texas having to deal with
implementation of child support systems. So I understand a lot
of the difficulties that go into that.
I do think that, you know, there are obviously some
concerns here about, you know, the timing and the length of
this and what the ultimate result and the product will be. You
know, the good news is I have spent a great deal of time in my
district in August, the district I represent in Texas 21, which
has Kerrville. And San Antonio, we have got some significant VA
facilities and a large number of veterans.
And the overwhelming sense from the folks that I talk to,
particularly in Kerrville and in Audie Murphy, is that where
things are already moving quickly with MISSION, they are seeing
some very positive results and they are seeing some significant
improvements in what is happening in terms of service, and we
are getting very--I am getting a very positive feeling on that
right now.
But the continued concern being raised, of course, is
everything we are talking about here in terms of how health
records and scheduling and all of these kind of work and how
much that interferes with the larger goals and initiatives of
what we are trying to do to deliver health and deliver services
to veterans.
Obviously, that is a statement and we are here to hear from
you all. But it is--I will just say, you know, one last point
on this. Right. In the time since this timeline began, right,
you know, this didn't exist. You know, 11 years ago, 12 years
ago this didn't even exist. That is both a statement of the
difficulty of the task, as technology has evolved so quickly,
and it is also a statement of can't we get this done. So many
great things have occurred in such a quick way.
And the one thing I would throw out there that I think is
worthy of discussion, the Army Futures Command, which I am
proud to represent in Austin, Texas has a co-located facility
that is located in downtown Austin, but with the Capital
Factory, which is a place in Austin, Texas where you bring a
bunch of entrepreneurs together to smash people together for
idea generation.
I would like to at least discuss the benefit of having, you
know, a VA component to something like that. It doesn't have to
necessarily be in Capital Factory in Austin, Texas, although I
would obviously advocate for that. But that kind of thinking
and approach to figure out how we can sort of bust through what
is a perpetual problem in government systems development.
I mean, let's just be honest. We all know it is a problem.
My dad worked in systems development in the Internal Revenue
Service in the 1980s. I kind of grew up around this and
understood it and saw it in terms of tax systems modernization.
It is hard and I get it.
But sometimes we get caught up in all of the buzz words of
interoperability and synergies and, you know, we can go down
the list of the words that will make up 80 percent of testimony
typically when we hear about these technologies. And I get it.
I understand it. But we get wrapped around those axles in my
opinion.
A couple of quick questions if you don't mind.
Dr. Davies, Dr. Kroupa, the scheduling system is obviously
very important to wait times. What impact will accelerating the
implementation of the Cerner scheduling system have on wait
times calculation and on the referral system?
Dr. Davies. We believe, and our experience has been that a
resource based system with a single calendar improves wait
times. That was our experience in Columbus, Ohio. And we
anticipate that benefit throughout the rest of the country.
Mr. Roy. Will that be experienced immediately?
Dr. Davies. No. In Columbus it took about 6 months to get
to an equilibrium where we had the system generated time stamps
kick in, and then after that we saw the improvement in wait
times.
So part of the issue was measurement of the wait times. The
other part of the issue is working more efficiently with better
tools.
Mr. Roy. Okay.
Mr. Cussatt, the Office of Information and Technology has
cycled through a lot of different project management processes
over the years. I understand you have a new one called Agile
DevOps. VA's record of IT capabilities versus IT investment has
been raised as a question over the years.
How is this new process different and what should we expect
from it?
Mr. Cussatt. So you are correct. We have, since about 2016,
adopted both ITIL practices and DevOps and Agile practices, and
we are actually in the process of renaming DevOps to DevSecOps
to show the security we build in.
So that coupled with our VIP acquisition process embeds all
the best practices for engineering and architecting and program
management into our programs, and we have been having some
success around that with over a 90 percent delivery rate on
time and within scope and budget for our programs.
Mr. Roy. Well, thank you very much.
Ms. Lee. Thank you.
I now recognize myself for 5 minutes.
Mr. Reinkemeyer, based on the recommendation made in the
VSE report, would the Inspector General be concerned about the
testimony we are hearing right now regarding the lack of
concrete plans? Why or why not?
Mr. Reinkemeyer. So, yes, we would. As was just stated,
there have been a number of IT processes over the years. They
all offer, you know, advantages and disadvantages and you could
compare and contrast them.
But essentially it boils down to having a discipline in
place to manage the system that you are developing. You have to
be able to balance the agility that you want with the
discipline to ensure that cost analysis, business cases,
testing, all of the required elements are in place.
And that is really what is boiled down to the findings that
we have had over the years in whatever system we are looking at
is, is there a sufficient discipline to ensure that all of the
key components of that system are managed and there is
visibility and there is oversight over all of those functions.
Ms. Lee. Thank you.
At a recent briefing, Mr. Windom, you indicated that the VA
expected to achieve a substantial out year savings in Fiscal
Years 2023 to 2027 due to an expected reduction in IT costs.
You specifically cited the declining costs of items such as
computers, servers, networking equipment as contributing
factors. I am concerned that these estimates do not consider
the need to keep pace with the evolving state of technology as
my colleague, Mr. Roy, pointed out.
Yes, it is true that a laptop that costs $2,000 today might
cost $1,000 next year. The government, however, should not be
planning to buy outdated technology as a cost savings. What is
the strategy? I mean, where did you get to that conclusion?
Mr. Windom. Ma'am, I think those remarks may have been
attributed to me by accident. I have done no analysis of out-
year equipment purchases. I would defer to my Chief Technology
and Integration Officer on where he thinks innovation is going
in the way of end user devices, whether it be as a service or
whatever.
So I have no recollection of ever making such a statement
because I have done no analysis in that arena.
Ms. Lee. Mr. Short, can you weigh in on that?
Mr. Short. Yes, ma'am. I didn't make that statement either,
but, yes, I will weigh in on that.
In that regard, it would be included in the Cerner cost
throughout the program more and more of the EHR functions,
scheduling functions, everything is included in the cost that
we have laid out for the 10 years. And as over time, all the
VistA pieces will be deprecated on the EHR side and OIT is
looking at the other portions of VistA.
So for many of the systems we have identified, there is
over 60 that we have identified outside of VistA that would
also go away through that 10 year transition, those would be
the ones that could be attributed to that, reducing the servers
and overall maintenance of those systems.
Ms. Lee. Okay. But let me be clear. So we are going to
start--you are going to start the scheduling process literally
from the east coast going west. But you are starting the EHR
from the west coast going east.
So how do you plan--like, you know, there is infrastructure
issues, obviously. And, by the way, you are not going to
start--you are not going to implement the EHR in Columbus until
2025 or 2026. So have you accounted for the infrastructure
needs with that sort of crisscrossing the country?
Mr. Windom. Ma'am, I eluded to it earlier. That is why it
is so important to get the infrastructure planned right. I
don't want to introduce to you costs that need to be brought
forward until we put the appropriate rigor in what is the, I
don't want to say the minimum level of infrastructure
investments, but the right level of infrastructure investments
to support just the scheduling deployment.
We are doing current state reviews just like we did on the
full implementation. And so we will have better visibility on
what actual infrastructure costs will be.
As you know in the pacific northwest, we discovered a
substantial refresh element that needed to be done in advance
of deploying the EHR. We want to put the same rigor, again,
that is why we are going to go Cleveland as a second pilot
because we believe Columbus has already been updated, upgraded
based on our assessments, and so it wouldn't be a true
representation of what is out there in the enterprise because
as you have indicated before, they have deployed the MASS
solution. So they are primed to receive the Cerner Millennium
Solution which would not be a representative example of the
environment.
So CSRs, ma'am, will give us that visibility where we have
turned on a CSR for Cleveland. And so we expect to have
realtime data for you in the coming weeks.
Ms. Lee. Okay. So we are expecting some cost estimates
around November. Is that included in this or is that separate?
Mr. Windom. Ma'am, the only remaining cost estimate we have
for you is infrastructure. We have got visibility on what we
believe the EHR--again, a separate labor force has to be set up
in Columbus. That is a cost that is separate from the licensing
to change management which is already incorporated in the
contract.
In addition, we think--as you heard, and if I may respond
to a question, we think we are bringing stability to this
because the pillar that leads scheduling will be under the
OEHRM umbrella, not under OIT, not under VHA. So there won't be
any mixed messaging provided to Cerner that they could
potentially capitalize on. We will be directing, with the
support of OIT and VHA, our partnership remains paramount. And
so we think we have a plan that stabilizes our deployment, vice
creates any additional risk.
Ms. Lee. I am over time.
I now recognize Ranking Member Banks.
Mr. Banks. Thank you, Madam Chair.
Mr. Windom, I think VA is making the right decision to go
live with the Cerner functionality at Seattle and American Lake
in October 2020 rather than partial functionality in April and
the rest later.
Can you elaborate on how that decision was made?
Mr. Windom. Sir, the one thing I have to keep reminding
myself, especially as a non-clinician, is that this is about
the end users. This is about changed management and user
adoption success.
And so I would rather defer that to Dr. Kroupa because that
is what our strategies are primarily revolved around, is making
sure we roll out capabilities in a way that maximizes user
adoption vice inhibits the education and training process that
we know.
So, sir, if I may defer to Dr. Kroupa.
Ma'am.
Dr. Kroupa. Certainly. Thank you.
So the decision to go to Puget Sound in October was really
based on the timeline that was needed to develop the workflows
and the functionality that is needed at a highly complex, two-
division, academic medical center, quite a bit more complex
than Spokane and needed more time for content development and
more time for configuration.
So it was based on the needs of the medical center and the
veterans there.
Mr. Banks. Okay. Good
Dr. Kroupa, as well, somewhere always has to go first and
Spokane is first. We have talked in previous hearings about the
normal productivity hit when a new EHR is installed. 30 percent
is typical. I know you are trying to mitigate that impact, but
do you think going live in Spokane in March with partial
functionality would make that productivity hit bigger or
smaller?
Dr. Kroupa. I just spent a week in Spokane last week
talking about this topic with them. I think we have a good
handle and getting a better handle on all the different impacts
that the functionality will have and just the basic
productivity hit that any EHR is going to do.
Truthfully, I don't think that most of the challenges we
are going to face are a problem because of the capability set.
They are getting, you know, a lot of capability, many things
are much of an upgrade from what they are currently getting in
CPRS. And they are ready for those workarounds.
So I think the capability set part of that is really a
minor component of the overall mitigation we need to do for
productivity for any EHR implementation, particularly an IOC
site where we are all learning as we go.
Mr. Banks. So when is the final decision point for Spokane
to go live in March?
Dr. Kroupa. Well, we will be making those decisions as we
go through different milestones. Obviously, we have testing
coming up. We will have training coming up. So we will be
working with VHA and OIT, and within--
Mr. Banks. But at what point--
Dr. Kroupa [continued]. --the program.
Mr. Banks [continued]. --at what point is it all systems
go, go live in March?
Dr. Kroupa. Well, I believe that we will be making that
decision in March right as we make sure that everything is
ready. But we will have milestones before that to look at that
timeline, if we have to alter that.
Mr. Banks. Okay. This next question is for Dr. Kroupa and
Dr. Davies.
As you know, I am concerned about the staffing vacancies,
physical infrastructure deficiencies and limited community care
capacity that already exists in Spokane. EHRM did not create
these problems, but the go live will make them more apparent,
even under the best circumstances.
What is VHA's plan to remediate those conditions before the
Cerner go live?
Dr. Kroupa. I can take that.
So, again, I spent last week in Spokane going through that
in detail, working with their front line staff to understand
their challenges, what they thought they needed. There is a
plan being put together to figure out--to put together and
describe the number of staff that they will need, what other
kind of mitigation policies we can use in terms of Telehealth,
bringing in providers from other sites, doing things remotely,
distributing some of their work remotely as well.
So we are developing that comprehensive plan and there is
active involvement of VHA and the visit and the facility in
that plan.
Mr. Banks. Anything else?
Dr. Davies. I would add that one thing we learned in MASS
implementation was that it made the scheduler's job easier, and
at one point our scheduler turnover had dropped quite a bit
from the baseline. So we hope and expect that those
improvements will bear out in the remainder of the
implementation.
Mr. Banks. Okay. Well, Dr. Davies, when staff visited
Spokane last month, they requested the data on staffing
vacancies and physical space deficiencies. The medical center
compiled it and sent it up to the Office of Congressional
Legislative Affairs. But we have yet to see it.
So I would like for you to provide that information by the
end of the week, if that is at all possible.
With that I will yield back.
Ms. Lee. Thank you.
I will now recognize myself for 5 minutes.
Mr. Reinkemeyer, first of all, we are expecting to see some
cost estimates in the next couple of weeks. What information or
data would be necessary to show that the VA had conducted a
proper analysis of how much the program will cost?
Mr. Reinkemeyer. Well, all of that should be outlined in
their new program management review process. But I would expect
to see the infrastructure costs, the system costs, the changed
management costs, really the life cycle costs of the system.
And if they have put a thoughtful plan together on that and
displayed that and made that visible for all to see, then you
could certainly have a discussion about whether it is the right
cost.
Ms. Lee. Thank you.
Would OIT have concerns about the reliance on future cost
savings in this case?
Mr. Reinkemeyer. I would have to--
Ms. Lee. See it?
Mr. Reinkemeyer [continued]. --we would have to take a
look. I mean, I think everybody at this table is well
intentioned in trying to move it forward.
Ms. Lee. Yeah. Mr. Cussatt or Mr. Windom, our understanding
is that the OEHRM is responsible for the hardware that is
needed for the deployment of the EHRM. The issue of funding and
planning for long term technology refresh cycle comes up every
EHRM briefing.
We are told that this is OIT's responsibility. So how are--
what is the plan to meet that responsibility?
Mr. Windom. Ma'am, I think based on the way the funding
profile is, is that you have entrusted OEHRM to fund those
related costs associated with EHRM efforts. I can tell you the
partnership with OIT is imperative to reach our objectives.
They have the contract vehicles in place in support of many of
these commodity purchases that we need to make. So we have
been, in fact, utilizing their subject matter expertise out in
the field, their contracting vehicles to support getting things
on contract in support of our timelines. That has been working
well.
And so I would defer to Mr. Cussatt for any additional
comments.
Mr. Cussatt. Yeah. I would agree with Mr. Windom. We have a
very close partnership to look at all of the sites that are
coming up for deployment, determine where they are in the
modernization that is needed to support the Cerner system.
We go ahead and pay for the things we already had planned
in the course of our normal modernization, and then we work
with OEHRM to determine the things that need to be accelerated
and done more quickly for that site. And then we work to cost
that out through the OEHRM program.
Ms. Lee. So is it safe to say that you will be responsible
for the likely refresh that is going to be needed at the
facilities that go live with the CSS early on, but then won't
get to the rest of the EHRM until the end of the deployment
cycle?
Mr. Cussatt. Yes, ma'am. We are responsible for the IT
underpinnings and infrastructure for all of the sites for
either CSS or Cerner to make sure they are ready for
deployment.
Ms. Lee. Okay.
I want to move on a little bit to management and
governance. And, Mr. Reinkemeyer, what key management concerns
did you find after looking at the VSE and what recommendations
did you make?
Mr. Reinkemeyer. So we made one recommendation and that was
really to ensure proper oversight and that the project
development process was followed.
As I mentioned earlier, there has been a number of
processes that have been in place over the years and they all
provide some value. It really comes down to leadership and
ensuring that they are following, ensuring that the T's are
crossed and the I's are dotted, ensure that there is a
discipline process that protects, you know, the taxpayer's
money and provides reasonably assurance that the system is
going to be developed.
One of the concerns that we had with VSE--and as I pointed
out, it is a very small project. So certainly it is not fair to
maybe compare it to the CSS system. But there is such a
constant shift in direction with VSE as I pointed out. For a
while it was the highlight, then it shifted to MASS, then it
went back to VSE and now it is ultimately with the Cerner
scheduling.
I think that is what I would highlight.
Ms. Lee. Okay. In June you told, Mr. Windom, you told the
Committee staff that OEHRM was only partially staffed. How many
positions have you filled and how many are open?
Mr. Windom. Ma'am, we filled 102 positions since we last
spoke.
Ms. Lee. Okay.
Mr. Windom. And we still have our primary goal of 265 to
285 as the outstanding goal. So we feel like we have about 150
plus more vacancies. But, again, I can't emphasize enough, we
want to bring the right people into the portfolio, which we are
doing, and with the right expertise. But we appreciate your
continued support. As we wrap up, we think we are making
substantial headway.
Ms. Lee. Great. Thank you.
I am finished. I would now recognize Ranking Member Banks.
Mr. Banks. Thank you, Madam Chair.
Mr. Cussatt and Mr. Short, continuing on the subject of
Spokane, I understand the data center is directly underneath
the kitchen and leaks have become almost routine. The only way
to solve it would be to relocate the data center or to relocate
the kitchen. And until that happens all anyone can do is put a
tarp over the servers.
Again, EHRM did not create this problem. What is the long
term plan to solve it?
Mr. Short. Mr. Banks, VHA owns the facilities. They have
the construction. It is one of the items that we laid out and
Cerner also called it out in the current state review, CSR,
that they did. And VHA is in the process of doing contracting
to resolve that.
For the record, we can come back and give you the plan that
they have.
Mr. Banks. All right. Anything else?
Mr. Cussatt. Yeah. OIT is working very closely with VHA to
look at all the data centers, the closets, make sure that their
current state, if they can't be changed by the time of the
rollout is within performance tolerance levels and can be
managed. But it is our aim and intent to work with VHA to
upgrade them for the long haul.
Mr. Banks. Okay.
Mr. Short, a recent slide deck provided to the Committee
indicates that the data interface for the Cerner scheduling
system for the initial sites has always been ``resolved.'' What
does that mean?
Mr. Short. That was a tracking document for--we have
several sets of interfaces we are tracking. One is for the EHR
going live, one is for CSS going live. So the view you were
looking at in that slide would be for the EHR going live.
So the resolve meaning on that slide was we don't need the
integration for the Cerner scheduling to VistA for it to go
live with EHR. We need it for CSS. Another slide would show you
that we have a plan for that. We are going to replicate the
integration that was done for Mass, with a couple of changes to
it, the same interface engine.
So resolve means we don't need it for EERH to go live and
we know what the plan is and design ins, and we have the
interface control document for the planning for the scheduling.
Mr. Banks. Okay.
Mr. Windom, one of your project metrics is a quote
``version 1 story,'' meaning an activity that must be performed
in order to deploy the EHR with initial capabilities. How many
of those are currently blocked and how many are blocked because
DoD and VA haven't made a decision?
Mr. Windom. Sir, I think the--your comments over the past
hearings revolving around the importance of DoD and VA working
together has created that synergy.
I can tell you we had 27 joint decisions that needed to be
made approximately 2 and a half months ago, and we are sitting
at a point where all but 8 have been made. So that progress is
fantastic.
So this is just the challenge of integrating 2 major
agencies with differing methodologies in many cases. I feel
like we have been moving progressively in the right direction.
I would also add because you made a comment about the FEHRM,
the FEHRM has been involved in that process and facilitating
those.
So I think the evolution that you wanted to see is
happening. It never happens as fast as you would like. But I
think it is happening and I think it is largely because of
Subcommittees like this, sir.
Mr. Banks. Thank you very much.
With that I yield back.
Ms. Lee. Thank you.
I would like to ask this for all the witnesses, and this is
with respect to governance and accountability.
With regard to the implementation of the scheduling
solution, what do each of you view the responsibility of OIT,
OEHRM and VHA?
Mr. Windom. Ma'am, I believe that under the Dep. Sec. I
report directly to the Dep. Sec. on behalf of matters as it
pertains to OEHRM. Scheduling is clearly a parameter and an
element of the contract we awarded to Cerner. We will be
modifying the contract because the deployment of scheduling
separately is a new scope item, but is a scope well within the
capability set that presently exists in the contract.
I believe the accountable person is me. O and VHA are vital
teammates. They are the ones who control the users, the
infrastructure, the network, the things that we have to ride on
and need to support our overall success. So that partnership is
inseparable.
And so that is the way that we are moving forward in our
strategies is OEHRM has the rose, if you will. But being able
to tap into the resources of OIT and VHA are critical to our
success, and I think those partnerships not only have evolved,
but continue to evolve in support of our ultimate mission
objective.
So that is how I would describe it, but I would defer to
OIT and Mr. Cussatt if he had any additional comments.
Mr. Cussatt. I certainly agree with everything Mr. Windom
said. A few things we do in OIT to hold up our end of the team
is we instituted; in light of the IG report we instituted a new
PMR process for all of our programs where we have some rigor
for reviewing those programs regularly.
We have a governance process that if there are any issues
that pop up during the PMRs, it goes up to the governance
process. And this is all overseen by our new quality,
performance and risk office that houses our chief risk officer
and our chief audit executive.
We have divested that office of any operational duties so
that they are sort of our internal conscience so to speak to do
some internal validation and verification and audit and make
sure that everything stays on track. And they can inform the
CIO and me as his deputy when things are off track, and the
governance board.
And, lastly, I will say I see it as our responsibility and
VA CIO's Office to build a very strong partnership with DoD
CIO's office. I co-chair with my counterpart, the principal
deputy in the DoD, Essye Miller, an IT steering committee. We
are going to meet actually this afternoon where we have a forum
to talk about interoperability issues. We want to break all
barriers down.
So I see that as part of our responsibility to build that
relationship.
Dr. Davies. Speaking for VHA, I would say that we represent
the end users and the interests of the schedulers and the
physicians and non-physician providers who are using this
system. Our strong interest is making this right, to make sure
that this works in the day to day workflow so that they can
serve the veterans' needs every day.
We do, ma'am, as you may know, maybe 60 million
appointments a year in VA. So this is right at the heart of our
mission delivery.
In that vein, our office is organized in 10N operations and
we have direct connections to the facility, network and
facility leadership. And we are very interested in the
organizational change of management, in the training, in the
preparation of the users for the go live date, and in the
continual improvement of using the information afterwards to
improve it.
So I see it as 2 steps. We need to get the information
system in place. And these guys are going to do that, and we
have a great team, you know, with working together with them.
Once that information system is placed, then we need to
train the users, especially the mid-level managers, to help
take the information, the new information that they now have,
and use it to continually improve care for our veterans and the
efficient operation of our system.
Ms. Lee. Thank you.
Dr. Kroupa. Well, I don't have too much to add because I
agree with Mr. Windom. But I see myself as, you know, I am a
clinician. I have been in VHA for over 30 years. So I strongly
feel that we are the servant for VHA to bring this
functionality to the frontline users and to veterans, including
VBA in that as well.
So we have definitely--I see myself as a champion for the
functional user, focused on this one mission so that they can
provide care to veterans.
Ms. Lee. Thank you.
Mr. Short. Ma'am, thank you for the question.
I agree with what other people have said. I would also add
that OEHRM has the overall end to end picture for the whole
program for each modernization, working with all these other
offices and with DoD to make sure that we provide the right
level of service and care to the veteran and for the clinician,
for VHA to maintain the facilities and improve the facilities.
So we put the EHR into the right facilities. We don't have
a leaking data center. We have the right data center. We have
the right facilities. And also to provide the staff for
guidance, direction testing, planning and validation to improve
that clinical care. And for OIT to continue to maintain the
systems, continue to facilitate our work with the system so
OEHRM and Cerner can modify the systems as necessary, and also
to provide access to those systems, and to also, as Mr. Cussatt
said, to really get that partnership with DoD for IT and cyber
to make sure that we can move through quicker and faster.
Ms. Lee. Thank you.
Well, I would like to thank all the witnesses for your
participation today, and I hope we can continue to work
together as this Subcommittee continues its oversight of the
technology modernization efforts.
Obviously, the EHRM project is a big concern and I don't
need to tell you how important and complex it is as you deal
with it on a day to day basis. But we are also looking at
numerous projects across the VA, most importantly, as Dr.
Davies said, to make sure we are providing the utmost care in a
timely manner to our veterans which is obviously where the
scheduling program comes in.
I think on my behalf, I mean, I think the biggest concern
here is we had a--you know, looking at the history of IT roll
outs in the VA and the fact that we had a scheduling pilot that
got rolled out and was successful and achieved the goals, and
then we are scrapping it obviously because it was included in
the contract with Cerner. As you eluded to, Mr. Windom, that,
you know, this interoperability and the lack of complexity in
terms of having another platform was a driving factor in making
that decision.
I will caution that my biggest concern with the EHRM
project is that you, all of you have the potential to really
revolutionize electronic health records not just for the DoD
and VA, but for all of health care providers across this
country.
And then as, you know, we are tasked with looking at how we
reduce health care costs across the country I think this
project, if rolled out and done well, will be a vital
determinant of that. And so when we look at making decisions
such as scrapping what was already done that was successful and
going with one provider, on developer. I have major concerns
with that, that we are basically developing a monopoly.
And so I would like to see as we move forward in this what
moves we are making to make sure that other vendors and other
providers have access to this type of technology without having
to make a contract with one provider moving forward. I am very
concerned about that.
And, you know, right now my biggest concern is we are
making these big decisions, yet we don't necessarily have a
handle on the costs or the plans. And I know that is
forthcoming. So we are looking forward to seeing that. We are
looking forward to the OIG facilitating and moving forward
because, clearly, the technology is there. It is obviously
individual--you know, we have it in the management and the
decision-making process is really what is going to determine
the success of this program.
And, you know, I can't stress the complexity of this and I
respect all the work that all of you do, and I continue to
stress that we are working together to make sure that this is a
success, most importantly for the veterans of this country.
So thank you all for your time and we will see you later.
Thanks.
All members have 5 legislative days to revise and extend
their remarks and include extraneous material.
And the hearing is now adjourned.
Thank you.
[Whereupon, at 11:20 a.m., the Subcommittee was adjourned.]
A P P E N D I X
----------
Prepared Statement of John H. Windom
Good morning Madam Chair Lee, Ranking Member Banks, and
distinguished Members of the Subcommittee. Thank you for the
opportunity to testify today in support of the Department of Veterans
Affairs (VA) initiative to modernize clinical scheduling by
accelerating implementation of the Cerner Scheduling Solution (CSS). I
am accompanied today by Dominic Cussatt, Principal Deputy Assistant
Secretary for the Office of Information and Technology, Dr. Laura
Kroupa, Chief Medical Officer for the Office of Electronic Health
Record Modernization (OEHRM), Mr. John Short, Chief Technology and
Integration Officer for OEHRM, and Dr. Michael Davies, Senior Advisor
to the Assistant Deputy Under Secretary for Health for Access, Veteran
Health Administration (VHA).
I want to begin by thanking Congress, and specifically this
Subcommittee, for your continued support and shared commitment to the
success of the Electronic Health Record Modernization (EHRM) program.
Because of your unwavering support, VA's mission of improving health
care delivery to our Nation's Veterans and those who care for them
while being a responsible steward of taxpayer dollars continues.
Background
VA currently manages clinical scheduling using the Veterans Health
Information Systems and Technology Architecture (VistA). According to a
VHA requested analysis, VistA scheduling is clinic-based, so the system
has one profile for each clinic in which a specific provider works.
Given that providers often work in five or more clinics, disparate
profiles prevent schedulers from viewing the provider's whole
utilization picture and efficiently deploying VA resources.
Additionally, its outdated user interface and manual process steps
create inefficiencies. VistA does not provide VA the requisite
functionality, usability, and overarching business benefits.
Therefore, in 2018, VA implemented the Medical Appointment
Scheduling System (MASS), a resource-based, commercial off-the-shelf
scheduling solution to replace the clinic-based VistA scheduling
system. The MASS pilot occurred at the Chalmers P. Wylie Ambulatory
Care Center in Columbus, Ohio, and showed that the resource-based
solution's intuitive user interface simplified scheduling processes,
increased scheduler productivity, and tracked provider utilization to
ensure efficient use of VA resources. Additionally, MASS standardized
reporting processes, increased visibility of available appointments,
and added greater functionality to support timely access to care.
VA's MASS contract agreement expires in June 2020. VA will use its
EHRM Indefinite Delivery/Indefinite Quantity (ID/IQ) contract with
Cerner to accelerate its CSS implementation, since the contract already
includes licenses to implement the CSS across VA's enterprise. To keep
capabilities in the hands of clinicians and standardize scheduling
processes across the enterprise, VA will leverage the architecture and
lessons learned from the MASS solution. VA is collaborating with key
stakeholders from the MASS implementation and Cerner teams to ensure
these lessons learned are incorporated in VA's new scheduling
initiative.
Like MASS, CSS is a resource-based scheduling solution that will
increase scheduling efficiency, provider productivity, and ensure
Veterans' timely access to care. The Chalmers P. Wylie Ambulatory Care
Center will serve as the pilot site for CSS, with Go-Live scheduled for
April 2020. The Columbus facility was specifically chosen due to the
site's expiring contract. The site assessment has been conducted and is
under staff review. Thereafter, the Louis Stokes VA Medical Center in
Cleveland, Ohio, will serve as a larger pilot site for CSS.
After this pilot, VA will replace the VistA scheduling system with
CSS on an accelerated timeline to facilitate the delivery of high-
quality health care to our Nation's Veterans. VA believes there is a
return on investment in productivity and efficiency by accelerating the
scheduling component from the EHRM effort. This implementation plan
will provide resource-based scheduling to VA facilities five years in
advance of full electronic health record (EHR) capabilities and allow
the VA to conduct current state reviews on the state of the
infrastructure to inform future year funding requests. Veterans and
end-users will benefit from an accelerated CSS implementation schedule
by:
Receiving a resourced-based scheduling solution that is
significantly more dynamic than the current clinic-based VistA system,
as CSS will provide enhanced views, reporting ability, and utilization
tracking;
Bringing state-of-the-market EHR capabilities across VA's
enterprise sooner; and
Building partnerships and leveraging training and change
management processes to aid in the full deployment of EHRM
capabilities.
Implementation Planning and Strategy
VA established a dedicated pillar within OEHRM to provide oversight
for CSS implementation. Government personnel and contractor staff will
support the scheduling modernization effort by providing expertise
based off the full EHRM effort in areas including deployment and change
management. The pillar will collaborate with end-users, VHA, (OIT),
Veterans Benefits Administration, Office of Technical and Integration
partners, and Cerner CSS implementation personnel to support the
transition to CSS.
Understanding that many transformations fail due to lack of
leadership buy-in or cultural resistance to change, VA and Cerner staff
will deliver on-site training in advance of CSS Go-Live, ensuring end-
user readiness and continuity of care for our Veterans. After the
Columbus CSS pilot, VA will refine its training and implementation
methodologies in support of anticipated full enterprise implementation
by 2025. Keeping in mind that front-line staff have important work to
do on behalf of our Nation's Veterans, CSS training will be conducted
on flexible schedules throughout the week, including weekends.
Site Readiness Activities
VA will proactively engage facilities across the enterprise to
prepare each site for the scheduling modernization effort. VA's
implementation process includes meetings with VA facility leadership
and staff, change management strategy, communications to end-users,
site assessments, configuration, testing, training, and Go-Live support
at each site. VA will look to internal and Cerner expertise to identify
requirements and transition sites to the new CSS platform.
VA has awarded a task order (TO) for pilot site survey activities
to gather the requirements for implementing CSS; VA will continue using
the ID/IQ contract structure, awarding firm-fixed price TOs as
requirements are validated. The Columbus and Cleveland pilots will
enable VA to better understand infrastructure requirements. VA will
integrate lessons learned and efficiencies gained from the pilots and
EHRM's efforts initial operating capability into future
implementations.
Funding Profile
VA plans to request bringing forward EHRM funds from the out-years
to support an accelerated CSS implementation. Planned expenditures will
support infrastructure modernization, accelerated training, and help
desk expansion services. VA is building a funding profile for
accelerated CSS implementation in collaboration with stakeholders and
will solidify the funding required to implement CSS at other facilities
based on factors identified during site surveys, including facility
size, complexity of legacy systems, and staffing requirements. After
the pilots in Columbus and Cleveland, Ohio, are complete, VA will have
a better understanding of the funding required to deploy CSS across the
enterprise.
VistA Scheduling Enhancements
In May 2014, VA developed VistA Scheduling Enhancements (VSE) as an
interim scheduling solution to fulfill patient scheduling needs until a
commercial scheduling solution could be implemented. VSE acts as a
bridge from legacy VistA to CSS, improving appointment reliability and
scheduling workflow functions until the CSS platform is fully in place.
In August 2019, VA's Office of Inspector General assessed VA's
management of the VSE project and recommended that VA improve project
management oversight so that project requirements are adequately
defined and supported before undertaking information technology (IT)
projects. VA concurs with this recommendation and VA's OIT is
implementing a new program management review (PMR) process that will
independently ensure IT projects are healthy and deliver the desired
outcomes. With the signing and issuance of a policy memorandum, dated
July 15, 2019, the new PMR process is now in effect. With this process
in place, VSE will continue to help VA providers achieve better
continuity of care for Veterans until their permanent scheduling
solution arrives.
Conclusion
EHRM's program efforts will enable VA to provide the high-quality
health care and benefits that our Nation's Veterans deserve, and CSS is
a vital component of the care delivery experience. VA leadership is
committed to successful implementation of CSS and believes that this
effort will improve our delivery of quality health care to Veterans. VA
will continue to keep Congress informed of milestones as they occur.
Madam Chair, Ranking Member, and Members of the Subcommittee, thank you
for the opportunity to testify before the Subcommittee today to discuss
one of VA's top priorities. I am happy to respond to any questions that
you may have.
Prepared Statement of Larry Reinkemeyer
Chairwoman Lee, Ranking Member Banks, and members of the
Subcommittee, thank you for the opportunity to discuss the Office of
Inspector General's (OIG's) oversight of the Department of Veterans
Affairs medical scheduling enhancement efforts. The OIG is committed to
serving veterans and the public by conducting oversight of VA programs
and operations through independent audits, inspections, reviews, and
investigations. Ensuring veterans have timely access to quality care is
a top priority and can only be accomplished through accurate and
efficient scheduling systems.
In August 2019, the OIG published the report, VA's Implementation
of the Veterans Information Systems and Technology Architecture
Scheduling Enhancement Project Near Completion.\1\ This audit detailed
how the Office of Information and Technology (OIT) and the Veterans
Health Administration (VHA) managed the implementation of VA's Veterans
Information Systems and Technology Architecture (VistA) Scheduling
Enhancement (VSE) project. The OIG determined that the VSE project
management team-which included OIT program and project managers and VHA
project managers-did not effectively manage the project to ensure
scheduling enhancements were adequately developed and met users' needs.
---------------------------------------------------------------------------
\1\ VA's Implementation of the Veterans Information Systems and
Technology Architecture Scheduling Enhancement Project Near Completion,
August 20, 2019.
---------------------------------------------------------------------------
BACKGROUND
VistA was designed as an electronic health record system with a
scheduling component. Since the 1980s, VHA has relied on the VistA
system to make and track patient medical appointments. The technology
underlying this legacy scheduling system that is used by VA medical
facilities became cumbersome, outdated, and unable to handle the
complexities and volume of VHA scheduling requirements as they
developed and expanded over time. The scheduling system was also not
designed to integrate mobile, web, and telehealth scheduling. In fiscal
year (FY) 2000, VHA determined that VistA should be replaced and
launched the Replacement Scheduling Application (RSA) project to look
for a commercial off-the-shelf (COTS) software program as a solution.
In April 2002, VA determined that no COTS software developers were
willing to make their scheduling application compatible with VA's
systems. The VHA chief information officer at that time decided to
significantly change the scope of the project from a COTS solution to
an in-house build of the scheduling application. In March 2009, a memo
from the under secretary for health to the acting assistant secretary
for information and technology stated that the RSA project had not
produced a single scheduling capability that VA could use, nor was
there an expectation that one would be made available. In May 2010, the
Government Accountability Office reported that VA spent an estimated
$127 million over nine years on its outpatient scheduling system
project and found that it did not implement any of the planned system's
capabilities and was essentially starting over.\2\
---------------------------------------------------------------------------
\2\ Government Accountability Office, Management Improvements Are
Essential to VA's Second Effort to Replace Its Outpatient Scheduling
System, GAO-10-579, May 27, 2010.
---------------------------------------------------------------------------
Development of VSE
In 2011, after analyzing alternatives, VA decided to pursue another
COTS solution that led to the development of the Medical Appointment
Scheduling System (MASS). Simultaneously, VA was working on its
strategic goal of mending its scheduling system under VistA Evolution,
a joint VHA and OIT program designed to improve the efficiency and
quality of veterans' health care. In 2014, VHA and OIT realized that
small enhancements to the VistA scheduling interface could not only be
made to meet its needs, but also could be done at a much lower cost and
with faster implementation than acquiring a new system. This joint
effort progressed into the in-house development of the VSE project, as
a temporary solution while MASS was being pursued.
In May 2014, the VSE project was launched and in July 2014 VA
awarded a firm-fixed-price contract to contractor Hewlett Packard
Enterprise Services LLC to design, develop, test, and support the
release and implementation of VSE with a cost of $4.1 million and a
completion date of November 2015. Ultimately, the cost increased to
$6.8 million due to additional software development, enhancements, and
a post-warranty support extension.
VSE was expected to update VA's legacy system with a modern
graphical user interface that resembles what a user might see on a
Microsoft Outlook calendar, while not changing any of the functionality
of the VistA scheduling system. Specifically, VSE would include
enhancements such as an aggregated view of clinic profile scheduling
grids, a single queue or list for appointment requests, resource
management reporting, and high priority or critical patches to the
VistA scheduling system.\3\ VSE was anticipated to reduce the time
schedulers took to enter new appointments by making it easier to see
care provider availability and appointment details.
---------------------------------------------------------------------------
\3\ VHA requires an aggregated view of clinic profile scheduling
grids to allow the user to view the resource's availability
collectively and decrease the time it takes to match an available
resource with the patient's request and improve the appointment
scheduling process.
---------------------------------------------------------------------------
Development of MASS
In November 2014, VA issued a request for proposals for the new
longer-term scheduling system, MASS, to replace the VistA scheduling
system. According to a VA fact sheet, MASS would enable proactive,
resource-management-based scheduling, which considers the availability
of staff, facilities, and equipment while also providing improved
transparency, and consistent implementation of scheduling policies and
directives. In August 2015, VA awarded an indefinite-delivery,
indefinite quantity contract for MASS to Systems Made Simple Inc. at a
maximum cost of about $624 million over a seven-year period. During an
April 2016 congressional hearing, VA officials at the time stated that
MASS was put on hold while VSE was being developed. However, in January
2017, the deputy secretary at that time directed that MASS pilot
activities resume at the Boise VA Medical Center (VAMC) in Idaho, with
the requirement that MASS would be used for scheduling veteran
appointments at the center by July 2018. In April 2018, MASS had been
successfully deployed at the Chalmers P. Wylie Ambulatory Care Center
in Columbus, Ohio, ahead of schedule.
In December 2018, VA reported to Congress on medical appointment
scheduling that the new Cerner contract includes an appointment
scheduling system component that will be rolled out across the VA in
conjunction with the electronic health records system over a 10-year
period. Cerner scheduling solutions are being coordinated by VA's
Office of Electronic Health Record Management. Due to concerns about
the length of implementation time for the Cerner electronic health
records system across VA, the Department intends to separate the
scheduling component within the Cerner contract and implement it on a
faster track to benefit all regions of the country. To mitigate the
risks, VA will not begin the scheduling component deployments until
after the full electronic health records system implementation is
achieved at two sites in the VA Northwest Health Network.\4\ The first
standalone scheduling component is planned for deployment in 2020 and
the last deployment is planned in 2023, according to the December 2018
report. However, with the decision for VA to go to a Cerner-based
solution, the MASS project will no longer be deployed to other sites.
The Cerner scheduling standalone component will replace VSE, MASS, and
the VistA scheduling system.
---------------------------------------------------------------------------
\4\ The VA Northwest Health Network (Veterans Integrated Service
Network 20) facilities are the initial operating capability sites for
implementation of the new electronic health records system.
---------------------------------------------------------------------------
INADEQUATE VSE PROJECT MANAGEMENT OF DEPLOYMENT AND IMPLEMENTATION
The OIG determined the VSE project team did not adequately manage
the development of the scheduling system due to deficient requirements,
insufficient testing and staff turnover. OIT approved the national
release and deployment of a version of VSE in April 2017. VSE suffered
delays from its original contract that ended in July 2016. In August
2016, the deputy secretary at that time implemented a remediation plan
for VSE to address usability issues that further delayed
implementation. Delays in deployment persisted until the final contract
modification ended in September 2017. As of August 2018, VA had
deployed the VSE to 157 of 160 locations according to a project analyst
for OIT's Enterprise Portfolio Management Division.\5\
---------------------------------------------------------------------------
\5\ According to a project analyst for OIT's Enterprise Portfolio
Management Division, VHA granted two permanent implementation waivers
for the VA medical centers in Columbus (using MASS) and Indianapolis
(implementing another commercial scheduling system about 15 years ago).
Also, the San Diego VA Medical Center remains partially implemented,
with an approved VHA waiver, but plans to fully implement VSE ``as is''
by the end of FY 2019.
---------------------------------------------------------------------------
VSE Requirements Were Inadequate
The audit team was told that VHA used high-level scheduling
blueprints to create simple requirements for VSE, according to VHA's
senior medical advisor for the Office of Veterans Access to Care, who
was the project's business sponsor.\6\ The senior medical advisor also
indicated that VA wanted a quick and simple ``stopgap'' or ``backup''
plan to MASS. The business sponsor also stated that MASS was the focus
of most of the team and VSE was a low priority. According to the
business sponsor, the VSE was supposed to address four enhancements:
---------------------------------------------------------------------------
\6\ The business sponsor, or customer/user, is responsible for
determining requirements, monitoring and approving changes to those
requirements, and accepting project increment deliverables. The
business sponsor is also required to validate and approve all project
requirements.
---------------------------------------------------------------------------
1.An aggregated view of clinic profile scheduling grids
2.A single queue for medical appointment requests
3.Resource management reporting dashboard
4.High priority/critical patches for VistA scheduling
The simple requirements were provided to the contractor and the
business sponsor validated and approved the requirements for the high
priority/critical patches. However, the OIG did not receive approval
documentation for the other enhancements. Thus, the audit team
concluded that the simple requirements were inadequate and may have led
to an incomplete analysis by the contractor.
One year after the initial operating capability testing, VHA
conducted a study of VSE to determine if usability issues existed, to
provide suggested changes to the user interface, and to help determine
future plans for the product.\7\ The usability study found functional
limitations, such as schedulers needing to toggle back and forth
between VistA, VSE, and the computerized patient record system to
create an appointment, and patient data was not being updated in every
location within VSE. The study concluded that VSE was not in a
deployable state, with bugs and defects that needed to be addressed.
This increased the amount of time schedulers needed to set veteran's
appointments and it was determined that VSE would be unable to solve
VA's problems related to tracking resources and availability. According
to an independent study required by the Veterans Access, Choice, and
Accountability Act of 2014, examples of resources include support
staff, exam rooms, and equipment availability.\8\
---------------------------------------------------------------------------
\7\ Initial operational capability testing is conducted in a cycle
within a project schedule for complex projects. These tests assess
system features and functionality.
\8\ P. L. No. 113-146 (August 7, 2014); McKinsey & Company Inc. A
Product of the CMS Alliance to Modernize Healthcare Federally Funded
Research and Development Center Centers for Medicare & Medicaid
Services. Prepared for VA.
Issues Identified in Initial Operating Capability Testing Delayed VSE
---------------------------------------------------------------------------
Implementation
The July 2014 contract for VSE included an optional task for
contractor support of initial operational capability testing, which was
exercised by VHA in August 2014. The task required the contractor to
execute an implementation plan, prepare for contingencies during the
release of software, and provide support of software deployment. In
June 2015, VSE was installed for initial operational capability testing
at the Charles George VA Medical Center in Asheville, North Carolina,
and at several of its clinics later in January 2016. According to VHA's
Human Factors Engineering (HFE) team, testing issues were reported
prior to its usability study, such as scheduling appointments through
VSE took twice as long as using the legacy system.\9\ Despite these
identified problems, the audit team found no information that any
efforts were made by VHA or the contractor to coordinate with the
Asheville medical center to develop a resolution plan, as required by
the contract. In July 2016, the Asheville VA Medical Center stopped
testing VSE due to the identified problems.
---------------------------------------------------------------------------
\9\ The HFE is a VHA office that conducts reviews of software to
determine deficiencies and areas of improvement from the aspect of the
system user.
---------------------------------------------------------------------------
In July 2016, the HFE team issued a report identifying 32
deficiencies associated with usability, software stability, limited
functionality, technology, training, and patient safety issues. Twenty-
six were related to functionality and six to training. Three of the 26
deficiencies were related to functions that were not delivered by the
contractor as required. Twenty-three functionality deficiencies should
have been considered in the original contract requirements but were
not. Some identified deficiencies included new appointments for
veterans that were not replicated across the interface grids and
comments placed in VSE were not being captured. In addition, usability
deficiencies included limitations on canceling and changing
appointments while software instability plagued the overall deployment
of the VSE project. Had the requirements been analyzed adequately, the
OIG determined there could have been a contract with better defined
requirements for VSE from the start, minimizing the types of problems
identified in the HFE usability study and those that schedulers
reported plagued the system. In August 2016, the deputy secretary at
that time issued a memo discussing the remediation plan for VSE and the
need to address the HFE-identified deficiencies. The remediation plan
required the VSE development team to standardize the current version of
VSE at five VA medical centers: Asheville, North Carolina; Salt Lake
City, Utah; Cleveland, Ohio; Hudson Valley, New York; and Chillicothe,
Ohio. The plan also required that the team fixed identified issues in
VSE and deliver up to two additional versions within six months.
Remediation efforts created additional development work, which further
delayed implementation of VSE, and still did not address all
deficiencies.
Project Staff Turnover May Have Affected Implementation
OIT and VHA did not have continuity in its management of the VSE
project. During the development of VSE from 2014 to 2017, key managers
and VSE project officials changed frequently. OIT's program manager
changed four times between May 2014 and May 2017. OIT's project
manager, who was also responsible for contracting officer's
representative duties, changed three times before being turned over to
a VHA project manager within the same period. Furthermore, the chief
information strategy officer, who was responsible for overseeing the
remediation period and ensuring successful coordination between OIT and
VHA during the critical redevelopment period, release, and
implementation of VSE, left VA and was replaced in November 2016. The
audit team determined the frequent turnover in these key management
positions could have impacted OIT's and VHA's ability to complete the
VSE project in a timely manner. The loss of project and program
knowledge may have delayed the development and implementation of VSE.
The audit team also found that VHA encountered difficulty in staffing
other positions critical to the VSE project. According to OIT and VHA
oversight briefings, personnel needed for the project included business
analysts, scrum masters, technical leads, testing managers, and
configuration managers.\10\ VA's solution to these staffing concerns
was to use contract employees and resources from the MITRE
Corporation.\11\
---------------------------------------------------------------------------
\10\ A scrum master is the facilitator for an agile development
team. Scrum is a methodology that allows a team to self-organize and
make changes quickly, in accordance with agile principles.
\11\ The MITRE Corporation is a not-for-profit company that
operates multiple federally funded research and development centers.
---------------------------------------------------------------------------
VSE Implementation Is Almost Complete
Delays in deployment began in July 2016 when all enhancements
should have been completed, and lasted until the final contract
modification ended in September 2017. According to the VHA project
manager, OIT approved the national release and deployment of a version
of VSE in April 2017. In May 2017, VSE project manager told the audit
team that no future developments were scheduled, and they would be only
focused on the sustainment of VSE. Therefore, the 23 additional
functionality requirements found by the HFE usability study were not
addressed. Starting in December 2017, any new requirements would be
completed under a new project called Sustainment of VSE. In February
2018, the VSE project manager told the audit team the three original
requirements from the initial contract were completed as of December
2017. However, the manager did not address the 23 outstanding
functionality issues identified by the HFE usability study.
Recommendation
The OIG made one recommendation that the VA assistant secretary for
information and technology and chief information officer should enforce
current required project management processes with improved oversight.
This should be executed to ensure project planning requirements are
adequately defined and supported before starting information technology
projects. At the time the OIG report was published, the VA assistant
secretary for information and technology and chief information officer
concurred with the recommendation and requested closure, as it has
implemented a new program management review process through a policy
memorandum signed on July 15, 2019. The program management review
process is designed to ensure information technology programs and
projects are delivered and sustain the intended outcomes. This
memorandum has been distributed to executive leadership team
correspondence leads. OIT's Office of Quality, Performance, and Risk
has begun its assessment of information technology projects through the
use of lightweight independent technical assessments. The OIG will
monitor OIT's progress and follow up on the implementation of the
policy memorandum to ensure it addresses the intent of the
recommendation.
CONCLUSION
Patient scheduling is critical to providing veterans with timely
access to medical care. The need to update the scheduling component of
VistA is instrumental to VA's efforts to achieve that goal. Although
VSE is a relatively small program and represented a short-term fix, it
only changed the look of screens. This seemingly small change was
expected to significantly reduce the time it took schedulers to
schedule appointments, however it failed to deliver on that promise and
it cost over $6 million. This is another example of the struggles VA
has in developing and managing information technology projects due to
inadequate requirements, insufficient testing, and staff turnover.
Having an effective program and project management structure in place
is essential to its information technology efforts.
Chairwoman Lee, Ranking Member Banks and members of the
Subcommittee, this concludes my statement. I would be happy to answer
any questions.