[House Hearing, 119 Congress]
[From the U.S. Government Publishing Office]
FROM RESET TO ROLLOUT: CAN THE
VA EHRM PROGRAM FINALLY DELIVER?
=======================================================================
HEARING
BEFORE THE
SUBCOMMITTEE ON TECHNOLOGY MODERNIZATION
OF THE
COMMITTEE ON VETERANS' AFFAIRS
U.S. HOUSE OF REPRESENTATIVES
ONE HUNDRED NINETEENTH CONGRESS
FIRST SESSION
__________
MONDAY, FEBRUARY 24, 2025
__________
Serial No. 119-6
__________
Printed for the use of the Committee on Veterans' Affairs
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
Available via http://govinfo.gov
__________
U.S. GOVERNMENT PUBLISHING OFFICE
59-864 WASHINGTON : 2025
-----------------------------------------------------------------------------------
COMMITTEE ON VETERANS' AFFAIRS
MIKE BOST, Illinois, Chairman
AUMUA AMATA COLEMAN RADEWAGEN, MARK TAKANO, California, Ranking
American Samoa, Vice-Chairwoman Member
JACK BERGMAN, Michigan JULIA BROWNLEY, California
NANCY MACE, South Carolina CHRIS PAPPAS, New Hampshire
MARIANNETTE MILLER-MEEKS, Iowa SHEILA CHERFILUS-MCCORMICK,
GREGORY F. MURPHY, North Carolina Florida
DERRICK VAN ORDEN, Wisconsin MORGAN MCGARVEY, Kentucky
MORGAN LUTTRELL, Texas DELIA RAMIREZ, Illinois
JUAN CISCOMANI, Arizona NIKKI BUDZINSKI, Illinois
KEITH SELF, Texas TIMOTHY M. KENNEDY, New York
JEN KIGGANS, Virginia MAXINE DEXTER, Oregon
ABE HAMADEH, Arizona HERB CONAWAY, New Jersey
KIMBERLYN KING-HINDS, Northern KELLY MORRISON, Minnesota
Mariana Islands
TOM BARRETT, Michigan
Jon Clark, Staff Director
Matt Reel, Democratic Staff Director
SUBCOMMITTEE ON TECHNOLOGY MODERNIZATION
TOM BARRETT, Michigan, Chairman
NANCY MACE, South Carolina NIKKI BUDZINSKI, Illinois, Ranking
MORGAN LUTTRELL, Texas Member
SHEILA CHERFILUS-MCCORMICK,
Florida
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
----------
MONDAY, FEBRUARY 24, 2025
Page
OPENING STATEMENTS
The Honorable Tom Barrett, Chairman.............................. 1
The Honorable Nikki Budzinski, Ranking Member.................... 3
WITNESSES
Dr. Neil Evans, M.D., Acting Program Executive Director,
Electronic Health Record Modernization Integration Office, U.S.
Department of Veterans Affairs................................. 5
The Honorable Seema Verma, Executive Vice President, Oracle
Health and Oracle Life Sciences, Oracle Corporation............ 7
Mr. David Case, Acting Inspector General, Office of Inspector
General, U.S. Department of Veterans Affairs................... 9
Ms. Carol Harris, Director, Information Technology and
Cybersecurity Issues, U.S. Government Accountability Office.... 11
APPENDIX
Prepared Statements Of Witnesses
Dr. Neil Evans, M.D. Prepared Statement.......................... 39
The Honorable Seema Verma Prepared Statement..................... 41
Mr. David Case Prepared Statement................................ 57
Ms. Carol Harris Prepared Statement.............................. 68
FROM RESET TO ROLLOUT: CAN THE
VA EHRM PROGRAM FINALLY DELIVER?
----------
MONDAY, FEBRUARY 24, 2025
Subcommittee on Technology Modernization,
Committee on Veterans' Affairs,
U.S. House of Representatives,
Washington, D.C.
The subcommittee met, pursuant to notice, at 3:01 p.m., in
room 360, Cannon House Office Building, Hon. Tom Barrett
(chairman of the subcommittee) presiding.
Present: Representatives Barrett, Luttrell, and Budzinski.
OPENING STATEMENT OF TOM BARRETT, CHAIRMAN
Mr. Barrett. Good afternoon. The first Technology and
Modernization Subcommittee hearing of the 119th Congress will
come to order.
I want to thank the witnesses for being here today to
discuss the Electronic Health Record Modernization (EHRM)
program at the Veterans Affairs (VA). Appreciate your
willingness to participate and help in this conversation.
Before we get to the business of today's hearing, I want to
first congratulate and welcome my fellow Midwesterner,
Representative Nikki Budzinski from Illinois, on becoming the
ranking member of this subcommittee.
We had a great meeting in my office a few weeks ago and
participated in a roundtable earlier already and really
appreciate your partnership in this effort.
Your position as ranking member is a testament to your hard
work for veterans and the goodwill you have built with your
colleagues on both sides of the aisle and look forward to
continuing to build that relationship together.
Our work on this committee will not always be easy, but our
mission is critical. That is why I am deeply honored to chair
this subcommittee and lead the House's oversight over VA's
Information Technology (IT) systems, infrastructure, and
modernization projects.
By way of background, I served 22 years in the United
States Army before retiring and becoming one of the thousands
of veterans in Michigan that receive their healthcare and
benefits from the VA. I personally go to the Battle Creek VA
Medical Center. I used the Montgomery GI Bill to pay for my
college. I understand what veterans in Michigan and across the
country need and expect from the VA.
I am going to use my leadership position here to make sure
that VA's IT systems are working for all veterans and employees
of the VA and ultimately making the VA a more effective and
efficient organization.
Whether you are making primary care appointments, filling a
disability claim, applying for education benefits, none of that
happens successfully without good, reliable, and secure IT
systems.
Perhaps the largest undertaking on this committee will be
oversight of the VA's Electronic Health Record (EHR) and the
efforts that are going into that today. It is the backbone of
VA's healthcare operations.
The Electronic Health Record Modernization program is VA's
project to replace its homegrown Electronic Health Record
system with a commercial, off-the-shelf EHR from Oracle.
This subcommittee has spent hundreds of hours over the last
several years conducting oversight over this program and
exposing, reviewing, and analyzing its issues and its problems.
While I am new to this conversation, I have spent
considerable time looking into this issue. I understand how
important it is for us to get this right.
These consistent problems include system outages, patient
harm, poorly functioning technology, budget overruns, reduced
patient volumes, and more.
Because of these problems, former Secretary McDonough put
the program in an indefinite pause in April 2023 to give VA and
Oracle time to fix a long list of problems with the system.
The Secretary decided right before Christmas that VA was
going to start the program up again in the spring of 2026 at
Ann Arbor, Battle Creek, Detroit, and Saginaw in my home state
of Michigan.
Two years ago, VA told this committee they had a lot of
homework to do before they put this system into another medical
center. Now VA is getting ready to turn in their homework.
I appreciate the opportunity to meet with both Oracle and
VA several times before this hearing to discuss the problems
that have plagued this system for years.
While I am not convinced that they have fixed all of the
problems, I want to hear today from VA and Oracle directly to
this subcommittee the progress that has come out of the last 2
years.
As the new chairman of this subcommittee, I am going to
examine the facts, consult my colleagues on this subcommittee,
and make my own judgments based on what progress has been made.
Here are some of the facts as they stand today.
First, VA is nearly 7 years into the original 10-year
contract and the Oracle EHR has been implemented in less than 4
percent of medical centers.
Second, according to the latest surveys, 69 percent of
users are dissatisfied with the system; 75 percent of users
believe the system does not maximize their efficiency. I told
someone earlier that the approval rating is less than that of
Congress, and that is a hard thing to do.
Third, Congress has not received a schedule nor an up-to-
date cost estimate to evaluate this program's current state.
The only independent cost estimate we can rely on for what it
will cost VA to implement this program is already 3 years old
and it was over $32 billion. That is more than double VA's
original estimate of $16.1 billion.
Given these facts, the VA needs to demonstrate how this
system has improved and explain why this program can succeed
before starting up again.
I have confidence in the Trump administration and Secretary
Collins to right the ship for veterans and their families, and
I intend to work closely with him to deliver results. It is my
commitment to this subcommittee and the stakeholders here that
I will be firm but fair in the oversight we provide.
Thank you all again for being here today. I look forward to
your testimony. I look forward to achieving the accountability
that all of us should expect.
With that, I yield to Ranking Member Budzinski for her
opening statement.
OPENING STATEMENT OF NIKKI BUDZINSKI, RANKING MEMBER
Ms. Budzinski. Thank you very much, Mr. Chairman. I want to
start out as well congratulating you on your chairmanship of
the Subcommittee on Technology and Modernization and to say
genuinely how much I look forward to working with you on this
issue.
In addition, I am working on this issue as the ranking
member because I think it is so important that we are ensuring
that veterans and VA employees have the access to state-of-the-
art technology that supports the delivery of healthcare and
benefits for our veterans.
I also want to thank the witnesses for being here today to
discuss the future of EHRM, this project, and for coming in a
couple of weeks ago to help me get up to speed on a lot of
these issues.
We had a robust discussion about the history and the
current status of this project, which I really appreciated, and
I look forward to working with all of you to support the
progress that has been made to ensure that the EHRM moves
forward in a safe and smart way.
In almost 2 years since the VA announced its reset of the
EHRM program we have worked in a bipartisan and bicameral way
with VA and Oracle to move the needle on this project.
In December, as the chairman mentioned, VA announced its
intention to resume go-live activities with the first
deployment happening in mid-2026.
I want to take this time to outline some of my concerns,
many of them that I have shared that the chairman has actually
already outlined, about the lack of progress being made by both
VA and Oracle on a series of issues that ultimately impact our
mission here, which is to address the betterment of veteran
health and safety.
Last month VA briefed my staff that there were still dozens
of outstanding recommendations from the VA's Office of the
Inspector General (OIG) and the Government Accountability
Office (GAO), recommendations that VA acknowledged it needs to
fix, and yet little has been made on progress on those issues.
It seems to me that these recommendations need to be
addressed before any other go-lives are allowed to occur. I
hope to hear from VA about how it intends to address these open
issues before its planned resumption of go-lives next summer. I
think that is important.
I am also concerned about the rollout of the 3b
bidirectional interface intended to fix major issues with how
Oracle supports VA's pharmacy operations.
I understand that 3b was deployed this weekend, but it
sounds like it is suffering from some of the same issues that
the system did during its initial go-lives.
The training is substandard and behind schedule, and staff
are concerned that the solutions will not meet their needs.
The 3b solution was meant to solve problems that were
identified at Mann-Grandstaff, the first site to go live in
2020. Yet here we are, almost 4 years later, and I am concerned
that this is still not fully ready.
The lack of these fixes has caused VA pharmacy to hire
additional people and use onerous workarounds to accomplish the
same tasks they did in Veterans Health Information Systems and
Technology Architecture (VistA). These fixes are desperately
needed, but they need to be done right and we cannot afford any
more half measures.
Also, I have some major concerns about how the recent
reduction in force actions at the Department are going to
impact VA's ability to move this project forward.
I know that Secretary Collins says these actions will not
impact veterans' access to healthcare and benefits, but I do
not see how cuts to a historically understaffed VA do not
impact veterans and specifically the implementation of this
program and our look ahead to these go-lives potentially in
2026.
I have questions about how many people in the EHRM
Integration Office were terminated under the probationary
employee purge, how many opted to take the ``Fork in the
Road,'' and how many others may have been purged through other
methods.
I also have questions about how the hiring freeze is
impacting EHRM. I expect the VA has a plan to compensate for
these vacancies in this crucial program and I need to hear
about that plan today.
My biggest concern as we continue overseeing the EHRM
program is ensuring that veterans' health and safety remain the
top priority throughout this implementation.
However, VA's track record has not always aligned with this
commitment, oftentimes prioritizing speed over execution.
VA rushed into this contract, accepting a false assumption
that it had to be on the same system to be interoperable with
Department of Defense (DOD).
It rushed into system configuration without doing the
standardization work needed to make the transition smooth.
It rushed to go live, even though we were in the middle of
a pandemic and all the VA staff were already overworked.
It then rushed to move on from its initial failures.
VA even allowed DOD to rush them into another go-live just
so DOD could finish its project.
We cannot allow VA to rush into the resumption of go-live
activities using the same old playbook they have used up until
this point. The Department must focus on improving their
protocols, training, and efficiency on the front end before
hastily implementing another go-live.
I want to say, though, we should be working toward this go-
live. I have all of these concerns. I think it is important
they be articulated. I do want to get to a point where we can
see this being successful, but I need to hear meaningfully how
these concerns are being addressed before we can do this.
The focus should be on improving staff training, for
example, to ensure the tools and skills to be able to use the
software most efficiently.
The focus should be on establishing a structured project
management timeline, something that I think the chairman has
also brought up, with benchmarks and deadlines that must be met
to push the project to the next step and eventually go live.
The focus should be on implementing a system that maximizes
efficiency within the VA and allows providers to have seamless
communication with other elements of the care delivery system.
We quite simply owe it to veterans and VA employees to
ensure that they have the resources and technology they need
for high-quality, safe, and effective care.
I look forward to this conversation.
Thank you very much, Mr. Chairman. I yield back.
Mr. Barrett. Thank you, Ranking Member Budzinski.
I would like to introduce our witnesses.
From the Department of Veterans Affairs we have Dr. Neil
Evans, acting program executive director of the Electronic
Health Record Modernization Integration Office.
Did I get all that right, sir? All right.
From Oracle we have the honorable Seema Verma, executive
vice president and general manager of Oracle Health and Life
Sciences.
From the Government Accountability Office we have Ms. Carol
Harris, the director of IT and cybersecurity at GAO.
From the Office of Inspector General we have Mr. David
Case, the acting inspector general of VA OIG.
I ask the witnesses to please stand and raise your right
hands.
[Witnesses sworn.]
Mr. Barrett. Thank you.
Let the record reflect that all witnesses have answered in
the affirmative.
Dr. Evans, you are now recognized for 5 minutes to deliver
your opening statement on behalf of VA.
STATEMENT OF NEIL EVANS
Dr. Evans. Thank you. Good afternoon, Chairman Barrett,
Ranking Member Budzinski, and distinguished members of the
subcommittee.
I want to thank Congress and this subcommittee for the
opportunity to testify today. VA recognizes and appreciates
your shared commitment to veterans and your support for VA's
Electronic Health Record Modernization efforts.
This testimony comes at a pivotal moment. With new
leadership at VA there is renewed commitment to ensuring that
every aspect of our EHR modernization effort is reevaluated.
Secretary Collins made it clear in his confirmation hearing
that the status quo is not acceptable. Every previous decision,
policy, and process must be thoroughly reviewed to ensure that
it aligns with our core mission--providing veterans with the
highest quality care through a system that works for them, not
against them.
Accordingly, VA remains committed to implementing modern
interoperable health information technologies across the entire
VA healthcare system.
The Federal EHR, which is the set of technologies we are
deploying through our Electronic Health Record modernization
efforts, works in concert with biomedical devices and other
critical health information technologies, and these together
will serve as key enablers behind VA's ability to deliver
comprehensive high-quality healthcare to veterans for many
years to come.
Between 2020 and 2022, VA deployed the system at 5 VA
medical centers, 22 community-based outpatient centers, and 52
remote VA sites that serve those clinical locations, for
example, call centers, telehealth hubs, and more.
In April 2023, veterans and staff had made it clear that
the system was not meeting expectations. With bipartisan
support from this committee, and consistent with
recommendations shared by GAO and OIG, VA announced a
comprehensive program reset, pausing further deployments of the
system and taking time to address issues that have been
identified by users at active sites.
VA also identified the reset period as an opportunity to
complete foundational enterprise work necessary for long-term
success, operating a far more centralized set of technologies
than VA's existing Electronic Health Record.
Finally, VA used the reset period to prepare for and then
successfully complete a joint VA-DOD deployment of the Federal
EHR at the Captain James A. Lovell Federal Healthcare Center
(FHCC) in North Chicago, Illinois, just under a year ago in
March 2024.
Relying on a data-driven approach to assessing progress, VA
demonstrated improvements in system stability, program
processes, veteran and VA staff experiences, and finally,
health system outcomes, including quality, clinical
productivity, and revenue management.
In light of that demonstrated progress, on December 20 of
2024, VA announced that it was going to restart Federal EHR
deployment efforts initially at the four aforementioned
Michigan facilities in Ann Arbor, Battle Creek, Detroit, and
Saginaw.
I should add this: VA acknowledges that there is still
improvement work to be done, as you will likely hear from GAO
and OIG, and we agree. Improvement efforts are not stopping. We
are simply, in addition to those improvement efforts, resuming
deployment work in parallel.
Examples of ongoing improvement efforts include migration
of the system to Oracle's cloud infrastructure for improved
performance of the system and access to Oracle's next-
generation technologies.
Establishment of a Federal EHR system baseline and
associated processes, which is a critically necessary step in
support of improved standardization across VA, which will help
lessen staff frustration, deployment delays, and cost.
Further training improvements to better support user
adoption.
Standardization and consolidation of user roles in the
system, which will simplify training and everyday use of the
system for VA staff.
Completion of crosscutting projects that we have called
``big rocks,'' which include reassessing system design choices
in order to deliver a better and more efficient user
experience.
There is much more.
Meanwhile, current state reviews, the initial activities
necessary to deploy the Federal EHR, have been in progress at
the Michigan sites mentioned earlier since the first week of
January, and those are going well.
We are also working closely with Veterans Health
Administration (VHA) health system leaders and with Oracle
Health to determine deployment plans beyond the Michigan sites.
These decisions will support the iterative development of a
revised integrated master schedule in addition to updated cost
estimates for this project.
Through the efforts of the recent program reset and through
our collaboration within VA with our end users, with
leadership, and with veterans, and outside of VA with DOD,
Federal Electronic Health Record Modernization (FEHRM),
oversight bodies, and our vendor partners, we are seeing
movement--yes, even momentum--in the right direction. By
sustaining these efforts our velocity should increase over
time.
VA remains optimistic to realize the successful
implementation of the Federal EHR across VA. I appreciate this
opportunity to testify and am thankful for this committee for
your commitment to serving veterans, and look forward to
working with you over the course of the 119th Congress.
We look forward to responding to any questions you might
have.
[The Prepared Statement Of Neil Evans Appears In The
Appendix]
Mr. Barrett. Thank you, Dr. Evans.
The written statement of Dr. Evans will be entered into the
hearing record.
Ms. Verma, you are now recognized for 5 minutes for your
opening statement on behalf of Oracle.
STATEMENT OF SEEMA VERMA
Ms. Verma. Thank you, Chairman Barrett, Ranking Member
Budzinski, and distinguished members of the subcommittee. Thank
you for the opportunity to speak with you today.
The VA's EHR Modernization program is at a turning point as
we begin work to restart deployments in Michigan. The
preparations will take place this year, with the deployments
occurring in early 2026.
I say turning point because we have several factors that
have come into alignment.
First, a new administration that is committed to
accelerating.
Second, we have completed countless optimizations and
improvements, including with pharmacy just this weekend, that
have improved productivity and are trending user satisfaction
in the right direction.
Third, Oracle is making a number of investments to make
implementations go faster and smoother, improving training and
support, while also investing in modernization of the EHR to
provide a state-of-the-art experience for providers.
Fourth, Oracle is migrating VA to our cloud to unlock a
slew of new technologies that VA can take advantage of while
enhancing cybersecurity.
We believe the program is ready to accelerate deployments
in 2026.
We are working closely with Secretary Collins, Dr. Evans,
and the VA team to figure out the right pace of deployments'
timing and schedule, even meeting at the VA just a few hours
ago.
As the committee considers the path forward, I think the
most important thing to recognize is that the program and the
EHR deployed at the first five sites is in a vastly different
state than it was nearly 3 years ago, before Oracle acquired
Cerner.
There have been many changes to address the challenges of
the past. My written testimony details--perhaps painfully for
those of you that read it all the way through--the amount of
change that Oracle has brought to the program.
It performs reliably without severe outages or frustrating
crashes for users. It has had more than 3,000 functional
changes made during the reset to make the EHR easier to use,
simplify workflows, and enhance functions like pharmacy.
Training has been overhauled. The new training program
worked very well and received high marks from the employees at
Lovell Federal Health Care Center in Chicago when it
successfully went live last year.
Revenue collections and productivity are returning and in
some cases even exceeding baselines from predeployment
averages.
Patient safety continues to be a top focus and we have
instituted several new procedures with VA to make sure safety
is always at the forefront.
Users filing tickets with potential patient safety concerns
have dropped 80 percent since our acquisition of Cerner.
The EHR has new opioid tools, registries, and clinical
decision support tools that enable greater patient safety,
close care gaps, and help improve outcomes.
Finally, when veterans receive care in the community their
health records can be accessed by community care providers and
updated to ensure VA providers have a complete view of the care
received outside the VA system.
Now, there are still challenges and work that we will
continue to do to improve the system to address the VA's unique
needs. We expect the same of VA in terms of focusing on
standardization, quick decision-making, and displaying a strong
commitment to adopting the new EHR.
All of this work will continue in parallel while we prepare
for the next round of deployments and none of the optimization
work will slow us down.
Looking into the future, as we move this system to the
Oracle cloud, we will have the ability to bring even more
modern features to the EHR and eventually upgrade it to our new
cloud-based EHR infused with Artificial Intelligence (AI).
The narrative about this program is largely negative. We
understand that first impressions last, and the first
impression in Spokane and the other initial sites was not good.
We are not Cerner that deployed at those sites. The
totality of updates, enhancements, investments, and innovations
to the EHR show that this is a dramatically improved system
from what was originally deployed in Spokane in 2020.
We are encouraged that everyone involved wants to see this
program on a path to success or get it back on course, as
Chairman Bost and others have said, because I know we all agree
that those who have made sacrifices for all of us deserve
modern technology to help make their experiences at the VA more
efficient, safe, smooth, and ultimately aid the providers that
serve them in delivering high-quality care that improves the
health of veterans.
Oracle stands ready and committed to getting this done on
behalf of our nation's veterans. Thank you.
[The Prepared Statement Of Seema Verma Appears In The
Appendix]
Mr. Barrett. Thank you, Ms. Verma.
The written statement of Ms. Verma will be entered into the
hearing record.
Mr. Case, you are new recognized for 5 minutes to deliver
your opening statement on behalf of OIG.
STATEMENT OF DAVID CASE
Mr. Case. Chairman Barrett, Ranking Member Budzinski, and
subcommittee members, thank you for the opportunity to discuss
the OIG's oversight of EHRM.
Since April 2020 we have published 22 reports on VA's
modernization program. More than half uncovered significant
patient safety concerns, such as problems with medication
management, pharmacy operations, and patient care coordination.
The others identified deficiencies in program and contract
management, including poor cost estimating and reporting.
We have remained focused on EHRM during VA's reset, because
an updated EHR has a central role in providing high-quality
healthcare to veterans.
I want to discuss four main unresolved issues previously
identified by OIG teams that VA should address as it emerges
from reset to the next wave of EHR deployment.
First, while VA has addressed many OIG-identified patient
safety issues, more work is needed to ensure that the veteran's
experience with a new EHR fulfills its promise of timely access
to seamless high-quality care.
For example, the new EHR has had serious issues with
handling scheduling changes and missed appointments. This
includes VHA allowing mental health staff at new EHR sites to
make fewer attempts to contact no-show patients compared to
legacy sites, creating a different standard of care between
sites.
VHA should address barriers created by software
deficiencies without compromising patient care and engagement
standards.
Second, in February 2024, I testified to this subcommittee
about pharmacy-related-patient-safety issues facing every VHA
provider. While VA has made progress on some aspects of the
pharmacy program, critical issues remain unresolved.
For the foreseeable future, legacy EHR site leaders must
have providers perform manual medication safety checks for
patients who have received care at a new EHR site due to
concerns with the automated process.
These manual safety checks are time-consuming and rely on
the vigilance of pharmacists and frontline staff. Facilities
using the new EHR require greater numbers of staff and
supplemental resources to manage the significant drop in
productivity.
Third, it is essential that the new EHR system be
consistently accessible. In September 2024, the OIG found that
VA procedures lacked sufficient controls to prevent, respond
to, and mitigate major performance incidents.
While we recognize VA has improved system reliability,
there are still incidents occurring every month.
Given the stress that the system will likely experience
during the next deployments in larger facilities, VA must be
ready to handle any future instability and equip providers with
the tools to continue working when the system is degraded or
inoperative.
Finally, VA must develop and maintain an integrated master
schedule to clearly track and project the program's cost to
completion. The 2026 deployments are on a much larger scale,
with a greater number of variables and tasks to plan and
execute. Also, VA's ability to train staff and resolve requests
for help may be challenged.
Additionally, there may be many developments over the next
year that could affect the schedule, and cost, and the veteran
and provider experience. A reliable master schedule will enable
VA to be fully transparent with Congress and their stakeholders
about what needs to happen for the successful rollout of the
new EHR in Michigan and other sites nationwide.
Our recommendations on the integrated master schedule and
project costs have been open for several years, paused
essentially because of VA's reset.
Given the restart, VA should get them done. Without them,
costs will be uncertain, risks may not be mitigated, and
efficiencies could be lost as tasks dependent on other actions
are not fully considered.
In conclusion, EHRM's success is dependent on VA's
transparency, careful planning, and the recognition and
remediation of patient safety issues.
Updating the EHR is critical for VHA's ability to deliver
healthcare in the future, and the OIG will continue to focus
our resources to provide VA with timely recommendations that
can help them and Congress ensure this transition succeeds.
Chairman Barrett, this concludes my statement. I would be
happy to answer any questions you or other members may have.
[The Prepared Statement Of David Case Appears In The
Appendix]
Mr. Barrett. Thank you, Mr. Case.
The written statement for Mr. Case will be entered into the
hearing record.
Now we will go to Ms. Harris.
You are now recognized for 5 minutes for your statement.
Thank you.
STATEMENT OF CAROL HARRIS
Ms. Harris. Chairman Barrett, Ranking Member Budzinski, and
members of the subcommittee, thank you for inviting us to
testify today on the EHRM program. This is VA's fourth attempt
at replacing its legacy Electronic Health Record system.
As requested, I will briefly summarize our recent work on
VA's progress to implement the system and the challenges it
faces moving forward.
As you know, VA provides healthcare services to roughly 9
million veterans and their families and relies on the legacy
system called VistA to do so.
In June 2017, the Department initiated the EHRM program to
replace VistA and deployed the new system to six of its medical
centers at a cost of about $12.7 billion.
The rollout of the system has been met with poor user
satisfaction, change management issues, and slow resolution of
trouble tickets, among other things. Given the magnitude of
user concerns, VA paused deployments in April 2023 to improve
the system and address those concerns.
On December 20, 2024, VA announced it would resume
deployments to four Michigan sites in mid-2026. This afternoon,
I will highlight four key points from our work.
First, VA lacks a strategic roadmap to get EHRM where it
needs to go.
During the 20-month deployment pause, a significant part of
VA's improvement efforts focused on the more than 1,500 system-
configuration changes needed to support user needs and
standardize workflows.
VA also closed many of its highest priority patient safety
enhancement requests.
These are all important accomplishments, but there are
still about 1,800 complex configuration changes and a number of
user-driven projects that need to be completed.
For example, referral management lacks functionality and
intuitiveness for users, and VA will be working to simplify the
existing process and standardize referrals across the
enterprise.
All this is to say that VA has been in reactive mode during
this reset period, putting out fires with no real plan for how
to move forward strategically.
Two pieces of information needed to inform such a plan are
the cost and schedule estimates. This leads to my second
point--that, unfortunately, Congress and VA senior leaders are
relying on estimates that are severely outdated and incomplete.
Existing cost estimates range from VA's $16.1 billion to an
independent one of almost $50 billion. While the latter is more
realistic, neither reflects the many changes and delays to the
program.
Also, planed system deployment at four additional sites
results in about 94 percent of VA's total medical centers
without the new system as of mid-2026.
With 2 years left on the Oracle Cerner contract, it is
vitally important that VA update these estimates to inform
decision-making.
Another key piece of information needed to inform a
strategic roadmap is the results of an independent operational
assessment, or an Independent Verification and Validation
(IV&V) test, which we recommended VA conduct in 2023.
An IV&V is an evaluation of the system end to end to
determine whether the system is operationally suitable. In
other words, it systematically catalogs these deficiencies of
the system and answers the question of whether or not the
system will realistically meet mission need.
This test is considered an IT best practice and one of the
reasons why DOD was able to successfully deploy the Oracle
Cerner system.
Now to my final point. VA's user surveys continue to
reflect general dissatisfaction with the new system. As of
September 2024, about 70 percent of respondents were not
satisfied with the system; about half continued to disagree or
strongly disagree that the system enabled them to deliver high-
quality care.
Seventy-five percent disagreed or strongly disagreed that
the system made them as efficient as possible. While there was
a slight uptick in satisfaction compared to 2022, these numbers
are still just too low.
Moving forward, it will be critical for VA to address these
issues along with our other eight open recommendations as soon
as possible. Doing so will position VA to more effectively
deliver a modern health record system as veterans deserve.
That concludes my statement. I look forward to addressing
your questions.
[The Prepared Statement Of Carol Harris Appears In The
Appendix]
Mr. Barrett. Thank you, Ms. Harris.
The written statement of Ms. Harris will be entered into
the hearing record.
We will now proceed to questioning, and I recognize myself
for 5 minutes.
Ms. Harris, thank you again for your testimony. Given all
that is happened with this program since its last cost estimate
in 2022, I know you mentioned this in your opening remarks, but
can Congress rely on the estimate of $32.7 billion over 13
years to fully implement the Oracle EHR at VA? If not, why not?
Ms. Harris. No, Congress cannot reliably rely on this
estimate. There have been too many changes to the program.
There obviously was this period reset over a 2-year period with
additional work being done and none of that has been accounted
for within the cost estimate.
That estimate needs to be updated before we proceed,
because, again, when you take a look at how much has been sunk,
12.7 billion already, and we have only deployed to six sites,
if you were to just multiply that across the 94 percent of
medical centers that have not had the same system, I mean, that
could be in the hundreds of billions of dollars potentially.
Mr. Barrett. Thank you.
Second question for you. The GAO has been publishing
reports about the EHRM program and making recommendations to VA
on how to improve it since 2020. You have been involved in each
of those.
How many recommendations have you made? How many have been
resolved? I know you said there are a number that are
outstanding. Can you give us an update on the number of
recommendations versus those that have been resolved so far?
Ms. Harris. We have made a total of 15 recommendations in
our previous work. We have an additional three that we have
with a draft report that is out for comment at VA at this time.
That would be a total of 18 recommendations. Of those 18, one
has been implemented.
Mr. Barrett. Which one was that?
Ms. Harris. That was one was related to, I believe, data
migration.
Mr. Barrett. Okay. Thank you.
Your most recent report said that the program has made
incremental improvements but there is much more work to be
done. Is that related to those recommendations or do you have
other thoughts on the incremental improvements that have been
made?
Ms. Harris. I mean, the incremental improvements have been
related to those system configurations that VA has completed,
so roughly 1,500 complex configuration changes.
I know that there have been more that are not complex that
have been made during this reset period. There is still a
significant number of configuration changes that need to be
made in addition to these user-driven projects, that those are
still outstanding, that need to be completed before the
deployment at Michigan in mid-2026.
Mr. Barrett. I know you mentioned what I would consider
lack of satisfaction or the reports as they surveyed those that
are interfacing with the system. Anywhere from low 70's to low
80's are seemingly frustrated or feel like the system is not
meeting their needs or obligations or desires.
Do you have any information as far as, like, an industry
standard outside of government or in other health systems as
far as the satisfaction among folks with electronic health
record systems?
Ms. Harris. We do have that information. I do not have that
with me at this time, but I am happy to work with your staff in
providing that.
I do believe that the industry numbers for the Oracle
Cerner system, it is higher than what we are seeing at VA, and
at DOD the numbers are a little bit higher than VA. Overall,
they are not super high compared to others, is my
understanding.
Mr. Barrett. Okay. Thank you.
In your opinion, does VA have a complete understanding of
all the problems with the system? Do they have a plan to
resolve them?
Ms. Harris. I do not believe they have a complete
understanding of all the issues. That is why this IV&V test is
so critical, because it is a systemic look at the system once
it is in the live environment.
When it is deployed, then the testers go in and they
observe the users actually use the system from an end-to-end
perspective, they catalogue all of those issues. That would be
a way for VA to identify the comprehensive list of issues and
then to prioritize from there.
They have not done that. They have done pieces of those
types of tests, but not from an end-to-end perspective.
Mr. Barrett. Quickly, I only have about 30 seconds left. I
know it is the position of VA and Oracle that, well, we have
got this in a live environment so we do not need that end-to-
end review.
What would you say to that position that they seem to feel
is testing it in a live environment?
Ms. Harris. I mean, I would disagree with that. It is an IT
best practice. It is also something that DOD did when they
deployed their Oracle Cerner system. It was one of the reasons
whey they were able to successfully deploy.
Mr. Barrett. Okay. Thank you.
I would now like to recognize Ranking Member Budzinski for
your questions.
Ms. Budzinski. Thank you, Mr. Chairman.
I had mentioned this in my opening. I am concerned about
any potential impacts with the recent announcements of some of
the positions, the workforce cuts to the VA.
Is that going to then affect the implementation of the go-
lives that we have coming up with the EHRM system?
I think my questions are really for Dr. Evans in this first
round.
Did you lose anyone in the probationary employee purge? If
so, how many?
Dr. Evans. Yes, we did lose several employees with regard
to folks who were let go who were in their probationary period.
I believe the number currently is eight.
Ms. Budzinski. Eight. Okay.
Then how many people in the EHRM Integration Office took
advantage of the ``Fork in the Road'' email that was sent?
Dr. Evans. 16 staff members signed up for the deferred
resignation program.
Ms. Budzinski. Okay. Then how many vacancies did you have
before these cuts started?
Dr. Evans. Our organizational chart is approved for
approximately 330 staff, and we are at approximately 250.
Ms. Budzinski. Okay. Based on that, can you kind of speak
to how confident you are that you can move forward successfully
with these staffing changes? Or how does that impact how you
are looking at the potential for success for a next round of
go-lives?
Dr. Evans. I think I would start by saying, first of all,
this is a priority. I mean, as we have heard here in the room
today, our successful implementation of the Federal EHR across
VA is a priority of the VA and is certainly a priority of
current VA leadership.
It is important to recognize that it is not just the
government staff of the Electronic Health Record Modernization
Integration Office who are going to be the folks who actually
make this happen.
I am super proud of every one of them and the work that
they are doing to move this program forward. They have been
putting in a tremendous amount of effort over the reset. Now,
as we are getting restarted in Michigan and beyond, they are
doing a tremendous amount of work.
We also rely on staff from the Veterans Health
Administration, from the Office of Information Technology, from
several of our vendor partners who work alongside of us and
really across in the field at the Veterans Integrated Services
Networks (VISNs). It takes many, many more than the folks in
the program office to successfully implement the EHR.
Ms. Budzinski. Okay. Thank you.
My next set of questions is really focused around the
restart that was put into place and then obviously this recent
news about going live with some sites in Michigan in mid-2026.
My question, just to start again with Dr. Evans, is just
can you talk a little bit about how confident you are, given
what you have shared and what has already really been shared
with the panel, that the VA and Oracle are ready to deploy this
system and that we will not hear the same concerns that the
first six sites raised?
Dr. Evans. Yes. I think good program management and sort of
a realistic approach to implementing--I mean, I think we should
step back and recognize that this is a really complex,
transformative change. We are changing effectively all of the
information technologies that support healthcare delivery at
our sites.
I feel confident that we are in a much better place and
completely agree with the fact that what we are deploying, the
product we are going to be deploying in Michigan, the set of
technologies and how they are configured to meet VA's needs, it
is an entirely different set. It is very different than it was
with the initial sites.
I do feel confident that we are entering Michigan having
learned a lot, having listened to our end users. That does not
mean that we can let up one bit with regard to our vigilance of
how do we deliver with excellence in Michigan.
Ms. Budzinski. I think maybe drilling down a little bit
more, like, are there metrics that you are using that give you
the confidence to say that?
Dr. Evans. Yes. We established a set of metrics that we
used to demonstrate our progress during the reset. They were
looking in four categories.
One was essentially how stable the system was, the
technical stability of the system.
The second category was looking at metrics around the
program processes that we use to support end users, how well
are we supporting the need for change and for adjustment to the
system and support for our end users.
The third was around the veteran and VA staff experience
around a host of pieces of the application.
The fourth was around the actual impact of the system on
health system operations, our revenue collections, healthcare
quality, et cetera. We have been tracking those metrics through
the reset and we will continue to track them as we move
forward.
Ms. Budzinski. I have run out of time, so I yield back. I
have other questions.
Mr. Barrett. Sure. Thank you. We will do another round
after this.
Mr. Luttrell.
Mr. Luttrell. Thank you, Mr. Chairman.
Ms. Harris, you look like you want to respond to that, what
Dr. Evans just laid out for us.
Ms. Harris. Sure. I mean, there has been a tremendous
amount of work that was done during the program reset period.
There is no doubt about that. There are a lot of very important
accomplishments.
At the same time, there is still a very large amount of
work that needs to be completed with regard to the continued
changes to the system, the configuration changes, as well as
the user-driven projects.
That, in combination with what they are doing in parallel,
which is getting these four sites ready to go live. I mean,
that is going to take a tremendous amount of resources. We know
within, like, the FHCC, for example, in Chicago that there were
some resources that were redirected from that system work to
get the sites ready.
Our concern is, in the absence of an integrated scheduler,
as Mr. Case has been talking about, as well as a cost estimate
to really know what it is really going to take to get to that
go-live, it is just a big unknown.
Mr. Luttrell. Thank you, Ms. Harris.
Dr. Evans, why did not the VA conduct an IV&V test or
assessment? Considering it worked for the DOD, I am curious
why.
Dr. Evans. Yes. It is a good question.
First of all, I think we have spent--we have had a lot of
attention in a very similar fashion to what an IV&V test would
do with our staff--and, frankly, staff from Oracle Health--
spending a considerable amount of time on the ground at our
live sites observing workflows from end to end and categorizing
where the issues are.
Those issues are the change requests that we have been
working through during the reset. You heard of the fact that
there are 1,800 still that we are working through at the
moment. Those are the change requests that have----
Mr. Luttrell. Would that assessment have problem solved
instead of doing it the way that we did conduct it? I only ask
that because we failed miserably. Would that assessment have
solved that problem?
Dr. Evans. I mean, I do not think that it would have----
Mr. Luttrell. We are, help me out with the numbers here, on
a $10 billion project, we are $12.5 into it, projected to go to
$50 billion.
Dr. Evans. Here is what I would say, is I think that with
regard to identifying the issues that we need to fix, I think
we have had an awful lot of help from our end users--and,
frankly, from our staff--in identifying what those issues are
that we need to fix.
I would welcome a third-party assessment of what additional
issues might be discovered. I do not think we are going to find
a significant set of surprise issues that we are not yet aware
of.
Mr. Luttrell. I would imagine there is a whole laundry list
of them, considering what the OIG just said.
Ms. Verma, how long have you been in your position?
Ms. Verma. I have been in my position for about a year.
Mr. Luttrell. Of course. Cerner--Oracle bought this project
from Cerner. That is the best way to say that, correct?
Ms. Verma. Oracle acquired Cerner.
Mr. Luttrell. Acquired it?
Ms. Verma. Yes.
Mr. Luttrell. Then I am sure they were very well aware of
the problematic issues that existed inside the EHR?
Ms. Verma. We have gotten more fully acquainted with those,
yes.
Mr. Luttrell. I am sure you are, considering where we stand
today.
What is your--because you said update costs. I am curious,
what is Oracle projecting the cost of activating the EHR, not
in 2026 but downstream in all 172 facilities, plus the
satellite campuses that you and I spoke about, Dr. Evans? When
everything is up and running what are we going with,
considering we are well over our original budget?
Ms. Verma. Just from a taxpayer perspective and former
government official to understand the need to really be very
careful with how we are using resources.
Just a couple of numbers. I heard the $50 billion estimate.
I think that is an outside group that we have not validated so
we do not agree with that estimate.
Mr. Luttrell. You cannot say you disagree with $50 billion
and it costs $49?
Ms. Verma. Agreed, but I do not think it needs to go to
that level. I think there are a lot of opportunities for us to
figure out how to run this program more efficiently.
The other thing I would say is that out of the $10 billion,
Oracle has received $5 billion. These are some of the other
estimates. Not all of that has gone to Oracle.
I do think that the way to make sure that we are using
taxpayer dollars efficiently, there is a couple of different
things that we can do.
Accelerating is really important. The longer the program
goes on, the more that it is going to cost.
There is also opportunities to do things around
standardization. Like we have heard a lot of discussion about
that. When we standardize, we also reduce the number of
additional configurations. Even if we can get done some of the
work around optimization, that also will decrease later on some
of the additional work that happens----
Mr. Luttrell. I am sorry, Ms. Verma. I apologize.
Mr. Chairman, are we doing----
Mr. Barrett. [Inaudible.]
Mr. Luttrell. Forgive me. I will circle back with you.
Thank you, Mr. Chairman.
Mr. Barrett. Thank you.
I recognize myself again.
Dr. Evans, IV&V you said essentially--not to paraphrase
what you said--but more or less that your users have accounted
for what that would do by utilizing the system and providing
change requests to that.
To me, not being an IT professional, that sounds a lot like
we are going to put it out there, and even though we know it is
not fully operating properly we are just going to wait for
people to tell us where it is broken and wait for it to then
roll in for the fixes that do that.
If we had planned this from the beginning and done that
thorough analysis that Ms. Harris had described, would it have
prevented that need for so many changes that are now
backlogging the system that are putting us in the position of
delay that we are in currently?
Dr. Evans. I think--so a couple things. When we look at the
DOD's experience deploying the EHR, they had a very similar
experience. Though they did complete an operational test, an
Operational Test and Evaluation (OTE) exercise to determine the
suitability of the system, they did not do that until after it
was already in production and was being used at several sites,
when they were having very, very similar problems to VA's
initial problems. That is point number one.
Point number two, I completely agree with you, end users
are one piece of the puzzle, but bringing experts in who can
watch and observe and provide insights that an end user
themselves might not be able to sort of provide from the
foxhole in which they are is valuable.
We have been able to do that with experts from the program
office, from our counsel, experts at the national level in VA,
from Oracle Health, from our research community who have very
carefully studied different parts of our implementation.
Again, I think we have learned quite a lot not just from
end user reports, but also from external bodies who have looked
at our implementation.
Mr. Barrett. Okay.
Ms. Verma, you said that a big portion of reducing costs is
accelerating the implementation of this program. While I
suppose I agree with you from a strictly cost standpoint, if we
were to just implement this everywhere maybe we could save some
money, but I do not think that you would have the confidence of
this committee--I am not going to speak for the other members--
but certainly from my own perspective given the track record
that is been out there so far.
Asking us to endorse the idea of accelerating this when it
has not proven to be adequate--and that is putting it in the
most diplomatic way I can--what can you give us to reassure
this committee that in fact accelerating is the appropriate
thing to do when the last time we implemented this we had to go
into an immediate freeze because it was so poorly implemented?
Ms. Verma. I think we are in a different place than where
we were 5 years ago, and even just the last 18 months. If we
look at the track record, even in respect to testing and
independent evaluation, if we look at the go-live in Chicago,
that has gone well.
If we look at even this weekend, I know time will tell, we
will get a better sense of the impact on our providers, I
think, if we just look at sort of in the recent history that we
have actually seen some good, successful----
Mr. Barrett. Would not you agree that the go-live in
Chicago was done with a significant assist from the Department
of Defense working in that regard as a spotter on the weight
bench, if you will?
Ms. Verma. I think that the DOD certainly helped. The VA
was there, the VA led that, and I think there were a lot of
lessons learned that will help them also take some of those
lessons in going forward.
There is still a lot of--if we look at the history in terms
of the impact that it has had, I mean, even on productively
there are some places where we have actually increased
productivity in some places.
We have also seen some of the work that we are getting in
terms of closing care gaps, some of the new tools that we have
been able to provide for providers on the front lines that have
actually helped them deliver more safe care in terms of
opioids, for example. The clinical decision support.
Those are things that I think are definitely helping.
Mr. Barrett. I heard that in your opening remarks and I
appreciate the focus on that. I am curious why we do not have--
to Mr. Case's point--we do not have a standardized schedule as
to when this is going to be implemented.
To Ms. Harris' point, we do not have that end-to-end review
and analysis of where the pitfalls and gaps are. If the answer
is to go faster, I do not think that a reasonable person would
view that that is the most appropriate thing.
I guess, with 15 seconds left, how can you reassure us that
that would be the case?
Ms. Verma. Look, we are committed to making sure that this
works. I think the history shows, the recent history, that
things have improved. The VA is paying for hosting, they are
paying for a lot of things that they are not using.
If we look at where we have been just over the last 12
months, I think things have started to turn around, they have
started to go well.
We are committed to getting the optimization work done as
well as in parallel trying to advance some of the work in terms
of going forward at the other sites.
Mr. Barrett. Okay. Our version of what good looks like
might be ought of sync with one another. I will be curious to
ask more.
I want to recognize Ranking Member Budzinski again for 5
minutes.
Ms. Budzinski. Yes. Thank you very much.
I actually want to kind of just build on what you were
talking about, about accelerated deployment.
What would it take for Oracle and the VA to be able to
accomplish that? If you could kind of further--both Dr. Evans
and Ms. Verma--if you could elaborate on that.
Ms. Verma. I think from my perspective one of the biggest
things that we need to do is standardize. If we get to sort of
a standard sort of configurations, that is going to really help
a lot.
Ms. Budzinski. I know you have said that a lot, not to
interrupt you, but I think one of the things that I have come
to understand and appreciate in some ways about the VA is every
clinic is different, every Community Based Outpatient Clinic
(CBOC) is different. The uniqueness of it, is that a bigger
challenge within the VA than it is perhaps with the DOD when
you say standardization?
Ms. Verma. If we think about the original goals of the
program was to create some standardization so that when
veterans go into one site they are getting a similar level of
care, a similar quality, a similar standard of care.
In order to do that, we need to make sure we are all
operating in the same way, and that starts with having the IT
configured in a way that is standard.
That, I think, that in and of itself will create a lot of
optimization, even if we think about a lot of the
configurations and a lot of the different requests, because I
think there is the expectation that we are going to do this in
a bespoke way.
That is actually contributing to more complexity and
increased costs, increased difficulties with training and
support if you are doing something different at every site.
When I say that I think we can go faster, that is
predicated on having standardization, that is key to be able to
go forward----
Ms. Budzinski. It is not--and, obviously, I believe in this
program, so I think that I believe in the standardization
concept. That is what we are moving toward. I am more asking
you the unique challenges within the VA to moving to something
that is standardized that is something a little different than
the DOD.
Ms. Verma. Yes. I think one of the things that--and we have
talked about this and been working on that--is how can we give
more support to the providers on the front lines, making sure
that they have adequate training.
I think that the work that we did in Chicago was a good
example of that. Not only did they have a lot of training in
advance, they also did elbow-to-elbow training so that when it
went live there were people circulating there that could give
them support. That is not something that should just happen on
the first day or the first 3 days. It still may take about even
a month later.
We find that a lot of the requests and a lot of the changes
that people are requesting, sometimes those are rooted in they
may not understand how to use the system. If we actually give
them more support and more training that they actually are not
going to request a change in the actual system. I think that is
also a piece of it, the standardization and more support, more
training.
Ms. Budzinski. Dr. Evans.
Dr. Evans. Yes. I think--and I think the original question
that you asked was--how do we feel confident about the ability
to accelerate and what does that look like?
I think I would start my remarks by saying that I just want
to--I mentioned it in my opening statement, but I would like to
underline it. VA is and agrees with the recommendation that we
do need a revised integrated master schedule and a life cycle
cost estimate regarding this program. Hundred percent agree.
When we made a decision to start our work in Michigan, we
made that decision very explicitly. We spoke to this committee
about that decision to say we are making a decision to start
moving toward Michigan with the understanding that we need to
in parallel be working on the schedule, the integrated master
schedule, for what things look like to get to Michigan, what
things look like to get to the sites thereafter, to the end of
this current contract, and to the end of our delivery.
We agree and are working on the schedule.
I think one of the things that I sense that you are trying
to get at, which I would very strongly agree, is that the
details matter when it comes to executing a complex project
like this. The details matter. That is what we capture in an
integrated master schedule.
We are at a very different spot in the program than when
this program began many years ago. And so sort of snapping the
chalk line and starting to build that schedule iteratively, we
have heard that recommendation from both GAO and OIG and we
will do it.
Ms. Budzinski. That is great. It was going to be what my
next question, is getting that schedule. It is great to hear
that that is something you would be open to.
Just because I have only have a few seconds left, can I ask
you really quickly, Dr. Evans, about the 3b pharmacy solution?
That was rolled out, I know I talked very briefly about it, but
can you generally just say how did it go this weekend?
Dr. Evans. It was rolled out this weekend. The block
upgrade, which is twice a year, we introduce significant
software upgrades to the Federal EHR. That occurs in February
and August every year. This block upgrade runs from the 21st to
the 28th. Much of it was implemented over the weekend. There
were a few small issues addressed this morning, but overall, it
is a little too early to tell, but it seems like things are
going well with regard to 3b.
Mr. Barrett. Thank you.
Mr. Luttrell.
Mr. Luttrell. Curious to hear what you just said about the
pharmacy, because Ms. Verma said it was going great, which it
kind of works in conflict here.
Can you elaborate a little bit more, Mr. Evans, on how that
pharmacy--when it was activated this weekend, because I want to
go back to Ms. Verma, because, again, I want to kind of unpack
the suitcase on this.
Dr. Evans. Sure. I think, probably as you and the committee
are aware, it has been a long journey to deliver this
particular improvement. When we first went live with the
pharmacy capabilities to support VA, and again, remember
pharmacy operations and VA are a little bit different than in a
typical commercial healthcare environment. VA runs both the
healthcare delivery organization as well as the pharmacy
fulfillment organization, both the pharmacy that is dispensing
the medication and the providers and staff who are ordering
those medications.
As we deployed pharmacy at the original sites, the pharmacy
community came and identified initially a set of seven issues
that they said were absolutely critical for us to deliver
before we went live at a level-1 high complexity site.
Of those seven capabilities, one of them happened to be
numbered three, and proved the most challenging. We had hoped
to have that delivered last February. There were some issues
with it and there has been a lot of work, including a very
different approach to how we did development and testing and
evaluation of 3b to get it delivered this week. I just--in the
interest of full transparency, as with any----
Mr. Luttrell. We appreciate transparency.
Dr. Evans. Yes. With any IT go-live, when something hits
the light of day, you might find something that you--you may
have done absolutely extensive testing and there may be some
small issue that you identify, which has been quickly fixed
this morning.
That is what I am talking about. I would say overall it is
great. I would agree that it is great that we are live with 3b
now, and that we are going to be able to see how that works in
the real world over the next week and the following weeks. That
is good news.
Mr. Luttrell. Ms. Verma, I know you are representing your
company and you are giving us kind of a grandiose expectations
of the success that is currently happening. We are in the
basement. We are in the basement. We are rock bottom. We
actually have failed totally because we are in pause.
Can you give me a statistical probability of success in the
next 2 years from Oracle? If it does not, if it is not
successful, the only person that is going to be sitting here
taking the punches is him. I want to hear from you and Oracle
if you think where we--I need to hear this out loud. Where we
are currently, when we go live in `26, and then running it
parallel with other facilities, correct, Mr. Evans? Is that how
you explained that to me in my office? Is that going to work?
Ms. Verma. Listen, we are committed to doing everything
that we can.
Mr. Luttrell. No, no.
Ms. Verma. I cannot----
Mr. Luttrell. Everybody says that that does not have a
plan.
Ms. Verma. Right. Well, I think we do have a plan. We
talked about a number of optimization projects that need to get
done. We are talking about how we implement in Michigan. We are
going back and looking at lessons learned and how we can
implement it in a way that supports end users with better
training, with better support.
Mr. Luttrell. You are highly confident Michigan is going to
fire up successfully.
Ms. Verma. I am confident, yes.
Mr. Luttrell. Give me a statistic.
Ms. Verma. I am not going to give you a specific number,
but I feel pretty confident about it.
Mr. Luttrell. Better than 50?
Ms. Verma. I think it is going to be better than where--I
think we are going to be building on all the improvements that
we have had over the, you know, with the optimization work and
some of the lessons learned in terms of training. We are also
putting in a lot of time and effort to figure out how can we
figure out how to optimize and how to make these go in a more
smooth way.
We are also building on a lot of the experience that we
have had, a lot of the challenges that we face. We have
actually learned from those, and so we feel comfortable going
forward and that this is going to be better.
I will say this, if we just look back, right, in the
recent, if we just look at Chicago, that has gone well, right?
That is the only implementation that was done by Oracle. That
has gone well. If we look at these recent code block releases,
this is not the only one that we have done. We have actually
done several. We did one in August. We did one last year. We
have seen that those have gone smoothly. If those had not gone
smoothly, I can understand that there would be some concern.
If we look at in the last year, Chicago has gone well. All
of the last two to three code block releases that we have gone
have gone fairly seamlessly. Yes, there is little bumps and
things like that that we have been able to get through, and
they have gone largely successfully.
If we look at the recent past and the turnaround in terms
of Oracle coming in, stabilizing the system, a lot of the
problems that we have had have been addressed. The stability
has been addressed. We have been able to do a good go-live. We
have also been able to upgrade the system in several different
areas successfully. Those areas, that is why I feel confident
that our next go-lives will go well and that we are capable of
accelerating.
Mr. Luttrell. Thank you. Mr. Chairman, I yield back.
Mr. Barrett. Thank you, sir. I will recognize myself again.
Coming to Michigan, 2026, we have level 1A complex facilities,
right? Ann Arbor and Detroit, correct? Have any of these been
rolled out in complex facilities up until now?
Dr. Evans. I can start with that. The James A. Lovell
Federal Health Care Center in North Chicago is a level-1
facility.
Mr. Barrett. Okay.
Dr. Evans. The short answer to that question is yes. I
think I will further answer your question that I am very, very
impressed with the leadership teams at all four sites in
Michigan. I have met personally with each of them as we are
getting things kicked off. In part in answer to Mr. Luttrell's
questions, one of the levels of confidence that I feel in
Michigan is this is as much about IT implementation, whether
the software works as it is the personal and human sort of all
of the staff at Michigan adopting, getting used to, working
through the change of adopting this new system. I am very proud
of the work that the teams in Michigan are doing to prepare. I
think that should also sort of raise our level of confidence
and success there.
Mr. Barrett. Okay. Doctor Evans, what are some of the
incomplete big rock projects that you described in your remarks
earlier?
Dr. Evans. Right. One of those is--an example of one of
those is what we call position standardization. The way the
Oracle health record, or what we call the Federal EHR works is
the user, depending on what type of user they are has a
different experience in the software, so whether they are a
social worker, primary care provider, rheumatologist, an
endocrinologist, which type of physician, nurse, et cetera,
there are more than 300 different roles configured into the
system.
We have learned at our sites that there were probably too
many roles that were chosen by VA when we started down this
path, and have been consolidating and standardizing what can be
done in those roles. We just completed standardizing the roles
for advanced practice registered nurses, geriatric and extended
care providers and social workers, and took what had been 28
roles, that is 28 different training programs, 28 different
sets of software capabilities to maintain, and compressed them
down to six.
We are now starting with the nursing roles, Registered
Nurses (RNs) and Licensed Practical Nurses (LPNs), there are
more than 50 nursing roles, and we anticipate compressing them
down quite a bit. Again, as we reduce complexity, it will
improve the sort of the efficiency of training and the user
experience as well. That is an example of one that is very,
very important that we are tracking toward making very
significant progress toward prior to the Michigan go-live.
Mr. Barrett. Okay. Are there any remaining big rocks that
are show-stoppers that if they are not done, it will inhibit
the ability to roll this out at the next phase in Michigan?
Dr. Evans. No. I would not say that they are show stoppers.
Mr. Barrett. Okay. Do you feel, right now, I guess I can
ask Ms. Verma this, is the system ready to go live today in
Michigan if we wanted to? Like, do you have confidence that we
could do this right now if we chose to? That is for you, Ms.
Verma.
Ms. Verma. I am sorry, I thought you said Dr. Evans. I
would say a couple things, right? If we look at the system that
we have today, we are making changes. We are doing
optimizations. I think that this year we should continue to do
that work around optimizations and so that when we do implement
with Michigan next year, which is what we are slated to do,
that that will go well. It should go better than the initial
five, because we will have identified what those optimizations
are and what those need to be, and so if we implement those
now, that that will make Michigan go better next year.
Mr. Barrett. Going back to when the initial rollout
happened, the number of facilities we have rolled out, north
Chicago, the whole thing and the pause, all of that, today you
do not have confidence that the system would be ready to go in
its current state?
Ms. Verma. I think that we can implement a system today
anywhere. However, there are some opportunities to improve the
experience, and I think we have agreed with the VA to do that
work and do the optimizations, do that before we go to another
deployment. That the system in and of itself, yes, that can be
deployed anywhere.
I mean, the reality is our EHR is being deployed all over
the world. The DOD is using it. From a technology standpoint,
yes, we can implement it. I think the VA and we agree that
there is some optimization work that would make that deployment
go in a much smoother better way and that it is worth making
those investments today.
Mr. Barrett. Okay. I am out of time on my question.
Ranking Member, go ahead.
Ms. Budzinski. Thank you, Mr. Chairman.
I wanted to zoom out again to the accelerated deployment
that we had talked about, that plan, and I wanted to
specifically get OIG and GAO's, if Mr. Case and Ms. Harris
could kind of--could you also share with us your thought about
the accelerated deployment idea by 2028 or 2029.
Mr. Case. Yes. To comment on whether it is feasible or will
it work, we need to see the schedule. We need to see the master
schedule. How do we get from where we are now to where it is
that they want to be, whether it is in Michigan or some
accelerated schedule. Show us the steps each way. How long is
it going to take? What are the risks that it will not work?
What is the critical path? How do you modify this if something
does not work or is not timely, and how much it will cost? To
really assess the viability of any acceleration, show us. Show
us the schedule. Show us the details, and then everyone in
Congress, VA itself, OIG, GAO can evaluate it and evaluate the
risks. I think that is really, from our perspective, one of the
issues.
One thing we have not talked about much here is, so far, is
our first recommendation of our first report was for VA to
assess what the impact was on productivity at sites, and how to
address that when it happens. Some loss of productivity is
expected, but how much, what is the rigorous analysis of the
loss of productivity, not just what happened at Lovell, but we
have got to be honest with ourselves. We are going forward at
four sites, including two complex sites, and how long will that
be and what are the efforts. We have not seen that analysis
yet, so we could close the recommendation.
Ms. Budzinski. Could I just, before we go to Ms. Harris,
the loss of productivity, could something like improving our
training get to addressing some of those concerns?
Mr. Case. That should be able to do it, ma'am, and we did a
report on the training at the early sites. It was insufficient,
and since then, my understanding is it has been improved, but
we have a new system. We have heard that again. It is a
different system now, and whether the new training that is
going to be done will be effective in the new system, that
remains to be seen, especially going forward at four sites.
Ms. Budzinski. Could I just ask you, Dr. Evans, because I
know we have talked about the schedule. Obviously Mr. Case just
emphasized that as well, but the importance of it, that sounds
like something the VA is committed to maybe--not maybe, but to
definitely putting together.
Dr. Evans. Yes, absolutely.
Ms. Budzinski. Did I hear that right?
Dr. Evans. Yes.
Ms. Budzinski. Okay. That is good. Can I ask you the same
question, Ms. Harris.
Ms. Harris. Absolutely, and before I respond to your
initial question, I just wanted to add on with regard to the
training, we have heard a lot from Ms. Verma regarding the
training, but the training really needs to--the ownership of
the training must reside at VA, and many of the pitfalls that
they had with the initial five sites was they sort of
relinquished that control to Cerner to do that training, when
VA should be responsible for it, because the VA users are the
ones that know the legacy system better than anyone else, and
what it takes to be able to train up on the new system.
My understanding--we made a recommendation that VA must
improve their change management strategies, no doubt. Training
is a huge component of that. My understanding is they are still
working on that, but that should be finalized and be very
detailed before they go forward with the go-live. In addition,
just with regard to the--are they prepared, I mean, they have
got--if they deploy in 2026 to Michigan, I mean, that is a huge
if for all the reasons that Mr. Case has stated, but you still
have 2 years remaining on that contract with 160 sites, it is
impossible. The answer is, no, they are not going to be able to
deploy to all of the sites within the remaining time of the
contract.
To get to Michigan, at least as Mr. Case said, we need that
integrated schedule, and if they do deploy, Mr. Evans did
mention--Dr. Evans mentioned something very important about
DOD, which is, you know, they did the IV&V after that initial
wave of--so it was multiple sites that they deployed to before
they did the IV&V.
It could be a very good test, I think, if they were to
deploy to Michigan to have that IV&V done at those four sites.
Again, an independent look, not VA, not Oracle Cerner to do
that, that review, so that we can identify in full--end to end,
what those issues are.
Ms. Budzinski. Just to go back to that point, because I
know we have talked about it a bit, but to just hear from
Oracle and the VA on the IV&V test, like, if there is some
added value to a third party helping us take a review, look
under the hood, I feel like I have heard there is not
necessarily opposition to that. That could be maybe helpful.
That is a question, I guess.
Ms. Verma. What I would say is there is a cost to that, so
that should be part of the calculus here. The second thing I
would say is if that we look at the more recent implementations
in these code blocks, the testing that we have done has worked
well, right? We have actually made changes to the testing
program. We have given them different environments to help
their internal process of testing, and it has actually gone
well.
It could be that an IV&V maybe a few years back would have
been helpful, but I think we are in a different place now and
the testing program that we have has worked well if we just
look at the recent past with the last two or three code block
releases. I think they have gone better, and I think if you are
going to go forward with IV&V, to think about the additional
cost of that as well.
Ms. Budzinski. Okay. Great. I yield back. Thank you.
Mr. Barrett. Thank you.
Mr. Luttrell. Thank you, sir.
Mr. Evans, what is the--I do not know the best way to say
this, but standard of success for Oracle and what they need to
meet in order to say, Hey, look, this--Michigan is good, we are
moving--which one is the next site? I mean, because we went
into the pause because the standards were not met. Are they
required to meet the standards that currently exist, or have we
increased those standards of success?
Dr. Evans. I think I mentioned earlier that we have worked
on some metrics that we are following, and actually, frankly,
with Oracle Health we follow a lot or many more metrics than
are actually on the dashboard of the nine core metrics that we
followed during the reset, and that we will continue to follow
as we move forward.
We have defined the thresholds for success for each of
those with the exception of one that we are finalizing now
after we did a burn-down of trouble tickets and are resetting
what the normals should be.
You may see that in a GAO recommendation that we need to
set our targets and thresholds. We have set eight of the nine.
The ninth will be set momentarily. That will be a measure of
how we measure success, but there are many others. Really to
me, what is----
Mr. Luttrell. Oracle needs to meet those nine standards----
Dr. Evans. Right.
Mr. Luttrell [continuing]. before we consider Michigan a
success and then we flex to the next facility?
Dr. Evans. Right. I think those standards are--I mean,
these are--some of the metrics are operational metrics, right?
It is how we measure quality of care at our facilities. It is
access to care for veterans. It is wait times.
Mr. Luttrell. All right. Let me break this down a little
bit further. My question is, at what point in 2026 when Oracle
activates Michigan, there is going to be those checks in the
box. Now, if they get three out of the nine, are they good to
go and that is considered a success, or do they have to have
all nine? Which I would say obviously you need all nine, but I
am asking you.
Dr. Evans. I mean, my preference is that we are meeting our
goals across the board. Now having said that, I would say it
is--we are deploying the Federal EHR in Michigan, right? This
is a partnership that VA and Oracle Health are going to need to
do this together to success--and frankly, with the local
leadership in Michigan, and so we--the success here is a shared
project. I would say that is number one. Number two, yes, I
would like us to see us meeting our goals across all the
metrics.
Mr. Luttrell. If--I am still trying to dig an answer out of
you guys. My concern is that we kick the can down the road. We
activate in Michigan and we only have a certain amount of
successes out of those nine like, Hey, that is good enough, we
are just going to go further and Michigan fails. Then here we
are all over again and we have to go into another pause and
projections well past $50 billion. Because of where we are
currently, there has to be a line-by-line check that says, We
are absolutely solid, Michigan is good to go, 100 percent, and
now we are moving on.
In my opinion, that is exactly where we are considering how
many dollar bills are going to this project. I do not know if
you can answer that question or not, Mr. Evans. Unless you want
to----
Dr. Evans. In short, I can say yes. System has got to be
stable. It is got to work for end users. It has got to work for
the operations of our medical centers, period. How we measure
that is something that I meet with the committee staff on a
monthly basis and we go through those metrics and we will
continue to review those as we lead up to Michigan, and
frankly, we are going to use those.
Mr. Luttrell. When you move out of Michigan, is there an
end date for Michigan where we activate here, we have got 2
months and then we are going?
Dr. Evans. We go live at Michigan, and then typically, we
are looking to see operations restored at the medical center in
the three-to four-month timeframe after a go-live.
Mr. Luttrell. Three to four months in Michigan alone.
Dr. Evans. I think--I mean, I think one of the things here,
as we talked about earlier, right? If we wait, we have a
challenge in front of us, right? We--the longer VA operates as
a healthcare system with two different electronic health
records, the more risk that we are taking on. We have to
balance the need to move forward toward a single electronic
health record to support an integrated national healthcare
system in VA. If we are going to move forward with the Federal
EHR, which is our plan, commitment, and desire, then we cannot
just wait for Michigan to be over and spend another year
deciding whether we were successful or not.
we are going to have to measure together as we move forward
our success month to month. It is one of the reasons why I meet
with this committee every single month. It is because we need
to be assessing our progress at that frequency. Because if we
do not start to do work at sites beyond Michigan, we will get
to the end of Michigan and we are going to have another 18
months to wait for the next go-live.
Mr. Luttrell. What if Michigan--I am sorry, Mr. Chairman, I
am over. You know where I was going with that.
Mr. Barrett. I will come back to you in a moment. Okay.
Ms. Harris, I want to go back to your earlier point.
Assuming we roll out in Michigan in 2026, that would give us
how many sites have been--have had the new electronic health
record implemented at that point?
Ms. Harris. 10 sites.
Mr. Barrett. 10 sites. How many would remain?
Ms. Harris. 160.
Mr. Barrett. 160. It is your opinion and your testimony
that it would be impossible to implement the remaining 160 by
May 2028. Is that correct?
Ms. Harris. That is correct, based on the previous track
record and also, based on the fact that all of these sites
have--are very bespoke. They are customized, and so, the change
management associated with each of these individual sites, I do
not think you can standardize that.
I think some might take longer than others, and it is just
very hard to predict, but collectively, within a 2-year time
period, considering it took them basically 6 years, 6 to 7
years to implement a 10 and you are talking 160, when you
compare, to me, there is no way they are going to get that all
done.
Mr. Barrett. Right. Okay.
Ms. Verma, obviously the next question for you is, are we
going to be through with the next 160 sites by May 2028?
Ms. Verma. I think there is a lot of variables here, right?
There is a technology piece. There is a training piece. There
is a people piece of this. There is also internal processes at
the VA. You know----
Mr. Barrett. I think a simple no would probably be the
easiest answer.
Ms. Verma. We have been working with the VA around
acceleration and how to do that, and there are multiple things
that need to be addressed in order for us to move faster and to
move quicker, and I think we are working toward that.
Mr. Barrett. Yes or no, is it your position that this can
be implemented by May 2028?
Ms. Verma. From a technology perspective, yes, but there
are other pieces of this that have to come along with the
technology.
Mr. Barrett. We live in the real world that we live in. We
do not live in the just hit send on the computer and it works.
Ms. Verma. Right.
Mr. Barrett. Understanding the real-world conditions that
we are in, understanding we do not have a schedule, to Mr.
Case's point, Dr. Evans, do you feel that there are any
possibility we can be implemented in the remaining sites by May
2028?
Dr. Evans. No.
Mr. Barrett. Okay. Mr. Case, do you?
Mr. Case. I do not see how it could happen. I would like to
see what the plan is if they were going to do it.
Mr. Barrett. Okay. Ms. Verma, do you feel that this is even
possible? I mean, can we even pretend anymore?
Ms. Verma. Like I said, we are responsible for the
technology standpoint, and we have said to VA from a technology
standpoint that may be possible, but there are other pieces
that go along, and we will acknowledge, right? That the VA,
whether it is standardization, whether it is integrations,
there is a lot of things that would need to happen to be able
to meet that timeline. We are committed to doing that.
However, that there is also other pieces of it and we would
need to work with VA on that, and we are having discussions
about that, about what the timeline should be and what the
schedule should be, and what are the things that need to happen
to be able to accelerate.
Mr. Barrett. Okay. I think even with acceleration, it is
impossible to expect that we can be anywhere close to that by
May 2028.
Now, Dr. Evans, to follow up on Mr. Luttrell's point,
assuming we are in Michigan for several months, I think he was
getting to the point of what if it is not good? What if it is
not working? What if the checklist is not adequate? What if we
are stuck in the position we were in the initial rollout? What
if these are not going as well as we would like? What happens
then? Do we go into another stand-down freeze? Do we keep
accelerating as Ms. Verma has suggested? What do we do then?
Dr. Evans. It is an excellent question. I would say this. I
do not--I think we have to succeed, and I think we should be
looking----
Mr. Barrett. I would have said that 7 years ago too. I
mean, we had to succeed then and we did not, leading us to
where we are now. I agree with you, we have to, but we have to
acknowledge the track record that led us to where we are now
for the reality in which we are facing.
Dr. Evans. Right. I understand that. I still believe that
we need to succeed and we need to be working toward Michigan,
with the understanding that that is what we must do. Your
question is a question of well, what would we do if we did not
succeed. I think we would be having very hard discussions right
here in this room, most likely, but where I am working with our
team, and frankly, where we are working at VA across the agency
is to say this is something we must do, and I will tell you
this.
In Michigan, they are enthusiastic. The sites are
enthusiastic in Michigan. One of the things that I think you
are well aware of in Michigan is that the sites in Michigan
work in a very interdependent fashion. Specialty care for
veterans who are receiving care in the Battle Creek or Saginaw
system often is delivered through Detroit or Ann Arbor. Lab
services are shared across the VISN. The idea of--the
opportunity to move from four separate instances of an
electronic health record supporting four separate VA medical
centers that are trying to operate more and more like one, they
look at the opportunity here to have a single electronic health
record for all four sites to support their sharing and care for
veterans in the Michigan market as a huge opportunity.
I think we should feel some level of confidence when the
sites are saying this is something we want, that we are ready
for.
Mr. Barrett. All right. Thank you.
Ranking member Budzinski.
Ms. Budzinski. I appreciate that. I wanted to actually
shift gears a little bit to AI as something that Oracle had
talked a little bit about in the written testimony. Can you
explain, Ms. Verma, how AI is being used in this instance?
Ms. Verma. A lot of the AI that we are going to be
deploying will come in sort of the modernization. I will say
from the outset with whatever AI that we do deploy is that
providers and physicians are always in control, right? This is
not AI doing things, but giving options and helping and
supporting, but decision-making about clinical care and all
workflows, there is a human in the loop. I would start out by
saying that.
I think what we have done to improve the her sort of
outside the VA but that can be applied and we are willing to
provide to the VA. For example, we have brought something
called our clinical AI agent, right? Which can--if the patient
agrees and the doctor's on board with it, you can listen to the
patient/doctor interaction and can generate a note. That is one
of the first things that we have brought to the market.
The second thing that we are bringing is a brand new EHR to
the market, and that has AI infused across all of it, so if you
think about a physician coming in to seeing a patient, today
and really any EHR, you have to do a lot of searching. Look at
their labs. Look at all their previous diagnoses. Especially
for a veteran where they have complex healthcare conditions,
and they have multiple co-morbidities, that is actually quite
complex. What AI can do is give them a summary of the patient.
It can detail here is all the different things.
The other piece that AI can do is it can help actually
provide guidance to the provider and say, okay, based on this
veteran's health condition, here are the care gaps. That is
actually--we are actually bringing that in today, but AI can
also--the provider has questions, bring me this, or it has a
specific question, it can actually ask the her that and that is
actually given.
The other thing that we can do with AI is we can identify
if there is research going on in the facility, and is this
patient a candidate for the research that is going on in that
facility as well? I would say that AI is also going to automate
a lot of the manual work that we see going on with providers,
especially in the area of our reimbursements and collections
and that particular area. There is a lot of manual work that
goes on with that today. Think about scheduling and
administration.
That is where I think by applying some of this modern
technology, there is also some efficiencies to be gained by the
VA as well, and it allows the providers to spend time with the
patient and not doing a lot of administrative work.
Ms. Budzinski. Is this--I mean, that type of innovation, is
that covered under the current contract, or will there be
additional cost associated with that?
Ms. Verma. There are some pieces, like our clinical AI
agent, which are in scope. In terms of the new EHR that we
have--that we are bringing to the market, we are committed to
providing that to the VA at no additional cost, so even though
our contract specifies Millennium, our leadership has agreed
that we would like to provide that to the veterans, because we
think that that is going to give the most optimal experience,
and it will improve care and quality for veterans, and that is
the most important thing, so we are committed to doing that.
Ms. Budzinski. How does Oracle look at quality and safety
as it relates to this issue of AI and veterans and patient
care?
Ms. Verma. Sure. Safety is a high priority. We have got
many systems and processes in place to assure that. I think the
most important thing is that when we are deploying AI in
anything that we deploy, there is always a human in the loop,
right?
This is supposed to be a tool. It is supposed to be an
assistant. AI is not a decision-maker. Everything that we
deploy, and even if we look at things like we are providing at
the--let us say the provider asks a question, we will not only
provide an answer to the question, but we will also identify
the source of the data.
Let us say we are providing a summary of the patient's
history. In the summary that we provide, we will actually
detail exactly where the data came from, so it is not like the
provider is getting guidance and we are getting a
recommendation, whatever you want to call it. They know exactly
where that came from. I think the most important thing that we
are doing is that the human is always in the loop in these
processes. That is how we are applying AI.
Ms. Budzinski. Will the system require informed consent
from the veteran before AI is used during their appointment?
Ms. Verma. Yes, and that is completely up to the VA in
terms of making sure that those consents and how they want to
put those processes in place. I think the most obvious one
today that we are already using is around our clinical AI agent
that generates the notes, and in those situations the provider
always asks the patient, are they comfortable with that.
Ms. Budzinski. Okay. Just going back to something I would
asked about earlier, not on AI, but related, Dr. Evans, when I
was asking early on about some of the VA cuts and thinking
about cost to staff, do you expect that the VA might have to
take on additional contracts to supplement for where some of
the staff might have been cut in order to make sure that we are
properly going live in March?
Dr. Evans. I think it is too early to tell that. We are
still looking to make sure that we are optimizing how our
current staff is deployed to meet the needs of the program.
Ms. Budzinski. Okay. I yield back. Thanks.
Mr. Barrett. Thank you. Mr. Luttrell.
Mr. Luttrell. Thank you, sir. Dr. Evans, is all of the
veterans' healthcare data being moved into the Oracle cloud?
Dr. Evans. Yes. Let me--so the--we have migrated data from
the VistA systems and actually every day are migrating data
from the VistA systems, our electronic health record, into what
is called HDI, Health Data Intelligence, it is an Oracle
product that will be hosted in the Oracle cloud.
That data then supports--is migrated into the Oracle
electronic health records support care delivery so that when
the provider logs in--one of the challenges when you move to a
new electronic health record is if you have to start over and
find all the information and repopulate the new record with the
information, it is a significant burden on providers and staff,
nurses and the like, to do that population of the electronic
health record.
VA made a significant investment in partnership with Oracle
Heath to migrate a significant amount of data from our existing
electronic health record, VistA, into the Federal EHR, which
should help ease the transition.
Mr. Luttrell. Which Oracle controls in their cloud?
Dr. Evans. It is VA's data.
Mr. Luttrell. Hopefully. God willing, that stays the same.
Does anyone--Ms. Verma, is the data that has moved from the
VA into the Oracle cloud, is it standalone and no one else can
touch it?
Ms. Verma. Correct. It is a standalone. All the data rights
and everything specify that that is controlled by the VA. It is
actually shared between the VA and the DOD.
Mr. Luttrell. Okay. Other organizations that work with
Oracle and work with Oracle cloud system, they cannot access
that data at all?
Ms. Verma. Correct. They do not have access to it.
Mr. Luttrell. It is completely standalone.
Ms. Verma. Completely standalone.
Mr. Luttrell. Okay. Thank you.
I yield back.
Mr. Barrett. Thank you.
Dr. Evans, you said it takes sites, on average, several
months, 3, 4 months maybe, to basically fully recover, if you
will, and get back to full-steam operations after implementing
this. Do you feel that the sites that have already implemented
this new electronic health record are fully recovered, fully
functional, and fully operational right now?
Dr. Evans. I think one thing I would clarify with that is
that what we have learned, thus far, is that the recovery
varies based on the specialty, based on where the care is being
delivered. For example, urgent care and the Emergency Room (ER)
recover their productivity almost near instantaneously, which
is important, because we cannot have volumes decreasing in
urgent care in the ER. Whereas in primary care, as an example,
in part secondary to perhaps some of the configuration
decisions with the electronic health record, in part secondary
to the complexity of primary care having to deal with a
patient's entire set of health issues, medication management,
for example, productivity, recovery lags in primary care, so it
is not exactly the same across the entire medical center.
Having said that, we are measuring productivity across the
entire medical center at each of our facilities. We have seen a
positive and appropriate trend at all facilities, and they are,
for the most part, at their pre-go live levels of productivity,
actually pre-pandemic, pre go-live levels of productivity.
Mr. Barrett. Spokane and the other facilities are nearing
pre go-live productivity marks across the spectrum?
Dr. Evans. Yes. I do not have the exact numbers right here.
That is on our dashboard. We share that regularly. It is north
of 85 percent across all facilities. Some are above 100
percent. We can certainly share those numbers.
Mr. Barrett. Okay. I know you said earlier that we must
succeed in this, and I fully agree with you. We have not,
though, given in the more recent discussion just now, given a
lot of consideration for the cost, and the cost of this is an
unknown at this point. I think yes, we have to--we have to make
this work, but we also have to be mindful and knowledgeable
about the cost that that is going to take. I think to the other
observer standpoint, that is something that remains an unknown.
I do not know if you can----
Dr. Evans. Yes, I can agree, and as I mentioned earlier, we
are committed to building out the integrated master schedule.
The integrated master schedule is the foundation for building a
life cycle cost estimate.
Mr. Barrett. How long until we can expect to see that?
Dr. Evans. I think, as I mentioned, our plan is to do that
iteratively, so the first phase that you will see is the
integrated master schedule for Michigan. What is happening at
this very moment in Michigan is what we call the current state
reviews that will inform the build of the integrated master
schedule by April for the Michigan sites. Then we will begin to
plot out the schedule beyond Michigan and build that integrated
master schedule then to the end of the contract.
We are accelerating that effort. I cannot give you an exact
date right now that we will be finished, but we understand how
important that is to be able to build the life-cycle cost
estimate. We are very aware of the importance of getting that
information to you and frankly others that are at the end of
the table here.
Mr. Barrett. If we are years, plural, into this, and we
still do not have the schedule and we are scheduling the
schedule, how can we expect to get that in a reasonable--just
to get our arms around it? I mean, our position here as members
of this committee is to look at the effectiveness of this, the
oversight of it, but also the cost of it, and are we delivering
a good value for the people that sent us here to spend their
money on their behalf, and we cannot even do that without an
analysis of what that is going to look like.
Dr. Evans. Yes. I mean, I know this--I do not want to
make--I am making a promise in the sense that I can promise you
that work is going on right now. I was just in meetings looking
at the analysis of cost to date, our projections of what that
cost looks like moving forward. We are actively working on the
life-cycle cost estimate, and I look forward to keeping you up
to date. We understand that it is important for you to get that
soon--sooner rather than later.
Mr. Barrett. Okay. Thank you. I appreciate the--everyone
here, to our witnesses for appearing today to discuss the
future of the electronic health record modernization program. I
understand committee members--sorry. Ranking Member Budzinski
for your closing remarks.
Ms. Budzinski. Thank you very much. I appreciate everyone's
time and the panel's discussion and the thoughtful questions
from my colleagues. I think one of the things that is very
clear is that we all share a lot of really deep concerns about
our next go-live date. I am encouraged to hear Dr. Evans
talking about the integrated master schedule and working toward
some real metrics that have been recommendations from the GAO
and the OIG to address some of these concerns.
That being said, I think it is really important that we are
not pressured into this accelerated--the VA, I should say, is
not pressured into the accelerated timeline of 2028 by anyone,
really just to avoid a potential contract extension, because
this really is ultimately about veterans' healthcare and making
sure that they are getting the health care that they deserve
and it is being administered to them. I think we share
collectively, I have a lot of concerns. I think we have some
steps to address some of those, and I look forward to
continuing this conversation. Thank you very much.
Thank you, Mr. Chairman.
Mr. Barrett. Thank you. Thank you for keeping me well-
behaved. Again, I want to thank our witnesses for appearing
today. Thank you for the testimony that you provided. I know
that many of you spent hours, plural, probably tens of hours,
hundreds of hours developing and analyzing this. As I stated at
the beginning, I am new to this committee and new to this
Congress and new to this issue, but I am invested in making
sure that we deliver the outcome that we all should expect.
I know you all understand when I say this is not just an IT
system. It has a lot of individuals and their health that
depend upon this. The productivity and the morale of the people
doing the work as well as the outcomes for health for our
veterans, it directly affects all of them.
The veterans that I represent in Michigan's Seventh
congressional district are going to be among the first to find
out if this system has fundamentally improved since the reset.
The facilities in Battle Creek and Ann Arbor, specifically, are
where the majority of veterans in my district go for their
healthcare. It is why I am concerned by the talk of speeding up
the pace of go lives considering the VA has only done one
partially successful go live in the last 3 years with
assistance, again, from the Department of Defense, and with the
complexity coming with the facilities in Michigan, it gives me
a great deal of concern for that.
It is far more important the VA does this right instead of
fast. The plan is to continue fixing the system's problem while
preparing to go live in Michigan, and that is something that we
need to be very mindful of. The subcommittee expects VA and
Oracle to show that they can get this right before veterans and
VA staff in Michigan have to figure it out for themselves.
Dr. Evans, I am encouraged by your remarks that the staff
in Michigan are very excited about the opportunity that lies
before them to integrate this together in the way that they
have been interoperably intending to do in Michigan for some
time now.
Thank you again for all of you for participating in today's
hearing. I want to thank the members of the committee for your
questions and remarks as well. I will ask unanimous consent
that all members have 5 legislative days to revise and extend
their remarks and include extraneous material. Without
objection, so ordered. With that, this hearing is adjourned.
[Whereupon, at 4:47 p.m., the subcommittee was adjourned.]
=======================================================================
A P P E N D I X
=======================================================================
Prepared Statements of Witnesses
----------
Prepared Statement of Neil Evans
Good afternoon, Chairman Barrett, Ranking Member Budzinski, and
distinguished Members of the Subcommittee. Thank you for the
opportunity to testify today about the initiative of the Department of
Veterans Affairs (VA) to modernize its electronic health record (EHR)
system.
I want to begin by thanking Congress and this Committee for your
shared commitment to Veterans and for your continued support of the VA
Electronic Health Record Modernization (EHRM) efforts. This testimony
comes at a pivotal moment. With new leadership at the VA, there is a
renewed commitment to ensuring that every aspect of the EHR
modernization efforts is reevaluated. Secretary Collins made it clear
in his confirmation hearing that the status quo is not acceptable.
Every previous decision, policy, and process must be thoroughly
reviewed to ensure that it aligns with our core mission: providing
Veterans with the highest quality care through a system that works for
them, not against them. Specifically, we are going to take a hard look
into previous EHR efforts. That process has begun. The VA is committed
to successfully implementing a modernized, interoperable Federal EHR
system across its enterprise. VA's implementation of the Federal EHR
system will provide a single, accurate, lifetime health record for
Veterans that includes their health records from the Department of
Defense (DoD). The new Federal EHR will provide a framework for
improved enterprise standardization of health care delivery, which will
positively impact patient care quality and safety. The Federal EHR will
support simpler integration of other modern health information
technologies and infrastructure to provide a more coordinated
experience for VA staff and clinicians as they care for Veterans. The
modernized EHR will support improved interoperability with the rest of
the American health care system.
VA's focus is keeping Veterans at the center of everything we do.
Veterans deserve high-quality health care, which means health care that
is timely, safe, Veteran-centric, evidence-based, and efficient. The
EHR is, and will remain, a key enabler of VA's ability to deliver the
comprehensive health care Veterans deserve. VA remains committed to
delivering an EHR that will support these goals. In addition, the
adoption of a single system used by VA and DoD will help simplify
health care delivery for providers in both Departments, benefiting
patients who receive care in both systems or who are transitioning from
DoD to VA for care. It will improve opportunities for collaboration and
joint operations between the health care systems.
VA paused deployments of the Federal EHR system after listening to
feedback from Veterans and staff who said the new EHR was not meeting
expectations. VA announced the EHRM Program Reset on April 21, 2023,
with goals of (a) addressing the concerns of the sites where the system
was live, (b) investing in foundational enterprise work necessary for
long-term success, and (c) preparing for successful system deployment
at the Captain James A. Lovell Federal Health Care Facility in North
Chicago, Illinois (Lovell FHCC), the sole exception to deployment
activities. A little less than a year later, on March 9, 2024, VA, DoD,
and the Federal Electronic Health Record Modernization Office (FEHRM)
realized one of those goals by successfully and jointly launching the
Federal EHR system at Lovell FHCC and at its affiliated clinical health
care sites.
Throughout the EHRM Program Reset period, VA took steps to
understand the issues, updated our contracts to better hold Oracle
Health accountable, made hundreds of improvements to the system, and
instituted a simpler and more effective process to address concerns
when they arise. It's important to review these lessons and ensure they
are learned and implemented. The VA is committed to delivering an
effective EHR that supports seamless care for Veterans and enables VA
employees to efficiently document and access important health
information. Specific program outcomes will include:
Ensuring Veterans Have Trust in the EHR system: Veteran
outpatient trust scores have increased at all Federal EHR system sites
since the beginning of the Reset period--reaching 93 percent at the
Columbus VA Medical Center (VAMC) in Ohio, which is an 11.6 percent
increase since Quarter 1 (Q1) of fiscal year (FY) 2023; 88 percent at
the Walla Walla VAMC in Washington state, which is a 4 percent increase
since Q1 FY 2023; 92 percent at the Mann-Grandstaff VAMC in Washington
state, which is a 3.5 percent increase since Q1 FY 2023; 85 percent at
the Roseburg VAMC in Oregon, which is a 5.2 percent increase since Q1
FY 2023; and 89 percent at White City VAMC in Oregon, which is a nearly
6.5 percent increase since Q1 FY 2023. In addition, Veteran trust at
the North Chicago VAMC has increased to 90.8 percent from 90.2 percent
since Federal EHR system deployment in Q2 FY 2024. These improvements
are the result of direct surveys of Veterans and their experience with
VA outpatient health care, and it is important to ensure that these
numbers of trust continue to increase.
Dramatically decreasing outages, which disrupt patient
care: Since January 2024, there has been a significant decrease in
outages for the Federal EHR system--with the system functioning 100
percent of the time for 10 of the last 16 months, and 99.8 percent of
the time or better in the remaining months. As of January 11, 2025, it
had been over 250 days since the Federal EHR system last experienced an
outage.
Decreasing the number of interruptions for clinicians,
therefore minimizing slowdowns for Veterans: The average user now
experiences near zero interruptions (freezes or delays, for example)
per day.
Increasing clinician and staff satisfaction: Clinician
and staff satisfaction with the Federal EHR system has increased--
including increases in agreement in employee surveys with the phrases
``the EHR is available when I need it'' and ``this EHR enables me to
deliver high-quality care.''
Launching the Federal EHR system successfully in Chicago:
During the Reset period, VA launched the Federal EHR system in North
Chicago because it is a joint facility with DoD. The facility saw a
rapid increase in productivity and use--outperforming previous
rollouts.
In light of these improvements, on December 20, 2024, VA announced
that it was beginning onsite planning efforts to deploy the Federal EHR
system to four Michigan facilities in Ann Arbor, Battle Creek, Detroit,
and Saginaw. Representatives from the EHRM Integration Office, Veterans
Health Administration (VHA), and the Office of Information and
Technology (OIT) compiled functional and technical metrics to determine
the best sequence for restarting site deployments. These metrics served
as data points to evaluate site readiness and were closely considered,
along with input from VHA, Veterans Integrated Service Network (VISN),
and VAMC leaders to determine selection and sequencing for these new
deployments. The final order and dates of system go-live events at the
sites will be determined by the findings of the current state reviews
(CSR) that are being completed as part of ongoing pre-deployment
activities. It takes more than a year to safely and effectively prepare
for a go-live.
The new VA leadership is rigorously reviewing the current State of
EHR and looking for ways to accelerate the previous timelines that were
committed to and will share with Congress what next steps are decided.
Based on lessons learned from previous deployments, the
preparations for going live will include new approaches to optimize
adoption and engagement for future users of the Federal EHR system,
such as the development of more ``Learning Labs,'' which proved
successful with our go-live at the Lovell FHCC, to allow end users to
practice their skills in a simulation environment prior to go-live.
Other improved or new approaches include improving onsite engagement
during and after go-live, offering informal sessions for sharing tips,
streamlining computer-based training, and further enhancing the
Refresh, Optimize, Adopt, and Reinforce (named ROAR) Initiative to
support continuous improvement, even after go-live events have
concluded.
VA leadership acknowledges there is still work ahead and is
committed to ensuring that we continue to listen to employees who are
using the Federal EHR system, that we use their feedback to make
enhancements to improve the system, and that we are ready and have a
solid foundation for upcoming deployments. The continuous improvement
efforts we have been focusing on during the Reset will continue
unabated while VA begins early stage deployment efforts in Michigan.
For example, VA identified several areas for improvement that are
important for long-term success, which required re-visiting decisions
made early in the program. These projects are more significant than a
simple fix and are being referred to as ``big rocks''--due to our
prioritization of these efforts and because of the complex scope of the
tasks. Some of these projects include further improving training for
new users; standardizing and consolidating user roles in the system;
and delivering pharmacy capabilities to improve coordination between
ordering clinicians and pharmacy professionals. Notably, an important
part of the planned pharmacy improvements, known as ``pharmacy 3b'' is
scheduled for release this month as part of a bi-annual software code
update.
Another improvement that will increase efficiencies and prevent
future delays is the development of the Federal EHR system baseline, a
tool that provides information on more than 2,300 functional and
technology components that make up the Federal EHR system and was
established as part of an enterprise-wide effort to better standardize
the delivery of health care in VA. Using the baseline to better conform
to national VA standards in advance of go-live will help avoid staff
frustration, deployment delays, and increased cost.
Ongoing improvements at existing sites and pre-deployment
activities at future sites can occur at the same time. We can, and are,
doing both at this time. VA is continuing to move forward with a
modern, commercial EHR solution in close coordination with our Federal
partners, including DoD and the FEHRM. This new Federal EHR system will
ultimately improve Veterans' health outcomes. Not only that, but the
new Federal system will also house Veterans' health care records in one
place from the first day they put on their uniforms to the last day of
their lives. It will empower Veterans to receive care that is more
seamlessly coordinated across the enterprise. It will help providers
more holistically understand injuries or illnesses that Veterans
suffered years ago, so that they can provide those Veterans with the
best possible care today. It has the potential to further streamline VA
operations and most importantly, it will improve the Veteran
experience.
Veterans are at the center of everything we do. Ultimately, our
goal is to deliver an EHR system that earns the trust of Veterans,
clinicians, and staff. This means a system that works efficiently,
enhances care coordination, reduces administrative burden, and will
improve health outcomes for Veterans. We are not simply continuing
business as usual - we are committed to getting this right. The
responsibility we carry is immense, and we will not rest until this
system delivers what our Veterans and providers truly need. With the
activities and improvements that are now underway, VA leaders are
optimistic about the success of our Federal EHR system optimization
efforts and the eventual full implementation of the system throughout
VA.
With the partnership of this Subcommittee, we look forward to
making the changes necessary to fulfill the vision over the course of
the 119th Congress. Furthermore, I extend my gratitude for your
commitment to serving Veterans with excellence. Together, we will build
a system that honors their service and improves their care for
generations to come.
This concludes our testimony. We look forward to responding to any
questions that you may have.
______
Prepared Statement of Seema Verma
Introduction
Chairman Barrett, Ranking Member Budzinski, and members of the
Subcommittee, thank you for the opportunity to speak with you today
about Oracle's work with the Department of Veterans Affairs' (VA)
Electronic Health Record Modernization (EHRM) program.
On Dec. 20, 2024, VA announced they were beginning the pre-planning
activities for deployments of the EHRM program, and pre-deployment work
has begun at four sites in Michigan: Ann Arbor, Battle Creek, Detroit,
and Saginaw. Pre-deployment work begins with a Current State Review
(CSR).\1\
---------------------------------------------------------------------------
\1\ A CSR is a thorough evaluation of a VA Medical Center's (VAMC)
current infrastructure, workflows, processes, and technologies that
interact with the EHR system. This includes technical aspects such as
hardware, network infrastructure, and existing software applications
and integrations, as well as operational workflows like patient
registration and clinical documentation. CSRs involve interviews and
discussions with key stakeholders across various departments at the
local site--administrative, clinical, and IT teams--to understand their
VA service scope and specific needs related to the EHR system.
Ultimately, with this information, Oracle provides actionable
recommendations to address gaps and ensure smooth implementation of the
EHR system. This could include updates to infrastructure, training
plans for staff, or adjustments to clinical workflows.
---------------------------------------------------------------------------
The CSR is followed by a period of approximately 12 months of site
preparation, training, etc. before a go-live. Oracle is working with VA
to compress this timeframe so that the time from starting CSRs and
doing pre-deployment work to the day of go-live is shorter. Depending
on how much we can compress this timeline, we anticipate the four sites
in Michigan to go-live sometime in the first half of 2026, hopefully
early 2026.
While we are excited to restart deployment work at these four
sites, we have encouraged VA to accelerate deployment activities and
expand the number of sites to deploy to so that we can more quickly get
this program back on course. Beginning CSR work now for additional
sites is critical so that, assuming the four deployments in Michigan go
well, we have new sites ready to deploy to later in 2026 and a
continuing pipeline of sites ready for deployments.
Acceleration
We are working very closely with VA to determine the best course of
action for accelerating deployments and building a schedule that will
enable this program to successfully deploy the new EHR across all of
VA's healthcare system. Maintaining the current pace will take decades,
which is not acceptable to anyone, and which is why we recommend the
path of acceleration. For acceleration to succeed, there is work we
need to do, and work VA needs to do. We are very encouraged from our
initial meetings with Sec. Collins and VA, as we do this important
planning.
Oracle recognizes there are high levels of skepticism for how a
program that has only deployed to six VA Medical Centers (VAMC) in six
years can rapidly scale up to complete 164 additional deployments in a
timely manner. This skepticism is driven by a largely negative public
narrative about EHRM that has not caught up to the current state of
affairs.
We believe this is due to lack of understanding of the significant
investments Oracle Health is making to accelerate and scale deployments
as well as the technological updates made to the EHR system, and the
significant improvement work done during the Reset. This testimony will
explain those updates and improvements in detail.
Oracle Health is fully committed to this mission, and as part of
our ongoing efforts, we have significantly enhanced and are overhauling
our deployment methods and tools. Leveraging Oracle's deep expertise in
engineering, automation, and scalable technologies, we are implementing
a suite of innovations designed to accelerate deployment timelines,
scale our efforts, and reduce resourcing dependencies. Key initiatives
include:
1. Streamlining Deployment Methodology: We are adopting a national
standard and a robust change management process that will then minimize
deployment activities designed to accommodate site-specific
customization requests. This approach ensures consistency, reduces
complexity, and accelerates deployment by allowing us to replicate
deployments across multiple sites with greater speed and precision.
2. Automating Testing: By investing in advanced testing automation
tools, we reduce manual testing efforts, shorten the testing phase, and
improve reliability. Automation ensures that each deployment is
thoroughly tested in less time, enabling us to support a higher volume
of simultaneous deployments without compromising quality.
3. Automating User Onboarding: Oracle is implementing sophisticated
user onboarding automation processes that streamline user provisioning
and access. By automating these critical tasks, we significantly reduce
manual effort, minimize delays, and ensure faster user access to the
EHR, allowing us to scale deployments with fewer resources.
4. Web-Based Data Collection and Automating EHR Configuration: We
are introducing a web-based tool to collect site-specific data more
efficiently. This tool allows us to gather the necessary information
from VAMCs quickly and more efficiently. Once the data is collected, we
can automate the process of configuring the EHR system to meet the
needs of each site. With our web-based data collection tools and
automation, we will speed up the overall deployment and reduce manual
work.
5. Automating Domain Refreshes: We are automating the process of
refreshing system domains. This means that when updates or changes need
to be made to the system, we no longer require manual intervention,
which can be time-consuming and disruptive. Automation helps keep the
system up to date with less effort, minimizes downtime, and frees up
our resources to focus on other tasks, allowing us to handle more
deployments simultaneously.
6. Leveraging Virtual Training: Oracle is scaling training
capabilities by investing in virtual training methods, including on-
demand learning platforms and interactive virtual environments. This
shift reduces our dependence on instructor-led sessions and enables us
to provide training to larger cohorts of users, significantly
increasing the number of simultaneous deployments we can support while
ensuring that all stakeholders receive timely and effective training.
7. Expanding Partner Ecosystem: To support the growing demands of
our accelerated deployment schedule, Oracle is engaging additional
strategic partners, including large enterprises, to supplement
staffing. These partners bring in specialized expertise and scale,
ensuring that we can maintain high levels of support and execution
across multiple deployment streams without overloading our internal
resources.
Through these strategic investments in automation, process
streamlining, and resource scaling, Oracle is positioning itself to not
only meet the demands of the current deployment schedule but to exceed
expectations by delivering faster, more efficient, and scalable
deployments across the entire VA network.
When the work of the Reset is combined with the successful
deployment of the EHR to Lovell Federal Health Care Center (FHCC) last
year, the completion of the entire Department of Defense (DoD)
deployment, the delivery of pharmacy and other enhancements, and the
technological improvements that we are bringing to the program to
enable more efficient deployments, it is clear we are not in the same
place in 2025 as we were in 2018 or 2019, and certainly not since
Oracle acquired Cerner in June 2022.
Consider the following:
1) The Millennium EHR that will be deployed at new sites is not the
same as was deployed at the first six sites. Oracle has made
investments that have made the EHR more stable, and Oracle has made
extensive updates, enhancements, and simplifications to the EHR.
a. At the current live sites, all these updates have been
incorporated into the EHR, and it is being used more
productively than ever before.
b. For example, during the Reset, more than 3,000 functional
changes have been completed for the EHR.
2) Significant improvements have been accomplished in testing,
training and change management practices.
3) As part of our ``Big Rocks'' projects, we are implementing
enhancements to address key areas for improvement identified by the six
live facilities. These upgrades--focused on pharmacy, referrals, Quick
Orders, and other aspects of the EHR--are designed to improve
productivity, drive standardization, and enhance usability.
4) We have greatly enhanced the process and procedures used for
patient safety.
5) VA, too, has made and is continuing to make significant strides
in its standardization project, decision-making and change management.
Moreover, technology has advanced, allowing Oracle to bring further
improvements to the EHR system.
1) Oracle is moving the EHR to the cloud, providing a base for
highly scalable deployments, making future updates to the system much
easier, and enhancing cybersecurity.
2) We are automating and scaling our deployment and testing
processes, which will reduce the amount of time from a CSR to go-live.
3) Oracle is making significant investments in modernizing the EHR
generally for all our customers. These updates are designed to improve
usability, reduce user burden, and enhance tools that help providers
increase the quality of care and improve patient outcomes.
4) Oracle is using the latest innovations in Artificial
Intelligence (AI) and integrating them into the EHR. When this program
started, nobody was talking about integrating AI into EHR's. The great
benefit to VA of utilizing a commercial provider is that they will be
the recipient of this innovation, and more, as we continue to drive
innovation in healthcare technology.
In this vein, as we drive toward a full deployment, we are working
with VA to bring new innovations to the system in 2025 so that veterans
can begin to see the benefits of the modernization even before their
local VAMC has deployed the new EHR. We intend to work collaboratively
with VA to explore the introduction of our Clinical AI Agent (CAA) to
the current live facilities this year and by early next year to all
VistA sites, as the CAA is EHR-agnostic. CAA is an AI-powered voice
assistant that provides context-aware summaries of a patient's history,
including diagnoses, medications, lab results, and past encounters. It
also provides ambient clinical documentation - capturing and
transcribing patient conversations in real-time, automatically
generating structured clinical notes.
We have offered to speed VA's entry onto our Qualified Health
Information Network (QHIN), which will enable every health system in
the country to share patient data with VA. And we have offered to
expand our Health Data Intelligence Platform (HDIP) from the current
live sites to VistA sites. HDIP integrates data from various sources
across the healthcare continuum, including the EHR, clinical systems,
and third-party data providers, helping to close care gaps by
proactively identifying missed interventions, enabling timely clinical
actions, and supporting better health outcomes for veterans.
With these advances and new offerings, and the work of the Reset,
the EHRM program is ready to accelerate deployments. We recognize of
course there are still opportunities for improvement and work to do -
by Oracle and VA - to optimize the system. This work can be done in
parallel with deployments and should not prevent us from taking
advantage of this current opportunity to move the program forward.
The remainder of this testimony continues in three sections:
Then and Now: Accomplishments
Opportunities Going Forward
Innovation and Modernization for the EHRM Program
Then and Now: Accomplishments
Since Oracle's acquisition of Cerner in June 2022, we have made
significant improvements to the technical performance of the system,
worked with VA to standardize and simplify care workflows, which are
supported by the EHR, enhanced training, worked with the current live
sites to improve productivity and revenue collections, and sped
delivery of system optimizations such as the critical pharmacy
enhancements, among other important updates described below.
Technical and Performance Improvements:
At the time of Oracle's acquisition of Cerner in June 2022, the EHR
system faced criticism for its performance. To this end, Oracle agreed
to new Service Level Agreements with higher financial penalties.
Immediately upon closing the acquisition and obtaining the keys to the
Federal Enclave data center in Kansas City that runs VA, DoD and other
federal users on the new Federal EHR, Oracle began applying our
technical expertise, engineering rigor and stronger resources to
stabilize and improve the system.
Our work has paid off and the EHR system is now reliably running
and consistently available for use without severe outages, incidents or
long pauses or crashes.
Part of ensuring a stable and performant system is strong testing
processes and early engagement with the user community. We have
instituted improved processes in this regard and are incorporating
earlier user testing during the development process, as most recently
demonstrated with Pharmacy 3b. This review process ensures that the
correct levels of testing, including regression testing, are performed
for each change.
Making the engineering changes such a high priority shortly after
the acquisition brought stability, which then enabled VA and Oracle to
focus on usability changes, simplifications and standardization.
Reset Accomplishments and Improving EHR Usability:
We are now nearly two years into the Reset, and while deployment
activities are beginning for new sites, the work of the Reset will
continue to prioritize optimization of the system for users. When VA
and Cerner originally configured the system per direction from VA's
National Councils, the effort at creating one system from 130 different
instances of VistA led to overly complicated workflows and more
``clicks'' and time spent in the EHR than providers were used to under
VistA.
During the Reset, Oracle and VA leadership visited the original
five live sites to understand specific suggestions from end users to
enhance the system. Oracle has made a total of 3,286 functional changes
to the EHR, updated 154 workflows, and added 119 new workflows through
Dec. 2024 based on feedback from users and a comprehensive review
during the Reset. A change is defined as the addition of a new item
that did not previously exist in the EHR or a modification to an
existing item. In addition, many issues identified by users did not
require a change to the EHR because they could be addressed with
additional training, which is ongoing.
This improved training as well as the simplifications and updates
made during the Reset are intended to improve the end user experience.
For example, working with VA we were given permission to make the
following workflow changes from those originally decided by VA's
National Councils in 2018 and 2019.
Acute Physician Track/Emergency: Collaboration between
appropriate VHA and DHA national groups and subject matter experts
(SMEs) to converge on a list of stroke PowerPlans (including
Telestroke) to increase efficiency and accuracy in caring for
critically ill veterans.
Ambulatory Core: Aligning search settings across all
applications provides consistency in fields and supports the end users
in selecting the correct patient, decreasing risk of error.
Oncology: Multiple improvements to Oncology workflow,
ensuring orders are activated on the correct encounter, documentation
is easily accessible and visible to the care team, and facilitating
communication efforts between team members to ensure veterans are
scheduled and medications are available for veterans to receive the
right care at the right time.
Case Management: Clinical Documentation Improvement (CDI)
mPage centralizes chart review to ensure clinical documentation is
complete for charges to be made, easing the burden on end users and
decreases time spent in navigating through various areas of the chart
to find pertinent information.
Women's Health: 16 PowerPlans were revamped to align
existing content with current clinical guidelines aiding providers in
providing consistent, up-to-date and evidence-based care for veterans.
New content for women's health was implemented that includes risk
assessments and standardized documentation for management and planning
using American Society for Colposcopy and Cervical Pathology (ASCCP)
guidelines. This data can now be viewed across the longitudinal record,
improving the efficiency and care coordination for women veterans
across the continuum.
Pharmacy: Automated removal of discontinued or voided
prescriptions from MMR eliminates the need for additional manual steps
as pharmacists work through the e-Rx queue.
In June 2024 Oracle collaborated with VA during five planning
sessions to define thirteen so-called ``Big Rock'' projects that will
help improve the user experience. These projects were selected by VA
and range from the Pharmacy 3b/3c work to improving PowerForms and
Quick Orders to standardization work and creating a new, data-driven
deployment schedule sequencing tool. Many of these projects will
address VA's unique needs.
More than half of these projects are well in progress. Two have
been completed, four will be completed by the end of May 2025 and four
are currently in process.
We are enthused that the Big Rocks work plan will address issues of
high user concern and make a significant difference in operations for
the currently live sites as well as improve adoption at future sites.
We believe this work can be done in parallel with efforts to deploy to
new sites.
In addition, during the Reset and continuing to the present, VA has
made changes as to its governance and standardization processes that
are critical to optimizing the system quickly for VA users. By actively
seeking and incorporating feedback from end users at live sites, VA has
ensured that system improvements align with their needs. The commitment
to ensuring end users have a voice has made sure that the optimization
projects deliver meaningful benefits to both end users and veterans.
Revenue Collections:
Across all live Oracle sites revenue collections activity in Fiscal
Year 2024 was uniquely challenged due to the Feb. 2024 Change
Healthcare cyber-attack, which caused claims processing to be
completely halted at all VA sites for most of the operating period.
Oracle sites were down from February through October while re-enabling
billing activity in VistA was prioritized. Before that, at the start of
Calendar Year 2024, the five live sites had seen improved performance
with collections to target reaching 100.3 percent of target from
January to March 2024.
VA and Oracle worked together to aggressively manage existing
backlogs to ensure the claims processing downtime didn't prevent
collections metrics to dip significantly in FY24. The two CPACs where
Oracle has live sites saw collections to target reach 100.1 percent
(West CPAC) and 94.4 percent (North Central CPAC) respectively in FY24.
That collaboration was further leveraged to facilitate expedited
billing activity in Q1 of FY25 once claims processing was safely
brought back up. Collections to target for the first 3 months of FY25
at Oracle-related CPACs were as follows: Oct: 70.8 percent (claims
processing still partially down), Nov: 442.6 percent, Dec: 347 percent.
Recently deployed optimization projects have allowed for improved
billing turnaround time, allowing for backlogs from FY24 to be managed
effectively to help drive the performance gains referenced earlier.
Productivity:
Productivity, in the context of healthcare, traditionally measures
the volume of patient care delivered within a specific timeframe. It's
a key indicator of how efficiently and effectively a healthcare system
operates. When transitioning to a new EHR system, it's common for
hospitals to experience a temporary dip in productivity. This is due to
a variety of factors, including decreasing non-essential services to
allow for the need for staff to learn new workflows, adapt to new
technology, and adjust to changes in processes. The commercial market
is driven by revenue and as a result clinicians are held accountable
for ramping back up to full productivity within a set timeframe.
Typically, it takes between three to six months for productivity to
return to baseline levels after an EHR conversion.
During the initial rollouts, VA took longer to return to baseline
levels compared to typical commercial clients or DoD. However,
measuring patient loads and productivity under VistA and Millennium has
been challenging. Productivity is measured by workload credit (or RVUs,
which are the national standard for measuring productivity, budgeting,
and expense allocation) in MCA Vera (fed by VistA). When VA originally
configured the workflows for the new Federal EHR, VA did not include
the necessary requirements to measure productivity in the same way that
it was measured in VistA. Oracle has recommended that VA engage on a
workload optimization project since 2022 to ensure this can be
accurately addressed moving forward. As a result, VA and Oracle have
initiated planning sessions to ensure they have the requirements in the
system to ensure comparable measurements.
The duration of time it took VA facilities to attain pre-go-live
productivity levels is longer than the industry average of 3 months.
This attainment of pre-go-live facility level funding can be measured
using Patient Weighted Work (PWW), which is a method used to adjust and
measure the workload or complexity of patient care based on specific
patient characteristics and needs.
Emergency Department (ED) measures are an area that, in many cases,
are exceeding pre-implementation productivity levels due to the
efficiencies the new EHR has provided ED staff, even with patient
volume exceeding pre-implementation levels at all sites. Launch-point,
a streamlined ED tracking board designed to enhance departmental
workflows by offering real-time access to veterans' records and
critical information, has significantly reduced Door-to-Doctor (D2D)
and total length of stay (LOS) and providers' productivity has
increased by 25-30 percent. The ED volume by site as of December 2024
compared to the baseline was:
Mann-Grandstaff: Baseline of 1218 and 1,632 in December
2024
Roseburg: 642 baseline, compared with 845 in December
2024
Columbus: 1,014 baseline compared with 1,576 in December
2024
Lovell FHCC: 1,499 baseline compared to 2,586 in December
2024
Addressing the following items, many of which are covered in the
Big Rocks projects, will enhance the return to productivity following
the implementation of the new EHR system:
1. Referral Management: The decreases in productivity are due to
several factors including insufficient staffing levels and backlogs
that preceded the go-lives. In addition, the referral process is not
standardized across the program and leads to differing requirements
from site to site and poor adoption of the standardized Oracle system.
2. Scheduling/National Baseline: Schedules are controlled during
go-lives to allow additional time for end users to adopt the system. To
return to baseline, the leadership must decide to increase the case
load to pre-implementation levels once the facility has adjusted to the
new EHR. There should be alignment at a national level of what
measurements will be assessed for return to baseline and pre-go-live
baseline must be measured in a consistent manner across facilities.
Return to baseline targets and timelines should be communicated at a
departmental level and sites should be held accountable at an
individual level. Resources are available to provide additional support
and training to users not meeting expected productivity levels.
3. Addressing Organizational Culture and Leadership Engagement:
Site leadership engagement varies from site to site and the lack of
strong leadership present and advocating for the program often reflects
lower adoption and productivity. Sites have reduced schedules during
go-live and no organizational directives are established for when
departments are expected to increase patient visits. Oracle recommends
enhancing training and support by increasing super user involvement--
highly trained individuals who act as subject matter experts (SMEs) and
system champions--during go-lives. This approach ensures hands-on
guidance and knowledge transfer, rather than relying solely on schedule
reductions with no defined timeline for returning to full capacity.
4. User Adoption and Education: Sites must not go live unless
recommended training percentiles are reached, and there needs to be
sufficient support staff through go-live and into sustainment to ensure
end users receive ongoing coaching.
5. Ensure Standard Operating Procedures: VA should ensure local
policies are aligned with national policies prior to implementation.
Examples of areas that have been historically impacted include Rapid
Response, Patient Movement, and Staff Scheduling. Another example,
Medication History and Reconciliation is a task that was not completed
by nurses in the legacy system. In Oracle, the recommendation is for
nurses to complete this task on or before the encounter so that
providers can complete their assessments in a timely manner.
6. Optimize Solutions: Several solutions can be utilized to
optimize the EHR for the unique needs of the VA. ``Big Rock'' projects
are an example of how VA has continued optimization of national
standards to gain efficiency and improve productivity in areas
including Quick Orders, Message Center and Referral Management.
Interoperability:
A key component of the EHRM program is improving interoperability
to ensure seamless care coordination across VA, DoD, and community care
networks. EHRM's interoperability advancements, such as the Joint
Health Information Exchange (JHIE) and Seamless Exchange, enable
automated, real-time data sharing, ensuring clinicians have immediate
access to comprehensive patient histories and that service members have
access to their complete records.
The JHIE is a secure network that shares health information. With
the JHIE, all health providers--whether at a DoD Military Treatment
Facility or from the TRICARE network--can securely access beneficiary
records and health information electronically.
When the EHRM program was envisioned, the focus was making sure
service members have a seamless record when transitioning to care at
VA. The use of the same system at DoD (where they call it MHS Genesis)
and VA (where they use the Cerner commercial name of Millennium)
ensures that the record is seamless.
Of course, with VistA having 130 different instances across the VA
healthcare system, interoperability just within VA is a challenge. EHRM
eliminates this problem by using one standard EHR across the entire VA
system, ensuring that no matter which VAMC a veteran uses across the
country, the veteran's record will be available - and the veteran
should receive the same standard and quality of care.
Since the inception of the EHRM program, interoperability with
commercial care providers for veterans has become critically important.
The VA MISSION Act enables veterans to receive care in their
communities if VA wait times are too long. When veterans go to the
community for care, their medical record needs to go with them and come
back to VA. The EHRM program ensures this is the case, as the Oracle
EHR is interoperable with 90 percent of community care providers.
Seamless Exchange, the use of which is being expanded with Code
Block 12, is a collection of record retrieval and reconciliation
services that collects external health data, compares it to a patient's
chart, and reconciles it for provider review. This allows a provider to
have complete and current data in their workflow for care decisions.
Seamless Exchange will reduce the amount of reconciliation clinicians
need to manually complete to do chart reviews. The addition of all
Seamless Exchange capabilities will help the VA realize additional time
savings in chart review.
Finally, Oracle is offering VA access to its future Qualified
Health Information Network (QHIN) under the federal Trusted Exchange
Framework and Common Agreement (TEFCA) at no additional cost, building
on these successes to provide a secure and innovative approach to
health data exchange and other potential data such as coverage and
benefits information. This will allow every health system across the
nation to share patient data with the VA, safely and securely.
User Satisfaction:
After each major code block update (twice a year in Feb. and Aug.),
user satisfaction is measured. VA conducts these surveys, which are
voluntary, and employees self-select into completing. Given that
deployments have been halted since 2022 (other than Lovell FHCC), the
user-base for surveys has not expanded beyond users at the initial five
sites where dissatisfaction from initial missteps runs high. Despite
these methodological limitations, the usability work and code updates
that have been implemented have shown that clinician and staff
satisfaction with the Federal EHR has increased each year since 2022--
including increases in agreement in employee surveys with the phrases
``the EHR is available when I need it'' and ``this EHR enables me to
deliver high-quality care.'' Oracle strongly believes that providing
our Clinical AI Agent and the many updates made during the Reset, as
well as completing the pharmacy enhancements, will continue to allow
for user satisfaction to improve.
Code Block 12 Upgrade:
On Feb. 21-23, 2025, the Code Block 12 upgrade will be implemented
for the EHR system. The Code Block 12 upgrade includes significant
improvements for the EHR system, including:
Pharmacy 3b/3c: See below section on Pharmacy.
Seamless Exchange: Oracle Health Seamless Exchange
aggregates external health data from multiple sources--such as national
and local exchanges--and deduplicates redundant information to create a
cleansed, comprehensive patient history. Seamless Exchange has been
piloted at the LaGrande clinic and is now being expanded to all ten
Walla Walla connected sites.
HealthShare Referral Manager (HSRM) via OPENLink
IntegrationPlatform (OIP): Rearchitect the current Millennium to HSRM
interface under a new cloud-based platform with key workflow, dataflow,
and process improvements, eliminating the manual workflow steps and
provide the clinicians with access to the data required to complete the
Referral steps in an efficient manner.
Image Viewer - Radiology: Image Viewer offers clinicians
the ability to view Digital Imaging and Communications in Medicine
(DICOM) images when configured with the Millennium Platform data base
to provide referential viewing. Users can also access the web-based
viewer when integrated with other application in CareAware Multimedia
Study Management, to provide referential viewing from various
workflows. This will be replacing SkyVue Distribution Viewer.
Financial Management Systems v8: This is a General Ledger
Accounting System, and the enhancement allows for First Party MCCF
Refunds to be issued to Veterans from any Millennium site.
LifeImage v1: ``Gatekeeper'' users can electronically
receive and send DICOM image studies from Community Providers/Systems,
eliminating the need to use CDs to receive or share images.
Managerial Cost Accounting v2: MCA is responsible for ALL
the cost accounting at FHCC for both DoD and VA. Oracle currently
provides MCA with data extracts on the VA side to support workload
capture, budgeting, various reporting requirements, congressional
inquiries, etc. For FHCC, additional DoD extracts are needed to provide
MCA with DoD data.
Medtronic PaceArt v1: PaceArt is a workflow solution that
compiles and manages patient's cardiac device data. The system
collects, stores, and retrieves data from programmers and remote
monitoring systems from cardiac device manufacturers. PaceArt captures
data from implant, in-clinic checks, and remote transmissions,
patient's data is generated and, it can be stored in the PaceArt data
base to be sent to EHR.
Medtronic PillCam v1: Medtronic PillCam gives the ability
for clinicians to perform capsule endoscopy and with this integration,
send results into Millennium.
Mental Health Suite (MHS) v2: MHS is used for creating,
editing, renewing patient intakes and treatment plans, and enabling
care teams to collaborate to ensure provider compliance with regulatory
guidelines. The MHS software was designed for psychiatrists,
psychologists, social workers, and anyone involved in the creation of
treatment plans for mental health patients.
Siemens Syngo v2: The functionality will provide the
ability for echocardiograms to be interpreted by a cardiologist at a
second EHRM VAMC, after the Echo is ordered, performed, and documented
at an initial EHRM VAMC site.
Pharmacy:
Overall, pharmacy operations in the six sites are stable. While we
recognize pharmacy staffing has increased, we are hopeful that the new
enhancements that have been provided in the recent Code Block release
will enable VA pharmacists to operate more productively going-forward.
The process in the new EHR for pharmacy is different than in VistA, but
it also provides patient safety features that VistA does not have,
which may take additional end user time but improves the standard of
care and safety for veterans. As an example, Medication Management
Retail (MMR - the outpatient pharmacy application) displays the
veteran's lab values within the clinical workflow of the pharmacist
allowing streamlined decision-making around medication dosing.
Seven million prescriptions have been filled using the new EHR at
the first five live sites from the VA Consolidated Mail Outpatient
Pharmacy since October 2020. This is in line with prescription fill
volumes under VistA.
At the Lovell FHCC, pharmacy adoption is going well too. Barcode
scanning (BCMA) for the month of Jan. 2025 at Lovell FHCC was 98.56
percent. This is better than the overall VA average of 97.7 percent and
supports higher patient safety, such as the Opioid Advisor Tool. Dual
beneficiary patients (those with both DoD and VA benefits) are all in
one system now for pharmacy. All prescriptions are on one medication
list regardless of whether VA or DoD is filling them. This provides
patient safety and efficiency benefits. Finally, Lovell FHCC is
continuing to use the conversion to the new Federal EHR to drive
electronic prescribing. In the last 30 days, 99.8 percent of
prescriptions have been authored electronically.
Reflecting back to the original deployment in Spokane, at that time
VA and Cerner had not adequately accounted for the unique ways that VA
operates its pharmacy. In choosing a commercial off-the-shelf EHR,
there should have been a recognition that the commercial market (and
even other government customers) does not operate their pharmacies the
same way as VA and adjustments should have been made earlier.
A typical Oracle Millennium commercial EHR system contains
functionality that enables the ordering of a prescription by the
provider (ordering party); the receiving pharmacy then utilizes its own
software for the dispensing of the medication (dispensing party). Our
original contract with VA for its EHRM program included these standard
capabilities.
However, after the Spokane and later deployments, it became
apparent that the baseline Millennium pharmacy capabilities originally
contracted needed to be enhanced to encompass the level of tight
integration required to meet VA's outpatient pharmacy needs because in
the VA healthcare system, VA is both the ordering party and the
dispensing party.
By the time of Oracle's acquisition of Cerner in June 2022,
frustrations with the pharmacy module of the EHR were very high. In
fact, it was the top concern we heard about in meeting after meeting,
both with VA and with Members of Congress. By August 30, 2022, VA had
put on Task Order the work necessary to customize the pharmacy module
to fit VA's unique needs. Seven key enhancements were contracted for,
and there was a three-year timeline for delivery. Oracle delivered the
first six and a half enhancements in a year and a half by Feb. 2024.
The last enhancement, which is 3b/3c, took longer due to changing
requirements from VA, but is still being delivered in Feb. 2025,
earlier than three years.
The 3b enhancement going in Code Block 12 over Feb. 21-23, 2025, is
an update that will enable VA pharmacists to modify a prescription and
have those edits return to the provider-facing application in the EHR.
Those edits will then flow through subsequent renewals of the
particular prescription. This requirement is unique to VA, and the 3b/
3c enhancement should enable pharmacists to work more productively. We
will of course work with VA to monitor the impact of 3b/3c to see if
any further action is necessary for pharmacy going-forward, but
delivery and implementation of 3b/3c completes the seven enhancements
that Oracle has provided to VA for the pharmacy module of the EHR.
Appendix A provides a summary of the first six enhancements and
other pharmacy updates that have been made, as well as various safety
features.
Training, Adoption and Support Services:
After our acquisition of Cerner in June 2022, it was made clear to
us that the training conducted for Spokane, and even later deployments,
was insufficient. Of course, with the deployment in Spokane, the
training challenges were compounded by deploying during the pandemic
and the associated stresses placed on healthcare providers at that
time.
Oracle invested significant resources in improving the training
program for new users as well as providing continued training for
existing users on the overall system and updates to it. With better
trained users, tickets and the need for support services generally
decrease.
At the live sites, we have conducted various onsite education and
optimization activities to help end users adopt existing workflows and
identify and execute configuration and workflow improvement
opportunities. In addition to observing how end users interact with the
new EHR and resolving their high-priority issues, we proactively
identify end users with poor performance experiences and collaborate
with VA to resolve these issues, including issues that are the
responsibility of VA.
In the last six months we have held over 2,000 training classes for
end users at the live sites. Specifically, we have conducted training
for Oncology and Long-Acting Injections at VISN 20 sites to improve end
user workflow adoption and experience. At White City and Mann-
Grandstaff, we worked with VA to identify and implement optimization
opportunities and educate users on workflows for Optometry. At Jonathan
M. Wainwright and Columbus, we executed similar projects for Audiology.
Oracle also utilizes workflow data to identify end users needing
additional training. Once identified, we send our staff onsite to
partner with the end users, assisting them in streamlining their
workflows, thereby reducing time spent in veteran charts and increasing
the time available to spend with veterans.
For provisioning tickets to give end users access to the system or
to specific roles, we have worked with VA to improve processes. In Jan.
2025, we successfully met the VA's target dates on 98 percent of the
provisioning tickets logged. Performance improvements have been
achieved with the transition to the Microsoft Edge platform, resulting
in a 51 percent reduction in full-page load time for Community Care
Coordination workflows.
Training for the deployment at the Lovell FHCC included new end
user adoption activities (e.g., Departmental Workflow Readiness
sessions and Learning Labs). These activities reinforced formal
training and provided participants with an opportunity to practice
their workflows through simulated scenarios in the VA Sandbox.
Learning Labs were created to bring together end users, with
support from super users (who are highly experienced users), provider
champions and informaticists to develop a comprehensive understanding
of selected respective service line workflows as a cohesive care team.
They were first piloted at Lovell FHCC in December 2023 for a small
group of 54 super users. Post event survey data showed that more than
ninety percent (92.1 percent) of respondents reported at least moderate
improvement in their preparation for go-live, with nearly two thirds
(65.8 percent) reporting great or exceptional improvement.
Based on the overwhelmingly positive feedback, the site and VA
asked to partner with Oracle to expand the use of Learning Labs for end
users prior to go-live. In close partnership with VA, we quickly stood
up an additional 55 sessions for more than 200 end users. Because of
the feedback from end users at Lovell FHCC, Learning Labs will be a key
activity to help future sites prepare to adopt the new system.
More than 70 supplemental training materials were provided to end
users prior to go-live to reinforce important training topics. In
addition to the new end user adoption activities for Lovell FHCC,
Oracle also made 36 early access computer-based training programs
available to super users to support their work and provided
supplementary surge training to dual hat users, pharmacists, and
pharmacy technicians just before go-live. Weeklong sessions were
conducted in November 2023 with the entire pharmacy operations staff,
working through various situations and workflows. Also, knowledge
transfer series led by pharmacists were held. Overall, Lovell FHCC
pharmacy staff provided a 9.82 out of 10 rating for the workflow
adoption training sessions they received.
Lovell FHCC:
The Lovell FHCC deployment is the first and only deployment that
has taken place under Oracle's ownership of Cerner. The previous five
sites that were deployed, were completed by Cerner. Oracle strongly
believes that the work done since the acquisition to improve technical
performance of the system, improve training, and work with VA to
simplify and standardize workflows and make other improvements enables
to us confirm, now that we are nearly one year after the March 2024
deployment, that it was successful.
The system at Lovell FHCC has been well-received by users, enabling
them to provide excellent care to veterans, active-duty military, and
their dependents. Compared to the original five live sites,
improvements in change management, training, and communications led to
notably higher adoption rates at Lovell FHCC. Oracle provided
significant in-person, onsite support for users and worked with Lovell
FHCC on specific areas where their workflows needed to be modified to
adapt to the new standard provided in the EHR.
In many cases, productivity at Lovell FHCC has returned to patient
volumes prior to the deployment. For example, providers are averaging
less than 21 minutes in the EHR, and nurses just under 4 minutes per
patient seen. In the Emergency Department, the pre-deployment baseline
monthly average of patients seen was 1,499, and the current monthly
average volume is far exceeding that at 2,856 patients. Similarly, the
Surgery Department is operating at 110 percent of baseline under the
new EHR.
The complexity of medical care at Lovell FHCC stems from its
comprehensive range of integrated services, making it a strong
predictor of success for larger VAMCs seeking to enhance care
coordination. Lovell FHCC offers a full spectrum of services, including
primary care, specialty care, behavioral health, surgical services,
rehabilitation, and long-term care, mirroring the broad scope of care
required at larger VAMCs. Key aspects that contribute to its complexity
and scalability include:
Integrated Military and Veteran Care: Lovell FHCC serves
both active-duty military and veterans, requiring seamless coordination
between DoD and VA systems, a challenge also present in large VAMCs
with complex patient populations.
Multidisciplinary Specialty Services: Offering
cardiology, orthopedics, neurology, and mental health care in one
facility allows for efficient referrals and comprehensive treatment
planning, a model that can be scaled for larger VAMCs.
Advanced Surgical and Rehabilitation Services: Lovell
FHCC provides both inpatient and outpatient surgical services, along
with post-operative rehabilitation, ensuring continuity of care, an
essential feature for larger VA health systems.
Behavioral Health & Substance Use Treatment: The
integration of mental health services, PTSD treatment, and substance
use programs aligns with the needs of high-risk veteran populations in
large VA hospitals.
The success of these multifaceted services at Lovell FHCC
demonstrates that with adequate resources, technology integration, and
strong care coordination, similar models can be scaled to larger VAMCs
to improve efficiency, patient outcomes, and overall healthcare
delivery.
Patient Safety:
Patient safety is the top priority for Oracle and VA in the
deployment of the new EHR. The EHRM program represents a transformation
in veteran healthcare delivery, aimed at improving care coordination,
reducing medical errors, and ensuring seamless interoperability between
VA, DoD, and community partners.
However, we recognize that concerns have been raised regarding
patient safety, from oversight bodies such as the Inspector General and
the Government Accountability Office (GAO), as well as anecdotal claims
of patient harm. Items such as the Unknown Queue \2\ have left the
misimpression that the Oracle EHR cannot be safely used at VA, even
though it is safely used at DoD and at commercial facilities across the
United States and worldwide.
---------------------------------------------------------------------------
\2\ Shortly after our acquisition of Cerner in June 2022, we were
made aware by media reports and a leaked Inspector General report of a
patient safety concern with the operation of the Unknown Queue (UQ).
Despite its unfortunate name, the UQ was not a bug, it was a backstop
to account for patient scheduling tasks to facilities or providers that
were not recognized by the system or entered incorrectly by the
provider. These scheduling tasks were not lost, rather they were routed
for manual review and processing, but employees were not trained to
monitor it.
By the time of this reporting on the UQ, the awareness and
training issues had been addressed and minimal numbers of orders were
entering the UQ. However, as the new owner we wanted to make certain
that we brought our technical and engineering expertise to the issue to
further reduce the chances of a provider entering an order incorrectly.
On August 1, 2022, we delivered to VA updates that alert a provider
when an order they entered could not be scheduled and requires
correction as well as a similar message to the provider in their
notification center. These alerts continue until the order is corrected
by the provider. These updates were provided by us at no cost to VA.
For the past two years, on average, only one scheduling task per site
per day is routed to the UQ, and it is not an issue of any continuing
concern.
---------------------------------------------------------------------------
In fact, the Oracle EHR is being used safely at VA too, and Oracle
continues its ongoing work to strengthen safety efforts, demonstrating
that the new EHR is not only safe but also continuously improving and
ready for nationwide deployment. The deployment at Lovell FHCC was
conducted with no patient safety incidents. In the time since Oracle's
acquisition of Cerner in June 2022, there has been an 80 percent
decrease in tickets reported as a safety concern by end users.
EHR safety is not a static goal--it requires continuous refinement
based on real-world clinician feedback, safety audits, and system
performance data. Oracle and VA have built a dedicated patient safety
infrastructure to rapidly identify, investigate, and resolve potential
issues. A key component of this approach is the Oracle Health VA
Patient Safety Team, which serves as the subject matter expert body for
patient safety. This team works closely with VA leadership, site-level
clinical teams, and national safety organizations to proactively assess
and mitigate risks.
Key safety initiatives include:
Real-time monitoring and response: A bi-directional data
feed between VA's ServiceNow (SNOW) and Oracle Health's Remedy system
ensures real-time tracking of patient safety tickets.
National safety governance: A structured patient safety
management model ensures stakeholder alignment across clinical,
operational, and financial domains to measure and mitigate risks.
Independent safety audits: As a demonstration of our
commitment to patient safety, Oracle commissioned a 3rd party to
conduct a risk assessment across high-impact clinical workflows
(oncology, emergency medicine, and perioperative care), identifying 74
risk areas--39 of which required configuration updates, 16 code
enhancements, and 19 governance policy changes.
Patient Safety Checkpoints: Oracle developed a structured
program that integrates the patient safety team into the Change Control
process to provide additional support to the government as they
adjudicate and approve the proposed changes. This initiative aims to
proactively assess and mitigate risks associated with proposed changes
in the system before they are acted on, ultimately ensuring the highest
standards of patient safety for the new EHR. Risks are assessed by
utilizing a library a curated patient safety controls, resulting in a
detailed analysis of identified risks and recommendations for the
selected change request.
The Oracle Health VA Patient Safety Team operates a Go-Live Patient
Safety Command Center, which runs 24/7 during deployments to oversee
and coordinate safety reporting and resolution in real time.
During each deployment, a dedicated Patient Safety
Manager is onsite, working alongside VA leadership and Oracle staff to
oversee patient safety risks.
National governance structures codified in VA Task Orders
ensure increased coordination between Oracle, VA safety agencies (NCPS,
IPS, VISN), and frontline clinical staff.
The new EHR has enhanced patient safety measures compared to VistA.
Key improvements include the implementation of medication safety
protocols and closed-loop medication documentation, which ensure
accurate medication administration. The implementation of bedside
barcode scanning has enhanced verification processes by allowing
clinicians to confirm patient identities and medication accuracy at the
point of care, thereby reducing the risk of errors. Furthermore, the
full deployment of automated medication dispensing cabinets streamlines
medication management, providing secure and efficient access to
medications while minimizing the potential for dispensing mistakes.
Together, these advancements create a safer environment for patient
care and improve overall treatment outcomes.
While early implementation challenges were identified, the system
has evolved and improved rapidly. These improvements, supported by
measurable reductions in safety risks, enhanced governance, and
accelerated issue resolution, demonstrate that the new EHR is not only
safe but also positioned for successful national rollout.
Patient Behavioral Health Flags:
The new EHR contains Patient Behavioral Health Flags that assist in
making providers aware of risks like suicide for veteran patients. In
Feb. 2023, Patient Behavioral Health Flags were updated to be included
in the Radiology and Lab components of the EHR. Flags were updated
again in August 2023 to extend their visibility across all users,
including schedulers using the registration function of the EHR. This
addition ensured behavior health flags are visible both upon
registration and between encounters to enhance coordination across a
veteran's entire care team. In 2024 they were extended even further in
the EHR to Pharmacy. Oracle Health continues to partner with the VA to
evaluate opportunities to improve identification and management of
Category 1 Patient Record Flags to enhance end user experience and
veteran safety.
Opportunities Going Forward
Oracle is committed to continual quality improvement and there are
opportunities for enhancement that we continue to partner with the VA
to address, but we do not believe any of these challenges should stand
in the way of proceeding with the four deployments in Michigan and
further accelerating deployments. The EHR system is in a vastly
different, improved place from the system of two to three years ago.
Due to the amount of time it takes from beginning pre-deployment work
to the actual go-live, we have the remainder of this year to continue
improving the system while VA makes needed operational changes.
Cybersecurity and Moving to the Cloud:
We believe there is wide recognition for the need to better secure
our veterans' healthcare data. The Government Accountability Office
(GAO) has repeatedly flagged VistA's decentralized architecture as
outdated and increasingly vulnerable to security threats. Transitioning
to a modern, cloud-based EHR will strengthen data protection, system
resilience, and overall cybersecurity posture, ensuring the integrity
of veterans' health information in an evolving threat landscape.
Within the U.S. Government federal space, Oracle holds a number of
DOD security accreditations and FedRAMP authorizations, and we are an
approved vendor under the Intelligence Community's Commercial Cloud
Enterprise (C2E) program and the DoD's Joint Warfighting Cloud
Capability (JWCC) program.
Oracle Cloud Infrastructure (OCI) was built with its foundation in
scalability and security, which is fully integrated with features such
as bastions for zero trust access, security zones for compartmentalized
workloads and integration of security across the Infrastructure, Data
base and Application Layers.
Moving the Federal Enclave to OCI is underway. The first phase will
be complete later this year, after which we will be able to start
integrating new features into Millennium for VA. One feature we expect
to integrate quickly is our Clinical AI Agent that reduces the need for
providers to spend time in the EHR and enables greater patient
engagement.
We anticipate the full migration of the EHR to the cloud to be
complete next year. Oracle has committed to making this move to OCI at
our expense.
Hosting the EHR on OCI will not only accelerate our ability to
increase capacity as we scale the number of deployments, but also
enable greater stability and reliability as the number of EHR users
grows.
Big Rocks:
As mentioned above, Oracle made several recommendations to VA
through the course of the Reset, including for how VA can: (1)
institute stronger governance controls through clearer escalation paths
for program decisions, such as those requiring cross-council consensus;
(2) enhance change control processes through closed loop communications
with end users and enforcement to standards; (3) standardize workflows
and healthcare protocols, such as referral management, workload
capture, and mammography; (4) improve system performance and
operations; (5) optimize end user engagement and communications; and
(6) advance workflow adoption and optimization.
Akin to our recommendations, VA stood up ten workstreams during the
reset period and onboarded several ``Big Rock'' projects, which are
specific initiatives aimed at improving the user experience,
efficiency, and outcomes. VA's efforts toward standardization,
establishing an effective configuration process, and creating playbooks
to ensure alignment to model workflows will help VA create one standard
of care across its healthcare enterprise and enable VA to provide
quicker answers when deviations from the standard EHR are requested.
Further, several of VA's ``big rock'' projects, such as position
standardization (i.e., ensuring every healthcare worker with the same
job title and responsibilities uses the EHR system in the same way),
referral management, and ad hoc folders (i.e., organizing documentation
that captures patient information in a standardized way), demonstrate a
commitment to achieving standardization across the VA healthcare
system.
Finishing the Big Rock projects is critical to the ability to
accelerate deployments.
Oracle Recommendations for VA:
To accelerate deployments and set up the EHRM program for success,
Oracle has made the following recommendations to VA:
1. Adopt a National Standard: Leverage known configurations that
are working well at DoD and the six live facilities to implement a
single, national EHR standard across all VA facilities. Standardizing
workflows will not only ensure veterans receive consistent, high-
quality care regardless of location but also reduce costs associated
with bespoke configurations at each VAMC. Additionally, a national
standard will streamline implementation efforts, improving deployment
velocity and accelerating the modernization timeline.
2. Advance Optimizations and Limit Third-Party Integrations: VA
should expedite execution of known optimization opportunities (e.g.,
workload credit, service connected / special authority compliance
changes) alongside deployment activity to improve system usability.
Further, VA should reduce reliance on third-party integrations to
simplify workflows, decrease costs and limit cybersecurity exposures.
3. Unlock Innovation and Scale by Migrating to the Cloud:
Transition from the on-premises systems to Oracle Cloud Infrastructure
(OCI) to enhance cybersecurity and system resilience. This will allow
for faster implementations and ability to adopt modern technology today
and in the future.
a. Migration to the cloud is underway. With timely cooperation
from VA and DoD for both infrastructure needs and necessary
government approvals, migration of the Core EHR could be
completed by 2026. This requires alignment to a predictable and
streamlined cybersecurity review for SaaS and cloud-based
technologies across VA and DOD to be quickly established.
4. Adopt Modern Technologies. Oracle Health is poised to deliver
several new technologies to the VA in 2025, including our Qualified
Health Information Network (QHIN), Clinical AI Agent (CAA), Health Data
Intelligence (HDI) Companion App, and Patient Portal. As these
solutions are EHR-agnostic, VA can deploy them across the enterprise
now, even before completing the transition from VistA to the Federal
EHR. This strategic approach will enable VA to immediately benefit from
enhanced interoperability, advanced clinical decision support, and
improved patient engagement--accelerating modernization efforts without
waiting for the full EHR conversion. This will help all providers and
patients benefit from new modern technology. We believe this will also
help with adoption of the new EHR as providers working in facilities
slated for a later deployment will start to understand how the new
technology will assist them.
5. Accelerate Federal EHR Deployments: Based on the current plan
and pace of deploying to four new facilities in VA's Fiscal Year 2026,
full deployment would take over 40 years. Instead, VA should build on
the success of the Lovell FHCC deployment and the optimization work
done during Reset to accelerate the deployment plan. Current State
Reviews (CSRs) should begin for additional sites so that after the
Michigan deployments there is not a time lag before the next
deployments can begin. In addition, VA should use a Veterans Integrated
Service Networks (VISN)-based deployment approach to scale, optimize
resource allocation, and ensure facilities within a network adopt a
common standard simultaneously.
6. Streamline VA Contracting: To address specific needs or
objectives with VA, task orders are awarded to Oracle Health under the
VA EHRM IDIQ contract with defined scope, deliverables and timelines
for a particular service or project(s) to be delivered. To streamline
management and improve efficiency, VA should consolidate the
approximately 35 active task orders. By merging overlapping tasks,
centralizing oversight, and standardizing reporting, the program can
reduce administrative burden and eliminate redundancies. Ultimately,
this will lead to cost savings, increased accountability, and a more
efficient path forward for the EHRM initiative.
Referrals:
One of the Big Rocks is Referral Management:
As part of the Big Rocks project, Oracle implemented
updates to the referral system including a new custom referral form,
functionality to enable sites to send one referral for multiple
diagnoses, added audit history and added functionality so all open and
archived referrals are shown in one comprehensive view.
These enhancements address VA's top requests and were
provided at no cost to the VA, including features like internal closed-
loop referrals.
Deployment Schedule:
The EHRM program has not had a proper deployment schedule since
before the Reset. Oracle is encouraged that VA is working with us to
prepare an integrated master schedule so that we all are working toward
the same goals and timeline.
Innovation and Modernization for the EHRM Program
While VA purchased the Cerner EHR, Oracle has made significant
investments in developing modern health care applications that are
available in OCI. Oracle's strategic investments in our product
offerings are closely tied to OCI. Many of our new products and product
features that we have brought to market, beginning in 2024 and all the
way through 2027, leverage OCI's capabilities. These advances help to
address areas that have proven difficult in the VA adoption of the new
EHR.
By migrating to OCI, VA can access these new state-of-the-art EHR
features, healthcare-specific solutions, advanced analytics, and
interoperability tools that are optimized for this cloud ecosystem.
Importantly, the VA does not need to be fully migrated to OCI to start
benefiting from these innovations--taking the initial step toward OCI
unlocks access to these capabilities, including:
Clinical Artificial Intelligence Agent (CAA) described
above.
Oracle EHR: Oracle is rolling out its new state-of-the-
art, modernized EHR in an iterative fashion, starting with a fully
featured Ambulatory EHR, with subsequent releases for specific
specialties and inpatient care. This will be introduced to our
commercial clients this year. The new EHR is designed to embed AI
across the entire clinical workflow to automate processes, deliver
insights at the point of care, and dramatically simplify appointment
prep, documentation, and follow up for physicians and staff. With
native integrations across a broad range of Oracle applications, the
EHR is also designed to help streamline information exchange between
payers and providers, support patient recruitment for clinical trials,
simplify regulatory compliance, and optimize financial performance.
Patient Administration System (PAS): PAS is focused on
improving the efficiency and effectiveness of patient management tasks,
including registration, scheduling, billing, and overall patient flow.
Patient Portal: The Patient Portal is a digital platform
designed to enhance patient engagement by providing individuals with
secure, easy access to their complete longitudinal health record from
their time in service, at the VA and in the community; and
communication tools with their healthcare providers.
Health Data Intelligence Platform (HDIP): The HDIP is a
comprehensive solution designed to aggregate, analyze, and derive
actionable insights from health data. This platform integrates data
from various sources across the healthcare continuum, including the
EHR, clinical systems, and third-party data providers. HDIP is
currently available at the six live facilities and can be integrated
with VistA, which would help VA close care gaps by proactively
identifying missed interventions, enabling timely clinical actions, and
supporting better health outcomes for veterans.
Oracle's Qualified Health Information Network (QHIN)
described above.
By utilizing a commercial off-the-shelf product, VA has access to
the latest technology, Oracle Health is delivering tangible cost
savings while ensuring VA has the tools needed for a more effective,
efficient, and resilient healthcare system for veterans.
Closing
We believe that when the totality of updates, enhancements and
innovations are considered, the EHR system must be viewed as a
drastically improved system from the system that was originally
deployed in Spokane in 2020.
Oracle is proud to continue working with VA to modernize its EHR
system, and we are confident that the EHRM program is ready to deploy
in Michigan and on an accelerated schedule for additional deployments.
We are steadfast in our mission to serve our nation's veterans
through this project. Thank you and I look forward to answering your
questions.
Appendix A - Pharmacy
Summary of the first six enhancements:
Enhancement 1: Toggle Prescription Synonym Visibility
Implemented Feb. 2023
This guides providers to order prescription/supplies
based on what is formulary and fillable through the VA/Consolidated
Mail Outpatient Pharmacy (CMOP). The intent is to reduce re-work
efforts by Pharmacist/Provider to adjust prescriptions after the
initial order entry.
Shortened the amount of prescriptions a provider sees by
almost 50 percent.
Enhancement 2: Optional Order Stop Date in Retail Med Manager
Implemented Feb. 2023
Keep ongoing medications for a patient on the active
medication list so providers and pharmacists continue to have better
visibility to the medications even when a new prescription is needed.
Maintenance prescriptions will no longer change to a
completed status once the legal date has been met. They must now be
manually completed/ discontinued.
Enhancement 3a: Display Legal Rx Expiration Date in Orders
Implemented Feb. 2023
Prescription Legal Expiration date will now display face
up to users in PowerChart. Prescriptions with 0 (zero) refills
remaining and expiration dates that have passed will display in red
text and will no longer be refillable regardless of refills remaining
but may be considered for renewal.
Visibility for providers when a prescription is no longer
fillable (past legal expiration date), to help identify when a new
prescription is needed.
This enhancement will add the prescription Legal
Expiration date to the Home Medications component and Orders profile to
display within the home medications, the patient's medication list and
Message Center.
Enhancement 4: Support mCDS Discontinue in Retail Med Manager
Implemented Aug. 2023
Reduce the step/clicks for pharmacy staff to discontinue
duplicate prescriptions within the drug interaction checking alerts
window.
Enhancement 5: Enable Orders Renewal Action on Retail Med Manager
Prescriptions
Implemented Feb. 2024
This enhancement allows providers to easily renew and
take other actions on outpatient pharmacy generated prescriptions for
consistency with provider entered prescriptions.
Enhancement 6: Optional Pharmacist Verification for Pharmacy
Technicians Refills
Implemented Aug. 2023
Increase Pharmacist efficiency (save Pharmacist time) by
removing Pharmacist verification requirement.
Enhancement 7: Request Refills from Power Chart to Outpatient
Pharmacy
Implemented Feb. 2024
This enhancement allows providers to perform a ``right
click'' refill action in Power Chart and transmit a refill to the
pharmacy for processing.
In the last two years, additional enhancements beyond these seven
were made as follows:
Task Order 31: Three Drug Image
This enhancement provides drug metadata (round, scored,
color, drug ID, imprint) in outpatient pharmacy workflows so that a
pharmacy user can accurately identify medications.
Task Order 31: Mobile Inventory Scanning
This enhancement helps monitor and control real-time
inventory and reordering processes by assisting with inventory and
reorder level updates through mobile scan-driven workflows during a
single adjustment step.
Task Order 31: E-Rx Monitoring Filling
This enhancement improves the E-Rx Monitor filter.
Electronic prescriptions are processed from VA and non-VA providers in
the E-Rx Monitor. Filtering the monitor allows for pharmacists to
segregate their labor pool daily and have a pharmacist focus on
singular aspects of workflow, for example community care prescriptions
from a non-VA provider.
Task Order 31: Weekly Multum Release
This enhancement increased the release cadence for Multum
content to move from monthly releases to release weekly, allowing for
increased delivery of drug content as it is updated.
Also in Feb. 2023, an upgrade was included outside of the seven
enhancements that quickened the search history for prescriptions in
Medication Manager Retail (MMR). Pharmacists use this search history to
research previously prescribed outpatient prescriptions and associated
activity. This upgrade reduced the average time of 15.3 seconds for a
search to 1 second, making it much easier for pharmacists to use.
Finally, all of the currently live sites benefit from pharmacy
safety features inherent in the Oracle pharmacy module, known as
Medication Manager Retail (MMR), that are not present in VistA:
In MMR, pharmacists can view VA and community care
prescriptions together, in a single provider view.
Pharmacists can see relevant clinical information and lab
values within the pharmacy application during prescription processing.
This capability informs proper prescription dosing without leaving the
order to go to another screen, unlike VistA.
The new EHR allows for improved communication between VA
pharmacists and Consolidated Mail Outpatient Pharmacy (CMOP)
pharmacists checking prescriptions. VistA does not have this capability
which can lead to prescriptions being sent back to the local VAMC for
clarification. This feature has been used nearly 34,000 times in fiscal
year 2025, which saves time in the fulfillment of prescriptions because
it reduces the chances of the CMOP canceling a prescription back to the
local VA facility.
The new EHR allows VA pharmacists to communicate
electronically with community care providers when requesting
prescription renewals. This is another net new capability, and it has
been used by VA pharmacists more than 9,200 times in fiscal year 2025.
This represents 9,200 phone calls not made to community care providers
by VA pharmacists. This enhances continuity of veteran care with
prescription medications and encourages prescriptions staying inside of
VA even when authored via community care.
Pharmacists and providers also benefit from the Opioid Advisor Tool
in the new EHR, which includes enhanced decision support, including
dose range checking alerts and the Opioid Advisor Tool. The Opioid
Advisor Tool allows clinicians to simultaneously check data from 47
state Prescription Drug Monitoring Programs (PDMP) and DoD facilities
to prevent improper prescribing of controlled substances. Previously
clinicians had to leave a patient's record and access PDMP data through
each state's website with different passwords for each site. The Opioid
Advisor tool has guided more than 4,360 modifications to opioid
prescriptions since October 2020. In these instances, the provider made
a different and beneficial clinical decision based on the information
the system provides. This information includes previous overdose
attempts and any history of suicidal ideation. This is a net new
capability that supports safer care of veterans.
______
Prepared Statement of David Case
Chairman Barrett, Ranking Member Budzinski, and Subcommittee
members, thank you for the opportunity to discuss the independent
oversight conducted by the Office of Inspector General (OIG) regarding
the development and deployment of VA's new electronic health record
(EHR) system. Since April 2020, the OIG has released 22 oversight
reports on VA's rollout of the new EHR system that identified critical
missteps and a lack of remediation for identified failings.\1\ Of the
93 recommendations issued to date, 32 have not yet been fully
implemented--with eight open for more than three years. Failure to
satisfactorily complete the corrective actions associated with these
recommendations can increase risks to patient safety and VA's ability
to provide timely, high-caliber care at the new EHR sites. Fully
addressing oversight recommendations could also help minimize
considerable cost escalations and delays in the upcoming
deployments.\2\
---------------------------------------------------------------------------
\1\ OIG reports may be found on the website at All Reports, with
those related to just the new health record system filtered to this
list of EHR reports.
\2\ While the OIG follows up with VA on open recommendations every
90 days, VA program officials can submit evidence of sustained progress
or the completion of corrective actions at any time to facilitate
closing recommendations.
---------------------------------------------------------------------------
The OIG recognizes the enormity and complexity of the work being
carried out by the Electronic Health Record Modernization Integration
Office (EHRM IO) and other VA entities to deploy the new EHR system for
the millions of veterans receiving VA care. In addition, OIG staff have
been engaging with VA personnel for more than five years at the main
EHR deployment sites in Washington, Oregon, Ohio, Illinois, as well as
other support locations, and have observed their unwavering commitment
to prioritizing the care of patients while mitigating implementation
challenges.
The statement that follows emphasizes the need to not only
implement recommendations but sustain change by fully addressing the
underlying problems identified in OIG reports. While some of these
reports reflect work from several years ago, they are still relevant
given their unimplemented recommendations--reflecting deficiencies that
have not been remedied during the nearly two-year-long reset pause and
could affect future deployments. For example, an April 2020 report is
highlighted to demonstrate that VA needs to do more work to ensure its
facilities and leaders receive guidance and resources prior to going
live to minimize impacts on VA provider and other personnel's
productivity as well as veteran safety.\3\ The OIG remains concerned
about the unimplemented recommendations from that April 2020 report
related to ongoing development and deployment operations. Additionally,
while some identified problems from OIG reports have been resolved by
VA, there is the risk that similar or new issues could emerge as the
system is deployed at much larger, more complex medical facilities.
With four facilities in Michigan anticipated to receive the new EHR
system next year, VA will still need to deploy it to over 100 other VA
medical facilities with hundreds of thousands of users. As the recently
concluded reset has led to new and updated system functionalities,
leaders must be prepared to train and retrain staff on the system and
swiftly manage any consequences from these updates that result in
compromised patient care and safety.
---------------------------------------------------------------------------
\3\ VA OIG, Review of Access to Care and Capabilities During VA's
Transition to a New Electronic Health Record System at the Mann-
Grandstaff VA Medical Center in Spokane, Washington, April 27, 2020.
This report focused on the EHR's initial capabilities and the potential
impact on patients' access to cares.
---------------------------------------------------------------------------
Though far from exhaustive, this testimony highlights several OIG
reports with unimplemented recommendations designed to enhance patient
safety and the health care that veterans receive from providers using
the new EHR. Remedying these issues, particularly in appointment
scheduling and pharmacy operations, is foundational to ensuring that
users accept the system and VA can deliver care safely and efficiently.
It also stresses the need for VA to make certain that the system is
stable and can handle future growth without the kind of outages and
service degradations previously experienced. Finally, it spotlights
long-term open recommendations regarding the lack of a master project
schedule as well as undefined infrastructure and deployment costs.
THE OIG DETERMINED UNRESOLVED SCHEDULING PACKAGE PROBLEMS MAY
NEGATIVELY AFFECT PATIENT EXPERIENCES AND FUTURE DEPLOYMENTS
For many veteran patients, their first and often most frequent
experience at a medical facility involves appointment scheduling.
Ensuring a smooth experience with appointment scheduling is a great way
for VHA to build trust with veterans, and giving its staff effective,
modern scheduling software can facilitate a more efficient workforce.
For those reasons, in 2021 and 2022, the OIG reported on difficulties
that employees experienced when using the patient appointment
scheduling package at the Mann-Grandstaff VA Medical Center in Spokane,
Washington, and the VA Central Ohio Healthcare System in Columbus.\4\
Among the findings were that VHA and EHRM-IO did not fully resolve
known limitations in the scheduling system before and after deployment,
leading to reduced effectiveness and increased risk of patient care
delays. Schedulers were forced to develop work-arounds for unresolved
issues and inaccurate data migrated from legacy systems. EHRM-IO
leaders did not provide scheduling staff with adequate chances to
identify limitations in the new scheduling system before
implementation, nor did leaders develop an efficient and transparent
method of handling requests for help and complaints (trouble tickets).
---------------------------------------------------------------------------
\4\ VA OIG, Care Coordination Deficiencies after the New Electronic
Health Record Go-Live at the Mann-Grandstaff VA Medical Center in
Spokane, Washington, March 17, 2022; VA OIG, New Patient Scheduling
System Needs Improvement as VA Expands Its Implementation, November 10,
2021.
---------------------------------------------------------------------------
While these very specific problems have been remediated, similar
problems could reemerge once VA begins large-scale deployments of the
system at new facilities. In 2024, the OIG alerted VA to the potential
that systemic, facility-level scheduling problems may be exacerbated at
larger, more complex VHA medical facilities.\5\ Among the issues raised
were the need for additional staffing and overtime, displaced
appointment queue functionality (described below), challenges related
to providers and schedulers sharing information, inaccurate patient
information, difficulties changing appointment types, and the inability
to automatically mail appointment reminder letters. Consequently, at
future go-live facilities, assessing staffing levels and overtime usage
before deployment and preparing staff with approved workflow best
practices may help to reduce employee stress and facilitate successful
adoption of the system.
---------------------------------------------------------------------------
\5\ VA OIG, Scheduling Challenges Within the New Electronic Health
Record May Affect Future Sites, March 21, 2024.
---------------------------------------------------------------------------
The Displaced Appointment Queue's Issues Can Impede Rescheduling
According to a March 2024 OIG publication, schedulers using the new
EHR are experiencing difficulties with what is termed ``the displaced
appointment queue,'' which at times resulted in patients not getting
rescheduled.\6\ That queue is used by scheduling staff to identify
appointments needing to be rescheduled if a healthcare provider has a
schedule change. Staff reported that the new EHR does not always route
appointments to the queue and that properly routed appointments
sometimes disappeared from the queue. EHRM-IO told the OIG it was aware
of the defects in the operation of the displaced appointment queue and
that updates in 2024 were intended to address them. EHRM-IO staff
stated that medical facilities received guidance informing schedulers
how to reschedule patients without using the queue and that the queue
was intended as a safety net. However, the OIG reported that the
defects in the operation of the displaced appointment queue made it an
unreliable safety net. The OIG oversight team could not definitively
identify how many patients were affected. However, because the problems
were not resolved, they could be amplified at larger VHA facilities,
given more staff will have more schedule changes requiring
rescheduling.\7\
---------------------------------------------------------------------------
\6\ VA OIG, Scheduling Challenges Within the New Electronic Health
Record May Affect Future Sites.
\7\ A separate issue affecting schedulers and care providers is
that they cannot easily share information about appointments, such as
notes explaining why an appointment was canceled, which was a function
in the legacy EHR system.
Some Previously Documented Scheduling Inefficiencies and Errors Persist
---------------------------------------------------------------------------
in Changing Appointment Types and Sending Appointment Reminders
OIG field work revealed that schedulers still face ongoing
difficulties in changing appointment types, and barriers remain in
automatically mailing appointment reminder letters. These deficiencies
may have been exacerbated by staffs' perception that training was
inadequate.
Changing Appointment Types. The OIG's 2021 scheduling report found
that VHA and EHRM-IO had not resolved many of the system and process
weaknesses identified by pre-implementation assessments and
workshops.\8\ One system weakness identified was the new EHR's
inability to change the appointment type (face-to-face, VA Video
Connect, or telehealth) for an existing appointment without canceling
the appointment and reordering a new appointment.\9\ This process
inevitably led to more burdensome work for schedulers and providers. In
2022, VHA planned on fixing this issue through a system update and was
finalizing guidance for schedulers' mitigation strategies. However, in
2024, schedulers from all five new EHR facilities confirmed that they
still need to cancel existing appointments and manually create new ones
when changing the type of appointment. Using manual processes could
have a much more significant impact at larger medical facilities.
---------------------------------------------------------------------------
\8\ VA OIG, New Patient Scheduling System Needs Improvement as VA
Expands Its Implementation.
\9\ VA OIG, Scheduling Challenges Within the New Electronic Health
Record May Affect Future Sites.
Appointment Reminders. Another weakness the OIG previously
identified in its November 2021 review was that the new EHR could not
automatically send reminder letters to patients for upcoming
appointments. While not required, veterans were accustomed to and
relied on these letters from the legacy EHR. The letters also reduced
``no shows'' and missed appointments.\10\ The automated mailing of
reminder letters is not a function within the new EHR, and during the
OIG's March 2024 review, the team determined that this system
limitation still existed. In November 2023, EHRM-IO had planned to
release an interface that would allow schedulers to automatically
generate the letters; however, the OIG understands the interface had
not been deployed as of February 2025 at the Captain James A. Lovell
Federal Health Care Center in North Chicago, Illinois, following that
facility's March 2024 implementation of the new EHR.
---------------------------------------------------------------------------
\10\ VA OIG, New Patient Scheduling System Needs Improvement as VA
Expands Its Implementation.
---------------------------------------------------------------------------
As of March 2024, facilities that continued to mail appointment
reminder letters had to manually print and mail them to patients, a
time-consuming process for staff. One facility with the new EHR is
estimated to have manually printed and mailed nearly 195,000 reminder
letters for its appointments in fiscal year 2023.\11\ The three
facilities using the new EHR have undertaken different solutions to
address this system limitation. Some now rely instead on other methods,
such as autogenerated text messages and emails or phone calls. VA
should expedite the release of the interface to all medical facilities
that are or will use the new EHR for those facilities wishing to
continue mail reminders, whether alone or in addition to other options.
---------------------------------------------------------------------------
\11\ VA OIG, Scheduling Challenges Within the New Electronic Health
Record May Affect Future Sites.
Training. The OIG team in March 2024 also identified issues with
schedulers feeling that training provided by Oracle Health was
inadequate.\12\ Some schedulers at new EHR sites rely on their own
local practices and guidance to supplement that given by Oracle Health,
and VA has provided facilities feedback on the supplemental training.
However, some of the facilities' locally developed work-arounds do not
adhere to VA's approved scheduling workflow processes, which can
contradict VA processes meant to standardize scheduling processes.
---------------------------------------------------------------------------
\12\ VA OIG, Scheduling Challenges Within the New Electronic Health
Record May Affect Future Sites.
---------------------------------------------------------------------------
The New EHR's Scheduling Errors May Have Contributed to a Patient Death
The OIG confirmed in a March 2024 report that a system error in the
new EHR resulted in staff's failure to complete the minimally required
scheduling efforts following a patient's missed mental health
appointment.\13\ While a letter was sent and calls were made on the day
of the missed appointment, staff did not complete the telephone calls
on separate days as directed. The OIG found that the patient's missed
appointment, although updated in the new EHR to no-show status, was not
routed to a ``request queue.'' As a result, schedulers were not
prompted by the system to conduct the mandated rescheduling efforts
meant to maximize opportunities to engage patients and not let them
slip through the cracks. The OIG concluded that the lack of follow-up
contact may have contributed to the patient's disengagement from mental
health treatment and, ultimately, the patient's substance use relapse
and death.
---------------------------------------------------------------------------
\13\ VA OIG, Scheduling Error of the New Electronic Health Record
and Inadequate Mental Health Care at the VA Central Ohio Healthcare
System in Columbus Contributed to a Patient Death, March 21, 2024.
---------------------------------------------------------------------------
On a larger scale, the OIG found that VHA was requiring mental
health staff at new EHR sites to make fewer attempts to contact no-show
patients than at legacy EHR sites. The standard operating procedure for
minimum scheduling efforts establishes a different standard of care
based on which EHR system is in use at a facility, which could result
in disparities affecting veterans' access to care. Scheduling is a
foundational element of any system that is designed to provide patients
with timely access to quality care. Yet the recommendation to the then
deputy secretary to monitor the new EHR's scheduling functionality, as
well as the recommendation directed to the then under secretary for
health to evaluate minimum scheduling effort requirements, are not yet
fully implemented. In sum, the new EHR's operation does not comply with
VHA's appointment scheduling policy.
THE NEW EHR HAS LONGSTANDING, UNRESOLVED PHARMACY-RELATED PATIENT
SAFETY ISSUES
In May 2021, after VA's first deployment of the new EHR at the
Mann-Grandstaff VA Medical Center, a pharmacy patient safety team under
the VA National Center for Patient Safety (NCPS) identified pharmacy-
related patient safety issues and staff concerns regarding the system's
usability. For example, updates to a patient's active medication list
were not routinely reflected at the patient's next appointment. The OIG
found that, despite being aware of users' ongoing challenges in 2021,
VA leaders elected to deploy the new EHR at four more VA medical
centers.\14\ Following subsequent deployment of the new EHR in April
2022 (more than a year later) at the VA Central Ohio Healthcare System
in Columbus, the OIG determined that patient safety and usability
issues identified by NCPS were still a factor in many of Columbus's
pharmacy-related patient safety incident reports.\15\ Although Oracle
Health and VA have since resolved some of those issues, the OIG remains
concerned, as described below, that the new EHR will continue to be
deployed at larger, more complex medical facilities before resolving
myriad known issues that remain related to prescribing medications and
medication safety.\16\
---------------------------------------------------------------------------
\14\ VA OIG, Electronic Health Record Modernization Caused
Pharmacy-Related Patient Safety Issues Nationally and at the VA Central
Ohio Healthcare System in Columbus, March 21, 2024.
\15\ VA OIG, Electronic Health Record Modernization Caused
Pharmacy-Related Patient Safety Issues Nationally and at the VA Central
Ohio Healthcare System in Columbus.
\16\ Oracle acquired Cerner in June 2022, changing the name of the
entity to Oracle Cerner and again to Oracle Health. The statement uses
Oracle Health for readability, while some events occurred under prior
business names.
---------------------------------------------------------------------------
Software Coding Errors Created Patient Safety Issues
EHR information is communicated between VHA facilities through
different channels, including the Joint Longitudinal Viewer (JLV) and
the Health Data Repository (HDR).\17\ For patients who receive care at
a legacy-EHR medical facility, the JLV application allows healthcare
providers to access a ``read only'' version of a patient's medical
record from both the legacy EHR, Veterans Health Information Systems
and Technology Architecture (VistA), and the new EHR.\18\ The HDR is a
data base that stores patients' clinical information, including
medications and allergies, creating a common repository of information
from both VistA and the new EHR.\19\ Every medication used in VHA is
assigned a distinct number, a VA Unique Identifier (VUID). The accuracy
and completeness of VUIDs and medication allergy information contained
in these systems is critical to supporting individual patient treatment
decisions.
---------------------------------------------------------------------------
\17\ JLV is a read-only web-based application for viewing patient
electronic health records from VA and community partners through a
customizable interface. JLV plays an important role in VA's transition
to the new EHR, as it allows users to see EHR data at other sites
regardless of the system in place. Because veterans are eligible to
receive health care at any VA facility, providers at all facilities
need accurate medication information. When a patient is prescribed a
medication at a new EHR site, that medication's unique identifier is
sent to the HDR. If that same patient seeks care from a facility
provider using the old system, and this provider enters a medication
order, a system software interface from the old system accesses the
medication's VA Unique Identifier from the HDR data base to perform a
safety check. This process, which relies on the accuracy of the
information in the HDR, verifies the medication being prescribed is
safe and compatible with any medications and allergies previously
documented in the patient's record.
\18\ The OIG uses the term ``legacy EHR'' to refer to Veterans
Health Information Systems and Technology Architecture (VistA), the
system used prior to the Oracle Health EHR product.
\19\ Va.gov, VistA Monograph, July 18, 2023. The VA Health Data
Repository (HDR) is ``a national, clinical data storehouse that
supports integrated, computable and/or viewable access to the patient's
longitudinal health record.''
---------------------------------------------------------------------------
A 2024 OIG report affirmed that an error in Oracle Health's
software coding resulted in the widespread transmission of incorrect
VUIDs from new EHR sites to legacy EHR sites.\20\ The OIG learned these
unique identifiers became inaccurate during their transmission to the
HDR when fills for certain prescriptions were processed through the
VHA's Consolidated Mail Outpatient Pharmacy (referred to as the mail
order pharmacy).\21\ In short, this error, now rectified, created the
potential for medication-related patient safety issues for patients
from any new EHR site who also received care at a legacy EHR site.
---------------------------------------------------------------------------
\20\ VA OIG, Electronic Health Record Modernization Caused
Pharmacy-Related Patient Safety Issues Nationally and at the VA Central
Ohio Healthcare System in Columbus.
\21\ The Consolidated Mail Outpatient Pharmacy is a centralized
automated pharmacy system comprised of seven pharmacies that provide
mail order medications to VHA patients. The OIG did not find any errors
on the part of mail order pharmacy staff or operations, and patients
received their correct medications.
Medication-Related Data Transmission Issues Pose Threats to Patient
---------------------------------------------------------------------------
Safety
The OIG learned that research into the cause of the VUID error led
to the discovery of other problems associated with the transmission of
medication and allergy information from the new EHR to the HDR. On June
15, 2023, the NCPS alerted VHA staff to data transmission issues and
errors, including missing, duplicate, or incorrect medication and
allergy information being transmitted. The consequences of inaccurate
medication information transmission to the HDR include
patients' medications that have been discontinued or
stopped by new EHR-site providers appear in the legacy EHR as active
and current prescriptions;
allergy warning messages not appearing when intended or
inappropriately appearing for the wrong medication;
duplicate medication order checks not appearing when
intended or inappropriately appearing for the wrong drug; and
patients' active medication lists having incomplete or
inaccurate information, such as missing prescriptions, duplicate
prescriptions, or incorrect medication order statuses.
VHA staff were told to remain aware that legacy EHR sites may have
inaccurate medication information for patients treated at both legacy
and new EHR sites. An EHRM-IO data leader told the OIG that EHRM-IO and
Oracle Health's original testing focused on data transmission from the
new EHR to the HDR, but no entity verified the data's accuracy when
accessed by legacy EHR users. Within the June 15 NCPS patient safety
alert, a series of mitigations were described to be employed by
frontline clinical staff at all legacy EHR sites and required that all
legacy EHR site leaders have medical providers perform these multistep
manual medication safety checks when prescribing new drugs for all
patients who had received care at a new EHR site at any time. These
manual safety checks are complex, time-consuming, and rely on the
vigilance of patients, pharmacists, and frontline staff.
Further, at the time of the June 15 notice, VHA could not determine
which patients were at risk of a patient safety event from the data
transmission errors, and therefore determined that all patients who had
been prescribed any medications at a new EHR site or had medication
allergies documented at a new EHR site were ``at risk.'' Per VHA data,
as of September 2023, approximately 190,000 patients had a medication
prescribed and 126,000 patients had an allergy documented at a new EHR
site. Approximately 68,000 patients were in both groups, totaling about
250,000 unique patients.\22\ In response to an OIG recommendation, VHA
has notified all patients affected by inaccurate medication data
transmitted to the HDR and informed them of the potential risk of harm
due to possible inaccuracies of their medication and allergy
information within the new EHR. However, patients have been advised to
bring their medications to each VHA visit so their providers have an
accurate inventory of current medications.
---------------------------------------------------------------------------
\22\ VA OIG, Electronic Health Record Modernization Caused
Pharmacy-Related Patient Safety Issues Nationally and at the VA Central
Ohio Healthcare System in Columbus. The data represent the most recent
update received by OIG from VHA of the number of unique patients who
have had any medication prescribed or any allergy documented at a new
EHR through September 29, 2023.
---------------------------------------------------------------------------
Despite these efforts, the OIG remains concerned that patients
served by a new EHR site who also receive care at a legacy EHR site may
still be prescribed contraindicated medications and that healthcare
providers at legacy sites are making clinical decisions based on
inaccurate data. For example, during the review of the HDR issues
described above, the OIG learned of a new EHR site patient with
posttraumatic stress disorder and traumatic brain injury with adrenal
insufficiency whose care was negatively influenced by inaccurate
medication data transfer from their new EHR site to the HDR,
contributing to the patient not being prescribed a critical lifesaving
therapy on admission to a residential rehabilitation treatment program
at a legacy EHR site.\23\ The legacy EHR site pharmacist's data from
the prescribing new EHR site did not include the patient's most recent
prednisone prescription. The patient realized they needed prednisone
after they began exhibiting unusual behaviors, but the nurse said there
was no prednisone on the patient's medication list. Eventually, the
patient was transferred to a local emergency room for care and
prednisone treatment was re-initiated. This example shows the
difficulty with completing numerous, accurate manual reconciliations,
particularly for patients with impaired cognition.
---------------------------------------------------------------------------
\23\ VA OIG, Electronic Health Record Modernization Caused
Pharmacy-Related Patient Safety Issues Nationally and at the VA Central
Ohio Healthcare System in Columbus.
---------------------------------------------------------------------------
The OIG continues to review VHA's efforts to comply with the
recommendation that they ensure legacy-site-EHR providers are aware of
mitigations needed for patients previously treated at a new EHR site,
as well as their efforts to monitor compliance with those mitigations.
The New EHR's Negative Effect on VHA Pharmacy Staff
The OIG determined that Columbus's chief of pharmacy prepared for
challenges during the system transition, such as the pharmacy staff's
increased workload due to the new EHR's operational inefficiencies. One
mitigation was to hire nine full-time clinical pharmacists, which
represented a 62 percent staffing increase, in order to reduce the
backlog and maintain timely prescription processing needs following the
April 2022 deployment.\24\
---------------------------------------------------------------------------
\24\ VA OIG, Electronic Health Record Modernization Caused
Pharmacy-Related Patient Safety Issues Nationally and at the VA Central
Ohio Healthcare System in Columbus.
---------------------------------------------------------------------------
A VHA leader stated that challenges with the new EHR's usability
also led to the creation of dozens of national and facility-level work-
arounds and educational materials for pharmacy personnel. Facility
pharmacy leaders also developed approximately 25 educational materials,
such as tip sheets, reference guides, and job aids, to further support
Columbus pharmacy staff. The OIG is concerned that the continued use of
numerous work-arounds and educational materials is overwhelming for
pharmacy staff to implement and may give rise to inconsistent
practices, which increase risks to patient safety. In addition, the new
EHR's usability issues contributed to staff stress about making errors
that could result in patient harm--concerns linked to pharmacy staff
burnout, low morale, and decreased job satisfaction. The OIG found that
following implementation of the new EHR, burnout symptoms for pharmacy
staff increased and the Best Places to Work score for pharmacy staff
decreased from the previous fiscal year.\25\ VHA pharmacy and patient
safety leaders told the OIG of a need for increased staff vigilance to
avoid patient harm. OIG oversight personnel believe this increased
vigilance is unsustainable by pharmacists and staff responsible for
clinical decision-making and may lead to increases in burnout and
medication-related patient safety events.
---------------------------------------------------------------------------
\25\ The OIG compared 2021 and 2022 facility All Employee Survey
(AES) results. A Columbus leader informed the OIG that VA launched the
2022 AES on June 6, 2022, 37 days after the new EHR's implementation at
the facility. ``Best Places to Work'' is a summary measure produced by
the Partnership for Public Service and is a weighted average of job and
organization satisfaction and likelihood to recommend VA as a good
place to work.
---------------------------------------------------------------------------
The OIG's recommendations from this March 2024 report to ensure
pharmacist staffing levels are addressed and to evaluate and remediate
the various technical and functional issues resulting from all these
work-arounds remain open as of February 20, 2025.
MAJOR PERFORMANCE INCIDENTS MAY HINDER USER ADOPTION AND TIMELY, COST-
EFFICIENT FUTURE DEPLOYMENTS
While the above reports describe the impacts of failings with the
system's programming and functions, the OIG is also concerned that the
contract between VA and Oracle Health and VA procedures do not have
sufficient controls to prevent, respond to, and mitigate the impact of
major performance incidents affecting the new EHR.\26\ Since 2020, the
system experienced hundreds of major performance incidents affecting
the medical facilities where the system was initially deployed. VA
lacked adequate controls to prevent system changes from causing major
incidents, to respond to those incidents uniformly and thoroughly, or
to mitigate their impact by providing standard procedures for outages
and interoperable downtime equipment. Further, although major
performance incidents can delay care to veterans, VA had no formal
process to link reports of these delays to these incidents. Ultimately,
the weak controls for handling major incidents stemmed from the
original May 2018 contract.
---------------------------------------------------------------------------
\26\ VA OIG, VA Needs to Strengthen Controls to Address Electronic
Health Record System Major Performance Incidents, September 23, 2024.
---------------------------------------------------------------------------
In May 2023, VA modified the contract with Oracle Health to
strengthen some requirements for addressing major incidents, but more
work is needed. These new requirements include a metric that outlines
monthly target percentages for the system to be free of incidents other
than outages, an increase in the target monthly uptime for the system,
and strengthened requirements for financial credits when problems have
not resolved within established time frames. Reliable system
performance and preventing incidents from happening were some of VA's
primary reset goals.
The OIG made nine recommendations, including for real-time data-
sharing related to potential problems in system operations,
prioritizing major performance incident response in a clear and
consistent manner, developing and enforcing response and other
performance metrics to hold the contractor accountable, requiring
sufficient detail in post-resolution reports, raising staff awareness
of procedures, acquiring appropriate backup systems for downtime, and
better identifying and addressing major performance incidents linked to
negative patient outcomes. The nine recommendations are currently open,
and the OIG has begun the follow-up process to assess VA's progress in
implementation.
Although the OIG recognizes VA's improving system reliability,
there are still incidents occurring every month. Separate from these
requirements, in August 2023, VA contracted with Oracle Health to
obtain a downtime viewer to provide an additional tool for clinicians
when the system is unavailable. Still, VA has opportunities to make
future contract changes that could help improve its management of major
incidents. Oracle Health could share real-time EHR incident data to
provide VA with greater awareness and enable quicker oversight action.
Detailed incident reporting would also help VA determine root causes
and prevent similar incidents from occurring. This is particularly
important as the new EHR system may be stressed by deployment in larger
and more complex medical facilities. Indeed, the planned 2026
deployments in Michigan will include, for the first time, simultaneous
deployments to complexity level 1 facilities.\27\
---------------------------------------------------------------------------
\27\ VA, Diffusion Marketplace, accessed February 18, 2025.
INCOMPLETE INFRASTRUCTURE ASSESSMENTS AND COST ESTIMATES CANNOT BE
---------------------------------------------------------------------------
REMEDIED WITHOUT A RELIABLE, HIGH-QUALITY PROJECT SCHEDULE
The OIG's oversight, which began before the system's initial
deployment at the Mann-Grandstaff VA Medical Center, focused on the
condition of VA's physical and information technology (IT)
infrastructure before system deployment. Two 2021 reports (published in
May and July) resulted from audits that examined cost estimates for
needed physical and IT-related infrastructure upgrades nationwide. For
the new EHR system to operate as intended, VHA facilities need these
infrastructure upgrades, but they are generally funded from different
sources. Because the life-cycle cost estimates for infrastructure
upgrades did not account for costs from all VA components' budgets,
some estimated costs were not included in mandated reports to Congress
from 2018 and 2020.\28\ Transparent and reliable cost estimates are
critical for Congress to make informed budgeting decisions. VA senior
leaders also depend on these cost estimates to plan program budgets,
approve acquisitions, and monitor program execution. The OIG determined
that both the existing physical and IT infrastructures were inadequate
for the new system at initial deployment sites. Pertinent life-cycle
cost estimates for infrastructure upgrades were also unreliable and
likely underreported by approximately $5 billion. However, these cost
estimates will not be reliable if VA does not develop and maintain an
integrated master schedule projecting the detailed activities needed to
bring the new EHR to its facilities.
---------------------------------------------------------------------------
\28\ VA OIG, Deficiencies in Reporting Reliable Physical
Infrastructure Cost Estimates for the Electronic Health Record
Modernization Program, May 25, 2021.
---------------------------------------------------------------------------
VA Has Not Developed a Reliable Schedule Enabling Deployment Planning
The OIG's 2022 audit of the EHRM program's master schedule found VA
lacked a reliable integrated master schedule consistent with their
adopted scheduling standards, which increased the risk of missing
milestones and delaying the delivery of the system.\29\ At the time of
publication, the OIG estimated that schedule delays could result in
about $1.95 billion in cost overruns per year and would undermine VA's
other modernization efforts on supply chain and financial management
systems. Given various inflationary pressures and the two-year pause on
deployments, that figure may understate the impact of cost overruns.
---------------------------------------------------------------------------
\29\ VA OIG, The Electronic Health Record Modernization Program Did
Not Fully Meet the Standards for a High-Quality, Reliable Schedule,
April 25, 2022.
---------------------------------------------------------------------------
To implement the program successfully within any proposed
timeframe, it is imperative that VA develop a reliable integrated
master schedule. GAO guidance, which the EHRM program office adopted in
its internal plans, states that a high-quality, reliable schedule
should be comprehensive, credible, well-constructed, and
controlled.\30\ This schedule is designed to cover the entire required
scope of work needed to successfully complete the program from start to
finish, including both government and contractor work. It is intended
to provide VA personnel with a road map to completion, track progress,
help identify potential problems and track their resolutions, and
promote accountability for assigned tasks. Further, it will help
determine more precisely the sum of financial resources Congress must
provide for project completion.
---------------------------------------------------------------------------
\30\ GAO, Schedule Assessment Guide, GAO-16-89G, December 2015.
---------------------------------------------------------------------------
Simply put, VA never completed a baseline schedule or an overall
schedule that fully integrated individual portions of the project. The
audit team found known tasks were not reflected on schedules and
longer-term actions had not been scheduled. Given the approach VA was
using in planning when the OIG completed this audit, VA would not have
a high-quality, reliable integrated master schedule until it starts
deploying the system to the very last facilities. While there may be
precise scheduling items for a facility that are not set until closer
to the actual deployment, there are many tasks and interdependencies
that VA can plan for currently.
Moreover, VA could not have relied upon any scheduling effort it
had conducted, since it did not engage in a risk analysis, which shows
how events would impact the likelihood the schedule could be met. Given
VA's announced intention to deploy the new EHR to larger facilities
next year, it must have contingency plans given that VA has experienced
numerous unexpected problems with the new EHR's functionality and
sustained drops in productivity requiring ongoing, resource-intensive
mitigations at new EHR sites.
Four of the OIG's six recommendations to EHRM-IO remain open. At
the time of this hearing, VA has not yet complied with internal
guidance by developing an integrated master schedule that meets
standards and makes certain that activities from all relevant VA
entities are included in the schedule. The department has also failed
to implement procedures for performing schedule risk analyses and make
contract language and program office plans (or other guidance)
consistent to confirm the appropriate individual is responsible for
developing and maintaining the integrated master schedule.
The two reports on physical and IT infrastructure recommended that
VA obtain an independent cost estimate for the EHR program's life-cycle
costs, which VA received from the Institute for Defense Analyses. In
September 2022, the Institute reported a cycle cost estimate of $49.8
billion, including $32.7 billion for a 13-year implementation phase
(including a three-year schedule adjustment reflecting the low
likelihood the new EHR would be deployed system-wide on schedule), and
the rest for sustainment.\31\ While the OIG has now closed those
recommendations related to obtaining the estimate, VA's stakeholders
should recognize that the Institute's cost estimates are not likely to
still be reliable given the delays and system changes of the
intervening years. Additionally, the Institute itself acknowledged that
its estimate did not have a high degree of certainty given the many
risks and uncertainties in the deployment schedule.
---------------------------------------------------------------------------
\31\ GAO, ELECTRONIC HEALTH RECORD MODERNIZATION: VA Needs to
Address Change Management Challenges, User Satisfaction, and System
Issues, March 15, 2023.
Deficiencies in Reporting Reliable Physical Infrastructure Cost
---------------------------------------------------------------------------
Estimates for the EHRM Program
This audit examined whether VHA's cost estimates met VA standards
and were comprehensive, well documented, accurate, and credible, and
whether they were reported to Congress. Some VHA medical facilities may
require significant physical infrastructure upgrades, such as
electrical work, cabling, heating, ventilation, and cooling to
successfully deploy the new EHR system.
VHA and the then Office of Electronic Health Record Management
(OEHRM) shared responsibilities for estimating and reporting physical
infrastructure upgrade costs.\32\ VHA developed the physical
infrastructure upgrade cost estimates, while OEHRM was responsible for
reporting all program life-cycle cost estimates to Congress in
accordance with the Veterans Benefits and Transition Act of 2018.\33\
It required quarterly reporting on the EHRM program's status, including
annual and life-cycle cost estimates and defined the program as any
activities to procure or implement the new EHR system. In early 2019,
VA's Office of General Counsel determined that physical infrastructure
upgrades must be funded from accounts specifically available for
construction-type purposes, such as VHA's nonrecurring maintenance and
minor construction funds. Given the pause of the program for the last
two years and the lack of a more specific deployment schedule, VA has
not yet produced evidence of sufficient progress to close the OIG
recommendation that VA disclose accurate costs for physical
infrastructure upgrades in program life-cycle cost estimates to
Congress.
---------------------------------------------------------------------------
\32\ In 2021, VA transitioned EHRM program management from the
Office of Electronic Health Record Modernization (OEHRM) to the EHRM
Integration Office (EHRM IO). EHRM IO has responsibility for all
recommendations originally assigned to OEHRM.
\33\ Public Law 115-407.
---------------------------------------------------------------------------
The EHRM Program Did Not Have Reliable IT Infrastructure Cost Estimates
In 2021, VA estimated the total program cost of $16.1 billion would
include $4.3 billion in IT infrastructure upgrades.\34\ Like the work
on physical infrastructure, the related IT infrastructure audit
examined whether OEHRM-developed cost estimates from 2018 and 2020 were
well-documented, comprehensive, credible, and accurate, and whether
OEHRM reported to Congress all IT infrastructure upgrade costs,
including future technology updates. The audit team found VA did not
include costs for critical program-related IT infrastructure upgrades
in the estimates reported to Congress during that period, effectively
underreporting program cost estimates by nearly $2.5 billion for IT
infrastructure upgrades that VA's Office of Information and Technology
(OIT) and VHA were expected to fund.\35\ Also like the physical
infrastructure costs, OEHRM officials stated they felt the omitted
costs were outside their scope of responsibility, but neither OIT nor
VHA reported these costs to Congress, despite VA and Government
Accountability Office (GAO) guidance requiring life-cycle cost
estimates to include all costs, regardless of source. The costs should
have been disclosed by OEHRM. VA did make changes to projected costs
starting in the November 2021 report to Congress, but because VA was
still developing the independent cost estimate, there was no certainty
the updates were reliable. Without all critical IT infrastructure
upgrade costs accurately presented, Congress lacks the comprehensive
picture of total program costs needed to make informed oversight and
investment decisions. Subsequently, based on the OIG's audit, VA's
reporting requirements were updated by the VA Electronic Health Record
Transparency Act of 2021.\36\ As of February 21, 2025, the
recommendations ensuring that (1) cost estimates align with VA policy,
(2) VA maintains full and complete cost accounting, and (3) complete
and updated costs are conveyed transparently to Congress remain open
because VA has not been able to identify all of the program's costs.
---------------------------------------------------------------------------
\34\ VA OIG, Unreliable Information Technology Infrastructure Cost
Estimates for the Electronic Health Record Modernization Program, July
7, 2021.
\35\ OIT is expected to fund some upgrades for the local area
network, end-user devices, phones, and Wi-Fi, while VHA is expected to
fund upgrades mostly for medical devices.
\36\ Public Law 117-154.
Facility and Staff Deployment Support Have Unidentified Costs and Risks
---------------------------------------------------------------------------
That May Continue in Future Deployments
Each VA medical facility that has deployed the new EHR has
experienced sustained drops in productivity and throughput.\37\
According to VA, its facilities have used strategies like temporary
staffing, increased use of community care, and finding efficiencies in
operations.\38\ The challenges associated with mitigating the
productivity drops at facilities during the training periods and after
deployment will be magnified at the larger, more complex deployments in
2026. In April 2020, before the first deployment, the OIG called on VA
to evaluate the impact on productivity during a deployment and provide
facility leaders with operational guidance and required resources.\39\
The recommendation remains open, as VA has not yet made sufficient
progress on this effort. The same is true of the OIG's recommendation
that VA minimize the number of mitigation strategies that facility
staff must employ to deal with decreased capabilities during the
deployment. Increased hiring efforts, temporary staff, and community
care utilization all have significant financial impacts that VHA
facilities must navigate. As discussed previously, the number of
additional pharmacy staff handling manual pharmacy operations at new
EHR sites has increased VA's payroll. There is likely to be an even
greater financial impact as VA may need to hire thousands of employees
to mitigate the drops of productivity at future deployment sites.
Without having finalized plans to deal with these issues, VA cannot
reasonably estimate deployment costs in addition to physical and IT
infrastructure needs.
---------------------------------------------------------------------------
\37\ VA OIG, Electronic Health Record Modernization Caused
Pharmacy-Related Patient Safety Issues Nationally and at the VA Central
Ohio Healthcare System in Columbus.
\38\ VA OIG, Review of Access to Care and Capabilities During VA's
Transition to a New Electronic Health Record System at the Mann-
Grandstaff VA Medical Center in Spokane, Washington.
\39\ VA OIG, Review of Access to Care and Capabilities During VA's
Transition to a New Electronic Health Record System at the Mann-
Grandstaff VA Medical Center in Spokane, Washington.
---------------------------------------------------------------------------
Other OIG reports have discussed issues impacting user acceptance
of the system from inadequate or insufficient training on the new EHR
to concerns with the process for resolving problems and requesting
assistance through ``tickets.'' \40\ These challenges impaired the
ability of contractor support staff to address end users' problems, led
to end users' disengagement, and increased patient safety risks. The
OIG generally found that EHR usability problems, training deficits,
interoperability, the need for post-go-live fixes and refinements, and
problem-resolution process challenges complicated VHA providers' work.
While these specific issues have been resolved, VA should be sure it
carefully monitors these points during future deployments. Further, as
the reset has led to new and updated system functionality, leaders must
be prepared to swiftly manage any consequences from these updates that
result in compromised patient care and safety. As the changes employed
during the reset are made to the entire new EHR system, this increased
alertness is vital not only at newly deploying sites but at all sites
that have deployed the new EHR to date.
---------------------------------------------------------------------------
\40\ VA OIG, Ticket Process Concerns and Underlying Factors
Contributing to Deficiencies after the New Electronic Health Record Go-
Live at the Mann-Grandstaff VA Medical Center in Spokane, Washington,
March 17, 2022; VA OIG, Training Deficiencies with VA's New Electronic
Health Record System at the Mann-Grandstaff VA Medical Center in
Spokane, Washington, July 8, 2021.
---------------------------------------------------------------------------
CONCLUSION
The OIG will continue to conduct rigorous oversight of VA's efforts
as it restarts deployments of the new EHR system with a keen focus on
patient care and safety, VA staff's ability to efficiently do their
jobs, and making the most effective use of taxpayer dollars. The OIG is
committed to providing impactful and practical recommendations that
flow from its oversight work to help VA efficiently deploy the new EHR
in a manner that improves veterans' safety, care, and experiences. As
our reports and testimonies over the last five years demonstrate, the
OIG has identified significant, unresolved deficiencies that have
thwarted progress on the new EHR and have contributed to patient harms.
It is incumbent on EHRM-IO, VHA and facility leaders, VA leaders,
and Oracle Health to ensure they are providing full transparency in
their communications with the veteran community, frontline VA staff,
and Congress. Effective program management is critical to the
successful deployment of the new EHR. Accountability established
through clear roles and responsibilities, meaningful metrics, and close
oversight with transparent reporting and swift remediation of any
identified issues will all need to be firmly integrated into future
efforts. Failures in any of these areas chance cascading problems that
put patients at risk, make it more difficult for VA personnel to do
their jobs, and perpetuate cost overruns and delays. Chairman Barrett,
this concludes my statement. I would be happy to answer any questions
you or other members may have.
Prepared Statement of Carol Harris
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]
[all]