[Senate Hearing 114-670]
[From the U.S. Government Publishing Office]
S. Hrg. 114-670
ACHIEVING THE PROMISE OF HEALTH INFORMATION TECHNOLOGY: INFORMATION
BLOCKING AND POTENTIAL SOLUTIONS
=======================================================================
HEARING
OF THE
COMMITTEE ON HEALTH, EDUCATION,
LABOR, AND PENSIONS
UNITED STATES SENATE
ONE HUNDRED FOURTEENTH CONGRESS
FIRST SESSION
ON
EXAMINING ACHIEVING THE PROMISE OF HEALTH INFORMATION TECHNOLOGY,
FOCUSING ON INFORMATION BLOCKING AND POTENTIAL SOLUTIONS
__________
JULY 23, 2015
__________
Printed for the use of the Committee on Health, Education, Labor, and
Pensions
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
Available via the World Wide Web: http://www.gpo.gov/fdsys/
U.S. GOVERNMENT PUBLISHING OFFICE
95-737 PDF WASHINGTON : 2017
----------------------------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Publishing Office,
http://bookstore.gpo.gov. For more information, contact the GPO Customer Contact Center,
U.S. Government Publishing Office. Phone 202-512-1800, or 866-512-1800 (toll-free).
E-mail, [email protected].
COMMITTEE ON HEALTH, EDUCATION, LABOR, AND PENSIONS
LAMAR ALEXANDER, Tennessee, Chairman
MICHAEL B. ENZI, Wyoming PATTY MURRAY, Washington
RICHARD BURR, North Carolina BARBARA A. MIKULSKI, Maryland
JOHNNY ISAKSON, Georgia BERNARD SANDERS (I), Vermont
RAND PAUL, Kentucky ROBERT P. CASEY, JR., Pennsylvania
SUSAN COLLINS, Maine AL FRANKEN, Minnesota
LISA MURKOWSKI, Alaska MICHAEL F. BENNET, Colorado
MARK KIRK, Illinois SHELDON WHITEHOUSE, Rhode Island
TIM SCOTT, South Carolina TAMMY BALDWIN, Wisconsin
ORRIN G. HATCH, Utah CHRISTOPHER S. MURPHY, Connecticut
PAT ROBERTS, Kansas ELIZABETH WARREN, Massachusetts
BILL CASSIDY, M.D., Louisiana
David P. Cleary, Republican Staff Director
Evan Schatz, Minority Staff Director
John Righter, Minority Deputy Staff Director
(ii)
C O N T E N T S
__________
STATEMENTS
THURSDAY, JULY 23, 2015
Page
Committee Members
Alexander, Hon. Lamar, Chairman, Committee on Health, Education,
Labor, and Pensions, opening statement......................... 1
Murray, Hon. Patty, a U.S. Senator from the State of Washington.. 3
Cassidy, Hon. Bill, a U.S. Senator from the State of Louisiana... 28
Franken, Hon. Al, a U.S. Senator from the State of Minnesota..... 30
Enzi, Hon. Michael B., a U.S. Senator from the State of Wyoming.. 32
Baldwin, Hon. Tammy, a U.S. Senator from the State of Wisconsin.. 34
Bennet, Hon. Michael F., a U.S. Senator from the State of
Colorado....................................................... 36
Whitehouse, Hon. Sheldon, a U.S. Senator from the State of Rhode
Island......................................................... 38
Witnesses
Kendrick, David C., M.D., MPH, Chair, Department of Medical
Informatics, University of Oklahoma, CEO, MyHealth Access
Network, Tulsa, OK............................................. 5
Prepared statement........................................... 7
Mirro, Michael J., M.D., FACC, FAHA, FACP, Past Chair, Medical
Informatics Committee, American College of Cardiology, and
Chief Academic/Research Officer, Parkview Mirro Center for
Research and Innovation, Fort Wayne, IN........................ 13
Prepared statement........................................... 15
Kibbe, David C., M.D., MBA, President and CEO, DirectTrust,
Senior Adviser, American Academy of Family Physicians,
Washington, DC................................................. 17
Prepared statement........................................... 19
Black, Paul M., MBA, President, Chief Executive Officer and
Director, Allscripts, Chicago, IL.............................. 22
Prepared statement........................................... 23
ADDITIONAL MATERIAL
Statements, articles, publications, letters, etc.:
American College of Cardiology, Heart House, Letter.......... 50
Response by David C. Kendrick, M.D., MPH, to questions of:
Senator Alexander........................................ 53
Senator Murray........................................... 54
Senator Enzi............................................. 55
Senator Casey............................................ 57
Senator Baldwin.......................................... 58
Senator Cassidy.......................................... 58
(iii)
ACHIEVING THE PROMISE OF HEALTH INFORMATION TECHNOLOGY: INFORMATION
BLOCKING AND POTENTIAL SOLUTIONS
----------
THURSDAY, JULY 23, 2015
U.S. Senate,
Committee on Health, Education, Labor, and Pensions,
Washington, DC.
The committee met, pursuant to notice, at 10:05 a.m., in
room SD-430, Dirksen Senate Office building, Hon. Lamar
Alexander, chairman of the committee, presiding.
Present: Senators Alexander, Enzi, Cassidy, Murray, Casey,
Franken, Bennet, Whitehouse, and Baldwin.
Opening Statement of Senator Alexander
The Chairman. Senator Murray will be here, but she has
asked me to go ahead and start. Welcome. This is a hearing of
the Committee on Health, Education, Labor, and Pensions.
Senator Murray and I will each have an opening statement.
Then we will have our witness testimony. We will ask you to
summarize your remarks in 5 minutes. Then we will have
questions.
I need to leave at 10:30 a.m. for just a few minutes to go
vote in an Appropriations hearing, and Senator Enzi will
preside during that time. Then I will be right back. I don't
want to miss anything from our witnesses. I read your
testimonies, very helpful.
Today's hearing is on information blocking. What does that
mean? Here is an example.
If I suddenly found myself at the Vanderbilt University
Medical Center emergency room and the doctors there wanted to
get my paperwork from the hospital and doctors that I usually
use, information blocking means there is some obstacle getting
in the way of my personal health information being sent to the
Vanderbilt University emergency room.
This could happen in a variety of ways. No. 1, could be
that my usual hospital refuses to share my information. No. 2,
the electronic systems at both hospitals don't talk to each
other. That could happen. No. 3, my usual hospital says it will
charge Vanderbilt a huge fee to send my electronic records. No.
4, my usual hospital might say it can't share them for privacy
reasons. That could be our fault here in the Federal
Government. No. 5, or my usual hospital won't send them because
they cite concerns about data security.
There could be a lot of reasons why I am sitting there in
the Vanderbilt University emergency room, and the doctors can't
get my medical records.
Since 2009, American taxpayers have spent $30 billion to
encourage doctors in hospitals to install electronic health
record systems through incentive payments to Medicare and
Medicaid providers. One of the chief goals was to improve the
electronic exchange of a patient's health information between
physicians' offices and urgent care centers and hospitals and
pharmacies.
By encouraging these practitioners to adopt these systems
and then encourage the manufacturers of the system to make them
able to communicate with one another, we call that
interoperability. That way, for example, I could have my
physician send my records easily to a specialist who could have
my entire medical history right in front of her before I even
had my first appointment. No more printing of papers that I
have to go to pick up and bring to my appointment, have added
to a paper file, and then reentered into the specialist's
computer.
The Federal Government even went so far as to certify
certain systems so physicians and hospitals generally expected
that the certified systems would enable them to transfer
records. But interoperability, this communication among systems
that is so critical, has been difficult to achieve. Information
blocking is one obstacle to interoperability, and I am
interested to hear today from the witnesses the extent to which
it is a problem, the extent to which the Government may share
the blame and the extent to which the Government can encourage
interoperability.
It could loosely be defined as intentionally interfering
with access to my personal electronic health information. Here
is what Reid Blackwelder, a family physician in Kingsport, TN,
told the New York Times earlier this year,
``We have electronic records at our clinic. The
hospital, which I can see from my window, has a
separate system from a different vendor. The two don't
communicate. When I admit patients to the hospital, I
have to print out my notes and send a copy to the
hospital so they can be incorporated into the
hospital's electronic records.''
Last year, I worked with the Appropriations Committee to
acquire a report from the Office of the National Coordinator on
information blocking. We received the report from Dr. DeSalvo
and her team, who did a good job describing the various kinds
of information blocking. The report described the sorts of
information blocking and the scenarios I outlined earlier.
It concluded that successful strategies to prevent
information blocking will likely require congressional
intervention, but this is not the only view of the practice.
Some view information blocking as rational, competitive
practices by for-profit businesses in a competitive healthcare
industry. In other words, why would a hospital or physician
network make it easy for a patient to go out of their network
when that is against their business interest, these persons
might say.
Senator Murray and I have created working groups within our
committee to deal with the various issues on electronic medical
records. We have also created a good working relationship with
Secretary Burwell, Secretary of HHS. In fact, we had breakfast
with her this morning and discussed this subject some.
Let me just give my own view about this. Based upon what I
have heard from doctors, hospitals, administrators, meaningful
use Stage 1 helped encourage adoption. Meaningful use 2 was a
mixed blessing. Meaningful use 3 is a whole other kettle of
fish.
The head of Mayo Clinic came to see me recently and said
that Mayo, which is unique in some ways, is budgeted spending
$1 billion over the next 4 years with 500 employees working on
it in order to put in place the requirements of meaningful use
Stage 3.
Another well-respected hospital told me that meaningful use
1 and 2 worked OK, but they were terrified by meaningful use 3.
My instinct is to say to Secretary Burwell let us not go
backward on electronic healthcare records, but let us not
impose on physicians and hospitals a system that doesn't work
and which they spend most of their time dreading.
Half the doctors are now paying penalties rather than
participate in electronic medical records. We want something
that physicians and hospitals buy into that help patients
rather than something that they dread.
One of the questions I will be interested in hearing the
witnesses talk about is the extent to which we might want to
slow down the implementation of Stage 3 of electronic
healthcare records, not with the idea of backing up on it, but
with the idea of saying let us get this right before we set 500
employees and $1 billion to work at the Mayo Clinic
implementing a system that is not right and that we then have
to change in 2 or 3 or 4 years, or that we impose on physicians
and doctors such an onerous system that we get a huge backlash
in 2 or 3 years and waste the effort that we have made today.
I took a little longer than I normally would in saying
that, but this is our fourth hearing on electronic health
records. We have got another one scheduled for September and
then another one in October. We intend to make whatever results
a part of our medical innovation legislation, which we hope to
finish by the end of the year.
This is a serious subject, both for Senator Murray and for
me and for Democratic and Republican members of the committee.
We are looking for advice about the right thing to do.
Thank you.
Senator Murray.
Opening Statement of Senator Murray
Senator Murray. Thank you very much, Chairman Alexander.
Thank you to all of our witnesses and colleagues who are
here with us today.
I believe strongly that when it comes to our country's
healthcare system, we need to keep moving forward, and that
means building on the progress made so far to continue
expanding coverage to our families, making healthcare more
affordable, and improving the quality of care patients receive.
This is, as the chairman said, our fourth hearing on
strengthening our country's health IT infrastructure, an issue
that is especially important to two of those goals--improving
the quality and the affordability of care. When patients and
providers have more convenient access to better health
information, they are more equipped to make truly life-changing
or even lifesaving decisions.
Hospitals and providers have made substantial progress over
the last few years toward expanding the use of health
information technology. Today, for example, 6 out of 10
hospitals report having used electronic health records to
coordinate with at least one provider outside their own
organization. That is a 51 percent increase from just 7 years
ago.
We have taken some important initial steps forward, but we
need to build on that progress and tackle a lot more challenges
to make sure that patients can get better, more efficient care
that reflects their medical histories. I am glad today that we
are focusing on one challenge in particular, the practice of
deliberate information blocking by some health IT organizations
which threatens to get in the way of the progress we need to
make to continue to improve our healthcare system.
It is important to make clear that there are some
legitimate reasons that a vendor might limit exchange of health
information--patient privacy or unanticipated technological
challenges. According to a recent report by the Office of
National Coordinator for Health Information Technology, there
is substantial evidence that some organizations are
intentionally setting up barriers between their systems and
other systems or overcharging or creating technical or legal
barriers to providers who want to access information through
the system they purchased, or both.
To me, these efforts to knowingly interfere with access to
patients' health information is completely unacceptable. We
have heard before in this committee how big a difference
medical records can make for patients and their families, and
because a strong national health IT system depends on different
electronic healthcare record systems being able to talk to each
other, we can't afford to have bad actors who prioritize their
bottom line over patients' best interests and block information
that hospitals, providers, and patients need to be able to
share with one another.
Instead, I am hopeful that members of this committee can
work together on ideas that would allow physicians to vote with
their feet when it comes to health IT and enable patients to
download and provide their medical histories in a standard,
easily transferable way.
Today I am going to be very interested in hearing from our
witnesses about these ideas and others, which would help move
us toward even more significant progress to strengthen health
IT and improve healthcare for the families and communities that
we all serve.
Thank you again to our witnesses. We look forward to your
testimony today.
Thank you, Mr. Chairman, for really focusing on a critical
topic.
The Chairman. Thanks, Senator Murray.
First, we will hear from Dr. David Kendrick, chair of the
Department of Medical Informatics at the University of Oklahoma
and CEO of MyHealth Access Network in Tulsa. MyHealth Access
Network connects more than 1,600 providers and patients through
an innovative, community-based health information system.
Dr. Michael Mirro is our second witness. He is the chief
academic and research officer at the Parkview Mirro Center for
Research and Innovation in Fort Wayne, IN, and is a national
leader in health information technology, serving as chair of
the American College of Cardiology Health IT Committee for 5
years.
Next we will hear from Dr. David Kibbe. He is the founding
president and chief executive officer of DirectTrust, a
nonprofit organization that helps and protects the rules and
regulations of those in the nationwide Health Information
Network. He is the senior adviser to the Alliance for Health IT
Innovation at the American Academy of Family Physicians.
And finally, Mr. Paul Black is here. He is president, chief
executive officer, and director of Allscripts Healthcare
Solutions, Inc., a leading electronic health record vendor. I
am told that Allscripts is now able to work with all major
electronic health record applications in today's market. Mr.
Black served as the EHR vendor Cerner's chief operating officer
prior to that.
Welcome. We look forward to your testimony. If you would
summarize your remarks in 5 minutes, then we will have a
conversation.
Dr. Kendrick, let us begin with you.
STATEMENT OF DAVID C. KENDRICK, M.D., MPH, CHAIR, DEPARTMENT OF
MEDICAL INFORMATICS, UNIVERSITY OF OKLAHOMA; CEO, MYHEALTH
ACCESS NETWORK, TULSA, OK
Dr. Kendrick. All right. Thank you.
Good morning, Chairman Alexander, Ranking Member Murray,
and distinguished members of the committee.
In previous hearings, committee members have asked how do
you define interoperability? From my perspective, every patient
deserves to have their complete longitudinal medical record
securely available wherever and whenever decisions are made
about their health.
Wherever includes places like the doctor's office or the
emergency room, between doctor visits when medications are
refilled, as well as at home with the patient or caregiver. I
think this definition is helpful because it reminds us that the
primary objective of interoperability is to better serve the
patient with high-quality, safe, and cost-effective care.
We have taken these principles to heart in Oklahoma and
have leveraged our ONC Beacon Community Award to create
statewide interoperability. MyHealth is a nonprofit that serves
as an entity most trusted by payers and providers to measure
quality, cost, and value in support of new healthcare payment
and delivery models.
This unique, trusted, third-party arrangement has
accelerated the adoption of value-based payment models in
Oklahoma, and interoperability has enabled providers to succeed
in these new care delivery models. For example, MyHealth serves
as the convening organization and data aggregator for CMMI's
Comprehensive Primary Care Initiative demonstration project in
which 265 Oklahoma primary care providers reduced Medicare
costs by 7 percent in 1 year.
These accomplishments have been hard fought, requiring more
than 5 years and $15 million to produce. By far, the most
significant barrier to success has been liberating accurate
patient data from practices, hospitals, and other
organizations. Generally, we think of data blocking as the
intentional interruption or prevention of interoperability by
one of two parties, the provider or the provider's EHR vendor.
Provider data blocking may have been an important challenge
early in the development of MyHealth, but it has quickly
receded as value-based payment models take hold. Thus, the
biggest challenge we face is liberating patient data from EHR
systems to make it interoperable.
While many EHR vendors work well with our customers and
with our organization to establish interoperability, we still
have so many specific experiences with inappropriate data
blocking and substandard data quality that we have created a
nomenclature to classify six common types. Below, we describe
one or more real examples of each type.
Type 1 is the golden rule. By far, the most common barrier
to interoperability is the high price charged by vendors to
implement and maintain interfaces, which commonly exceeds
$10,000, and it is not uncommon to see charges of $30,000 to
$40,000 per practice, regardless of practice size, to do
something that meaningful use supposedly required.
Type 2 we call the Hotel California. There is a component
of the EHR certification program called data portability that
is intended to help providers change EHR vendors if they like.
Unfortunately, few vendors appear to offer this functionality
as intended, and so we say that the customers of these vendors
have a Hotel California problem. They can check out other EHR
products anytime they like, but their data can never leave.
Type 3 is inexplicable. We recently were told by a large
EHR vendor's project manager that, ``We just don't do CCDs,''
which is the patient summary of care document, ``because
they're not in our DNA.''
The CCD is required of all certified EHRs, and no amount of
arguing thus far has been able to convince this project manager
that his vendor could do this.
Type 4 we call garbage in and garbage out, and this simply
relates to the quality of the data we are getting in these
standard CCDA files. We are finding uncoded lab results,
uncoded medications. When the data is not coded properly, it
can't be duplicated and essentially becomes useless for other
providers to use in care.
Type 5 we call EHR at the center of the universe.
Increasingly, we are hearing from large health systems that
their EHR vendor provides all of the interoperability that they
need without the community HIE. This is because the vendor
provides convenient connections to other systems that use their
product.
However, providers in critical services using other EHR
products are excluded, leaving a big hole in the
interoperability of the community and the health system. Most
concerning about this mistaken belief among health system
leaders is that it subverts interoperability at the community
and State level, creating instead a corporate EHR network for
interoperability which is not subject to the trust arrangements
and policies of the community.
Type 6 we call the bait and hidden switch. In this type of
data blocking, the vendor achieves certification with one
feature set and then either hides or eliminates functionality
when the EHR is actually deployed in practice. By filing a
complaint with ONC's surveillance program, we recently
discovered that one of the Nation's largest ambulatory EHR
vendors had actually hidden functionality that they used to get
certified and were instead charging tens of thousands of
dollars to practices for an additional product to do the same
thing.
We have since used this knowledge to connect many practices
at no cost. Unfortunately, the vendor in question still has not
been compelled to communicate this information to all of their
customers.
We have also observed a certain fear among providers that
filing complaints with ONC against their vendor will produce a
backlash in the form of poor service or increased fees. Most of
the time our participants ask that we, the health information
exchange, file the complaint to preserve their relationships
with their vendors.
My recommendations to improve the flow of data for
interoperability are, No. 1, expand HIT certification to test
interoperability in the field. Don't just test it on the bench,
but also test it once it is deployed to prove that it works.
No. 2, further emphasize the ONC certification surveillance
program. No policy has ever been successful without at least
the threat of enforcement.
No. 3, consider implementation of a lemon law for EHRs to
protect providers and other organizations stuck in long-term
contracts with vendors who do not perform.
No. 4, require certain clauses of EHR contracts to be
transparent, especially the interoperability clauses that can
hide these data blocking issues.
No. 5, continue to expand value-based payment models, which
inherently encourage interoperability among providers.
And No. 6, although it may sound controversial coming from
a board-certified medical informaticist, we don't need any new
standards right now. We need to make the ones we have work. We
should continue research and development, but please don't
layer in new standards.
HITECH called upon ONC to establish the governance for the
nationwide Health Information Network, but perceived
limitations on ONC's authority have left this entire segment of
our critical national infrastructure without governance. This
vacuum is being filled by various consortia and collaborations
of vendors and large provider organizations.
ONC's authority should be made clear in three groups.
The Chairman. If you could wind it up, Dr. Kendrick.
Dr. Kendrick. Sure. Those who receive care, provide care,
and pay for care should be organized into governance.
[The prepared statement of Dr. Kendrick follows:]
Prepared Statement of David C. Kendrick, M.D., MPH
Good morning, Chairman Alexander, Ranking Member Murray, and
distinguished members of the committee. I am pleased to offer you my
testimony on a topic that has consumed nearly all of my professional
life--the modernization of our healthcare delivery system, and lately,
the information blocking that is getting in the way of progress.
I reviewed the previous healthcare IT testimony to this committee
and noted that a very important question has consistently been asked by
committee members: ``How do you define interoperability?''
My definition of interoperability is that every patient deserves to
have their complete, longitudinal medical record available wherever and
whenever decisions are made about their health. ``Wherever'' includes
places like the doctor's office or the emergency room, between doctor
visits when medications are refilled, as well as at home with the
patient or caregiver and on smartphone when the patient travels.
I think this definition is helpful because it reminds us that the
primary objective of interoperability is to better serve the patient.
It also provides us with a convenient test for any health IT
implementation or policy. ``Does my EHR ensure that my patients have
their records from my clinic available no matter where they seek
care?'' is a fundamental question all doctors should ask about their
systems.
Another important test is ensuring that each patient, and
population, is receiving high value care--the primary ingredient
required for successful value-based payment models. In order to do
this, quality and cost measures must contemplate ALL of the relevant
data on each patient. Since the average patient sees more than 3
different doctors each year, and the average Medicare patient sees 7,
this almost always means aggregating data from multiple doctors and
hospitals in order to get an accurate picture of individual and
population health. To do otherwise would be like assessing the record
of a football team based only on the performance of the left tackle.
We in Oklahoma have been hard at work transforming our healthcare
delivery system. MyHealth Access Network is a non-profit health
information exchange organization serving more than 4 million patients
and connecting 275 organizations, including doctors, hospitals,
pharmacies, tribal health systems, payers, employers, home health,
hospice, long term care, State and local agencies, and many others.
MyHealth received a Beacon Community Award from ONC in 2010 which
encouraged and enabled us to build the Oklahoma approach to
interoperability.
MyHealth has established effective health information exchange with
dozens of EHR's, administrative systems, and payer claims systems, and
consolidates this data into a single record for each patient to ensure
that their complete medical record is available wherever, and whenever
needed for care. In addition, MyHealth serves as the entity most
trusted by payers and providers to measure quality, cost, and value in
support of new healthcare payment and delivery models. This unique
Trusted Third Party arrangement has accelerated the adoption of value-
based payment models in Oklahoma, and has enabled providers to succeed
in these new care delivery models.
For example, under the Beacon program, MyHealth demonstrated
significant improvements in critical ACO success measures: preventable
admissions and ER visits for asthma, COPD, and congestive heart
failure. MyHealth technology has been shown to improve care transitions
by reducing wait times for access to specialty care by 2/3 and
significantly reducing the total cost of care for transitioned patients
in the Medicaid population.
MyHealth has also served as the Convening organization and data
aggregator for the Oklahoma implementation of a CMMI pilot project
called the Comprehensive Primary Care initiative (CPC), which includes
local commercial payers as well as Medicare and Medicaid. These
multiple payers have partnered to implement a value-based payment and
practice transformation program in primary care practices. In the first
year of the program, Oklahoma's 65 CPC practices reduced Medicare costs
by 7 percent, prompting Secretary Burwell to seek commentary on the
potential expansion and permanent implementation of the CPC model in
the latest CMS Physician Fee Schedule.
These accomplishments have been hard fought, requiring more than 5
years and $15M to produce. By far, the most significant barrier to
success has been liberating accurate patient data from practices,
hospitals and other organizations.
Generally, we think of data blocking as the intentional
interruption or prevention of interoperability by one of two parties:
the Provider or the Provider's EHR vendor.
Provider data blocking may have been an important challenge early
in the development of MyHealth but it has quickly receded as value-
based payment models take hold. The Comprehensive Primary Care
initiative, the rise of several ACO's, and important moves by
commercial payers such as Blue Cross and Blue Shield have all combined
to convince providers in Oklahoma that value-based payment models are
the present and the future. CMS further endorsed this thinking with
their announcement in January. As providers recognize that their
success in these new models of care and payment is dependent on having
their patients' comprehensive data available wherever it is needed for
decisionmaking (even if that is a competing organization), the
provider-driven barriers to interoperability tend to melt away.
Thus, the biggest challenge we face is helping our willing provider
members to liberate the patient data from their EHR systems to make it
interoperable. We have so many specific experiences with inappropriate
data blocking and substandard data quality that we have created a
nomenclature to classify the six common types. We have had some success
in solving these issues, but many remain unresolved. Below we describe
each type of data blocking in the context of a real event.
Before reviewing the examples, I would like to point out three
things. First, many EHR vendors work well with their customers and with
our organization to establish interoperability. Second, until recently,
we have been left with few options to address most data blocking
issues. Recently, we have become active users of the ONC Certified EHR
Technology Surveillance program, filing complaints after we have
exhausted all other efforts to work with the vendor and the provider to
implement interoperability. Finally, the cases below are examples of
the types of issues we have experienced, but these issues arise in most
other communities as well.
type 1: the golden rule
By far the most common barrier to interoperability, exorbitant
interface and maintenance costs cause many small practices and
hospitals (and some large ones) to forego participation in HIE or at
least providing data to the HIE. The EHR Certification requirements do
not set parameters for the fees that vendors may charge for
interoperability, so this is a very common barrier.
In our experience, typical, acceptable interface costs are below
$2,500 per practice. However, several well-known vendors charge $10,000
or more per practice, regardless of practice size, and some charge more
than $30,000 to $40,000, which for many practices in Oklahoma amounts
to $3-$5 per patient seen for an entire year.
Other vendors charge per patient. One vendor in particular has,
until recently, charged more than $2.00 per patient per year, which
added nearly $1M in cost to large health systems and prompted an
avoidance of the standard interface approaches with that vendor. Asking
a CFO to pay $1M extra just to provide competitors with access to see
their patient data seems to be a guaranteed way to keep the health
system to participate in HIE--and this is exactly the effect that this
per-
patient per-year fee model has had.
When we question vendors about the exorbitant cost of interfaces,
we are often told they are technologically complex and labor-intensive.
While this may be true, the complexity is usually a result of the
vendor's own decisions about architecture and their implementation of
the meaningful use interoperability requirements. In addition, it is
difficult to recommend to our participating providers that they pay
these fees when a number of well-known EHR vendors have been extracting
the data from their customer's practices, de-identifying it, and
selling it for years. Certainly this process is more technologically
complex than making a standards-based interface for clinical data.
type 2: the ``hotel california''
There is a component of the EHR Certification program called ``data
portability'' that is intended to help providers to change EHR vendors
if they like. Vendors are required to enable providers to create a
batch export of their patient records in a standardized format. It is
also a very helpful capability for interoperability. Unfortunately, few
vendors appear to offer this functionality as intended, and so we say
that the customers of these Vendors have a Hotel California problem--
they can check out other EHR products any time they like, but their
data can never leave.
MyHealth has filed complaints about this issue and our initial
complaints have been found to have merit by ONC and the ACB, but no
specific timelines have been provided. Thus, doctors using these
Vendors' EHR products are facing pressure to meet Meaningful Use by the
end of the year without a clear idea of whether, or when, the product
will enable them to do so.
type 3: the inexplicable
In some cases, the reasons for data blocking are not clear, and do
not seem to be linked to any specific technology limitation or business
driver. Often, given time, the real motivation behind the data blocking
will become clear and it usually resolves to a vendor-or provider-
driven decision about cost.
For example, during the install of a major comprehensive EHR
product in one of our largest health systems, we were told by the
Vendor's project manager that ``We don't do CCD's, they're just not in
our DNA''. We pointed out that their product was Meaningful Use
certified, implying their ability to produce a CCD (a Patient Summary
of Care file), and, in any case, this was their customer's request.
Despite an hour of questioning, the project manager remained unfazed
and simply continued to repeat ``we don't do CCD's, they're not in our
DNA''.
This issue remains today, despite the fact that we now get CCD's
from other instances of this vendor's product. We were forced to build
and maintain five different HL-7 (an older, less robust protocol) feeds
to replace the missing CCD. The missing data from this health system
means that their patients are at higher risk of Adverse Drug Events,
duplicated testing and imaging (and radiation exposure).
type 4: garbage in, garbage out
All certified EHR's are required to produce Patient Care Summary
records according to a common format, but many of them fail to include
the proper structure, clinical content, or standard codes. We have
never seen a completely correct Patient Care Summary despite processing
millions of them.
Poor data standardization and quality prevents data from being
combined with other records on the patient, creating a messy and often
inaccurate chart riddled with duplications. Further, this prevents the
calculation of metrics and care gaps, as well as quality measures,
compromising the safety and accuracy of clinical decision support, and
undermining the success of value-based payment models.
type 5: ehr at the center of the universe
Increasingly, we are hearing from large health systems using
certain EHR systems that their EHR vendor provides all of the
interoperable information they need. These vendors have done an
excellent job of implementing interoperability with other health
systems using their EHR products. However, this interoperability does
not extend beyond the specific vendor's customers, excluding
independent providers and small hospitals, pharmacists, ancillary care
services and long term care, etc.--all of which play a critical role in
the health of patients.
Most concerning about this belief is that it subverts
interoperability at the community and State level, creating instead a
corporate EHR network for interoperability, which is not subject to the
trust arrangements and policies of the community.
type 6: the ``bait and hidden switch''
In this type of data blocking the vendor achieves certification
with one feature set and then either hides or eliminates functionality
when the EHR is deployed in a practice or hospital.
We pursued interoperability for nearly 4 years with one of the
Nation's largest ambulatory EHR vendors, but were told repeatedly that
we must purchase their proprietary ``HUB'' product. In addition to a
base cost of $40,000, the HUB carries an additional monthly service fee
of $50-$100 for every provider in every practice--more than doubling
the cost of HIE services.
Recognizing that 2014 EHR Certification required them to produce
the Patient Care Summary files for interoperability, we filed a
complaint with ONC, which was forwarded to the Accredited Certifying
Body (ACB) for the Vendor, who apparently forwarded the message to the
Vendor.
Almost immediately we received an email from the Vendor indicating
that they would no longer work directly with our HIE. We were quite
surprised and concerned, but fortunately, within a few hours, we
received an email from an executive with the Vendor. Realizing that the
certification challenge was credible, the executive offered some new
information, unknown to any other practice or HIE in the country, as
far as we can tell. It turns out that instead of requiring the purchase
and implementation of the HUB product, the vendor could make a simple
``configuration'' change to enable the data to flow out of the system.
We immediately requested that the configuration change be made in our
participating practices, and by 10 a.m. the next day we had data
flowing from three practices, at no additional cost to the providers.
This EHR Vendor product had passed certification testing with the
configuration switch turned on, but turns it off by default for every
installation of the product. Until we filed this challenge, no amount
of direct questioning of the Vendor support, sales, and implementation
staff revealed the existence of this ``switch''.
The follow-on story is also informative. This feels to us like an
important product defect that would be communicated to customers in any
other industry. Since most of these installations were funded with tax-
payer dollars, it would seem that the commitment to transparency would
be even greater. Unfortunately, the Vendor does not appear to be
communicating this information to their customers, and continues to
offer the expensive HUB as the only way to get data out of their EHR
system. So, we have shouldered the burden of informing the public and
relevant stakeholders, including our HIE colleagues. We have assisted
several States in making this configuration switch work for them, and
we are happy to do it--but we continue to ask why a formal and
transparent communication process is not being required.
Of particular concern to me has been the clear reluctance on the
part of the practices to file or participate in the filing of
complaints against their EHR vendors. Several times, affected practices
have requested MyHealth to file the formal complaints on their behalf,
but expressed fear that filing directly could prompt retribution from
the EHR vendor. I have been surprised to find intimidation of providers
by their vendors, whether real or perceived, playing such a significant
role in the data blocking issue.
recommendations
I have several suggestions to address and prevent the issue of data
blocking.
1. The HIT Certification program is the strongest lever available
to ensure Vendor alignment with success of the Nation in achieving the
optimization of health and quality of life for all Americans. I
recommend a tuning of the initial certification and an expansion of the
ongoing surveillance program.
a. Initial and ongoing certification recommendations:
i. Current testing: EHR's are currently certified based
on testing in an ideal, laboratory environment.
ii. Expanded Certification Testing: We recommend that
certified EHR's have their interoperability functions tested in
the field with each deployment of their product in order to
maintain certification. In this way, the product can be proven
to be interoperable before the ``keys'' are handed over to the
provider. This would specifically address Information Blocking
Types 2, 3, 4, and 5 above. We are happy to expand on ways this
can be accomplished cost-effectively as requested.
iii. Meaningful Use 3 Proposed requirements: It has been
proposed that Certified EHR's must enable providers to choose
standard clinical documents and schedule them for automated
delivery to specific locations. Preservation of this
requirement is critical to achieving broad interoperability.
b. Certification Surveillance program recommendations: As
noted above, most progress on addressing the data blocking
issue has come through successful use of ONC's Certified Health
IT Surveillance program. We propose expansion and amendment of
this program in the following ways:
i. More prominent role: The ONC Surveillance program can
be very important in solving these issues, but it must become a
more prominent component of the Certification program to have
maximal impact. Doctors and hospitals need to know how and when
to use it, and Vendors need to know what to expect from it as
well.
ii. Increased Transparency:
1. When a complaint is found to have merit, and
changes have been requested of a specific vendor, the
customers deserve to know the details and timing for a
solution. Posting the adjudicated issues and timelines
will enable providers to plan and increase trust among
the provider community.
2. An annual Health IT Surveillance report has been
produced by the Accredited Certifying Bodies, but the
report is only accessible via a Freedom of Information
Act request--creating an additional barrier to
transparency and therefore trust.
iii. Whistle-blower protections: As described above, many
providers, especially small clinics and hospitals, are
concerned about backlash from their vendors that filing a
complaint may generate. Whether these fears are warranted or
not, it's important to create an environment where these
providers can feel safe in airing their concerns.
iv. Independence from Conflict of Interest: Currently, the
Accredited Certification Bodies, who are responsible for the
Certification program, are also responsible for executing the
Surveillance program. This could be perceived to be subject to
conflict of interest, since the ACB generates nearly all its
revenue from the Vendors that it certifies. For example, an ACB
perceived to be overly strict by Vendors could lose its
Certification business.
c. Lemon law for EHRs: On more than one occasion in our
community, providers have invested heavily in an EHR product,
only to discover that it does not meet their needs. One small,
financially strapped hospital in Oklahoma recently fired their
EHR vendor when it became clear that the Vendor would not meet
2014 EHR Certification--which would prevent the hospital from
meeting its Meaningful Use obligation. Unfortunately, the EHR
vendor sued the hospital for breach of contract.
An appropriately crafted Lemon Law could help to prevent
these kinds of issues.
d. Transparency in contracting: The contents of EHR vendor
contracts are among the best kept secrets in America, and the
signatories are often bound by strict non-disclosure
agreements. Certain elements of these contracts, and
specifically those pertaining to interoperability, should be
made transparent to customers and other healthcare
stakeholders.
2. Payment model incentive alignments provide the strongest
incentive for providers and hospitals to support and enable
interoperability. In particular, the expansion of value-based payment
models are prompting providers to look beyond the walls of their
organization for the patient information they need.
a. In the short term, a process measure for interoperability
should be employed to help providers and other stakeholders
gauge their progress in achieving appropriate levels of
interoperability. We have defined several measures that could
be of use and would be happy to share them.
b. CMS and other Federal partners such as the DoD, VA, and
HRSA should begin to place more value on Clinical Quality
measures derived from a comprehensive record of the care each
patient receives, rather than from a single EHR or site of
care. This will further encourage provider participation in
meaningful health information exchange, and will significantly
improve the accuracy of the quality measures being reported. In
Oklahoma, the commercial payers and Medicaid have already
recognized the importance of this approach to value measurement
and are proceeding to implement it.
c. Support the development of regional data aggregation such
as HIE's and the implementation of whole-patient quality
reporting. These are important infrastructure elements that are
needed to support the kinds of measurement described above for
value-based payment programs, and also to ensure that patients
get comprehensive, safe care no matter where they seek it.
3. Standards: It may be controversial for me to say this as a Board
Certified Medical Informaticist, but we have plenty of standards--we
need to focus on correctly implementing the standards we have right now
and monitoring their performance. R&D on new standards should continue,
but they should undergo rigorous testing before becoming a part of the
certification or meaningful use requirements. The ONC Standards
Advisory hits the mark well on this issue.
4. Governance: This is perhaps the most critical issue limiting the
impact of the tax-payers $30B investment in health IT. In the original
HITECH act, ONC was called upon to establish the governance for the
nationwide health information network. Now, more than 6 years later,
that governance still does not exist, due in part to interpretations of
limitations on ONC's authority. Thus, there is a vacuum in governance
for this critical component of America's infrastructure--and that
vacuum is being filled by various consortia and collaborations of
vendors and large provider organizations. In order to rapidly advance
health IT and interoperability, ONC's authority should be made clear,
and I believe strongly that the correct perspectives to include in that
governance are:
a. Those who receive care (patients, special population
representatives);
b. Those who deliver care (providers, public health); and
c. Those who pay for care (payers, employers, governments).
Thank you for this opportunity to share my experiences and offer my
advice. The progress made to date is tremendous, and I am confident
that with your guidance, health and healthcare in America can become
the best in the world.
The Chairman. Thank you very much.
Dr. Mirro.
STATEMENT OF MICHAEL J. MIRRO, M.D., FACC, FAHA, FACP, PAST
CHAIR, MEDICAL INFORMATICS COMMITTEE, AMERICAN COLLEGE OF
CARDIOLOGY AND CHIEF ACADEMIC/RESEARCH OFFICER, PARKVIEW MIRRO
CENTER FOR RESEARCH AND INNOVATION, FORT WAYNE, IN
Dr. Mirro. Chairman Alexander, Ranking Member Murray, and
members of the committee, thank you for the opportunity to
speak today on this important issue of information blocking and
the unforeseen problems that have been created and the possible
solutions that will help improve patient care.
My name is Michael J. Mirro. I am testifying today on
behalf of the American College of Cardiology, a 49,000-member
medical society that is the professional home for the entire
cardiovascular care team.
I am board certified in internal medicine, cardiovascular
disease, clinical cardiac electrophysiology, and geriatrics. I
have focused the majority of my clinical work on cardiac
implantable devices in patients with serious heart rhythm
problems, and my clinical investigation historically has been
on new and advanced technologies to enhance their function.
I have worked also in the deployment of informatics tools
since 1995, assisting in the refinement and development of
decision support tools to improve point of care of patients
with congestive heart failure.
In private practice, my partners and I were early adopters
of electronic health records, and these systems were
implemented before the HITECH passage. They were complemented
by the fact that they had user-centered design as opposed to
software-centric design, which now occur commonly.
Additionally, many of the current systems lack clinical
utility and create substantial practice inefficiencies and
reduce quality of the patient-physician interaction during an
office visit.
I first became aware of information blocking when my
colleagues in other private practices of cardiology adopted
EHRs and were forced to spend substantial resources to
interface with their health system's EHR. Those practices would
have been able to better financially plan if the cost had been
disclosed during the contracting.
Transparency of additional hidden costs or fees within
contracts with EHR vendors should be evaluated. Many contracts
between providers and EHR vendors include a gag clause, which
prevent providers from speaking publicly about the problems
associated with their EHRs. EHR vendors should not be allowed
to include such clauses in the contract.
The delay of information sharing is another form of
information blocking. I once had a patient who was admitted to
the emergency room in cardiac arrest. Because of a delay in
receiving his cardiac history, data critical to the care of
this patient, it resulted in an unfortunate outcome. The
patient experienced complications. There was an emergency
cardiac catheterization, subsequently resulting in a prolonged
hospital stay. However, he did survive.
Rapid and secure exchange of health information is
critical. In some cases, it could mean the difference between
life and death.
Data fluidity should mean not only that information reaches
the provider, but the data is transmitted quickly and securely.
Many EHR vendors provide the functionality needed but require
the user to purchase their IT products to make the elements of
EHR interoperable. Like other products, such as consumer
electronics, you are able to connect, but you must buy a
company's specific products in order to do so with ease.
The ramifications of technology in healthcare that we are
not able to communicate are serious, resulting in decreased
care quality and stunting improvements in population health.
EHR vendor products should be universal and connect with other
EHRs offered by different companies.
Another advantage of the free flow of data is to empower
patients in their care decisions. One of our recent projects
has been to establish a way for patients who are remotely
monitored with implantable cardiac devices to receive their
data.
Each element of the four companies that make these
implantable devices are available. However, we have to require
interface between each vendor in our EHR system. We have to pay
for that interface to be created.
Health IT vendors and providers should be incentivized to
establish networks for patients to monitor their cardiac
devices particularly, empowering them to actively participate
in their healthcare decisions. In addition, adoption of public
data standards should be expected and supported in the best
interests of patients so the patients can receive their data.
Many information blocking problems stem from the financial
incentives of the EHR companies to obstruct data. The HITECH
Act, along with implementation of meaningful use program, has
improved data sharing and data liquidity to some degree.
With that stated, the unintended consequences of meaningful
use is that the systems were designed to facilitate charge,
capture, and revenue cycle management and focus less on
clinical data utility. Although meaningful use programs have
brought favorable results within the context of data transfer,
many of the requirements set forth in the program are
unattainable.
Recognizing that only 11 percent of physicians have
attested for Stage 2 meaningful use, I recommend, in concert
with the ACC, that Stage 3 meaningful use be delayed in its
entirety.
In addition to what I have discussed, the College has
called for many of the same actions recommended by the ONC's
report of April 2015, and these include the following:
No. 1, strengthen in-field surveillance of health IT
certified by ONC;
No. 2, constrain standards in implementation specification
for certified health IT;
No. 3, work in concert with HHS to improve stakeholder
understanding of HIPAA provisions provided related to
information blocking; and
No. 4, work with CMS to coordinate healthcare payment
incentives and leverage market drivers to reward
interoperability and discourage information blocking.
In closing, I commend you, Chairman Alexander and Ranking
Member Murray, and your excellent staff for gathering us today,
taking this initiative to accomplish specific goals related to
interoperability and information blocking. Furthermore, I
applaud the collaborative bipartisan approach and thank you
again for the opportunity to speak here today.
I will look forward to the discussion.
[The prepared statement of Dr. Mirro follows:]
Prepared Statement of Michael J. Mirro, M.D., FACC, FAHA, FACP
summary
ehr vendor contracts
Transparency of additional (or hidden) fees should be
evaluated.
EHR vendors should not be allowed to include gag clauses.
data fluidity
Patient information should reach the provider without
delay in a fast, secure manner.
EHR vendors' products should be universal and connect to
other EHRs offered by different companies.
Health IT vendors and providers should be incentivized to
establish networks for patients to monitor their devices and to empower
them to actively participate in their health decisions.
Adoption of public data standards should be expected and
supported in the best interest of patients.
meaningful use stage 3
Stage 3 of the Meaningful Use program should be delayed in
its entirety.
endorsement of actions outlined in onc's april 2015 report to congress
on information blocking
Strengthen in-the-field surveillance of health IT
certified by ONC.
Constrain standards and implementation specifications for
certified health IT.
Work in concert with HHS to improve stakeholder
understanding of the HIPAA provisions related to information sharing.
Work with CMS to coordinate health care payment incentives
and leverage other market drivers to reward interoperability and
exchange and discourage information blocking.
______
Chairman Alexander, Ranking Member Murray, and members of the
committee, thank you for the opportunity to speak today about the
important issue of information blocking, unforeseen problems that have
been created, and possible solutions to help improve patient care.
My name is Michael Mirro and I am testifying today on behalf of the
American College of Cardiology, a 49,000-member medical society that is
the professional home for the entire cardiovascular care team. I am
board certified in internal medicine, cardiovascular disease, clinical
cardiac electrophysiology, and geriatrics. [In addition to seeing
patients, many who suffer from multiple chronic conditions, I also
serve as Chief Academic Research Officer at Parkview Health System in
Ft. Wayne, IN where I manage over 90 clinical trials.] I have focused
the majority of my clinical work on cardiac implantable electronic
devices in patients with serious heart rhythm problems and clinical
investigation into new and advanced technology to enhance their
function. [I have worked extensively on remote monitoring of cardiac
devices and electronic messaging patients their data from their
individual device.] I have worked in the development of health
informatics tools since 1995, assisting in the refinement of clinical
decision support software to improve point of care quality related to
congestive heart failure.
The private practice that my partners and I owned was an early
adopter of electronic health records. These systems, implemented before
HITECH's passage, had a user-centered clinical design, as opposed to
the software centric certified EHR systems of today. Additionally, many
current systems lack clinical usability and thus create substantial
practice inefficiency and reduced quality patient-physician interaction
during an office visit.
I first became aware of information blocking when my colleagues in
other private cardiology practices adopted EHRs and were forced to
spend substantial resources to interface with their health system's
EHR. These practices would have been able to better plan financially if
these costs had been disclosed at the outset. [Fortunately, the
practice was in a financial position to absorb these costs, but many
other practices are not.] Transparency of additional (or hidden) fees
within contracts with EHR vendors should be evaluated. Many contracts
between providers and EHR vendors include gag clauses which prevent
providers from speaking publicly about problems associated with EHRs.
EHR vendors should not be allowed to include such clauses.
The delay of information sharing is another form of information
blocking. I once had a patient admitted to the emergency room in
cardiac arrest. [The patient was a truck driver from out of State.]
Because of a delay in receiving his cardiac history, data critical to
his care was not available in a timely fashion. The patient experienced
a complication during the emergency heart procedure resulting in
prolonged illness. [The support of electronic messaging of standard
clinical summaries is a critical issue with respect to quality and
safety of patient care.] Rapid, secure exchange of health information
is critical and in some cases can mean the difference between a patient
living and dying. Data fluidity should mean not only that information
reaches the provider, but that the data is transmitted quickly and
securely.
Many EHR vendors provide the functionality needed, but require the
user to purchase their health IT products to make the elements of the
EHR interoperable. Like other products such as consumer electronics,
you are able to connect, but you must buy a specific company's products
to do so with ease. The ramifications of technology in health care that
are unable to communicate are serious, resulting in decreased care
quality and stunting improvements in population health. EHR vendors'
products should be universal and connect to other EHRs offered by
different companies.
Another advantage of the free flow of data is to empower patients
in their health care decisions. One of my recent projects was to
establish a way for patients to remotely monitor their implanted
devices. Each element of the four devices available in the market had a
different vendor, requiring us to contract with four different vendors
and pay four different set-up costs to allow patients to accomplish one
task. Health IT vendors and providers should be incentivized to
establish networks for patients to monitor their devices, empowering
them to actively participate in their health decisions. In addition,
adoption of public data standards should be expected and supported in
the best interest of patients.
Many information blocking problems stem from the financial
incentives of EHR companies to obstruct data. The HITECH Act, along
with implementation of the Meaningful Use Program, has improved data
sharing and data liquidity. With that stated, the unintended
consequence of Meaningful Use is that systems were designed to
facilitate charge capture and revenue cycle management and focus less
on clinical data and usability. [The importance of exchanging a
clinical summary document has been enhanced by this program, but we
need surveillance of individual vendor behavior.] Although the
Meaningful Use program has brought favorable results within the context
of data transfer, many of the requirements set forth in the program are
unattainable. Recognizing that only 11 percent of physicians have
attested to stage 2, I recommend, in concert with the ACC, that stage 3
of Meaningful Use be delayed in its entirety.
In addition to what I have discussed, the College has called for
many of the same actions recommended in the Office of the National
Coordinator's April, 2015 Report to Congress on Information Blocking,
including:
No. 1: Strengthen in-the-field surveillance of health IT
certified by ONC. [The ACC feels strongly that a program such as this
is needed and that ONC would be the appropriate entity to administer
such a program. ONC could hire an outside contractor to affirm
compliance--similar to what CMS has done with the Meaningful Use
program.]
No. 2: Constrain standards and implementation
specifications for certified health IT. [This committee has debated
whether the Federal Government or the private sector should establish
common standards, and the ACC believes it should be a combination of
both. Medical specialty societies are well-equipped to engage in the
creation of these standards, while the Federal Government is needed to
oversee enforcement of the standards.]
No. 3: Work in concert with HHS to improve stakeholder
understanding of the HIPAA provisions related to information sharing.
[HIPAA is outdated and in many cases is actually an impediment to
patient care. The ACC would encourage the committee to reevaluate HIPAA
in its entirety--including its successes and failures--and whether all
aspects of HIPAA remain appropriate given today's technology.]
No. 4: Work with CMS to coordinate health care payment
incentives and leverage market drivers to reward interoperability and
discourage information blocking. [As with my example given earlier
about creating a mechanism to remotely monitor devices, this is proof
that when coupling providers with innovative companies, we can improve
the well-being of our patients and reduce costs.]
In closing, I commend you, Chairman Alexander and Ranking Member
Murray, and your excellent staff for gathering us today and taking the
initiative to accomplish specific goals related to interoperability and
information blocking. Furthermore, I applaud your collaborative,
bipartisan approach. Thank you again for the opportunity to be here
today. I look forward to the discussion.
The Chairman. Thank you, Dr. Mirro.
Dr. Kibbe
STATEMENT OF DAVID C. KIBBE, M.D., MBA, PRESIDENT AND CEO,
DIRECTTRUST, SENIOR ADVISER, AMERICAN ACADEMY OF FAMILY
PHYSICIANS, WASHINGTON, DC
Dr. Kibbe. Good morning. Chairman Alexander, Ranking Member
Murray, and distinguished members of the committee, thank you
for the opportunity to share my thoughts today.
Direct exchange was designed by an ONC-led consortium of
over 50 organizations to replace paper-based mail, fax, and
courier services with secure electronic messaging between users
of different electronic health record applications and personal
health record applications. Direct messaging is very similar to
electronic mail, email, in that a sender can compose a message,
attach a file or files, and send the package over the Internet
to the address of a recipient.
Because transmissions are encrypted and identity validated,
this method of sharing data is ideally suited to the handling
of personal health information, which needs to be protected at
all times.
The use of Direct exchange has grown very rapidly. There
are now over 300 electronic health record vendor products that
are certified by ONC as Direct enabled, and over 50 health
information exchanges nationwide provide Direct exchange
services. DirectTrust members have provisioned nearly 1 million
Direct addresses and accounts in the healthcare industry,
enabling Direct exchange at over 40,000 healthcare
organizations.
Over 30 million Direct messages have been exchanged in 2014
and 2015 so far in support of transitions of care and care
coordination. The Indian Health Services, the Veterans
Administration, the U.S. Postal Service, and the Centers for
Medicare and Medicaid Services all have Direct implementation
programs underway.
Despite this success, information blocking by healthcare
provider organizations and their EHRs, whether intentional or
not, is still a problem for some providers and their relying
parties. Persisting information blocking problems include local
electronic health record and provider policies that are not
standard, EHR product design and/or implementation flaws, lack
of or inadequate product and service support, and high pricing
for interoperable-enabled software products and services.
Time allows me to provide you with just a couple of
examples. EHRs capable of Direct exchange should accept all
trusted inbound messages and attachments. However, some
electronic health record companies' products require that an
incoming Direct message be accompanied by a particular
attachment.
No attachment or not the particular attachment, the inbound
message is discarded and dropped, often without letting the
sender know that that has happened. Clearly, this is
frustrating to relying parties, but it is also not the original
intent of Direct exchange, which supports transport of messages
without attachments and with many different kinds of file
attachments.
Another example. Believe it or not, although ONC certified
to send and receive Direct messages, some EHR vendors' products
are unusable in the field, as has been noted. They lack an
inbox or a message compose button or some other key component
that allows the user to easily compose messages and send them.
In my opinion, the responsibility for assuring secure
interoperable exchange resides primarily with the healthcare
organizations, not with EHR vendors and not with the
Government. Healthcare provider organizations must demand
collaborative and interoperable health IT tools from their EHR
vendors to make patient-
centered care routine and ubiquitous as a practice throughout
all communities in the United States.
However, there are roles for Government, and I want to list
a few that we find of high importance. Government can continue
to shed light on these problems and work with trade groups,
standards, and policies organizations, as you have done with
these hearings that I think have been extremely valuable
already.
These hearings are setting high expectations for
interoperability of electronic health records and other
applications and especially for those that have benefited from
the subsidies of the meaningful use programs.
ONC can bring better and improved EHR certification
processes forward beyond the testing laboratory so that the
EHR's usability of interoperability features in the field
becomes part of the public record and can be used in purchasing
decisions, as my colleague has already mentioned. It is very,
very important to make that public record available.
Congress can accelerate multiple Federal agency uses and
demand for open standard-based interoperability of health IT
with private sector communities and providers in order to
remove the demand and lessen the use of fax, e-fax, mail, and
courier. The savings in documentation and time and cost for
these Federal agencies, as well as their partners in the
private sector, would be enormous in a very short period.
Finally, CMS should continue to link the use of certified
EHR technology to participation in value-based purchasing
programs. Doing this makes interoperability and collaboration
across multiple organizations in multi-vendor environments and
with patients directly financially rewarding to providers and
their health IT vendors.
Demand for collaboration and interoperability is best
driven by underlying business models and business cases
supported by good regulation and oversight.
Finally, I would just like to say I think the memo has gone
out. Collaboration and interoperability of health IT is
mandatory. It is no longer nice to have. We still have some
kinks to work out and some people to deliver that message to,
perhaps a second round.
Thank you.
[The prepared statement of Dr. Kibbe follows:]
Prepared Statement of David C. Kibbe, M.D., MBA
summary
My name is David Kibbe and I serve as the president and CEO of the
non-profit trade alliance DirectTrust, and also as senior advisor to
the American Academy of Family Physicians, the physician membership
organization representing over 100,000 of the Nation's family
physicians, residents, and students.
Because of the added privacy, security, and identity layers of
Direct exchange, this method of sharing of data between providers using
different EHRs, and between providers and patients, is ideally suited
to the handling of personal health information which must be protected
at all times.
Interoperable Direct exchange has grown rapidly since becoming a
required feature of EHR technology certified by ONC in 2014. There are
over 300 EHRs that are certified as Direct-enabled, and over 50 HIEs
nationwide provide Direct exchange services. DirectTrust members alone
have provisioned nearly one million Direct addresses in the health care
industry, enabling Direct exchange at over 40,000 health care
organizations. Over 30 million Direct messages have been exchanged in
2014 and 2015 so far in support of transitions of care and care
coordination. The Indian Health Services, U.S. Postal Service, Veterans
Administration, and the Centers for Medicare and Medicaid Services all
have Direct implementations under way to replace mail, fax, and efax
communications between these Federal agencies and providers in the
private sector beginning later this year.
While it is true that interoperable health information exchange has
made great progress in the past 2 years, information blocking by health
care provider organizations and their EHRs, whether intentional or not,
is still a problem for some providers wishing to use Direct exchange.
Persisting information blocking problems include: Local EHR and
provider organization policies; EHR product design and/or
implementation flaws; lack of or inadequate product/service support;
and high pricing for HIE-enabled software upgrades.
In my opinion, the responsibility for assuring secure interoperable
health information exchange resides primarily with the health care
provider organizations, not the EHR vendors, and not the government.
However, there is a role for government to encourage and incentivize
collaboration and interoperability. Among the actions that government
can take should be: To continue to shed light on these problems; to
bring better and improved EHR certification processes forward beyond
the testing laboratory; to accelerate Federal agency use of and demand
for open, standards-based interoperable HIE with private sector
providers and provider organizations; and to continue to tie more
robust ONC EHR certification and use of certified EHR technology to
participation in Value Based Purchasing programs.
The root causes of information blocking are not technological or
due to a lack of standards for interoperability or EHR capabilities for
interoperable exchange. As noted in the ONC Report to Congress on
Information Blocking of April 2015,\1\
---------------------------------------------------------------------------
\1\ http://www.healthit.gov/sites/default/files/reports/
info_blocking_040915.pdf.
``While some types of information blocking may implicate
these technical standards and capabilities, most allegations of
information blocking involve business practices and other
conduct that interferes with the exchange of electronic health
information despite the availability of standards and certified
health IT capabilities that enable this information to be
---------------------------------------------------------------------------
shared.''--(Emphasis added.)
Therefore, attempts to redress information blocking must address
the unwillingness of some providers and their EHR partners to share and
exchange data, and not just the specific technical problems that may be
encountered in making exchanges run smoothly and reliably. In my
opinion, that unwillingness originates in the current business models
of the health care industry in general, wherein fee-for-service payment
creates disincentives for sharing of health information and rewards
information hoarding, or at least the delay of timely information
exchanges. Changes to these payment incentives could do much to reward
business models where collaboration and interoperability are highly
valued, and would create conditions for the technological capabilities,
standards, and infrastructure for interoperable health information
exchange now in place to be put to much better use.
______
opening remarks
Chairman Alexander, Ranking Member Murray and distinguished members
of the committee, thank you for the opportunity to share my thoughts on
problems that impede the sharing of health information between and
among parties authorized to access such information, now often referred
to as ``information blocking.'' I will offer some near-term suggestions
to help improve upon the current situation.
My name is David Kibbe and I serve as the president and CEO of the
non-profit trade alliance DirectTrust, and also as senior advisor to
the American Academy of Family Physicians, the physician membership
organization representing over 100,000 of the Nation's family
physicians, residents, and students.
DirectTrust's 150-plus members are a vibrant community of service
providers, health IT vendors, and health care organizations dedicated
to the use of interoperable, secure, standards-based health information
exchange via the Direct standard, as well as other vendor-agnostic
technologies.
Direct exchange was designed to replace paper-based mail, fax, and
efax transmissions of health information with secure electronic
messaging between users of different software applications, like EHRs.
Direct messaging is very similar to electronic mail, or email, in that
a sender can compose a message, attach a file or files, and send the
package over the Internet. Both sender and receiver need to have Direct
addresses that usually have the format
[email protected]
Practice.com, supplied by Health Internet Service Providers, or HISPs.
The word ``direct'' in the address signifies that both the message and
attachments are encrypted end-to-end, and that the identities of both
parties have been validated.
Because of the added privacy, security, and identity layers of
Direct exchange, this method of sharing of data between providers using
different EHRs, and between providers and patients, is ideally suited
to the handling of personal health information which must be protected
at all times.
Interoperable Direct exchange has grown rapidly since becoming a
required feature of EHR technology certified by ONC in 2014. There are
over 300 EHRs that are certified as Direct-enabled, and over 50 HIEs
nationwide provide Direct exchange services.
DirectTrust members alone have provisioned nearly one million
Direct addresses in the health care industry, enabling Direct exchange
at over 40,000 health care organizations. Over 30 million Direct
messages have been exchanged in 2014 and 2015 so far in support of
transitions of care and care coordination. The Indian Health Services,
U.S. Postal Service, Veterans Administration, and the Centers for
Medicare and Medicaid Services all have Direct implementations under
way to replace mail, fax, and efax communications between these Federal
agencies and providers in the private sector beginning later this year.
DirectTrust members have significant experience with
interoperability testing and the problems that can impede Direct
exchange information flows. Indeed, DirectTrust is something of a
laboratory wherein these problems are routinely identified,
investigated, and usually solved. Here are some of our collective
observations on information blocking from an ``on the street''
perspective.
examples of information blocking
While it is true that interoperable health information exchange has
made great progress in the past 2 years, information blocking by health
care provider organizations and their EHRs, whether intentional or not,
is still a problem for some providers wishing to use Direct exchange,
as well as for these providers' clinical partners who want to be able
to exchange Direct messages and attachments with them.
Persisting information blocking problems include:
Local EHR and provider organization policies. For example,
an EHR might require that an incoming Direct message be accompanied by
a particular attachment type. No attachment? The inbound message and
its files are discarded, often without letting the sender know. Which
is very frustrating to relying parties. Clearly this was not the
original intent of Direct exchange, which supports virtually any kind
of file transmission, with or without an attachment.
EHR product design and/or implementation flaws. For
example--believe it or not--although certified to send and receive
Direct messages, some EHR vendors' products lack an ``inbox'' or
``compose'' button, or other key component needed to allow the user to
compose messages, attach files, and so on.
Lack of or inadequate product/service support. If an EHR
customer can't get service assistance for their product's
interoperability functions, this inhibits or delays information
exchange set up and implementation for providers seeking to use
interoperable health IT.
High pricing for HIE-enabled software upgrades. While some
vendors include the costs of upgrading from Stage 1 to Stage 2 features
and functions, including Direct exchange capability, others make the
new features a new cost that practices must bear. Clearly, this hurts
the smaller practices more than it does the bigger institutions.
Registration and ``whitelisting'' requirements for message
exchange. Making exchange partners register with the practice's or
hospital's EHR in effect discourages EHR users from engaging in
standards-based interoperable HIE. It's a little bit like having a
phone that requires each caller to fill out a complicated form and
``apply'' to be able to reach you before you'll accept their call.
``HIPAA doesn't allow.'' Perhaps the most significant
problem of all is faced by patients and consumers trying to use Direct
exchange to access their medical records, only to be told that HIPAA
won't allow them to do so. Patients and consumers ought to be able to
be full participants in Direct exchange and partners with their
providers in health information exchanges.
the role of government to encourage health
information exchange
In my opinion, the responsibility for assuring secure interoperable
exchange resides primarily with the health care provider organizations,
not the EHR vendors, and not the government. Health care provider
organizations must come to realize that acting in the best interest of
patients is to assure that health information follows the patient and
consumer to whatever setting will provide treatment, even if that means
in a competitor's hospital or medical practice. And they must demand
collaborative and interoperable health IT tools from their EHR vendors
to make this routine and ubiquitous as a practice in every community in
the United States.
However, there is a role for government to encourage and
incentivize collaborative and interoperable health information
exchange. Among the actions that government can take to help overcome
the kinds of continuing problems I have mentioned above should be:
To continue to shed light on these problems, and work with
trade groups, standards and policies organizations, and others to set
expectations for interoperability of EHRs and other applications
certified as interoperable, especially those that have been federally
subsidized within the Meaningful Use programs. Let's ``Finish what we
started before moving to more complex solutions that may or may not
work.''
To bring better and improved EHR certification processes
forward beyond the testing laboratory, so that the utility and
usability of interoperability features of ONC certified EHR products in
the field becomes part of the public record, and can be used in
purchasing decisions. Collaboration and partnership with non-profit
trade groups to achieve this goal would be advisable.
To accelerate Federal agency use of and demand for open,
standards-based interoperable HIE with private sector providers and
provider organizations, thereby removing reliance on paper-based mail,
fax, efax, and courier for these Federal programs.
Examples include Veterans Health Administration referrals to and
from private sector medical practices and hospitals; Veterans Benefits
Administration health information exchanges with private sector medical
practices and hospitals; the use by Medicare, Medicaid, and State
agencies of interoperable HIE for communications with private sector
providers and provider organizations for limitation of fraud, payment
adjudication, claims attachments requests, and other administrative
transactions now done via fax and mail.
To continue to tie more robust ONC EHR certification and
use of certified EHR technology to participation in Value Based
Purchasing programs, wherein interoperability and collaboration across
multiple organizations in multiple-vendor environments is financially
rewarding to providers and their health IT vendors. Demand for
collaboration and interoperability is best driven by underlying
business models and business cases supported by regulation and
oversight.
summary
Information blocking is a persistent and real problem faced by
providers, provider organizations, and patients who wish to share and
exchange health information between and among parties authorized to
access such information, and to use that information to improve quality
and care coordination.
Progress is being made, and, at its root the causes of information
blocking are not technological or due to a lack of standards for
interoperability or EHR capabilities for interoperable exchange. As
noted in the ONC Report to Congress on Information Blocking of April
2015,\1\
``While some types of information blocking may implicate
these technical standards and capabilities, most allegations of
information blocking involve business practices and other
conduct that interferes with the exchange of electronic health
information despite the availability of standards and certified
health IT capabilities that enable this information to be
shared.''--(Emphasis added.)
Therefore, attempts to redress the root causes of information
blocking must address the unwillingness of some providers and their EHR
partners to share and exchange data, and not just the specific problems
that may be encountered in making exchanges run smoothly and reliably.
In my opinion, that unwillingness originates in the current business
models of some health care provider organizations, and the health care
industry in general, wherein fee-for-service payment creates
disincentives for sharing of health information and rewards information
hoarding, or at least the delay of timely information exchanges.
Changes to these payment incentives could do much to reward business
models where collaboration and interoperability are highly valued, and
where the technological capabilities, standards, and infrastructure for
interoperable health information exchange now in place would be put to
much better use.
The Chairman. Thank you.
Dr. Black.
STATEMENT OF PAUL M. BLACK, MBA, PRESIDENT, CHIEF EXECUTIVE
OFFICER AND DIRECTOR, ALLSCRIPTS, CHICAGO, IL
Mr. Black. Thank you.
Chairman Alexander, Ranking Member Murray, distinguished
members of the committee, thank you for the opportunity to
share my perspectives on the critical topic of the impediments
to health data exchange and the best ways to address them. It
is a true honor to be here.
You have my written statement, which I will summarize in my
remarks today.
My name is Paul Black, president and CEO of Allscripts, the
largest developer of health information technology, including
electronic health records, revenue solutions, population
health, and information exchange services.
More than 180,000 physicians, including those in 45,000
ambulatory practices, 2,700 hospitals, and 13,000 post-acute
facilities utilize Allscripts solutions to connect clinical and
business operations within their organizations and throughout
their communities. We employ 7,000 team members and have
offices in 16 different States, including Illinois, North
Carolina, Vermont, Georgia, Massachusetts, as well as people
working in all 50 States.
Congress and the American people have wisely made the
investment in the advancement of health information technology
toward a vital goal--confirming that this country's citizens
are receiving the best care. Information exchange across vendor
platforms and care settings is now required to meet that goal.
Tomorrow, connected healthcare networks won't be built by
one company or by technologists alone, but by all of us.
Allscripts decided years ago to invest in an open approach to
connectivity. Our corporate vision is granted in our dbMotion
connectivity EMR-
agnostic platform, and our philosophy, which has led to the
development of a large network of external certified software
developers who build apps based on our open application program
interface, or APIs.
While current narrative on interoperability is often
negative, we note that there are many examples of providers who
have established community-wide connectivity and who are
connecting patients to their records, preventing disease and
saving money.
Thankfully, with today's technologies, changing EHRs isn't
necessary in order to provide medical professionals with access
to information, though not all stakeholders have embraced our
position that replacing systems is an inefficient way to
establish data exchange. There are many factors that need to be
tackled ultimately to build an open, connected network of
health.
No. 1, we need to maximize the standards development,
building on progress to date and encouraging adoption of
standards-based approaches by everyone. Congress' role is to
give stakeholders guidance on what is expected and create
reasonable timelines.
We need agreement on the so-called ownership of patient
data. We need a way to identify each individual patient in the
system, which we think is very vital. And last, we need greater
transparency in the areas of interoperability and within health
IT, and we must achieve this interoperability transparency.
It is true that not all stakeholders seem to be equally
motivated to make information liquid. Sluggish exchange largely
stems from one massive gap, the lack of a strong business case
for interoperability in healthcare. The payment system that has
been in place for decades does not motivate them to create an
interconnected healthcare environment.
Recent legislation, such as the replacement of SGR and
reform, such as ACOs, are great steps, and continuing to think
about the relationship between payments and care coordination
can only strengthen this imperative. We encourage Congress to
allow recent laws and regulations to play out before additional
interoperability legislation is passed.
The same advice applies to standards development work.
There should be some time allowed for the fine-tuning of
existing standards, such as Direct, even when we explore new
ones. Congressional attention would be best served in directing
ONC to drive consistent adoption in implementation of the
standards rather than focusing on the need to create new ones.
Last, Mr. Chairman, I note that you said previously that
the best way to solve the problems around interoperability
would be for the health IT industry to do something itself. I
share this view. We have a real obligation here, along with
other provider organizations that we support.
I feel strongly that this is doable, and I challenge all of
my colleagues today to continue working together with us, with
you, the provider stakeholders, the ONC, and the patient
community that have so much to offer in this conversation until
we have achieved success.
Thank you again for the opportunity to be here.
[The prepared statement of Mr. Black follows:]
Prepared Statement of Paul M. Black, MBA
Chairman Alexander, Ranking Member Murray, distinguished members of
the committee, thank you for the opportunity to share my perspectives
on the critical topic of impediments to data exchange and the best ways
to address them across the health system. It is a true honor to be
here.
My name is Paul Black, and I serve as the president and chief
executive officer of Allscripts. Allscripts is the largest developer of
health information technology for this country's healthcare providers,
including Electronic Health Records, revenue cycle management software,
and population health and information exchange services. More than
180,000 physicians, including those delivering care in 45,000
ambulatory practices; 2,700 hospitals; and 13,000 post-acute care
facilities and homecare agencies utilize Allscripts solutions to
connect the clinical and business operations both within their
organization and within their community. We employ 7,000 team members
and have offices in 16 different States, including Illinois, North
Carolina, Vermont, Georgia, and Massachusetts, as well as people
working in all 50 states.
I was invited here today to speak about interoperability and
concerns about information blocking, and as more independent doctors
use our software to treat patients than any other commercially
available product, I'm pleased to share recommendations with you on
this topic. This is important for two reasons: if a stakeholder were to
intentionally get in the way of information exchange, (1) it would be
bad for patients, and (2) it could be anti-competitive. Period.
Congress and the American people have wisely made an investment in
the advancement of health information technology, all oriented around
one goal: ensuring that this country's citizens are receiving the best
possible care--both from a quality and cost perspective. Robust, open
information exchange across a multitude of vendor platforms and care
settings is critical to ensuring that we meet that goal for America's
patients. An increased level of transparency and cooperation is needed
to meet this challenge--health information technology developers,
caregivers, employers, payers, pharmaceutical companies, health systems
and the government must all work harder together to solve this problem.
Tomorrow's healthcare networks won't be built by one company alone, or
even by health information technology developers alone, but by all of
us.
Allscripts has been working with healthcare professionals across
the spectrum of care for many years during a period in which health
care and health IT have evolved at a tremendously rapid rate. The
changes that have been required have been challenging--they have
disrupted systems that have been in place for decades. But we realize
that innovation arises from disruption, and we have embraced it.
Several years ago, Allscripts made a decision to invest in an OPEN
approach to connectivity--one that is grounded in our dbMotion
connectivity platform and a philosophy which has led to the development
of a large network of certified software developers outside of the
company who build apps based on our open APIs. From North Shore LIJ--
the largest private integrated delivery network in the country--to
thousands of independent, single provider practices who make up the
backbone of care in this country, we partner with physicians and other
professionals nationwide who are taking this opportunity to innovate
with us.
And while the narrative on information exchange is largely negative
in conversations in Congress and in the media, it is important to note
that there are many examples of providers who have worked through the
process of establishing connectivity and are making it work. These
providers are changing lives by preventing disease and saving money.
Organizations like Holston Medical Group, which has offered to connect
all providers in NE Tennessee and SW Virginia and is already working
with Allscripts to facilitate data exchange between 25 different EHR
systems used by two hospitals and 1,200 physicians in more than 50
groups (either already connected or in process). University of
Pittsburgh Medical Center, which has set up a connected network of 22
hospitals, 4,000 physicians, imaging centers, labs and others using
dozens of different health information technology systems. Citrus
Valley Health Partners in California, Baylor in Texas . . . that's
another 1.5 million patient lives, and the list goes on. In fact, while
it is clear there is still effort required, our clients demonstrate
every day that information exchange can lead to quantifiable and
demonstrable improvements in care delivery.
It is true, however, that today not all stakeholders in the
healthcare industry seem to be equally motivated to make information
liquidity a reality. While the money spent through HITECH and other
congressional investments have helped the industry to realize
measurable benefits from the rapid adoption of electronic health
records--an important success that shouldn`t be overlooked--clinical
data exchange is not where it needs to be. There are many factors that
need to be addressed for us to ultimately be successful:
We need to expand the standards development process,
building on the real progress underway with guidance from government
and allowing the private sector to continuously develop, adopt and
modify new standards;
Key constituencies, such as public health registries,
labs, State health information exchange organizations and others who
are not following available standards in their work, should be required
to do so;
State laws and regulations must be harmonized,
particularly those related to privacy and security, patient consent and
other similar topics;
Legal and liability concerns among providers about how the
data will be used outside of patient care must be addressed;
We need to get beyond the focus on how data is transmitted
and agree on what and how data is stored;
Activation strategies are needed to increase use of health
IT by patients and their caregivers, while also generating
accountability for their health outcomes;
We need a national patient matching strategy--a way to
identify each individual patient. This is a real challenge to both
robust data exchange and patient safety, and Congress needs to stop
blocking progress on this critical issue; and
Finally, generally, greater transparency around
interoperability and health IT among virtually all stakeholders must be
achieved.
Beyond all that, though, the sluggish progress we're discussing
today most closely stems from one critical deficit: the lack of a
strong business case or a true market driver for interoperability.
At the end of the day, healthcare in most environments is a
business where margins must be considered and the bills paid, and the
current payment system simply does not provide appropriate financial
motivation for providers to truly be invested in creating an
interoperable healthcare environment; this is especially true given
that the burden of cost falls to them almost exclusively.
Healthcare providers are genuinely committed to providing the best
care they can to patients, of course, but in many instances, the common
reality of running on only a few days' cash-flow often trumps loftier
goals. Much as CMS policy has already had a marked impact on hospital
re-admission rates by associating them with payments, creating a direct
relationship between payment and data exchange would have the same
result. This could be the strongest step taken to create a genuine
imperative for interoperability.
H.R. 2, the Medicare Access and CHIP Reauthorization Act of 2015
(MACRA), is a good start in the right direction, but Congress needs to
ensure that alternative payment models envisioned in this reform are
rolled out appropriately. The good news is that the expansion of
delivery reforms is already motivating accelerated electronic data
exchange progress. We see this in ACOs, and demonstrations like the
Comprehensive Primary Care Initiative, which Allscripts supports as the
technology provider for a very sizable percentage of the participants.
Simply put, they create a use case for health IT that focuses on
clinical value and less on what level of visit they can code. We have
already seen real change result within our client base from new
approaches at CMS and within the commercial payer space, and I expect
that will accelerate as MACRA is implemented.
For this reason, given the volume of new programs that have been
and are being rolled out along with Meaningful Use Stage 3, which we
expect to push the industry further in terms of interoperability, we
encourage Congress to allow the impacts of these recent changes to play
out further before additional legislation is passed specific to
interoperability. There is an opportunity to see what adjustments
providers make in response to the new payment models and what steps
they start taking to maximize the new revenue opportunities.
Generally, the same recommendation applies to standards development
work--it is important that there be time for maturation and the fine-
tuning of elements that are already being embraced by the industry (for
example, Direct and CDA), and there is no need to toss aside approaches
that are working. This doesn't preclude exploring new and innovative
approaches in an appropriately transparent manner, but the work done
with standards development is not intended to have a lifetime of 2
years or 5 but longer than that so it's important to move thoughtfully.
I do understand the eagerness of Congress, the Administration and
industry stakeholders to move rapidly because everyone is keen to see
the results, but looking to standards as a panacea for the challenges
still ahead of us will only result in usability complaints from
providers as immature technologies are mandated by the government.
Congressional attention would be best served in directing ONC to drive
greater standards adoption and consistency of implementation of those
standards, rather than focusing on the need for all new standards.
Further, it is important to witness what innovation comes from the
private sector, generally, in response to the recent legislative and
regulatory activity, as well as client requests. There are exciting
technologies and services in development now and on the product road
maps for the next several years based on what our clients have
requested of us, and I think we can all agree that we want to avoid a
prescriptive, heavy-handed statutory or regulatory mandate in which the
government becomes the de facto product manager for our industry as a
whole.
Another important consideration in this conversation about
information liquidity are the physician practices (small and large) and
independent hospitals who have been pressured to move off of their
current Electronic Health Record system--Allscripts in some cases--to
one used by the large enterprise health system in their area. Sometimes
the change is compelled through conversations about referrals, for
example; threats not to include people in data networks; or even just a
steady drumbeat of pressure, and it's often done under the auspices of
increased interoperability.
In actuality, with today's technology, changing systems just isn't
necessary in order to provide physicians and other medical
professionals with access to the information they need. The rip-and-
replace strategy emphasized by some in the industry is many years
outdated given the advanced data exchange capabilities that are out
there. Allscripts' dbMotion platform provides an advanced semantic
engine that aggregates and normalizes all clinical content across a
connected community into a single view, accessible within whichever EHR
the provider uses, to enable them to find relevant information quickly
while with the patient. This technology is in use across numerous
communities in the United States and overseas, including the entire
country of Israel, and in each environment, it's connecting dozens of
different vendors successfully and directly changing the care decisions
being made because of the additional information that's available.
Many people have termed what I just described--the pressure to
change systems--as data bullying; others, data blocking because one
involved party isn't committed to establishing connectivity between
current systems and in some instances, will even put up indirect
roadblocks. This raises what I believe to be a fundamentally important
issue--what, exactly, is the definition of data blocking? The ONC
report on information blocking stated that it occurs when persons or
entities knowingly and unreasonably interfere with the exchange or use
of electronic health information, but it also notes that the extent to
which such information blocking is impeding the effective sharing of
electronic health information is not clear because much of the evidence
is anecdotal and difficult to interpret. This is an issue that really
must be addressed before even implicit data blocking can be addressed.
An additional factor at play is the commoditization of data that is
occurring everywhere within the industry. Through our partnership with
our clients, one thing has become clear. Healthcare is mirroring a
trend seen virtually everywhere in business--attempts to access and/or
control data are driving many of the dynamics that are being discussed
today. The topics that are raised in the meetings I have with clients
every day are all about the power of data. ``Big data'', population
health, personalized medicine, quality-driven reimbursement and
information exchange--each a conversation about data and its enormous
potential. Until there is greater clarity regarding the so-called
``ownership'' of the data, this will continue to be a significant
factor in negotiations around interoperability.
I will note, too, that this committee's use of its oversight
authority has had important effects already in driving undesirable
behavior out of the industry, and we encourage continued attention in
the coming years as health information technology is used not only as
envisioned within the EHR Incentive Program but also for other
important purposes, such as population health and personalized
medicine.
Last, Chairman Alexander has said previously that the best way to
solve the problems around interoperability would be for the Health IT
industry to do something itself. I share the view that we have a real
responsibility here, along with the provider organizations that we
support, and I feel strongly that this is doable. I challenge all of my
colleagues to continue working together with us, with you, the provider
stakeholders, the ONC and the patient community that have so much to
offer in this conversation until we have achieved success.
Thank you again for the opportunity to be here today.
Senator Enzi [presiding]. Thank you.
We will go to Senator Murray.
Senator Murray. Thank you very much.
Dr. Kendrick, you have personal experience working with
providers in Oklahoma to make sure they are able to exchange
electronic health information. I regularly hear from providers
about information blocking, and HHS reported to Congress in
April that in many cases, they have limited tools to find out
when information blocking is actually occurring.
To make matters worse, we have heard testimony that in
order to use health IT, providers often have to sign contracts
that include gag clauses, which prohibit them from disclosing
technical problems or unfair pricing. Health IT is certified by
the Federal Government because it stores and transmits
information that is used to save lives, and users should be
able to report safety problems.
In your experience, can you talk to us about how easy it is
to address unfair and unreasonable barriers to information
exchange?
Dr. Kendrick. That is a big question. I like to refer to it
as ``death by 1,000 cuts'' because there are so many ways that
the free flow of information about patients for their benefit
can be blocked.
The motivations are various. On the provider side, as
several of the panelists have recommended, moving to value-
based payment models really does a nice job of aligning the
incentive of the providers in the patient's favor to make sure
that their record is available wherever and whenever it is
needed.
On the vendor side, however, there are lots of other
motivations at play, and to keep it to less than an hour, I
would say that perhaps the most critical thing we can do is to
get governance for the nationwide Health Information Network
correct because until we do, Congress will be dealing with this
issue year after year after year.
By getting governance correct, I mean getting a body that
includes those who receive care, those who deliver care, and
those who pay for care to help guide this network. They don't
need to have the technological knowledge. They can be supported
by vendors and so forth, but that is where you are going to
get--that is the eye--that is the right perspective to have in
this.
They will be able to set policies, because those three
perspectives matter. They will be able to set policies that
ensure that those contracts are transparent, that ensure that
the standards that we have today can be used and that we move
when we are ready to the next set of standards.
Senator Murray. Do you think we need some tools or
protections to ensure that individuals can safely report on
unfair market practices without fear of repercussions?
Dr. Kendrick. Yes, absolutely. It has only been within the
last year that we have begun to leverage ONC's surveillance
system, surveillance program for EHR certification. It has made
a world of difference.
In almost every issue we have filed, because we don't want
to waste people's time, and we really--we exhaust every
opportunity with the vendor and the provider to get the data
moved. Almost every issue we filed has been found to be
credible and has brought into question an EHR vendor's
certification.
Senator Murray. OK, thank you. That is very helpful.
Dr. Kendrick. Sure.
Senator Murray. Dr. Kibbe, let me turn to you. You
testified that it is important for providers to have good
information about how electronic health records worked in the
field. In particular, you noted that information from this type
of real-world testing could improve the purchasing decision of
providers.
Dr. Mirro also testified that ONC should support a
strengthened in-the-field surveillance program and that
technology developers must disclose additional fees they charge
for sharing information between providers. I was actually glad
to see the administration propose to include those two policies
in their 2015 certification rule, but what types of information
do you expect in-the-field surveillance to cover?
Dr. Kibbe. Yes, so thank you.
My comments would very much piggyback on David's comments.
The confusion in the marketplace now among providers and
provider organizations with respect to what their products are
supposed to deliver in terms of interoperable exchange and
particularly via Direct is a big problem, and there is no place
where provider organizations can go to see how a particular
product operates in the field.
I mentioned that some of them are very good. Their products
and features are familiar. People can begin to use them without
any training. There are others where they don't even have an
inbox for the messages to be sent to, and obviously, they can't
operate very well if they don't have those features.
That is the kind of information that the marketplace needs
to make better decisions, not necessarily because they want to
switch their products to another product. They don't
necessarily want to do that. They want their product to be
better.
Senator Murray. Right.
Dr. Kibbe. This would put pressure on the marketplace and
end that confusion.
Senator Murray. OK. I really appreciate that response, and
I need to go join Senator Alexander at Appropriations markup.
We are going to keep this hearing going, but I appreciate
everybody being here, and I will be back shortly.
Senator Enzi. Thank you.
Since Senator Cassidy also has to go to another meeting, I
will defer my turn for questions to him.
Statement of Senator Cassidy
Senator Cassidy. Thank you, Senator Enzi.
Dr. Kendrick, you are very explicit, saying that there are
companies which hide functionality in order to basically bill
for the same thing. You do not mention the company. Which
company is it?
Dr. Kendrick. This particular company was eClinicalWorks.
Senator Cassidy. eClinicalWorks. Mr. Kibble, you also spoke
specifically of companies blocking and imply that it was not
justifiable. Which companies are they?
Because I meet with these companies, and they deny it. They
are saying we are not doing it. It has been a little bit hard
to get people on the record so that when I meet with them I can
channel you into that discussion. Mr. Kibble, can you kind of
comment on who you see the offenders as?
Dr. Kibbe. That is Dr. Kibbe. But that is OK.
Senator Cassidy. I am sorry. I can't see your thing. I
apologize.
Dr. Kibbe. That is all right. Yes, I think that it is no
secret in the industry that there are two companies in
particular, electronic health record companies, very leading
companies, Epic and eClinicalWorks, which both have over the
course of the years, even before Direct exchange, developed
their own proprietary messaging systems.
One of the reasons why they have perhaps found it difficult
to adopt Direct for their customers is because of their
business model. I respect that business model, but I also feel
that it is important for them not to create the problems in the
marketplace that Dr. Kendrick has mentioned.
I would also mention that eClinicalWorks is a member of
DirectTrust, and we are working with them on this problem.
Senator Cassidy. I met with Epic, and they deny information
blocking. They say that they no longer require transfer fees
and that if there is an issue of information blocking, it is
because, oh, the hospital in Montana doesn't know that the
hospital in Baton Rouge is a member or da-da, da-da-da, and I
can make a very plausible argument.
What should I ask them next time if I am in the room with
them, and you were to give me the question to ask?
Dr. Kibbe. Yes, I would ask them what are their customers
telling them, asking of them? Because I know for a fact that
their customers are putting pressure on them to be more
transparent and easier to work with other vendors' products.
Senator Cassidy. That does not necessarily imply a business
model which is information blocking. It could just suggest that
the technology of their software is not meshed with the
technology of the others.
Is that a fair statement?
Dr. Kibbe. Yes, and I would not say that either one of the
companies that I have mentioned has a business model based on
information blocking. That is not what I am saying.
I am saying that they have in some cases implemented Direct
exchange in ways that benefit their customers, but not
necessarily exchange partners outside of their own customer
base.
Senator Cassidy. Dr. Kendrick, you are the guy, frontline
man, with all these FPs. What would you add to this?
Dr. Kendrick. It would be nice to have a metric of
interoperability, some way to measure, for example, in what
percent of a patient's visits was there a complete record
available from all sources. If we had that, then we could
actually put a number to it.
We don't right now. I would really encourage us to get to
that point because then you could really quantify. I can say
that, yes, sometimes the provider is user involved, but
generally when value-based payment models come to town, that
shifts.
With the vendors, we have noticed some change in the last
few months--probably as a credit to these hearings and as a
start.
Senator Cassidy. Let me interrupt because I am almost out
of time. Dr. Mirro, you spoke of something. I am a physician. I
still practice, and I have a young resident working with me now
who points out some literature that an intern spends 40 percent
of her time documenting and only 8 minutes per day per patient.
That just blows my mind, 40 percent documenting, 8 minutes
per day per patient. The patient should feel cheated because
the patient has been cheated.
You mentioned, Dr. Kendrick, let me bounce off the two of
you, that we should have no more standards. I am thinking we
should have a standard that the electronic medical record is at
least as time efficient as a paper record because, otherwise,
we will not have 15 minutes with a patient, we will have 6 or
7.
Dr. Kendrick. One option I have raised is that----
Senator Cassidy. Dr. Mirro, I haven't heard from you.
Dr. Kendrick. Oh, go ahead.
Dr. Mirro. Yes. Actually, well stated. I could tell you
that in the field so that the fundamental problem is these
systems all suffer from the fact that they are built on
administrative datasets. Their meaningful use has really been a
catalyst for charge capture revenue cycle management, what the
systems are based on, and not clinical data capture.
A lot of the systems then have to be morphed or redesigned
to collect clinical data.
Senator Cassidy. I am out of time. Let me just interrupt.
If we separated billing or charges from clinical data, that
could make efficiency and make it more work for the patient?
Dr. Mirro. Also user-centered design. I think the systems
are built by software developers, and they are easily usable by
a software developer, but they are not easily usable by
clinicians. We need to have more user-centered designed built
in the system.
The vendors do know this and are working for solutions.
This is not volitional, but they do know who their customer is,
and their customer is usually the CFO of the health system.
Senator Cassidy. Got you.
Dr. Mirro. It is not a patient. It is not a doctor. It is
the CFO. That is who they are selling to. That is why they are
designed the way they are.
Senator Cassidy. Thank you very much. I yield back.
Senator Enzi. Thank you.
Senator Franken.
Statement of Senator Franken
Senator Franken. Thank you, Mr. Chairman.
I am going to pick up a little bit where Senator Cassidy
started. Dr. Kendrick, you noted in your testimony that the
biggest obstacle that your members face in trying to share
patient data across payers and providers is liberating data
from their EHR systems.
I have asked a number of my constituents about information
blocking. Is it happening? Why? Who is to blame? I have gotten
very different responses, depending on whom I am talking to.
A common theme I picked up on is that those healthcare
providers who are using the same EHR don't think information
blocking is happening, even when they are part of competing
health systems, and the information blocking example No. 5 that
you have aptly named ``EHR at the center of the universe''
helps explain this.
If one EHR vendor is designing a technology that is not
completely compatible with other systems and using its market
clout to pressure providers to buy its product, doesn't this
raise some serious legal concerns about anticompetitive
behavior?
Dr. Kendrick. I am no lawyer. I am just a doctor.
Senator Franken. Humble.
Dr. Kendrick. I find it ironic that large vendors who claim
huge amounts of interoperability are primarily exchanging data
with themselves, with other installations of their own product.
There needs to be some truthiness to that.
Senator Franken. I want to ask you another question that
was kind of in your answer to Senator Murray. Because your
testimony also highlights the new healthcare payment and
delivery systems, those models, the success that they have had
in incentivizing the coordination of information and improving
health, the coordination of patient care, and in eliminating
market incentives for providers to hoard their patients' health
data because they are being paid for outcomes.
Based on this experience with the Comprehensive Primary
Care Initiative that you write about, a multi-payer medical
home model being tested by CMS's Innovation Center, how can we
focus more on these value-based reforms instead of just
compliance with meaningful use?
Dr. Kendrick. I am glad you asked that. Earlier, the
comment was made that perhaps we should delay Stage 3 of
meaningful use, and I want to make sure it is clear that there
are actually two programs.
There is the meaningful use program, which affects the
doctors, and there is the EHR health IT certification program,
which moves forward the technology. I would certainly say that
the technology needs to continue to move forward, but that the
rapid deployment of these value-based payment models may well
help to accomplish the things that were intended in Stage 3 of
meaningful use anyway but will do so with the full and vested
interest of the providers being onboard.
I would not want to slow down progress on the technology
side because the EHR vendors certainly have some ground to
cover, especially to support value-based payment models. If we
can get the accelerated adoption of value-based payment models,
the providers will be very much in support of interoperability.
Senator Franken. In Minnesota, we like value-based models
because we are good at them, and we kind of lead the country in
that. I want to keep incentivizing--I want to keep implementing
those as smartly, as intelligently, but as quickly as we can
and robustly as we can.
We were getting to talking about standards, and I think
there is some dispute here about whether the industry does its
own standards or whether the Government dictates these
standards, but can't we put in standards for operability,
interoperability?
What is the dispute here? I mean, why? Can I hear both
cases? I know I am out of time.
Mr. Black. I will take a swing at that. The Government has
put standards in place, which we have been talking about today,
about what is required inside of, if you will, a packet of
information that goes back and forth from an Allscripts system
to a Cerner system to an Epic system. Those standards actually
have a great deal of information that make the exchange of that
information possible and make that important to be consumed by
the caregiver.
What we are working on and what I think that the debate is
about is, there is an electronic exchange of information. If
people are blocking that, punish them. They should not do that.
The standards allow us, even if someone is trying to block
that, for us to auger in and get that data out anyway, and then
we can liberate or emancipate, or whatever the word is going to
be, to go get that data. We actually do that each and every
day.
We connect to over 350 different electronic medical record
systems in the United States. ONC certified a lot of systems to
be certified for MU2--MU1 and MU2. We connect to over 350 of
them in practice today.
That is an important thing to get the connectivity done.
Once you have them connected, however, you have to put the
context of the way it was connected in practice.
Senator Franken. Sorry to cutoff your answer, but I am way
over my time, and if there is a really short response to that?
Dr. Kibbe. Yes, one very short response is finish what we
have done before we do something new. I think that is what you
are hearing pretty loud and clear here.
We have got standards. We have infrastructure for the use
of those standards. It is working. It could be better. Don't go
off and do something entirely new until we have got that job
done.
Senator Franken. OK. Thank you.
Sorry. Sorry, Mr. Chairman.
Statement of Senator Enzi
Senator Enzi. It is OK. I would remind the Senators that
you can submit questions. These people have agreed that they
would handle ones in writing. That is part of testifying, and
that gives us an opportunity to ask even more technical
questions that might bore the entire audience. I am used to
doing accounting hearings.
[Laughter.]
I understand that. This whole topic fascinates me. I have
been on the High-Tech Task Force for a long time, and when we
first started talking about interoperability, we talked about
Australian railroads. How, when you get to the middle of the
country, you have to change from one train to another because
the railroad tracks are different sizes where they meet up in
the middle.
We wanted to make sure that with health IT that the
railroad tracks would meet, and people would have access to
information. I have kind of a science fiction version of what
we need to achieve in all of this, and I picture the day when I
would be able to have a card in my wallet that has every bit of
my medical information on it, every X-ray, every MRI,
everything.
That would be a lot of data, but I have watched what the
data transition has been. I would even have access to this card
so that I could record trips that I went on or bug bites that I
got or falls that I have had so that any doctor that is taking
care of me would have access to all of that information. It
wouldn't be a matter of them getting a hold of somebody to have
the record transferred.
I once fell down and had a bad ankle, and I got an X-ray.
They said there wasn't a problem, but a few days later, I was
in Wyoming, and I was still having a problem. So, I went to the
doctor, and they said, ``Well, we will have to do an X-ray.''
I said,
``Oh, no, no. I had an X-ray.'' They said, ``Well, it
would be too hard for us to get that one. We will have
to charge you for another one.''
I have a whole series of questions here that I am not going
to have a chance to ask. I will be submitting questions.
Dr. Kendrick, I was particularly interested in your
comments about the need for a lemon law, and I won't have you
expand on that at this moment.
Instead, I would like to know what you see as some of the
incentives now for information sharing across providers and
settings and how the market responds to that? How do you see
vendors responding to these market forces?
Dr. Kendrick. I think the message we get from our
providers, as we try to connect them--in fact, I was on the
phone with a physician group just yesterday kind of arguing
with their vendor about the $40,000 fee and the 9-month
timeline to get the interface built. I think the providers have
bought in. They are ready to do value-based payment models, but
they can't get their technology to keep up with them in moving
that direction, which is why I cautioned against slowing the
certification process.
Even if MU3 was to slow down, the certification and the
improvements in technology need to continue because I do think
that the vendors have a big job. They have got a huge job, and
they are critically important.
At the end of the day, they are serving those who deliver
care and receive care and need to meet their requirements.
Senator Enzi. Dr. Mirro and several of you mentioned the
need to have Stage 3 delayed. There is a security component
here. Do you think your patients understand what data you have
access to and what the protections might be for them and their
personal health information?
It is a difficult balance in some of these situations, but
what risks do you see?
Dr. Mirro. Certainly, the privacy and security concerns are
of utmost importance to patients and that we certainly do
everything we can to protect their personal health information.
A lot of my work is focused on delivering content to patients
and in a secure fashion, specifically from remote devices, as I
had in my testimony.
Patients are concerned about the privacy security. There is
the encryption and secured file transfer protocols that
certainly, as of today, seem to adequately protect the
patient's information and particularly when we transmit this
data to what we call a personal health record, where the
patient can virtually store all of their information and access
it.
Right today, we do have adequate security privacy. Could it
be better? Absolutely. I am not a cybersecurity expert. I am
just a doctor. There are people working on this, and I think
that everyone is concerned about the privacy security.
Dr. Kibbe. My concern is data hackers will have Senator
Enzi's collection of data before I will have it or the patients
will have it at this rate. Security is a big problem, and it
has to be dealt with on a very, very strong basis going
forward.
I do think we are doing a pretty good job in terms of
transport. Direct exchange is encrypted and identity validated
before the exchange can go back and forth. That is not as much
of a problem.
The issue around Stage 3 meaningful use, and I would agree
with Dr. Kendrick on this, is that we don't want to put any
barriers to innovation in health information technology in the
standards development. We do--and I am speaking for the
American Academy of Family Physicians now, we do want to delay
Stage 3 meaningful use until we get the merit-based incentive
payment system reorganized to go forward because there should
be alignment between those payment--those value-based payment
systems and whatever happens with respect to further
development of meaningful use.
Right now, I think your providers are saying, wait a
minute, there is almost no relationship anymore between many of
the objectives and metrics in Stage 3 meaningful use and the
payment systems that we are being asked to comply with and to
do well under.
Dr. Mirro. I just wanted to make one other comment. If we
had update the HIPAA Act and unique patient identifier, that
would help. Actually, not just the unique patient identifier
but have some two-factor authentication, such as facial
recognition or retinal scan. Patients, we will clearly know
that we are dealing with the patient. We have that matched
adequately.
The vendors spend a lot of effort and resources in
probabilistic matching, which is also a problem with any kind
of health information exchange.
Senator Enzi. Thank you. My time has expired.
Senator Baldwin.
Statement of Senator Baldwin
Senator Baldwin. Thank you.
I want to thank the Chair and Ranking Member for this
entire series of hearings. They have been very instructive.
I also want to say that as a fellow member of the
Appropriations Committee, I withheld going down there. Please
excuse me when I depart immediately following the questions and
I may submit some for the record in addition.
I absolutely share the Chairman and Ranking Member's
commitment to finding the way to address deliberate data
blocking. There are a couple of questions that I wanted to get
into, starting with Dr. Kibbe and Dr. Kendrick.
In your testimonies, you suggest that we need to increase
transparency for all stakeholders to help address some of the
problems that we see with data blocking and promote
interoperability. I wanted to get a little bit more granular
about how we create this transparency.
What exactly needs to be measured and reported by
healthcare providers, as well as obviously by EHR vendors, for
us to determine if data blocking is occurring or if progress is
actually being made?
Dr. Kibbe. I will take the first crack at that. One thing
that was very, very useful when e-prescribing was being
integrated into electronic health records, and what we are
seeing now is analogous to that in some ways. We are seeing
health--Direct information messaging and transport of files
integrated into electronic health records.
One of the things that would be very helpful, which was
done with e-prescribing, is for people to actually see the
software. Is it immediately easy to use? Is it familiar? Are
pieces lacking? Et cetera, et cetera.
This was done with e-prescribing largely because of the
industry's backing of that. I have been a proponent of getting
our industry to do the same thing. That is one example.
Senator Baldwin. Transparency with technological
capabilities and software is part of----
Dr. Kibbe. Yes.
Senator Baldwin. OK.
Dr. Kibbe. Being able to see what you get and what you
would buy if you used it.
Senator Baldwin. Dr. Kendrick.
Dr. Kendrick. There are three things.
No. 1, the section of contracts dealing with
interoperability needs to be transparent, at least that part,
because there are so many hidden things in those contracts.
No. 2, the process around surveillance should be
transparent, at least when a complaint is found to have merit.
Everybody is innocent before proven guilty in that model.
And No. 3 is, I alluded to earlier, we need a good metric
for interoperability, a measure or two, a number that we can
look at. We have got a couple of numbers we use in our health
information exchange, which are a little bit complex to go into
here, but I would be happy to share details on those.
Senator Baldwin. If you would, I would love to have you
answer that in more detail because obviously, transparency and
metrics is going to be helpful. I suspect in some of the data
gathering that you are talking about, we are going to be
comparing apples to oranges, and we will really need to figure
out a way to make the data that is, we hope, provided in a more
transparent way relatable to one another.
There has been a number of folks who have talked about how
a move to value-based payment models is going to improve the
very topic that we are talking about. I wonder if you could
just take a moment, given that there are acknowledged instances
where it is the healthcare providers that are unwilling to
share the data because of economic incentives.
If you could sort of walk through why our current fee-for-
service system discourages and other similar business models
discourage health systems from exchanging data, and whether in
our payment system reform that is ongoing and we certainly will
receive a lot of future attention, what Congress could be doing
to help address that aspect of data blocking?
Dr. Kibbe. In a word, risk. Because if you don't have risk
for patient outcomes, then in a fee-for-service model, we
duplicate tests. We don't coordinate care very well because we
are not responsible or held responsible for the costs that are
incurred when the handoffs are made poorly.
Senator Baldwin. In fact, sometimes rewarded for that?
Dr. Kibbe. We are sometimes rewarded. In most communities
in America, the healthcare system is multi-vendor. There are
Allscripts and Epic and Cerner and eClinicalWorks, and all
those different vendors are out there. There are people who
don't use electronic health records at all, like home health,
for example, or long-term care facilities may not have those.
What happens in a community that is starting to do
accountable care is they recognize they have got to connect
with all those people. Therefore, collaboration and
interoperability becomes a must, not a nice to have.
Dr. Mirro. Could I comment on that just for a second?
Because we are in a fee-for-service system, the systems are
designed around charge capture, as I mentioned, in
administrative datasets. Whereas, if we would go to a value-
based purchasing model, now the system is really focused on
clinical data capture. They become much more usable by
clinicians.
The nurses and doctors are all struggling on the usability
issues of these systems, and the reason is it is a reflection
of the payment model.
Senator Baldwin. Thank you. I will submit additional
questions for the record.
The Chairman [presiding]. Thanks, Senator Baldwin.
I am going to wait until the end. Senator Bennet will be
next.
Statement of Senator Bennet
Senator Bennet. Thank you, Mr. Chairman. I appreciate the
hearing very much and the testimony of all the witnesses.
It is obvious that there is a lack of adequate oversight
when it comes to the adoption of interoperable electronic
health records. Ultimately, it is the providers and patients
who suffer the most.
In Colorado, I have heard both rural and urban providers
complain about the cost, the time, and compliance issues they
have had with electronic medical health records and the
vendors.
Dr. Mirro, in your testimony, you discuss the lack of
governance around who is in charge of making sure that these
electronic health records are interoperable. You also discuss
our inability to take action against vendors who make
interoperability difficult for providers and patients.
To you and the other panelists, who should be in charge of
this process, and what additional steps should Congress take to
make that happen?
Dr. Mirro. I think ONC should provide that surveillance
oversight, and Dr. Kendrick gave some examples of that. We need
to have more transparency about the surveillance system and
which vendors are really on the watch list, if you will, so
that health systems and clinicians can connect. You see what is
going on.
The system that is in place just needs to be strengthened.
I don't think we need to develop a new system. More
transparency on the process and also exposure of the few bad
actors.
Because I think, far and away, the majority of vendors are
actually all trying to do the right thing. They are definitely
concerned about patient safety and personal health information
and protecting that and behaving the right way. We could do
more to just encourage that.
A surveillance system that is working and functional would
help improve that.
Senator Bennet. Does anybody else have a view?
Mr. Black. We are also getting together as a vendor
community where we are building our own standards for the
patient identification issue that is important for all of us.
Identifying the correct patient is a big piece of making sure
that as a clinician, I actually trust the source system that is
sending me clinical information about somebody.
If I have a doubt about the fact that it is really Bob
Jones, I probably won't administer the care that I think they
deserve based on the information I received from a source that
may not be 100 percent verified or trusted.
We are getting together through different organizations in
order to come up with a set of standards by which we can
identify at a national basis the patient, and we are doing that
independent of any regulations or, excuse me, independent of
any Government oversight.
We just think it is a problem that has to get solved. It is
a problem that has been punted down the road, and we are saying
it needs to be done.
Dr. Kendrick. I run a nonprofit health information
exchange, and we have tremendous success in addressing some of
these issues. Our governance is those who receive care, those
who deliver care, and those who pay for care.
Everyone else is supportive to those three types of
stakeholders. Those perspectives, I believe, are the ones that
matter and the ones that should be guiding these efforts.
Senator Bennet. On behalf of the providers in my State and
those that all of you represent who face a multitude of
requirements and compliance issues around electronic health
records, what do each of you think is the most important thing
for us to understand from the provider's perspective when
crafting and making changes to the current health IT policies?
Mr. Black. I spend a lot of time on the road with our
clients, and they are very concerned about the comment that was
made earlier in these proceedings about the amount of time that
they spend entering data into the system. Was that data
important about the care of the patient?
If it is administrative data, if it is data that be
compliant with a set of administrative things that were, if you
will, ``pushed down by the Government,'' that is where they
complain. It is not always apparent to them that these things
that we are collecting are important.
There is a lot of quality information that we are
collecting that are important. It may not be important to their
specific practice, but it is important to that population that
they serve both from a quality metric standpoint, but also from
a future research basis, which, depending upon who you talk to,
which constituent you are working with, that may not be as
obvious to them at that time.
Dr. Kibbe. Yes, what we hear is it has been too much, too
fast, with not enough time to reflect and to digest the things
that we have learned about what works and what doesn't work.
That is really the bottom line.
That, and focus on the issues that are really important and
are relevant to future value-based payment systems. Those are
roughly, in order of importance, interoperability, the ability
to report on outcomes and report on quality, and third,
security.
Those are the three critical issues that we need to pay
continuing attention to before we go off and do a whole lot of
other things that may be very worthwhile, but are too much, too
fast, and without enough resources and time to digest it.
Dr. Mirro. As a physician who does provide care to
patients, I can tell you the usability issues, as Paul pointed
out, are huge in every system. We have to get that right, and
it is because there is too much time in documentation to
satisfy meaningful use. It has deteriorated the quality of the
office visit for the patient.
We are spending more time with documentation. You are
turning very highly trained clinicians into clerical people,
basically. That is what happens.
One of the unintended consequences of this, shortage of
healthcare in rural America. I live in a rural area. The health
system I work for has seven hospitals, but we are very close--
we are in farm area, and the older physicians in these rural
towns are just retiring at a rapid rate. It is accelerated by
just this very fact.
We need to do usability testing in some way in the
certification process.
Dr. Kendrick. I would just add one thing. I completely
agree with my colleagues. The one thing I would add, though, is
in these new models of payment, physicians are increasingly
being judged, and rightfully so, by the satisfaction their
patients have with their care. It might be time to have
satisfaction scoring for the vendors.
Senator Bennet. Thank you, Mr. Chairman.
The Chairman. Thank you.
Senator Whitehouse.
Statement of Senator Whitehouse
Senator Whitehouse. Satisfaction scoring for the vendors,
they would love that.
Thank you all for being here, and thank you for this
hearing, Mr. Chairman and Senator Murray. I really appreciate
it.
A quick question to begin with. Dr. Kendrick, you mentioned
hidden things in the contracts. There were reports to ONC that
gag clauses in some of the vendor contracts prohibit the
provider from actually complaining about what is wrong in the
contract.
Can anybody on this panel see a single reason why those
contracts should have gag clauses in them?
Mr. Black. I cannot.
Dr. Kendrick. No.
Senator Whitehouse. Four noes. OK, let the record reflect
that.
Thank you very much.
Mr. Black. You are welcome.
Senator Whitehouse. I am concerned about the meaningful use
program having kind of outlived its original usefulness, and to
the question of how it might be redirected, there are two
things that I hear a lot from my home State. One is that there
has been precious little support for health information
exchanges compared to support for and testing of doctors and
whether they have got the equipment running on their desk
right.
That makes it kind of a bank shot, to use a pool metaphor,
at supporting health information exchange when, in fact, if you
have really vibrant health information exchange, it becomes an
almost inevitable part of anybody's business model because it
is working well.
I think we have got it upside down between the extent to
which we support health information exchange versus the extent
to which we put mandates and responsibilities and equipment and
credits and everything onto doctors' desks.
The second is that two of the key transactions in the
healthcare equation, one is the loop between skilled nursing
facilities and hospitals, which is a very unproductive loop.
The second is patients who have significant behavioral health
issues for whom necessarily their behavioral health provider is
their medical home because it is their behavioral health
provider who makes it possible for them to negotiate the rest
of the system.
Leaving out behavioral health and leaving out skilled
nursing facilities seem like very significant oversights.
Starting with Dr. Kendrick, I would like to ask you to comment
on--I know you run an HIE, so I am particularly interested in
yours. We have got a really good one in Rhode Island. Laura
Adams runs CurrentCare very, very effectively, and we are
pretty much out front with you guys on this.
I would be interested in the panel's thoughts on those two
questions. If you could just keep them brief and fill in later,
I have got another question I would like to fit in, too.
Dr. Kendrick. OK. I will try to be brief on these. One,
transitions of care are critical. Agreed.
Senator Whitehouse. Particularly those skilled nursing
facility and behavioral health omissions.
Dr. Kendrick. Yes, particularly those. One thing we need to
recognize, at least as a clinician, when I write an order to
transition a patient from one place to another, that is
generally the action that makes it happen. That is an order.
Currently, we don't track that as an order, right? We fax
something somewhere or we send a secure email somewhere, but I
don't know what happened to it after it got sent. I don't know
that it got--the patient showed up. I don't know that the loop
got closed and the record got back to me.
I would recommend that we shift our thinking around
transitions of care to start thinking about them as an order
that gets tracked with a status, just like a prescription or a
lab test or anything else.
Senator Whitehouse. Should we move our focus more toward
health information exchange support, as opposed to----
Dr. Kendrick. That would be the appropriate clearing house,
so to speak----
Senator Whitehouse. For that?
Dr. Kendrick [continuing]. For those orders, I would
suggest.
Senator Whitehouse. Got it.
Dr. Kendrick. Your second question, your second question
was about--remind me.
Dr. Mirro. Behavioral health.
Dr. Kendrick. Oh, behavioral health. On behavioral health,
we have actually----
Senator Whitehouse. Why are they outside of meaningful use
when they, in fact, are the medical home for----
Dr. Kendrick. Right.
Senator Whitehouse [continuing]. That set of patients?
Dr. Kendrick. We have most of the community mental health
centers in our State connected or connecting to the health
information exchange, despite the fact that we have to jump
through incredible hoops to meet the provisions of 42 CFR that
permit--that prevent certain data from moving----
Senator Whitehouse. We are doing it, too, and it is a
nightmare, and it is not supported by the meaningful use
program.
Dr. Kendrick. It is really--if you could fix something,
please fix that.
Senator Whitehouse. OK.
Dr. Kendrick. Because those patients are getting cheated.
Senator Whitehouse. I am about to be out of time. Let me
offer two things. One, I share the experience that Senator
Bennet had of going around my State and having doctors'
offices, having hospitals, having skilled nursing facilities,
having people say, ``Oh, my God, what a nightmare to go to
electronic health record.''
If you ask them the next question, ``Would you go back?''
They all say, ``Oh, my God, no. Thank God, we got through
that.''
Mr. Black. Right. Right.
Senator Whitehouse. ``But this is really a big, big
improvement for us.'' While it was painful, it was--I am
getting a lot of nodding heads from the panel as well.
The second thing is that I heard what Mr. Black said about
the quality reporting and the burden of that. That is really
important, and I would suggest to you that it has a patient
component as well.
If we load up the system, as I think we have, with a whole
broad array, a multiplicity, a babble of quality reporting,
then it creates a huge burden for the doctors who have to meet
the regulatory burden. Also it blunts or blurs any kind of a
public signal that doing badly on the quality reporting ought
to on occasion.
Somebody comes up with a really bad quality report on
something, and the next thing they are saying, ``Yes, but I got
these other five really good quality reports.''
Mr. Black. Right.
Senator Whitehouse. The whole thing just turns into a mess.
I would encourage you and ask it as a question for the record
to make recommendations on how we could simplify the quality
reporting so that it has real consumer impact, in addition to
any questions about undue regulatory burden.
Mr. Black. There are a lot of quality ratings that are out
there today. There are star ratings that are out there that are
actually pretty effective at being able to do that. You can
rate the institution. You can rate the provider. You can rate
the organization.
There is actually a fair amount of that data that are out
there, and that data wouldn't come across again had you not
done the broad mass adoption of these systems. That is what MU1
did.
While it was tough, while it was hard, while $30 billion
was spent, there is no way we could have these conversations
about any of the rest of these things, including HIEs, if
they--they have to connect to something. They have to connect
to the electronic medical record that is in that doc office or
that hospital.
To me, that was a great amount of money that was spent, a
lot of time and effort. But you hit the promise of this, which
is that you can interconnect these. You can get data. You can
get quality information. And you can rate them.
Over time, because of a consumer focus, there will be a lot
of transparency on these data that are going to force
additional changes in behaviors from providers and
organizations in order to become more relevant to consumers.
Senator Whitehouse. Thank you, Chairman.
The Chairman. Thank you, Senator Whitehouse.
Mr. Black, who owns my personal healthcare information?
Mr. Black. You do.
The Chairman. Thank you.
No. 2, is there a consensus or are there surveys about the
actual percentage of time that physicians spend on
documentation? Anyone know of that?
Dr. Mirro. I could probably answer that. Actually, I am
part of a Federal grant that is coordinated by the University
of Nebraska, and there are several sites. Duke University,
University of Nebraska, Christiana Healthcare in Rhode Island
is participating also.
The Chairman. Do we know what the percentage is?
Dr. Mirro. Yes. It is, at least from that experience, the
data that we have, preliminary data because we are in the midst
of this grant, we are spending well over 15 minutes per office
visit. We have a 15-minute office visit. There is quite a bit
of variation between individual clinicians, right?
We have some high users that maybe will only spend 5
minutes documenting.
The Chairman. So 30 percent?
Dr. Mirro. At least 30 percent.
The Chairman. Do any of you know of any other surveys,
reputable surveys about the time doctors spend?
Dr. Kendrick. I can't give you anything reputable, but I
can tell you my department, my academic department runs the
electronic health record system for the university. The
majority of notes are signed between 10 p.m. and midnight.
The Chairman. Yes.
Dr. Kendrick. That is the busiest 2-hour window for
documentation.
Dr. Kibbe. It is a real problem.
The Chairman. Mr. Black, of the people you work with, do
you have any idea?
Mr. Black. It depends on the specialty type. In some cases,
people will say they are actually faster with electronic
medical record, and other people will say it takes them more
time because of the charging of the other documentation that is
provided, and they actually are doing it versus, as was said
earlier, perhaps a clerical staff.
The Chairman. There has been some discussion about delaying
meaningful use 3, about whether it is a good idea, whether it
is a bad idea, whether to delay part of it, whether to delay
all of it. I would like to ask each of you if you would
summarize for me what you think the answer to that question is.
Assume you were Secretary Burwell and you had it within
your power, and you are looking ahead, and you are saying,
``OK, we have two big rules coming down in September or so,''
which are going to establish the rules of the game, which will
be implemented, which will go into effect, what, in a couple of
years, something like that?
Mr. Black. Correct.
The Chairman. You have Mayo Clinic, as I mentioned earlier,
saying, ``all right, we are going to spend $1 billion over the
next 4 years, 500 employees, to finish the job.'' You have
others saying, ``we are terrified by the prospect'', and you
have half the docs not--paying penalties rather than
participating.
What would you move forward with, and what would you slow
down in not to stop electronic healthcare records, but to gain
the largest amount of buy-in and most effective use of it so
that providers look forward to using it rather than dreading
it?
Then if you--and then after you are through, if you have
time, if you would send me your written suggestion for that,
that would be helpful to me. Dr. Kendrick, can we just go down
the line?
Dr. Kendrick. Sure. I indicated earlier that meaningful
use, in my mind, is separate from the EHR certification
program, and I very much would recommend against slowing down
the EHR--the advances in the EHR health IT certification
program.
There are some--there were important gaps----
The Chairman. Because vendors need to be--and others need
to be able to rely on that and go ahead and do their work on
that. Is that right?
Dr. Kendrick. Because we still--we have all heard we need
improvements in workflow, and we need improvements in other
components. In particular the next round of certification
requirements, as proposed, fix some specific holes in
interoperability that were in the last round.
The Chairman. OK.
Dr. Kendrick. That I would proceed with. I tend to agree,
though, that when major programs through MACRA and others are
coming down the pike, it would be great if we could synchronize
those efforts, have one set of measures, have one approach.
I would only add one additional item, and that is these new
payments models, this approach we are taking depends heavily on
assessing value correctly. That means measuring value
correctly. Our current architecture will measure value in a
single practice, and the same patient can appear in several
different doctors' measure, right?
What we care about as a nation is what is the health of an
individual, and is it being improved? To do that, we have to
measure at a higher level with the patient's complete record
available.
I recommend that that needs to be a part of things going
forward, or we will miss the boat on value-based payment, and
we will be arguing over measures that aren't valid.
The Chairman. Dr. Mirro.
Dr. Mirro. As I testified, we want to delay Phase 3
meaningful use in its entirety. Phase 2 only----
The Chairman. Does that mean including the certification
standard? So you disagree with Dr. Kendrick on that?
Dr. Mirro. I actually do agree with Dr. Kendrick on the
certification side. On the eligible provider side, I am
referring to, we definitely need to delay Phase 3 meaningful
use. Phase 2 we only have 11 percent of eligible providers
attesting right now.
We have to actually analyze the impact of Phase 2
meaningful use on improvement and care and its impact within
the efficiency of the health system because we have created
incredible inefficiencies, as we have just discussed, with
usability and workflow issues. We really have to digest the
impact of Phase 2 meaningful use and continue to strive to
simplify the requirements.
The Chairman. Dr. Kibbe.
Dr. Kibbe. Yes, we would recommend an immediate moratorium
on Stage 3 until Stage 2 is fixed. I agree with my colleagues
in the sense that there are parts of Stage 2 that need to be
fixed and improved so that more providers and eligible
hospitals can participate in Stage 2 before we go on to
anything called Stage 3.
In particular, the certification needs to be focused on and
made better. Interoperability needs to be focused on and made
better. The quality reporting we have talked about needs to be
made better, and the security side of Stage 2 also needs to be
fixed.
The Chairman. Fix Stage 2 before we move on to Stage 3?
Dr. Kibbe. Indeed.
The Chairman. Mr. Black.
Mr. Black. It is important that there is a bunch of
information that is going to be gained out of these electronic
medical records that are going to be further extended by MU3. I
would suggest that we keep on the path we are on. We think it
is going to be in the 2018 timeframe. That is a long time from
now.
I need time to prepare for that, and so the sooner that we
can get the specifications. We have to be done with those and
get certified prior to these folks rolling them out. For me, it
is important to get advance notification.
There is a lot more patient information than patient
benefit, the consumer benefit as a result of some of the
regulations that I have seen around MU3. I also think from an
interoperability standpoint, there is a lot of API exposure
that we welcome the rest of the organizations to be exposed to.
The Chairman. API exposure?
Mr. Black. Yes, sorry. There are requirements in there for
all vendors to expose more of their application programming
interfaces. Allow third parties to innovate on top of platforms
to allow other people to be able to go in and extract data out
of their systems. That is what we do today.
The Chairman. Build that for----
Mr. Black. That is a piece of the proposed legislation that
we would support as well. We think it is good for America to
have access at a detailed level to electronic medical records
versus being reliant upon a specific vendor to give that to
you.
The Chairman. Thank you.
If each of you could, even if it is just 1 page, just say
if I were Secretary Burwell, here is what I would do this fall,
that would be helpful to me and to other Senators.
Senator Cassidy wanted to ask another round of questions,
and he knows a lot about the subject, and I thought that was
good. Why don't we--I will go first to Senator Murray and see
if she has questions, then to Senator Cassidy and if Senator
Whitehouse has some. Then we will conclude the hearing.
Except, Senator Enzi, you may have----
Senator Enzi. No, no, I have learned as much as I can
handle right now.
[Laughter.]
Mr. Black. Any accounting questions?
The Chairman. All right. We will start with Senator Murray.
Senator Murray. I would just submit any additional
questions, but I do really do appreciate this panel and the
information you are providing us. It has been a lot. We
appreciate it.
The Chairman. Thank you.
Senator Cassidy.
Senator Cassidy. Mr. Black, when we spoke earlier, and you
mentioned how you all have apparently your product, semantic
interoperability.
Mr. Black. Yes, sir.
Senator Cassidy. Wherever you go as a patient, there is an
integration of the blood pressure in a standard format. You
also mentioned in an earlier conversation that one person may
mention weight in kilos, and the other person in inches, and
the other person in feet and inches. How can you get--I am just
asking, how do you get semantic interoperability?
Because we are told we have to have a standard----
Mr. Black. Right.
Senator Cassidy [continuing]. That everybody use the same
standard before we can have that and yet, apparently, you have
accomplished. I am asking how can that be?
Mr. Black. Yes, even inside of--we attached, as I said,
over 350 different electronic medical records today. Even
though you may have one client who has one electronic medical
record vendor, they actually may have different instances of
that.
Let me give you an example. We have one client, very large
client, who has 44 different instances of the same electronic
medical record vendor established across 186 different
hospitals. You would think that by us plugging in and trying to
determine a male/female status, that would be a very easy
interface.
In fact, because of the flexibility of the system, there
are 25 different ways in which as you plug in, you read that
information, it actually comes back to us in a different
manner. We have to take that information and harmonize that
information to make it say either male or female.
It sounds very easy, but I am just telling you, the
information systems the way that they are out there, the way
that they are designed, the way that they are implemented can
vary widely even inside of just one.
Most places I go to have anywhere from 300 to 700 different
systems. Our largest client today has 100 electronic medical
records----
Senator Cassidy. With that said, are you able to harmonize
without--apparently, you are.
Mr. Black. We are.
Senator Cassidy. Because you have semantic
interoperability. There is some way to do a workaround aside
from having everybody sit down and say we are going to call it
meters, or we are going to call it----
Mr. Black. Right. We take it, as I say, God is on the side
of computers, and we are going to figure it out. There are
algorithms you can say to read this SNOMED file and convert it
to CPT. Take the CPT file, convert it to ICD-9. Take this ICD-
9, convert it----
Senator Cassidy. Got you. How do you define data blocking?
Because you mentioned in your testimony, we must, and so how
would you?
Mr. Black. I would define it by somebody who knowingly is
out blocking the data. The definition of that would be very
difficult to enforce, but I do believe that there is very--a
lot of people are not knowingly doing it. But to the extent
that we had practices in the past, which this committee
actually because of the questioning that you had, actually got
some suppliers to change the way that they charge for that.
That is extraordinarily important.
The light that you are shedding on this has actually been
very well received----
Senator Cassidy. Let me ask each of you if you can submit
for the record your definition of data blocking because that is
important to us going forward.
Next, on the House side, there is a bill, part of 21st
century is a portion which says that if someone is found to
data block, that they would be decertified. If you are a doctor
and you just paid all this money and now your product is
decertified, it is actually the physician that loses.
If it is a small practice in Iowa or in Bunkie, LA, you are
stuck. Is there some way to enforce a you can't data block and
if you do, you are busted without busting the doc? Dr. Kibbe?
Dr. Kibbe. Yes. Part of the problem there is we are kind of
in an all or nothing State right now, right? We need a better
certification program so that in-the-field usability of these
products is testable. If there are problems, they are available
early, and the vendors and the users of those products have a
chance to fix it.
Senator Cassidy. That doesn't help those who have already
purchased?
Dr. Kibbe. No, it would because it would expose the vendors
whose products are having problems or in usability early enough
that they would have a chance to correct that----
Senator Cassidy. OK. Let me go back to Dr. Kendrick. Dr.
Kendrick, you mention that there are these gag orders. I had to
press you to tell me that it was eClinicalWorks that is getting
certified with a package which they then pull out and hide and
then charge $10,000 in order to have access to.
If you are under a gag order, you can't really say, ``Oh,
my gosh. I am paying 10K a year for something which I should be
getting for free.''
How do we--you could have a gag order. You are doing a beta
test, and you can't report that it is a dog. Do you follow what
I am saying?
Dr. Kendrick. Yes, in my written testimony submitted, it
says that we are finding a fair amount of almost intimidation
among the folks we work with, the practices we work with, that
they are afraid to file these things themselves. Maybe because
of a gag order, maybe because of experience. Certainly when we
filed one complaint recently and it got forwarded to the
vendor, we got a note almost immediately from the vendor saying
that they wouldn't work with us anymore.
Thankfully, an executive at the corporation decided it was
worth that this really could be an issue. I would say in terms
of--transparency has got to be there, and in terms of dealing
with the next----
Senator Cassidy. Should we outlaw gag orders? Should we
have legislation that says, ``my gosh, if you have got a dog,
you can say I heard it bark''?
Mr. Black. Please.
Dr. Kendrick. I think we should.
Senator Cassidy. OK.
Dr. Kendrick. I would just add that in terms of dealing
with the decertification, if you take away that threat, then
there is no reason for an organization to comply, and I would
suggest that perhaps the EHR vendor that becomes decertified
should be compelled to reduce their prices to their practice--
--
Senator Cassidy. How do we hold the doc harmless if she has
bought a product which is now decertified? How do we keep her
from being punished? Do we have a money back guarantee?
Dr. Kendrick. That is where I was going is maybe the EHR
vendor has to reduce its fees or charge no fees until they are
certified again.
Senator Cassidy. We would allow a grace period for the
physician to continue with the decertified record?
Dr. Kendrick. We have to make the data portability real. I
need to be able to strike my data and move on to another
vendor.
Senator Cassidy. OK. Thank you for allowing a second set of
questions.
The Chairman. Thank you for your participation.
Senator Whitehouse.
Senator Whitehouse. Senator Cassidy, put me down as a
cosponsor----
[Laughter.]
Senator Whitehouse [continuing]. On your legislation to
restrict these gag orders, and we will work together on that. I
am having a very enthusiastic agreement from the entire panel
on that point.
I had just one specific question since I get another round,
and this has to do with Dr. Kendrick's written testimony with
respect to the part of ONC's certification program that
requires data portability. The quote I have from your testimony
is that few vendors appear to offer this functionality as
intended, and I would like you to have the next couple of
minutes to elaborate on what you meant by ``few vendors appear
to offer this functionality as intended.''
Dr. Kendrick. The reason we came across this functionality
is because it is another option for interoperability.
Unfortunately, Stage 2 of meaningful use required that the
standard document be created and that the provider be able to
export it, but not that the exporting of that document be
automated.
Many vendors buried that export three buttons down, three
menus down, and force a manual process. Until the next round of
EHR certification comes out and hopefully fixes that, we are
stuck with other options that are written into the policy. One
of those is data portability.
Senator Whitehouse. Just so I understand it, what we would
like to see--what CurrentCare has accomplished in Rhode Island
is that the record is kind of automatically populated with
data. If you go out and have a CAT scan or if you pick up a
pharmaceutical or get a test of some kind, it loads more or
less automatically----
Dr. Kendrick. It moves where it needs to go automatically.
Right, right.
Senator Whitehouse. Exactly. What you are saying is if that
doesn't happen, somebody has to page through several----
Dr. Kendrick. Right.
Senator Whitehouse [continuing]. Screens and then hit
``send'' before it----
Dr. Kendrick. What happened is, we discovered the data
portability clause, which is the requirement that a batch
export be possible of these CCDA files, these standard files,
and we can do that on a nightly basis and use it.
Unfortunately, when we have gone to now several vendors to
say where is your batch export functionality, they can't
demonstrate it to us, and many have to come in and actually
rewrite code to make it happen. We filed a complaint, and the
first one of those has been found to have merit.
It did require some pretty significant conversation between
ONC and the certification body because, remember, the
certifying body tests, ONC writes the test. The question was
whether the test adequately tested that functionality, and so
things slipped through the door.
Senator Whitehouse. Got it. OK.
Thank you, Chairman. It is a technical issue, and it can
drive people into a state of stupor.
Dr. Kendrick. Sorry.
Senator Whitehouse. It is also really, really important,
and the panel has been terrific.
The Chairman. We need to see if we can come to some
consensus here on the committee about what we think about this.
Senator Murray and I had breakfast with Secretary Burwell
today, and we talked a little while about it. Of course,
Secretary Burwell and the President are very interested in
changing the way physicians are paid, value-added, and you are
saying there is a relationship between that and better use of
this.
Those are relevant things, but there are big decisions to
be made. If any adjustments are to be made, they probably need
to be made this fall, and I don't want to recommend adjustments
that are not the right adjustments to make. We want to complete
our work on this by this fall and understand that there are
steps. If there are four or five steps that can be taken to
make the electronic healthcare system work better and if they
require legislation we can deal with that early next year.
If it is something the Secretary can do, we can talk with
her about it and let her do it. That would be much more
preferable to do it that way because it could get done more
rapidly.
Let me ask, this is an off-the-wall question, and it may
sound like a really stupid question. I am old enough not to
worry about things like that. Should we be defining--are we
being too ambitious with interoperability?
Should we say that we should have interoperability to just
an extent? If what we are really talking about is making sure
that information, let us say about me, is available at all the
places where I might have gotten healthcare.
What if a rural doctor somewhere where I might have gone
after a fishing trip says I will send you--if you are going to
go see your doctor next week, I will fax it to him? I will
FedEx it to him. I will keep it according to a certain standard
that you tell me to keep.
Why in this age of communication where we have many
different ways of getting information to people, when the usual
time that you need information about your healthcare, I would
assume, is a scheduled appointment with a physician or a
hospital, where a doctor in some other place would have time
either to fax the information he or she has or to Fed Ex it by
the next day or UPS it?
Or maybe, maybe we just say that interoperability ought to
apply to this much information and these kinds of data, and it
is too ambitious to try to go beyond that. It is just we just
can't do that in a system as big and diverse as ours. Is that a
really ridiculous question, or is there an easy answer?
Dr. Kibbe.
Dr. Kibbe. Let me--there is an easy answer. Because doctors
and the hospital systems and medical practices don't want to go
outside their electronic health record to exchange health
information. They want to work within that electronic health
record interface.
The Chairman. Oh, I don't know that is true. I know some
very good doctors who keep a separate drawer filled with
written information about their patients.
Dr. Kibbe. They may do that. When the information flow is
required to go from one place to another, you really don't want
to go to a separate Web browser over here or go to a fax
machine over here. That is inefficient. That is extra
documentation. That is extra cost for everybody.
One of the things that I meant in my fixing of the Stage 2
meaningful use is that we have an infrastructure, we have
standards, and we need to fix those and make them a little
better.
For example, a Direct message ought to be acceptable with
any kind of attachment, not just a CCDA. A PDF document, a Word
document, a file image. Those ought to all be acceptable means
of transfer of documentation from within the electronic health
records' current standard for exchange.
Mr. Black. I don't think you spent $30 billion, we did. You
have to take it to the next level. Otherwise, you are going to
have just a great, big, huge community and country full of
silent information that is not expandable, and it doesn't
really benefit the patient.
In order to get that information to an interoperable manner
through an HIE, whatever it has to be, in order for it to be
used, though, by a clinician, you have to have all the
information about that patient, irrespective of which EMR that
they are in across the community.
Then when I log in to my patient and my patient is in front
of me today, I want to see that there is additional information
in the community about that patient that is going to change my
mind. At the University of Pittsburgh Medical Center today, we
sit on top of Epic and Cerner, and when they click on the
community view, that clinician, 60 percent of the time, makes a
different clinical decision because of the fact they saw the
totality of the patient's record, not just what was inside of
EMR 1 or EMR 2.
The Chairman. OK.
Mr. Black. We now have 100 of them attached to that, I can
promise.
Dr. Kendrick. The issue--I would zoom out just a little bit
and say, or a lot and say that this--the health of America and
the success of our industries in the world depends on solving
this issue appropriately. If I were to compare healthcare to
banking, it is about 1965 in healthcare.
We just don't have electronic transactions like we need
them. We have lots and lots of labor force dedicated to things
that haven't been done manually in banking for decades, and we
won't be able to compete on the world stage as a nation and
have a healthy workforce if we don't solve this.
The efficiency gains from interoperating electronically far
exceed the paper processes.
Dr. Mirro. I agree with my colleagues wholeheartedly, and
we cannot have enough interoperability. Remember, we are trying
to have a patient-centric electronic record. It is actually
about the patients, and that is we want to get to that stage
where the patient is holding all their information, and we have
patients that are using a highly functional personal health
record so that their data is stored in a PHR, and that could be
accessible from any site, even rural America.
The Chairman. I thank all of you for coming. I would ask
once again if you could send me in 1 or 2 pages of what you
would do if you were Secretary Burwell if faced with decisions
about these two rules that come down in September about
electronic healthcare records. I would appreciate it very much.
The hearing record will remain open for 10 days. Members
may submit additional information for the record within that
time.
Thank you for being here. The next HELP Committee hearing
will be Wednesday, July 29.
The committee will stand adjourned.
[Additional Material follows.]
ADDITIONAL MATERIAL
American College of Cardiology, Heart House,
Washington, DC 20037-1153,
June 16, 2015.
Hon. Lamar Alexander, Chairman,
HELP Committee,
U.S. Senate,
Washington, DC 20510
Hon. Patty Murray, Ranking Member,
HELP Committee,
U.S. Senate,
Washington, DC 20510
Dear Chairman Alexander and Ranking Member Murray: The American
College of Cardiology (ACC) is a 49,000-member medical society that is
the professional home for the entire cardiovascular care team. The
mission of the College is to transform cardiovascular care and to
improve heart health. The ACC leads in the formation of health policy,
standards, and guidelines. The College operates national registries to
measure and improve care, provides professional medical education,
disseminates cardiovascular research and bestows credentials upon
cardiovascular specialists who meet stringent qualifications. The ACC
also produces the Journal of the American College of Cardiology, ranked
No. 1 among cardiovascular journals worldwide for its scientific
impact.
The ACC has a vested interest in complete interoperability of
health information technology not only because of its diverse
membership of cardiovascular care team members including physicians,
nurse practitioners, nurses, and practice administrators, but also
because of its operation of five hospital-based, one outpatient, and
two multi-specialty clinical data registries.
The College would like to applaud you and your respective staff for
taking the initiative and working to accomplish specific goals related
to interoperability of EHRs. The College appreciates the opportunity to
provide input and encourages you to address these pertinent issues.
The ACC views the following as key priorities that should be
addressed related to EHR interoperability:
vendor data blocking
Issue: The ACC has been on the record with the Senate HELP
Committee in bringing the issue of ``vendor data blocking'' to the
forefront and the College is appreciative of the committee's
responsiveness and eagerness to address this issue. The ACC views
vendor data blocking as one of the largest barriers to EHR
interoperability. EHR vendors charge exorbitant fees to transfer data
from hospital to hospital or hospital to physician office, undermining
the very purpose of EHRs. Many times, hospitals are in a better
financial position to incur these costs. Physician practices, which are
typically smaller and have fewer resources, are not in the position to
absorb these costs.
Example: For each patient, cardiologists are often required to
reference several tests to obtain a complete understanding of a
patient's condition. These required tests are sent to various labs,
each of which operates its own separate EHR system, often administered
by different vendors. In order to fully exchange information, EHR
vendors charge physician practices upwards of $20,000 to fully
interface with each lab's EHR system. While this is usually a one-time
fee, many physician practices cannot absorb these unexpected startup
costs. In order to provide appropriate and effective levels of care to
their patients, these providers face fees to interface with necessary
ancillary systems to facilitate the transfer of data between settings.
Once the connection is established, there are often additional charges
for the exchange of information. The College feels these exorbitant
fees must be brought under control.
Solution: The ACC acknowledges that an initial fee to establish a
connection could be appropriate. Our concern lies with the amount of
fees these vendors have arbitrarily established. Perhaps a solution
could be for vendors to work these fees and others into the initial
agreement signed with physicians, including (but not limited to)
bundling open application programming interface (API) costs into the
overall maintenance fees. This would require vendors to be up front and
transparent with their pricing both at the time of purchase and
throughout the use of the implemented EHRs and the peripheral elements
included in these contracts. Additionally, it would be ideal to know up
front the costs associated with purchasing interfaces to exchange with
another vendor's EHR. Penalties should also be established for vendors
whose actions prohibit the exchange of data under any circumstances,
which leaves the practice without options to solve the problem. The ACC
looks forward to working with the committee to determine the most
appropriate way to address this issue.
effective ehr standards
Issue: The Office of the National Coordinator for Health
Information Technology (ONC) has attempted to establish effective
common EHR standards since the passage of the Health Information
Technology for Economic and Clinical Health (HITECH) Act in 2009. The
Certified EHR Technology (CEHRT) criterion (2011, 2014, and the
proposed 2015 criteria) aims to set a floor for certification to avoid
stifling innovation while still working to require EHRs to meet the
specific needs of clinical settings. With the current EHR standards in
place, clinicians not only face continual challenges exchanging the
simplest elements of data between EHRs--that have all met the EHR
standards in place at the time of their certification--they also face
basic usability issues. Despite these issues, there is still a widely
felt concern that if effective common EHR standards were to be
established, they would be too prescriptive and would stifle
innovation.
Example: As a part of certification, EHRs are tested to meet
varying criteria and specific definitions. The criteria are tied to
components of the Meaningful Use program such as computerized order
entry, secure messaging, and e-prescribing. Definitions address other
items of the Meaningful Use program as well, such as how a ``base EHR''
is defined, along with other items such as how patient health
information is captured and how to import, calculate, and report
clinical quality measures. This is in addition to base requirements
relating to privacy and security, accessibility-centered design, and
safety-enhanced design. Once the EHRs are certified and implemented,
many times data received by a certified EHR from other certified EHRs
populates in inappropriate fields or the data is received in a format
that is unusable. For example, a clinician may receive a chart mapping
a patient's blood pressure rather than individual data points. Another
example is that clinicians in the outpatient setting frequently refer
their patients to a hospital across the street from their office for
procedures. The inpatient setting, however, often uses a different EHR
and the different systems cannot communicate. When patients are
admitted to the hospital, clinicians have to print out their notes and
send a copy to the hospital so the notes from the clinic can be
incorporated into the hospital's electronic records for the inpatient
setting. This information is often scanned and inserted into the
hospital's EHR as a PDF and is therefore far less usable. Thus, in
order to truly achieve health information exchange these providers and
their small clinics are forced to incur additional fees to replace
their outpatient EHR vendor to match the hospital's system and make the
records interoperable.
Solution: The ONC should provide a clearer path to certification
that includes an enhanced focus on usability and interoperability.
These standards could include the ability for systems to connect with
multiple Health Information Exchanges (HIEs). The most important aspect
of a standard is that they be clinically relevant and useful, as would
occur if the standards were created in cooperation with specialty
societies such as the ACC. Through its rigorous process of creating
clinical guidelines, societies such as the ACC are well-equipped to
make these specific determinations as to what standards need to be
applied and how they should be applied. In addition to adjusting the
certification criteria, thorough testing must be performed not just of
the EHR itself but in exchanging information with other EHRs and other
actors in the health IT sphere such as HIEs and registries. This can
lead to the higher level of bi-directional data exchange that we need
in order to achieve the true benefits of health information exchange.
post-certification surveillance of ehr systems
Issue: Since the passage of the HITECH Act in 2009, the Federal
Government has invested over $30 billion in EHRs. Currently, no
programs exist to ensure that existing EHRs are functioning properly.
Implementation of a post-certification surveillance program of EHRs
would add value to the Federal Government's already substantial
investment and set the Nation on a path of complete interoperability of
EHRs.
Solution: The ACC requests that ONC or the HHS Secretary conduct
post-certification surveillance of EHRs to properly evaluate what
elements are effective and what elements are not working with respect
to basic usability and interoperability functionalities providers
require of EHRs. This includes the removal of contract gag clauses to
enable documentation by the Federal Government of any data portability
issues and to provide for further transparency in pricing. It should be
clearly stated that the burden for upgrades would pass to the EHR
vendors rather than physician practices or hospitals. Additionally, a
quarterly report from the Federal Government summarizing the
surveillance findings would further aid in fixing usability and
interoperability issues of CEHRT. The ACC applauds CMS for launching
the initiative to collect feedback via email from patients, clinicians,
and others whose health data was stymied.
reevaluation of hipaa and security of data
Issue: The ACC operates five hospital-based, one outpatient, and
two multi-specialty clinical data registries within a suite of
registries collectively known as the National Cardiovascular Data
Registry (NCDR). As a result of the Health Insurance Portability and
Accountability Act (HIPAA), hospitals and health systems within which
the NCDR conducts business require security contracts to transmit data.
The ACC understands that certain measures must be taken to comply with
HIPAA and ensure data security. However, HIPAA has resulted in overly
risk-averse interpretations of an almost 20-year-old law that was based
largely on paper data storage. This in turn has created unnecessary
demands from multiple layers of compliance officers with several layers
of review which may not actually be relevant or afford the best
protections in a digital, mobile-enabled environment.
Example: Compliance officers from larger health systems and
academic medical centers require NCDR to complete over 40 pages worth
of security questionnaires that are unique to their own institutions.
It may be possible for larger vendors with large numbers of staff to
accept this as a cost of doing business, but for society-operated
quality improvement programs and startups, these practices are
extraordinarily burdensome and stifle innovation by creating barriers
that only the largest entities can reasonably overcome.
Solution: The ACC has been on the record requesting the
reevaluation of the Health insurance Portability and Accountability Act
(HIPAA) and its appropriateness in a 21st Century digital landscape.
Technology has changed substantially since HIPAA was originally adopted
in 1996. The ACC urges Congress to convene a hearing to reevaluate the
role of HIPAA, including its successes and failures and whether all
aspects of HIPAA remain appropriate given today's technology.
delay of meaningful use stage 3 in its entirety
Issue: The Centers for Medicare and Medicaid Services (CMS)
released a notice for proposed rulemaking on March 20, 2015 outlining
the third and final stage of the Meaningful Use Program to be in place
starting in 2018. The proposed changes increase thresholds for
objectives and measures to an unattainable level in an aspirational
attempt to achieve greater care quality through the use of health
information technology.
Example: The Health Information Exchange objective (#7) of the
Stage 3 proposal requires program participants to provide or retrieve a
summary of care record when their patient moves to or from their care,
and calls for the participants to incorporate summaries of care from
other providers into their EHR using the functions of certified EHR
technology. This is required for a certain percent of transitions that
is far too high given the existing problems outlined in previous
examples of this letter and the lack of solutions currently in place.
In full disclosure, other issues exist with this objective and the
other seven objectives proposed.
Solution: The College has provided comments to CMS on this proposal
outlining our concern with the overreaching requirements. In light of
these concerns, the College has called for a delay in the
implementation of Meaningful Use Stage 3 in its entirety. Delaying only
certain parts of Meaningful Use Stage 3 would cause further confusion
around the program and lead the government to veer off the current
course of reducing complexities of the program. Given the lack of
participant data available from Meaningful Use Stage 2 coupled with the
data exchange issues that already exist, it is not feasible to
implement the increased demands of the program in 2018. Time is needed
to reevaluate the issues participants are facing in Stage 2 of the
program and to develop and enact solutions.
The ACC applauds you and your respective staff for taking the
initiative to accomplish specific goals related to interoperability of
EHRs and commends you for your collaborative approach. On behalf of the
entire cardiovascular care team and the patients who we serve, the
College appreciates the opportunity to provide input on these concepts
and encourages you to address these very pertinent and closely
connected issues. For additional information on the perspectives of the
ACC, please contact Charles Cascio ([email protected]) and Lucas Sanders
([email protected]).
Sincerely,
Kim Allan Williams, Sr., M.D., FACC, FAHA, FASNC,
President.
______
Response by David C. Kendrick, M.D., MPH, to Questions of Senator
Alexander, Senator Murray, Senator Enzi, Senator Casey, Senator
Baldwin, and Senator Cassidy
senator alexander
Question 1. If you were Secretary of the Department of Health and
Human Services and were faced with decisions about the Meaningful Use
Stage 3 rule and the 2015 Edition Certification rule, what would you
do?
Answer 1. I would pause the roll-out of the Meaningful Use Stage 3
program until the measures and expectations placed on providers can be
synchronized and coordinated with MACRA, PQRS and other related
programs which have the potential to create massive duplication of
effort and confusion. By coordinated, I mean the following:
1. The various programs (PQRS, MACRA, and MU) will all be unified
from a timing, target, scope, content, and measurement perspective. To
the extent that the programs are not fully integrated, at least all
required measure and utilization reporting should be synchronized
across the various Federal programs.
2. CMS should continue to fund and maintain the creation of
measures and the measure bundles that support consistent valid
implementation of the measures through the Quality Data Model. This
work is foundational to the assessment of value and assuring that the
U.S. tax dollars are being spent wisely in healthcare. Without this,
all previous investments in changing healthcare will be in jeopardy as
we will have a system for which we cannot measure the impact,
preventing iterations toward improvement.
3. The Federal partners in healthcare delivery should be subject to
the same requirements for reporting and interoperability as every other
organization--a majority of the care funded by DoD, IHS, and VA is
provided in the private sector and the private sector cannot deliver
high value care and services without appropriate and timely access to
accurate data about the patients they are treating. The Federal
partners should participate in the Health Information Exchanges that
are relevant to the care of their patients. The current approach to
interoperability, through a federated connection, is not responsive
enough to be relevant to clinical care. It can take several minutes to
retrieve a single patient record, by which time the provider has
usually moved on to the next patient. BY participating instead in the
State and regional level HIE efforts, the Federal partners will be
assured that their patients are getting the same benefits from
interoperability as the other patients in the community, including
fewer adverse drug events, reduced duplicate lab testing and imaging,
and improved coordination of care. Surely the America's active duty
military, veterans, and tribal members deserve the same chance at
receiving high quality care as any other American and active
participation in local/regional Health Information Exchanges is an
important step.
4. An important concept in Stages 2 and 3 of meaningful use is the
notion of the transition of care (ToC) from one setting (such as a
hospital) to another (such as a long-term care facility).
Unfortunately, CMS' original approach to this measure, which required
that a secure email be sent to the next provider in the transition was
not successful and CMS has recast the requirement in the Amended
Meaningful Use Stage 2 regulations that were released on 10/6/2015. The
new requirement is much improved because it allows the sending/
referring provider to send the patient records by other electronic
means, such as to a Health Information Exchange. However, the burden of
ensuring that the receiving provider actually looks at the data on the
referred patient remains on the shoulders of the sending/referring
provider, who is actually penalized in the measure if the receiving
provider fails to check the patient record. This is not the way
healthcare works. Once the sending/receiving provider has made the
current records available to the receiving provider, his or her
obligation has been met. It is the obligation of the receiving provider
to check the records on the patient before making diagnostic and
treatment decisions--not the referring provider. This may seem like a
minor point but in reality this single suggested policy change could
have the most impact of anything else that could be done to drive
interoperability as well as improving healthcare. On the other hand,
without it, the incentives for providers to demand the complete,
interoperable medical record on each patient they see will remain low.
Please consider this alteration to the Transitions of Care policy in
the amended MU2 and certainly in MU3.
When a combined MU3/MACRA/PQRS strategy rolls out, ideally there
will be a single program for providers to work with, one that is
designed and tested with significant provider, patient and payer input
and can align well with the current drive toward multi-payer value-
based payment models such as the Comprehensive Primary Care initiative.
However, I would push forward with the 2015 Edition Certification
program, and especially the interoperability components and the new
process of certifying that the EHR actually works in the field as it
did when being certified ``in the lab''. We must get beyond the
interoperability limitations and loopholes in the 2014 Certification
program which are enabling EHR vendors and occasionally providers to
skirt their interoperability responsibilities. Elements that should be
emphasized in the 2016 Certification program and launched immediately
include:
1. Data interoperability will be sufficient to enable the owner of
the EHR to elect to participate in HIE and configure their system to
export data of their choosing structured and coded to current
standards, based on an automated trigger of their choosing and deliver
it securely to a location of their choosing. Appropriate triggers would
include the referral of the patient to another provider, completion of
an encounter note, sign off on a new lab or imaging test result, and
other clinical events that should be communicated with other members of
the patient's care team. Appropriate locations for the exported data to
go would include a health information exchange, secure email, public
health program, or registry.
2. Data export formats will include standard message types (HL-7
2.x) as well as document types (HL-7 3.0's CCDA 2.x standard) and
documents to support external evaluation of quality and value (QRDA1
files with complete coded data), etc. At a minimum, all file & message
types required for MU, PQRS, MACRA, and any other mandated reporting
programs should be able to be exported at no cost and delivered as
above in 1.
3. The interoperability described above should be rigorously tested
and validated by the Certification Bodies both in the development lab
at the EHR vendor (where they are tested today) AND in the field when
the product is deployed in a provider practice or hospital (where it is
not tested today, and most commonly fails to meet interoperability
expectations).
4. Vendors should be required to support credential exchange
standards such as SAML and OAUTH which enable doctors' and other
providers' workflow to move seamlessly among different vendor products
in the course of treating each patient and conducting their other care-
related work such as quality assessment and utilizing decision support.
This is critical to enabling innovation to continue in the Health IT
space, even though a few large vendors claim large swaths of the
territory.
senator murray
Question 1. You testified that you have personal experience working
with providers in Oklahoma to make sure that they are able to exchange
electronic health information. I regularly hear from providers about
information blocking, but the Department of Health and Human Services
reported to Congress in April that in many cases they have limited
tools to find out when information blocking is occurring.
Answer 1. Yes, that is correct--this is a very complex issue, and
it takes expertise in the field of medical informatics, focused
attention, and a desire to fix the problem, all of which are rarely
available in busy medical practices. In addition, one must have access
to the vendor product at the practice level to identify the issues.
There are dozens of ways that an EHR vendor can (intentionally or
unintentionally) prevent interoperability from occurring. Further, many
practices I encounter are actually too afraid to report their vendor--
fearing repercussions such as poor service, higher bills, or delayed
responsiveness to critical issues. Any of these can put a practice or
small hospital out of business and this fear is clearly limiting the
filing of formal complaints. I listed six common ways data blocking can
occur in my testimony, but there are many others.
Since it is impossible to think of all the ways that data blocking
can occur, effective surveillance and enforcement is required. Ideally,
ONC should receive sufficient funding to provide visible and active
enforcement of the requirements in the Certified Health IT program in
order to protect the tax-payers' investment in health IT. These funds
should be used to make the existing ONC Certified Vendor Surveillance
program more robust. The success that MyHealth has had in identifying
and pursuing remediation with vendors whose products fall short of the
required capabilities has been limited to only a few vendors and has
required inside knowledge of the workings of ONC and the Accredited
Certifying Bodies as well as significant time focused on the tracking
and reiteration of our complaints. Furthermore, the critical knowledge
we gained about unlocking an EHR vendor system through the Surveillance
program has not been publicly communicated--rather that burden has
fallen to our small non-profit organization. While we are happy to help
our peers and others seeking interoperability it would seem that ONC
should be tasked with publicizing the results of Certified Vendor
Surveillance activities just as they publish the list of certified
products.
The Accredited Certifying Bodies, the private organizations
contracted by ONC to implement the Certification program, are also
currently tasked with conducting the Surveillance program. Because
these ACB's are dependent on the vendors they certify for most of their
revenues, there seems to be a potential conflict of interest created by
the assignment of the ACB's to also conduct the product Surveillance
program. Whether driven by conflict or not, the current Surveillance
activities seem much less substantial than they should be. The 2014
Surveillance report by the ACB's (which has not been made publicly
available as far as I can tell) was disappointingly thin on details and
action, especially given the tenor and rate of complaints that we hear
from providers regularly.
Question 2. This committee has also heard testimony that, in order
to use health IT, providers often have to sign contracts that include
``gag clauses,'' prohibiting them from disclosing technical problems or
unfair pricing. Health IT is certified by the Federal Government
because it stores and transmits information that is used to save
lives--users should be able to safely report problems without facing
any sort of repercussions.
Answer 2. Agreed. These gag clauses are certainly a part of the
problem. We recently surveyed our member organizations to gather the
data on the costs they faced when connecting their EHR to MyHealth's
Health Information Exchange. The reports ranged from $0 to $40K +
ongoing monthly fees. This was concerning, but even more concerning was
the fact that most of the providers and other organizations we surveyed
reported that they could not divulge the cost due to contract
restrictions (gag clauses).
Question 3. What tools and protections are needed to ensure that
individuals can safely report on unfair market practices regarding the
use of health IT, without fear of repercussions?
Answer 3. Expanding and encouraging ONC's certified technology
Surveillance program (as described above) is very important. Publicly
reporting significant findings of this surveillance program is also
very important. Like any good enforcement program, whistle-blower
protections should be put into place, and cases should be promptly and
professionally handled. In our experience, even when a vendor is forced
to correct a shortcoming in their technology, the remediation time
window is so long (months to more than a year) that it adversely
affects the practice and other stakeholders anyway, and no compensation
is offered to the affected practices or entities while the remediation
is being performed. In some cases, the practices have been forced to
pay the additional ``product'' fees the vendor originally requested
even though we know the vendor's product has been found to be defective
and the obligated fix is working its way through the system.
A successful Surveillance program must have teeth--that is, ONC
must have the power and a mechanism to withdraw certification if
necessary. In theory, this is currently possible, but in reality this
is rarely done because of concern for the impact of decertification on
those providers that use the decertified products. This creates a ``too
big to fail'' scenario with EHR vendors. These implications should be
addressed clearly in the Certification program. There are several ways
to address this:
1. Require the decertified vendor to continue to provide services
to the practices but discount their fees to offset the Medicare or
Medicaid penalties the practice will incur by not meeting Meaningful
Use. If the loss to the provider exceeds the feeds then actual
financial compensation could occur.
2. An insurance product could be required of Certified Vendors (or
even made available to practices) that would enable the providers to
change vendor products should the Certification of their existing
systems be withdrawn or should the vendor go out of business. This
could be a means of supporting the ``lemon law'' I proposed in my
previous testimony.
senator enzi
Question 1. What is the greatest barrier in obtaining access to
data from the perspective of vendors?
Answer 1. I suppose that the EHR vendors would cite the complexity
of the MU Certification process, the cost of connecting their products
to HIE's in various communities, and the highly variable levels of
sophistication among the HIE vendors and HIE organizations. While I
agree that these elements all play a factor, one simple fact makes me
question the real significance of these issues. It has been known for a
long time (decades in some cases) that many of the large EHR vendors
aggregate data from their customers and resell de-identified data to
researchers, pharma, and other organizations interested in big
healthcare data. Most practices and hospitals sign agreements with
these vendors that contain fine print that grants the EHR vendor the
rights to that de-identified clinical data. Very few providers, in my
experience, are aware of this practice. This common practice, which
requires all of the same activities as interoperability, makes me
skeptical about vendor complaints that interoperability is ``too hard''
or ``too expensive''.
Question 2. Is this challenge technical in nature, rooted in
custom-designed EHR's? What, in your opinion, would be the simplest way
to address this problem?
Answer 1. There are two potential sources of data blocking: (1)
Provider data blocking and (2) Vendor data blocking.
Provider data blocking: As noted in my testimony, with the rise of
value-based and risk-sharing payment models, provider opposition to
interoperability seems to be melting away, because the incentives to
have the complete medical record available wherever and whenever needed
for care are now shared among the patient and all providers they choose
to involve in their care.
Vendor data blocking: Unfortunately, these incentive realignments
do not seem to have affected Vendor data blocking. In our recent
experience, the greatest barriers to broad interoperability are EHR
vendor policies and business practices around interoperability, not
technical challenges. While it is true that many of the largest EHR
vendors are still running highly proprietary code written in the 1980s
and 1990s, and some are still using database products from the 1970s,
the basic capabilities needed for appropriate levels of
interoperability have existed for a long time and frankly work quite
well if they are used.
As described above, most vendors are quite capable of extracting
clinical data, standardizing its structure and exchanging it with other
organizations as evidenced by decades of resale of de-identified
clinical data to third parties by many EHR vendors. Despite this
extensive experience, EHR vendor capabilities for data exchange and
quality of the exported data remains quite poor and non-compliant with
the Certification Criteria. Thus, one can only conclude that the
correct incentives are not in place to warrant these vendors to
participate fully in interoperability.
The simplest solution to this problem is iterative, not sweeping--
this is, it builds on the work already begun rather than ripping and
replacing with some other new, potentially more expensive approach.
1. As recommended above, the 2015 EHR certification program should
be allowed to proceed, because it contains a number of provisions that
should increase the interoperability capabilities of the EHR vendors
and improve the Surveillance program.
2. The Certified EHR Surveillance program at ONC should be
adequately funded, as noted above. In particular, vendors should be
required to demonstrate that each install of their product is
interoperable by making a connection to a standard test server operated
by the Certification Program or perhaps a local health information
exchange organization certified by ONC, and transmitting the required
standard documents on a few dozen test patients. This level of testing
would provide assurance to the practice or hospital and their patients
that the EHR product has the capabilities and is properly configured
for interoperability. This concept has been referred to previously as
``field testing'' for interoperability and would go a long ways toward
assuring that the technical capabilities are not a limitation.
3. Making the patient, rather than the practice or provider, the
frame of reference for quality and value measurement: By emphasizing
(and even requiring) the use of interoperable data on each patient in
quality reporting and other value assessment, CMS and others can shift
vendor and provider incentives away from data blocking toward improved
health data interoperability.
Making this change to focus measurement on the patient is also
critical for the success of value-based payment models. New value-based
payment models for healthcare are showing some promise that savings can
actually be achieved while quality is maintained or improved. For
example, CMMI's Comprehensive Primary Care initiative has shown 7
percent and 5 percent cost savings over the last 2 years in Oklahoma.
However, all value-based payment models depend completely on having the
ability to measure Value accurately.
Unfortunately, those of us practicing medicine and working with
clinical information daily have come to recognize a significant
methodological flaw in CMS's current approach to value assessment which
calls the validity of most of today's value assessment into question.
Although an exhaustive discussion of this is not possible here (happy
to provide more details if there is interest), the basic issue is that
the current measurement approach uses the wrong frame of reference.
Just as adopting the correct frame of reference is important to solving
basic physics problems, so too is adopting the proper frame of
reference critical to correctly measuring value in healthcare.
Unfortunately, most current CMS projects adopt the doctor or the
practice as the frame of reference for measurement, when in reality,
the proper frame of reference is the patient. To choose any other frame
of reference than the patient is to risk repeatedly counting the same
(and especially the sickest) patients over and over as they see
multiple doctors and hospitals. In addition, only by centering
measurement on the patient can the impact of team-based care and
effective care coordination be measured and reported accurately and
without methodological issues. We have demonstrated this approach to
measurement in Oklahoma as part of the successful Comprehensive Primary
Care Initiative, and it makes a significant difference in the results.
The achievement of these three items and especially item #3 will
create a marketplace whose natural tendencies are toward, rather than
away from, interoperability.
Question 2. When you work with vendors, providers, and ONC to
resolve issues, do you see there being a need for a more defined
process to facilitate those interaction?
Answer 2. The CEHRT Surveillance program was originally conceived
to provide such a defined process but until the last year or so has not
been an active program. ONC appears to be making strides in
strengthening the Surveillance program but funding for the effort seems
limited. As recommended above, a strong, transparent, and responsive
Surveillance program with real ``teeth'' will go a long ways toward
reducing EHR data blocking. We would be happy to contribute specific
experiences and suggestions for improving and expanding the
surveillance program.
senator casey
Question 1. In the absence of a straightforward method of patient
matching like a national patient identifier, health IT uses complicated
algorithms to match patient records from multiple locations and
providers. Algorithms include data such as name, date of birth,
address, and social security number. Pennsylvania has a significant
Amish population and many Amish people don't have Social Security
numbers or phone numbers. Additionally, many older female patients
share a Social Security number with their husbands and many are widows,
making matching of their records challenging. As a result, hospitals
like Geisinger in Danville, PA have had to develop unique and
complicated ways of matching patients that are not always highly
reliable.
How does lack of reliable method of patient matching contribute to
difficulties in sharing information? Are there additional sub-
populations for which patient matching is especially difficult? What
are the implications for care of the patient? What, in your view, is
the best solution?
Answer 1. Assuring accurate patient identity is critical to the
health and well-being of the patients we serve. Unfortunately, the
political will to implement a nationwide patient identifier has not
existed since the passage of HIPAA authorized it. Nevertheless, there
are many things that can be done to address this issue. In particular,
electronic master patient indexes (eMPI's) have become important tools
for resolving identity at regional and State levels, and many HIE's
around the country have these tools and several offer them to local
doctors, hospitals and other providers as a source of truth against
which patient registrations can be searched. In addition, State and
regional eMPI's can provide active feedback to providers about patient
identities that may have errors or need resolution. By correcting data
at the source, the accuracy of patient identity throughout the entire
healthcare system is improved. We have seen that this approach can be
quite successful, with close to 99 percent of all patient identities
accurately matched between organizations without human intervention.
The remaining patient identities must be matched to the correct patient
with human intervention. In the case of a State or region, this is
manageable. However, it is probably premature to attempt to construct a
national eMPI because the trust does not yet exist at a national level,
and further the number of manual interventions required would be in the
millions, requiring significant staff.
Our experience at MyHealth confirms the concerns outlined in the
question about specific missing data (like social security numbers). In
addition, we often see patients with rapidly changing addresses, phone
numbers, and even last names. We carefully monitor the identity of
twins, triplets, quadruplets, etc, because they often have very similar
first names and identical last names, dates of birth, addresses and
phone numbers. Having a single medical ID does not necessarily help
with these kinds of issues. A single medical ID could be helpful--it is
after all just one more data point on the patient. However, in our
experience, even with a national identifier, all of the logic and
capabilities of an eMPI will still be required to accurately maintain
identity.
For regions/States which have implemented eMPI's, the challenge of
maintaining accurate patient identity is greatly reduced. In order to
optimize this approach, some regions have included State registration
data such as that from State DMV's or identity from credit agencies
such as Equifax and others.
This does leave the challenge of maintaining patient identity on a
national level, but a strong regional/State-level eMPI infrastructure
will enable the optimization and exchange of patient identity at a
national level as well. By optimizing identity where patients receive
most of their care (near their homes), the issue of national patient
identity becomes much more manageable. As States and regions work with
one another to exchange interoperable health data, they are addressing
these issues of identity management as well, so this issue may already
be on its way to solution.
senator baldwin
Question 1. In your testimonies, a number of you suggest that we
need to increase transparency for all stakeholders to help address some
of the problems we see with data blocking and to promote
interoperability.
How do we create this transparency; what needs to be measured and
reported by providers and vendors for us to determine if data blocking
is occurring or if progress is being made?
Answer 1. Creating transparency is important. I would offer three
ways of achieving transparency of the data blocking issue, and suggest
that these approaches are also essential to curing the data blocking
issue:
1. Further emphasize (and fund) the ONC Certification Surveillance
program with included language about transparency of findings and
public communication of results.
2. Establish a nationwide measure of interoperability. This metric
is constructed using existing data sources and would be based on the
premise that every patient deserves to have their complete,
longitudinal medical record available wherever and whenever it is
needed for decisions about their healthcare. This single metric can be
applied to specific types of clinical encounters and events to track
and monitor our progress as a nation toward interoperability. A few
months of research would be required to rigorously validate the metric
but it will provide enormous insight and motivation toward
interoperability once available. We have prototyped such a measure and
it seems quite feasible to roll out broadly within a year.
3. Measure value at the correct frame of reference: the patient. As
noted above, value-based payment models must be based on value
assessments that take into account each patient's entire experience of
care, not just the pieces of experience that occur in each practice or
with each individual doctor. If payment policy shifts to reward
reporting metrics based on multi-sourced/complete data on each patient
rather than just individual practice perspectives, sophisticated
interoperability will spread quickly because it will have a clear
business case for providers and vendors.
senator cassidy
Question 1. How do you define data blocking?
Answer 1. I would define data blocking as any factor, whether
intentional or unintentional that prevents a patient from having their
complete, longitudinal medical record available wherever and whenever
they or their healthcare team needs it for decisionmaking about their
health. Admittedly this is an outcome-based definition of data
blocking, but I find it helps avoid the loopholes and semantic issues
that have thus far enabled vendors and others to thwart attempts at
true healthcare data interoperability. Unintentional causes of data
blocking should be identified and pursued for correction, which can be
greatly facilitated by the creation of an Interoperability Metric as
described above. Intentional causes of data blocking need to be
identified (again, using an Interoperability Metric) and brought to
light using smart policy (to align incentives away from data blocking,
for example) and enforcement of regulations (such as the ONC
Surveillance program).
Question 2. In the context of certification, how do you penalize
the vendor without penalizing the physician?
Answer 2. As described above in more detail, it makes sense that,
because the EP or EH has purchased a product that has been
``certified'' by a government agency, it is not fair to penalize them
for subsequent failures of that product, including the potential
revocation of the product's certification. The purchase and
implementation of an EHR product is a massive, expensive and complex
task for any healthcare organization, and is usually second only to
(and often exceeds) the cost of their physical facilities. In today's
healthcare marketplace, the EHR is central to the success or failure of
the organization.
For these reasons, it seems logical that significant consumer
protections and risk mitigation should exist for purchasers of these
systems. Among these should be:
1. Requirement that a vendor who loses their certification, even if
only temporarily, must cover any MU penalty losses to their customers,
first through discounts on licensing fees, and then through future
service or actual monetary compensation.
2. Establishment of insurance products to mitigate the extreme cost
and market share loss experienced by practices and hospitals who must
make an unexpected EHR vendor change. Practices or hospitals could then
require that their vendor be ``bonded'' against such an unexpected
event, similar to the way risk is mitigated in the construction
industry today.
3. Make Vendor contracts, including as much as possible the Total
Cost of Ownership, transparent, and encourage a competitive, consumer-
driven marketplace which will ultimately be much more conducive for
success as well as continued innovation.
4. Be very careful not to ``over design'' the requirements of
certified Health IT (and the required uses of that IT)--the
requirements should be written to focus on the desired outcome rather
than prescribing exactly how the outcome should be achieved. If we
eliminate the potential for innovation then we have failed utterly in
our efforts to improve health and reduce costs.
Unfortunately, many EP's and EH's are being penalized now for not
meeting MU requirements even though they invested in Certified health
IT. Putting in place policies that ensure that health IT is an open
marketplace with sufficient transparency to enable the consumers (in
this case doctors and hospitals, but eventually patients as well) to
drive the success or failure of the vendors will be the best and most
permanent fix.
Question 3. Should ``gag clauses'' in contracts be made illegal?
Answer 3. Yes. I can't think of a single way that these gag clauses
improve patient care, reduce costs, or improve the practice of
medicine.
[Whereupon, at 11:50 a.m., the hearing was adjourned.]
[all]