[This Transcript is Unedited]
NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS
SUBCOMMITTEE ON STANDARDS AND SECURITY
May 27, 2004
Hubert H. Humphrey Building
Room 505A
200 Independence Avenue, SW
Washington, D.C.
Proceedings by:
CASET Associates, Ltd.
10201 Lee Highway, Suite 160
Fairfax, Virginia 22030
(703)352-0091
List of Participants:
- Simon Cohn, M.D., Chair
- Donna Pickett
- Vivian Auld
- Karen Trudel
- Judith Warren
- Jeffrey Blair
- Steven Steinde
- Stanley Huff
- J. Michael Fitzmaurice
- Suzie Burke-Bebee
TABLE OF CONTENTS
- E-Prescribing: The Providers’ Perspective
- Planning for Next Meeting
P R O C E E D I N G S (8:45 a.m.)
DR. COHN: Time to get started. I want to call this meeting to
order. This is the third day of three days of hearings of the
Subcommittee on Standards and Security of the National Committee on Vital and
Health Statistics. The committee is the main public advisory committee to
the U.S. Health and Human Services national health information policy.
I am Simon Cohn, Chairman of the subcommittee and the National Director for
Health Information Policy for Kaiser Permanente and a practicing
physician. I want to welcome fellow subcommittee members, HHS staff, and
others here in person. I also want to welcome those listening in on the
Internet. I want as I always do remind everyone in attendance to speak
clearly and into the microphone. You notice I am speaking very closely
into the microphone, because if you get very far away from it, no one can hear,
so just a word.
This morning, we continue our work regarding e-prescribing standards.
We start the day with presentations and discussions with providers, and I want
to thank you for joining us this morning. After the morning break, we
will have an open microphone session, and this will be followed by subcommittee
discussion and next steps. We will adjourn by 12:45, I think
realistically sometime between 12 and 1.
I want to emphasize, this is an open session. Those in attendance are
welcome to make brief remarks if you have information pertinent to the subjects
being discussed. As we commented, there is also a time for open
mike. Finally, for those on the Internet, we welcome e-mail and letter
comments on the issues coming before the subcommittee.
With that, let’s have introductions around the table and then around the
room. We ask for those on the National Committee, if there are any
conflicts of interest related to any of the issues coming before us today, in
your introduction would you so publicly state.
(Whereupon, introductions were performed.)
DR. COHN: Welcome, everyone. Before we go any further, I just
want to take a moment and acknowledge our Vice Chair, Jeff Blair, and Maria
Friedman, our lead staff, for what I think has been a tremendous set of
hearings so far. So you really have the appreciation of the committee for
your tremendous work. Thank you so much.
Jeff, do you have any comments you would like to make in setting the frame
for the day?
MR. BLAIR: Just a very brief couple of statements in
perspective. These hearings are part of hearings that are being conducted
over several months to help the committee, the NCVHS, gather information that
would help us in the evaluation, selection and recommendations of health care
information standards for e-prescribing.
This activity was directed by Congress, that the NCVHS should perform these
evaluations and recommendations within the Medicare Prescription Drug
Improvement and Modernization Act. There is a work plan for those of you
who might be interested in all of the different stakeholders that we plan to
listen to, in accordance with the law. That is at the NCVHS website, or
you can contact either Maria or Marietta Squire for copies or myself; I’ll be
happy to send them to you also. The work plan is a living document.
It changes in accordance with various things, but at least it will give you
some idea of what the law says that we are following, also at least our planned
schedule, which again may change.
Other than that, I just want to very much thank Maria and Simon and all the
support that we have gotten for these hearings, and thank the testifiers who
are about to testify to us this morning.
DR. COHN: With that, why don’t we start with our first panel.
Jean Narcissi, good to see you. Thank you for joining us.
DR. NARCISSI: Thank you. My name is Jean
Narcissi. I am the Director of Electronic Medical Systems for the
American Medical Association. It is my pleasure to appear today on behalf
of the Subcommittee on Standards and Security of the National Committee on
Vital and Health Statistics. I would like to thank you for the
opportunity to testify.
Most approaches to addressing the problems in the current health care
delivery system include a reference to the need for information driven health
care, and a commitment to improving the quality of care. Electronic
prescribing is a key component of the quality equation.
During the Medicare Prescription Drug Improvement and Modernization Act of
2003, MMA, debate, the AMA supported provisions to create uniform national
standards to help facilitate the dissemination of e-prescribing. We are
hopeful e-prescribing can achieve its promises to improve patient safety and
increase administrative efficiency.
My statement will summarize the views of the AMA on electronic prescribing
standards. The following comments will address the questions we have been
asked to discuss before the subcommittee, as well as other issues of concern to
the AMA.
In response to the first question, what are the best standards or code sets
to meet the requirements of the law. Although the concept of electronic
prescribing appears simple, broad-scale electronic prescribing as defined in
MMA barely exists in the ambulatory setting. Some hospitals and large
practices can electronically communicate with onsite pharmacy via closed
systems, but two-way, interoperable and secure systems currently are used only
in extremely limited circumstances. There seems to be a lack of focus on
the necessity for vendors to incorporate the specific needs of the practicing
physician in the ambulatory care setting.
Some of the technology
developed to date has been delivered at the desktop for hospital facilities in
large group practices. Standards for advanced e-prescribing systems with
roaming or wireless capabilities are needed in the ambulatory setting as well
as the larger facilities in order to reduce errors that may occur.
Development of a single set of standards appropriate for all health care
delivery situations will be a significant undertaking.
From the physician perspective, standards for electronic prescribing must
take into account the wide variety of clinical settings and specialties.
We urge NCVHS to recommend standards that are flexible and scalable, in an
effort to encourage adoption from small to large health care organizations, and
low to high volume prescribing physician specialties.
Core electronic prescribing standards must allow for basic stand-alone
electronic prescribing platforms that permit small practices to meet the
regulatory requirements without an undue financial burden. The standards
should also provide for the larger, more complex group practices and health
systems. This flexibility will allow physicians to consider critical
factors such as clinical quality, safety, efficiency and integration with
existing management software and electronic medical record systems when making
an investment.
The standards for e-prescribing are also deficient in the area of medical
vocabulary. In addition, the AMA believes there is no single solution to
meet the vocabulary needs. The AMA also believes that the terminologies
supporting patient data should be comprehensive.
For example, one of the suggested e-prescribing nomenclatures is Rx
Norm. Rx Norm is a clinical drug nomenclature produced by the National
Library of Medicine in consultation with the Food and Drug Administration, the
Department of Veterans Affairs and Health Level 7 standards development
organization. Rx Norm provides standard names for clinical drugs and for
dose forms as administered. It provides links from clinical drugs to
their active ingredients, drug components, and some related brand names.
To the extent available from the Food and Drug Administration, national
drug codes for specific drug products that deliver the clinical drug are stored
as attributes of the clinical drug in Rx Norm. However, the scope of the
nomenclature needs to be broadened to account for other details such as
packaging sizes, flavorings or medications, et cetera. The standards
should accurately capture what the doctor ordered. Rx Norm may be
suitable for the VA environment, but it may not be sufficient in the private
sector at this time.
In addition, interoperability with many clinical terms is also an
issue. For example, some terms may be used differently in a hospital
setting than an ambulatory environment. Standards currently exist that
support the sharing of prescription information to some extent, as do
vocabularies that describe the drugs prescribed. However, standards need
to be enhanced where necessary, as well as support vocabularies that clearly
define the intent of the prescription. Improved vocabularies and
standards are needed to enhance quality and efficiency, and to facilitate
interoperability between the various electronic systems involved in the
e-prescribing process.
Unifying state prescription form standards, establishing a consistent
position level drug vocabulary, and standardizing formulary information are
among the highest needs. AMA believes that demonstration projects should
be required in order to prove that all the proposed standards will work in
multiple similar and dissimilar environments. In addition, the
demonstration projects should address the work flow issues and establish the
business rules in order to impose no undue burden on the physicians.
For example, a physician should be able to efficiently submit a secure,
authenticated prescription electronically in real time to a payor or pharmacy
benefit manager for eligibility and formulary adjudication. Then the
prescription should be forwarded to the pharmacy. The physician needs the
ability to communicate with the PDM and the payor and pharmacy to indicate if
the patient has failed previous therapy, has a condition or allergy which makes
the preferred drug an unacceptable alternative, et cetera. Whatever the
scenario, the physician should never have to leave the electronic prescribing
environment to complete a prescription.
Furthermore, the AMA believes that demonstration projects must also show
achievable financial models for appropriately funding the acquisition of the
technology, training and support for electronic prescribing in the various
clinical settings.
In response to question two, which of the standard code sets do you
use. Standards are vital for any transactions that involve more than one
system. In the world of e-prescribing, there are many interchanges
between different physician systems, patient health records, pharmacy systems,
payors and pharmacy benefit management systems, public health records, clinical
decision support systems, et cetera.
Currently there are two messaging standards in the U.S. that support
e-prescribing functions. These two standards are the National Council for
Prescription Drug Programs, NCPDP script standards, and HL-7 standards.
E-prescribing is a two-way communication between physicians and
pharmacies. That is what the script standards address. The script
standard was created specifically to facilitate the electronic transfer of
prescription data between pharmacies and physicians. It was designed as a
computer to computer transaction that does not involve faxing of paper
scripts. If the message would be printed, it could be read, but computer
to computer transactions minimize human errors and are more efficient.
Patients do not need to carry a piece of paper into the pharmacy and then
wait 20 or 30 minutes to get the prescription filled. It is being filled
while the patient is in transit to the pharmacy. The script standard
supports ordering new prescriptions, prescription changes, refills of existing
prescriptions, prescription fill status notifications, as well as
cancellations.
Other areas that script standards should address are eligibility, formulary
information and medical history. Although script is a message format,
work flow also needs to be considered. The script standard should
potentially do much more of clear business requirements and the needs for all
parties involved are defined.
For example, script allows a physician to send a prescription message to
the pharmacy, but it currently does not account for prior authorization.
Today, prior authorization is necessary to fill many prescriptions, and this
must be done by either a phone call or a fax before some prescriptions can be
transmitted.
Work flow issues need to be resolved between the physicians, the pharmacy
benefit managers, payors, vendors, and then incorporated into the standards for
efficient real time transactions.
HL-7 is a messaging standard that is used for a variety of administrative
and clinical transactions, including medication therapy, patient ID, clinical
notes, lab results, et cetera. While most commercial pharmacy
transactions are using the script standard, HL-7 has been used in some
installations. NCPDP and HL-7 are now engaged in a coordination effort,
but more work still needs to be done.
HL-7 and NCPDP are both ANSE accredited standards development organizations
responsible for the development of new pharmacy and health care related
standards. The AMA believes that that they should continue to advance the
standards and encourage the participation of the appropriate entities in order
for the standards to be more work efficient, more reliable, and less prone to
abuse than the paper or fax prescriptions.
In response to question three, what are the strengths and weaknesses of the
standards and code sets that you use. As I previously stated, from a
physician perspective, standards for electronic prescribing must take into
account the wide variety of clinical settings and specialties. We urge
NCVHS to recommend standards that are flexible and scalable, in an effort to
encourage adoption from small to large health organizations and low to high
volume prescribing physician specialties.
Core electronic prescribing standards must allow for basic stand-alone
electronic prescribing platforms that permit small practices to meet the
regulatory requirements without an undue financial burden. The standards
should also provide for the needs of larger, more complex group practices and
health systems. This flexibility will allow physicians to consider
critical factors such as clinical quality, safety, efficiency and integration
with existing management software and electronic medical record systems when
making an investment.
With each health care organization being unique, flexibility to customize
and implement systems to scale is critical. Most importantly, the
physician-patient relationship must be enhanced, not impaired, by this new
technology. Incorporating electronic health records in prescribing
systems in a physician practice without disrupting patient care is
critical. For physicians to widely adopt a new technology, it must allow
for a safer and more efficient way to treat patients, whether it is a new
imaging technology or information technology. These principles remain
constant.
Moreover, time is a precious commodity for physicians. The standards
for electronic prescribing must be developed in a manner to provide vendors
with a platform to create technology solutions that save time for physicians
and their staff. Time saved can be refocused on patient care, providing
physicians with a tangible clinical and financial incentive for investment.
Response to question four, is nationwide adoption of these standards and
code sets necessary. The AMA believes that a true private sector approach
to standards development for e-prescribing is needed, with the federal
government participating in the standards development process. However,
we urge that the maintenance and modifications in the standards not get
hindered by an extensive rulemaking process, similar to what has been
experienced with the HIPAA administrative transaction standards. The
federal government can also play a useful role in helping to identify any gaps
in the standards.
The AMA believes that there should be unification of varying state
regulations concerning the proper format of a prescription. Currently
there is a wide variety of state regulations, most of which are directed toward
the same objectives, but differ because they were developed separately.
In addition, the standards, terms and structures used by formulary information
providers are also needed to be unified. Without standardization, there
will continue to be an increase in complexity and cost of e-prescribing
systems, since vendors must keep up with varying regulations and vocabularies.
The AMA also believes that there should be further enhancement and testing
of Rx Norm, as stated previously. Enhancement and testing of the national
drug reference terminology is being developed for the Veterans Administration
as a reference standard for medications to support a variety of clinical,
administrative and analytical purposes.
In addition, national drug codes work well for pharmacy purposes, but they
are too granular for physicians’ use. As a result, many prescriptions
today are transmitted in free form text, resulting in a re-entry of potential
errors at the pharmacy, as well as lost opportunities for clinical decision
support. Therefore, much work is still necessary in order to provide a
consistent physician level drug vocabulary.
Prescribing system drug dictionaries also need to be consistent, so that
specifications of allergy groups, drug interaction groups, et cetera, are
interoperable between two different applications that use different commercial
dictionaries. Once agreement has been reached on a vocabulary, it should
be incorporated into the definitions and requirements of the NCPDP script
standard.
In addition, AMA believes that all vocabulary and coding systems referenced
for use in the e-prescribing standards should have an open updated process, and
any interested paries should be eligible to submit proposals for additions and
modifications. In addition, a responsible panel or committee of experts
that are representative of a broad cross section of the relevant stakeholders
should maintain the vocabularies.
The AMA does not believe that it is necessary for all the vocabulary
developers to be ANSE accredited. However, the organization maintaining
the code sets should insure continuity and efficient updating of the system
over time.
In response to question five, what are the e-prescribing standard code set
gaps. Clinical quality improvement and patient safety are significant and
important drivers for physician adoption of new technology. Information
technology’s greatest potential lies in these areas, and we believe electronic
prescribing presents great promise for improved patient care. However,
new technology must also be reliable.
In the fall of 2003, a significant and unforeseen computer hardware failure
in the Maryland Medicaid program left one thousand pharmacies without the
ability to check eligibility, formulary and claim information, leaving
prescriptions unfilled for several days. No technology is immune from
failure. Standards must require failsafes in any new system to prevent
the introduction of new health care errors. Similarly, standards must be
employed that minimize the electronic transmission of false positive clinical
information to physicians, such as clinical irrelevant drug-drug interaction
alerts.
The AMA urges the NCVHS to make recommendations in the context of lessons
learned when implementing the administrative simplification provisions under
the Health Insurance Portability and Accountability Act. The critical
factor in the protracted implementation of electronic transactions and code
sets rule has been the inability of the provider community to upgrade their
practice management and billing software in a timely manner.
CMS had the most difficult task of trying to resolve intergovernmental
differences from across the federal government in the addendum to the
electronic transactions and code sets rule. The additional time that it
took to resolve those differences left inadequate time for the various vendors
to work with their customers, the provider and the payor communities, to
achieve timely compliance with the new rule.
Vendors need to see a commitment by the government before creating or
adapting existing systems to meet new standards. Although HIPAA was
signed into law in August of 1996, all the transaction standards have still not
been fully implemented, and that is nearly eight years.
Another example I wanted to share with you regarding factors to consider is
that since 1998, the Drug Enforcement Association has met with the AMA and
others to develop an encrypted electronic prescribing system for schedule two
and other controlled substances. The DEA has yet to issue a notice of
proposed rulemaking to begin implementation. That is six years.
Pilot testing for electronic prescribing vendors should be required before
HHS releases a final rule. Vendors should be factored into the
regulations somehow and encouraged to bring products to market that assist
physicians to comply with the statutory requirements ahead of any
deadlines. Staggered implementation dates should be considered as
pharmacies and pharmacy benefit managers must test systems up and running to
allow physicians to send test prescriptions that comply with the
standards. Physicians must rely on their vendors to provide them with the
tools necessary to comply with the electronic prescribing program. Strong
government leadership is critical to rapid and seamless conversion.
In response to question six, what are the barriers to the development and
adoption of these standards. Development and pilot testing are essential
to determine the standards work. For technology related products, this is
done by interested groups of early adopters, and there are relatively few of
them. Current surveys vary in definitions and results, but it is
estimated that percentage of physicians currently using electronic prescribing
is in the single digits.
The AMA is in the process of conducting a survey on physician use of
technology in their practice. It is a joint study by the AMA and
Forrester Research. However, results of the survey will not be available
until mid-July. We are collecting information on a number of topics,
including e-prescribing. We will keep the committee informed of our
findings once the survey is finalized.
I have already mentioned some of the barriers, such as interoperability,
work flow issues, inadequate messaging and vocabulary standards. In
addition, the AMA believes that the effective and affordable user
authentication will be a key enabler of this business growth, providing a
foundation for the high level privacy and confidentiality that are essential in
the health care industry.
The authentication process must be simple, quick and reliable, and it must
be flexible enough to authenticate users across the complex network of health
care websites. The AMA has been working with Verisign Incorporated to
deploy a form of electronic identification called the AMA Internet ID that will
help protect physician and patient privacy and confidentiality when they use
the Internet to send and receive medical information.
The AMA Internet ID will uniquely identify physicians over the Internet,
providing a more reliable identification technique with passwords for secure
Internet transactions. AMA Internet ID functions in the online world in the
same way driver’s licenses, passports or other trusted documents function in
the paper world.
Gold standard Multimedia has recently entered into an agreement with
Verisign to use the AMA Internet ID to authenticate participating physicians
conducting electronic pharmacy transactions in the Florida Medicaid
program.
In response to question seven, what incentives or other suggestions should
the government consider to accelerate the development and adoption of
e-prescribing standards. Finally, to successfully implement voluntary
electronic prescribing in the Medicare program, NCVHS and HHS must be fully
aware of the future Medicare environment. By law, electronic prescribing
standards must be in place by April 1, 2009.
At the same time, CMS actuaries predict approximately minus five percent
reductions each year in Medicare reimbursements to physicians from 2006 to
2012, with a slightly lesser cut in 2013. Concurrent with these cuts, the
cost of care for patients are likely to continue to grow at a pace that exceeds
inflation. This means that by 2014, after eight years of reductions,
physicians will be paid about 40 percent less than in 2005, while practice
costs will have increased significantly.
Although matching grants have been authorized to help the adoption of
electronic prescribing, funds have not yet been appropriated. In this
financial environment, it will be extremely difficult for physicians to
allocate the resources necessary to invest in new technology, unless it
provides an irrefutable, tangible benefit to their patients and practice.
To this end, careful and deliberative standards development is critical to
widespread adoption and achievement of electronic prescribing’s promise of
improved efficiency, patient safety and health care quality.
Thank you for this opportunity to present the views of the American Medical
Association.
DR. COHN: Thank you very much. We will take questions after
everyone has testified. Patricia Hale, you are next. Thank you for
coming.
DR. HALE: Thank you. My name is Dr. Patricia
Hale, and I am here representing the American College of Physicians. I
wanted to give you a little bit of background about myself, so that you can see
that I am bringing several different things to the table.
One thing is, I have been in practice for 16 years in New York State and
the Adirondacks, in a rural practice. In that practice, at least twice we
have tried to implement e-prescribing and have failed to be able to do
so. So I have been very interested in this area. I see the safety
and the benefits and want to bring it to my practice, but have had difficulty
doing that.
In addition, in my role in the American College of Physicians, I
participated as co-chair in the design and implementational work group for the
electronic prescribing towards maximum value and rapid adoption. I also
was working in that group to try to help develop ideas on how we can help
implementation, and also the development of functional prescribing standards.
I wanted to go over a few general comments from the American College of
Physicians first, and then I wanted to answer the specific questions that were
asked.
There are really five key areas that the American College of Physicians
feels are very important in this area of electronic prescribing. One is
that these systems need to be extremely usable. They have to be
functional in a physician office, and work in the work flow of physicians’
offices in a multitude of settings.
In addition, we strongly believe that interoperability is critical. I
need to be able to have all the patient information in my hands when making a
prescription, and right now in my office, my patients are on medication from
another physician that I don’t even know about. So we need
interoperability to bring you that information in a usable form, and also to
bring other information like patient history and other things that can happen
with decision support.
We need freedom of choice for not only the physicians on what medications
are the best to use, but also the patients and physicians on what pharmacies to
use. We need incentives that will help bring this to reality in the
practice of medicine, monetary incentives, but also incentives in terms of
improving the work flow and patient care and safety.
Then the last point is, we feel strongly that this is just a step towards
what we really need, which is a full electronic health record that can be
interchangeable and used in multiple systems and have interoperability.
So we feel that this is a step towards that, and everything that is done should
be geared toward being able to roll forward to the electronic medical record,
and e-prescribing systems that are in place should be able to move toward that
without any undue difficulty.
What are the key needs and barriers for wide-scale e-prescribing
adoption? There are several points that I would like to make on this
topic. The first is that we need a universally accepted prescribing
classification in coding nomenclature. I am going to go over that a
little bit more when I am answering the specific questions.
We need the standards to be very simple and easy to implement, and they
need to be adaptable and scalable. They need to be scalable all the way
to an electronic health record without difficulty. The systems need to be
transparent to the physician and patient. They need to enhance our
relationship, and not distract them or make it more difficult.
The systems need to help us overcome the acquisition costs. We need
incentives and other methods to help us be able to implement these in our
practice. We need pilot testing, because many of the situations where
these have been tested just don’t match the situations that many physicians are
in. There aren’t very many situations that I have seen that are anywhere
near the practicing environment that I work in, a rural environment with
multiple small practices, multiple small health plans and pharmacies that may
or may not be able to receive a fax. E-prescribing standards must conform
to the HIPAA security standards and the DEA requirements for controlled drugs.
What are the key features and functionalities of an effective e-prescribing
system to make it advantageous for physicians? There are several points
on this. One is that e-prescribing must prove itself speedy and
efficient, and it must be integratable with existing electronic health records
that physicians may have in place, but also integratable, to be able to use
decision supports such as Peer, which is a decision support tool developed by
the American College of Physicians. There are other knowledge bases as
well, and we need to have these available to improve our decision making
process when we are making a prescription, so it is a better job of doing the
prescription, not just a different way of doing the prescription.
We again need to have choices on what medications are the best ones to use
for the patient, and not have anything that obstructs our choices and make it
more difficult to make one choice over another, depending on other reasons
outside of what is the best choice for medication.
We also felt strongly, as was well described by AMA, that there needs to be
final authorization already in place when we do the prescribing process, and
that there is no difficulty in waiting for telephone calls and finding out that
the prescription is not accepted. It needs to be all part of the
prescription process, one smooth process.
We also feel that the systems need to be clear of commercial bias and not
force physicians into a certain kind of decision. However, we do feel
that it is very important that physicians have the information available for
pricing, so that we can make the best decision for the patient.
To specifically answer the questions, the first question, what are the best
standards and code sets to meet the requirements of the law, we feel that Rx
Norm, the ANSE PADP script and the NDC for pharmacy medication details are
important. I also think that the HL-7 and CCR are also important in terms
of messaging, in order to be able to do decision support and communication
between various health care systems, so that there is an accurate health
record.
Which of these standards and code sets do you use? ACP as a whole
obviously doesn’t use any particular ones, but in terms of our practicing
physicians who are using electronic prescribing systems, most of them are using
the NCPDP script standard.
The other problem that we see is that there is a great variety in the drug
databases or dictionaries that are used, so we have physicians using Moltem,
MediSpan and others. So there is a great difference across the physician
population in that. HL-7 is used for messaging and some of the more
sophisticated systems.
What are the strengths and weaknesses of the standards and code sets that
you use? Well, in terms of NCPDP script, it has different proprietary
codes for every unique product and for each pharmaceutical manufacturer.
We feel that this needs to be changed so that there is a single one. We
think this is important, because in terms of interoperability and ability to
change systems, if a physician is using a particular system and has to change
to another one, it will communicate with another system. If there is a
difference in these, there is a constant naming difference between the
different systems, that is going to make it more expensive, more difficult and
less timely to do those communications.
We also feel that the NDC is too finely grained and not really useful for
the physicians. We feel that there are multiple competing proprietary
drug vocabularies, as I said before. They are not necessarily compatible
with each other, and it makes it very difficult to use varying systems and
having to communicate.
The script standard is a messaging standard, and it needs to be
supplemented with standards for various message segments, both format and
vocabulary. Right now, most script messages have the drug and free text,
and SIG likewise, and those are not helpful when you are trying to use decision
support and other tools in the prescription process.
Script presently only covers prior authorization messages between the
pharmacy and the prescriber. We also feel that it is very important that
that loop is completely finished, all the way through the health plans and
PBMs, so that the decision can be made at the time of prescribing.
We also feel that the formulary and any other information must be presented
in a neutral way, that doesn’t force the physician to go down a certain path or
take more steps to actually prescribe a drug because of preferences of the
health plans or other forces.
In terms of the Rx Norm, we feel the Rx Norm is incomplete. While Rx
Norm does allow specifications of a particular drug’s ingredients, dosage and
form, this new system does not go far enough in allowing a physician to specify
critical details of the drug of choice for the patient’s medication.
Specifically this includes if a drug should be provided in a compliance
packaging form like certain steroids have to be done in dosage forms, whether
certain allergic ingredients such as gluten must be avoided, and what flavor or
other kinds of things for child prescriptions are necessary. We think
that it is very important to include dosing by weight and body surface and
other things, which are also going to be important for the pediatric
population.
There is a need for unification of varying state regulations as well, not
only in format and prescriptions, but also there are great differences in
ordering controlled substances. We feel that the HL-7 and CCR need to
work together to get messaging standards along with a script standard, so that
we have complete flow of messages between all the systems, and they can
communicate with all the other resources that can help us in the decision
making process such as communication with decision support and communication
with other entities that have prescribing information on patients, including
medication histories, which also can be used for a decision making process.
Question four. Is nationwide adoption of these standards and code
sets necessary? Nationwide adoption of the right standards and code sets
is necessary if we are to have a shot at cost effective, portable,
interoperable systems. Now that there is the physician side and pharmacy
side both armed with information technology, there will be more times when side
A is talking to multiple different company products from side B. Without
standards, it is going to be very expensive to build, rebuild and re-implement
technology for each new vocabulary, and very difficult for physicians to change
systems when they need to.
Number five. What are the e-prescribing standard code set gaps?
Standards that reconcile different state requirements for what must be on
prescriptions is one. Another is standards for format, vocabulary and
structure of formulating information, both the upfront periodically updated
data and the actual transactions and alerts. The standards for the doctor
level drug name and form for ordering convenience also needs to be
standardized, and avoid using free text, and only use it if necessary.
Standards for at least some elements of the SIG need to be enhanced and
developed to help drive compliance checking and reduce errors in
dispensing. Unified standards for messaging between all entities involved
in the prescribing process and health information using information networks,
HL-7 and CCR.
What are the barriers to the development of the adoption of these standards
and code sets? We feel that time and effort needed to develop the
detailed and unifying standards that are going to cover everything that is
necessary as a major difficulty. That is what we have seen in HIAA, and
that is what is going to be seen here. Trying to develop standards that
will work in the various multiple practice settings from hospital settings and
hospital groups and large medical systems, all the way down to individual
practices in rural areas is going to be difficult.
There is difficulty in overcoming the proprietary drug dictionaries,
getting everyone to agree to standards where people can share information back
and forth seamlessly. The upfront support that is needed to insure what
is developed will actually be accepted by everyone and be implemented is also a
problem.
ACP feels that the effort should be done similar to the way that the LOINC
standard became successful in lab tests. This is a good example of how
many of these things were overcome.
Number seven. What incentives or other suggestions should the
government consider to accelerate the development and adoption of e-prescribing
standards and code sets? We feel very strongly that grants for pilot
studies to look at implementation in multiple practices is very critical, and
to develop methodologies that can be shared with physicians to specifically
guide them in implementation in all these practice settings.
We think the grants that support the formation of regional health
information networks are critical. In fact, where I am practicing now,
that is the answer we hope to use in order to be able to bring e-prescribing to
practices like mine. Once we develop a network, we are going to be able
to get the integration with the health plans and pharmacies, so that
implementation will be possible.
We think that there are other inducements that are coming in the
future. Medicare pay for performance and reimbursement for differential
for proper usage of the system are all things in the future, but that is not
where things should start.
Then there should be incentives for physicians to help them move all the
way to the goal, which is a fully electronic health record, encouraging from
the very start that that is the end goal, and that e-prescribing is just a step
towards that.
In summary, the ACP greatly appreciates the opportunity to provide its
input on this very, very important endeavor. We feel that electronic
prescribing is certainly workable, and it is an important national system that
should be adapted, because it is going to benefit not just physicians and
health systems, but patients and all of us.
ACP lends its hearty support to this worthwhile endeavor, and is willing to
actively participate in e-prescribing and national health information
infrastructure pioneering efforts in any way we can.
Thanks.
DR. COHN: Thank you very much. David?
DR. KIBBE: Thank you, Dr. Cohen, and all of you
for the opportunity to testify before the NCVHS Subcommittee on Standards and
Security concerning e-prescribing standards.
I am a family physician. I am here on behalf of the 93,700 members of
the American Academy of Family Physicians. I’d like to acknowledge and
thank the more than 200 practicing family physicians who have supplied me with
feedback and comments regarding prescribing through various mediums, including
our listserves and direct contact at meetings in the past month, as i was
anticipating giving the testimony.
I would also like to acknowledge and thank my colleagues at the AMA and the
ACP for their very detailed comments regarding these standards. We are in
agreement with 99.9 percent of what has been said in the past 40 minutes
regarding these standards, and in particular many of the details, such as the
need to better define a common nomenclature for e-prescribing. So I am
going to take a little liberty here and abridge my comments and perhaps shorten
them, so that we will have more time for questions during these
proceedings. I ask your indulgence in that. Our written testimony
is more detailed.
The Academy is very interested in issues of health information technology,
particularly in seeing software and hardware that is compatible and
interoperable across health care settings. While the promise of
e-prescribing is important, we are concerned that standards adopted for
voluntary e-prescribing as a result of the Medicare Prescription Drug and
Modernization Act not hamper the larger functions or progress of the electronic
health record for our members.
Clearly, e-prescribing is complicated because of the several parties that
are usually involved. These include the patient, the prescribing
physician, the patient’s health plan, the pharmaceutical reference entities,
the pharmaceutical benefits managers and the pharmacies. There may be
others that I haven’t mentioned.
in order for all of these parties to communicate, standards will be needed
to allow connectivity between electronic health records and a variety of other
parties. Family physicians are eager to see a fully functional electronic
health record that allows communication within the broad community outlined
above, because there is a great potential for increased safety and
productivity for such a system, and they have identified this themselves in
their practices, particularly those who are using some form of an early
generation of e-prescribing now.
E-prescribing can offer significant savings in time, effort and money for
medical practices, physicians and their patients. These savings are known
to the committee and result from several work flow and information process
improvements such as fewer callbacks from the pharmacy or PBM due to improved
legibility and completeness of prescribing information, decreased telephone
time with patients, pharmacies and health plans regarding prescriptions. I have
had family physicians tell me that they have now discovered that their
telephone systems work as a result of putting in e-prescribing as part of their
health record. Increased needs of storage and access to patient
prescription information within the electronic health record of course is
another benefit.
These types of gains in efficiency and timeliness and accuracy of
e-prescribing may also significantly enhance and are related to patient
safety. I think it is very important to understand that family physicians
generally tend to couch their safety, quality and efficiency issues
together. They don’t always separate them out. They see them as
being integral.
Family physicians really care about the context in which e-prescribing
occurs. The best context for the patient is one in which a comprehensive
record of patient health information is available to the prescribing family
physician at the time the prescription decision is made. Clearly, refills
and renewals may not be as complicated, and may not require as much patient
information to review. However, safety, quality and cost considerations
all favor that e-prescribing be done within the comprehensive practice setting,
for considerations such as previous drug experience, comorbidity, drug-drug
interaction and even the patient’s economic situation can be assessed prior to
ordering the medications.
Family physicians clearly would prefer that e-prescribing take place within
the context of a fully featured electronic health record, which acts as the
central nervous system of a patient’s health care, and not through single
purpose or sliver applications that provide an electronic connection to the
pharmacy or PBM, but do not assist the clinician by providing access to other
information required for quality clinical decision making. Picture in
your mind the busy family physician with three or four PDAs hanging from his or
her white coat, and you have the idea of where this is going right now, and it
is not a good place.
Small and medium sized ambulatory care offices, where the majority of care
is delivered in the United States, must also be able to easily implement any
e-prescribing standards developed. Many family physicians feel strongly
that e-prescribing should be based on choice, not coercion. They do not
favor governmental mandates that would force physicians or their practices to
engage in e-prescribing, but believe this clinical tool should be voluntary and
based on its attractiveness to patients and providers. Doctors should not
be forced to adopt technological solutions that are not affordable or which
lead to unfair incentives for patients or practices to utilize a particular
commercial pathway for medication fulfillment. Such forced adoption,
either through governmental or payor mandates, holds the possibility of
exacerbating the already precarious financial situation which many primary care
physicians in small and medium sized medical practices find themselves, as has
been pointed out by my colleagues in earlier testimony this morning.
Therefore, e-prescribing must be free to physicians and patients.
Physicians should not have to pay for e-prescribing through transaction fees or
be forced to acquire technologies offered by health plans, pharmacies, PBMs or
their subsidiaries or agents, or forced to surrender their data that can be
commercially used later on. To allow third parties to determine what
technology is required in a medical practice would redirect valuable capital
that might more appropriately be spent on other technology purposes, such as
electronic health records and other upgrades. This is a very, very
important point among our physicians, that they bring up over and over again.
Physicians, we believe, are in the best position to determine their
practice management and clinical software needs. Similarly, physicians
should not be forced to report practice data to any of these parties who might
later use it for commercial benefit or profit. We have a term for this
whole issue, called data stewardship, which we think is extremely important to
continue to explore.
In any e-prescribing function, family physicians want to be able to see all
of a patient’s prescribing history from as many sources and databases as
possible in real time and when prescribing new medications. This is
really basic clinical common sense. It is an important safety issue,
particularly for Medicare beneficiaries, two-thirds of whom have multiple
chronic conditions. It has been mentioned here that often, physicians
don’t know of other physicians’ prescribing actions, and this is a real safety
issue that needs to be addressed here.
Generic and brand product names are often similar and confusing, especially
to the patient on multiple prescriptions. Most existing electronic
health records are not able to exchange drug lists from doctor to doctor.
We feel the continuity of care record may be an important standard in helping
to address that. But such fragmentation of the health care system often
leads to errors of duplication and prescribing of medications, and the promise
that e-prescribing holds out is that it will correct this problem. But it
will only do so if physicians have access to the full patient prescribing
history in any given community.
I am going to end here. Thank you very much for the opportunity to
give the testimony. I hope we will have more to say during your
questions.
DR. COHN: Thank you very much.
MS. SCHUMAN: Thank you very much for letting me
take this opportunity to testify before you on behalf of the American Nurses
Association, and also to represent nursing from a variety of perspectives, and
to represent those other providers who have a lot to contribute to this
discussion, and who represent the needs of many patients as well.
My background is that I have over 30 years of experience as a registered
nurse. I have been a certified nurse practitioner for 20 of those years,
and I have also, in addition to all of that, practiced and managed in several
of the situations in which I am going to offer examples today of areas where
the complexity of e-prescribing is even more complex than we have talked about
today. Let me say first of all that the ANA is
supportive of e-prescribing and of many of the comments that have been brought
here this morning. But I am here to hopefully broaden your perspectives
on the complexity and the applicability of e-prescribing beyond the physician
community and to other types of settings other than those that have been
traditionally considered as either acute care or ambulatory care.
The first two days of testimony are focused around pharmacies and
pharmacists, group and individual physician practices, CMS and other payors,
drugs and pharmaceuticals, adverse drug events, and the alphabet soup of coding
messages and communications methods. However, I think you may have been
missing the perspectives out e-prescribing from groups such as dentists and
dental practices, psychiatric and mental health practitioners, integrative
health practitioners, infusion services, pain management and ambulatory surgery
centers. In addition, I think there are other perspectives from public
health practitioners and organizations, from practitioners in school and
college health centers, home care, long term care, skilled nursing care,
assisted living facilities, and certainly last but not least, patients.
I believe that there are insights that can be gained from looking at case
studies and looking at examples of situations in all of these areas where
people are receiving prescriptions or are receiving care. In the process
of doing this, we need to look at the participants in the process, we need to
look at the settings and the work flow that goes on within those settings and
how it affects e-prescribing. We can describe minimum data set
components, we can identify regulatory issues and other concerns, and we can
inform the necessary educational system content, because there are systems
involved here, and people need to learn how to use those systems as well.
Let me give you some examples. There are several vulnerable
populations out there that gain health care on a regular basis from folks other
than physicians, and sometimes from physicians in these settings as well.
K through 12 school based clinics — often we know that in certain parts of the
country and in certain urban settings in particular, the school based clinic is
providing a huge amount of health care to the youth of that community.
Contraception, treatment of sexually transmitted diseases, unfortunately; many,
many children come into the school based system these days with chronic
conditions that have to be managed and articulated with those health care
providers out in the community, and they present many unique confidentiality
issues.
If we move up the ladder in terms of age. we look at college health.
There are similar confidentiality issues with that. There are many access
issues. Where is the care being delivered? I used to run a college
health center that had a student population of over 20,000 students. We
provided care about 65 hours a week. That still wasn’t a 24/7
operation. What is the mechanism for how the care is being communicated
and is continuous with other providers in the community?
Who is paying for the services in the college health population?
Furthermore, we know that they are a population that is very transient, that
they move at least twice a year, as you probably are all painfully aware at
this time of year.
Two things that school based clinics and college health have in common is
that that is not their core business, to provide health care. It is not
the core business of an educational institution to provide that care.
They may have limited resources around which to do that. We have heard my
colleagues talk this morning about the difficulty of implementing some of these
kinds of systems in locations where they are pretty much resource poor.
When we move further up the chain, we can look at reproductive health
issues for all adults. Those are provided in many types of clinics or
through many types of providers beyond the typical physician provider.
Again, often people seek that kind of care because they want particular levels
of confidentiality. There may be controversial issues there, there may be
timeliness issues there, and those need to all get factored in to the
e-prescribing process.
If we look at mobile clinics, nurse run clinics, clinics for homeless and
migrant workers, again we are talking transient populations. How do we
get to a point where anybody has any idea what might have been prescribed for
that individual from one location to another?
Then we go to the gerontology client population. I would suggest that
most of you around this room probably belong to an organization that was for
families with elder adults with chronic conditions. You know what it is
to make sure that your older family members are receiving the kind of care that
you want, that they are getting the kinds of medications that you expect them
to get, that they are taking them when they are supposed to be taking
them. We all know that they have a variety of means around which they try
to access those prescriptions.
There are many advantages to e-prescribing for that geriatric
population. However, we also know from their own computer literacy
standpoint, that for every elderly person who has learned how to use the
computer, there are many more that have not. So to even hep them
understand the concept that you can provide a prescription to the pharmacy of
their choice, and they don’t have a piece of paper in their hand, it might be
difficult for some.
We can also talk about self pay and indigent care. Those again are
looking for the best way to get their health care provided. They may be
going to a variety of urgent care settings and emergency rooms where it again
becomes difficult to keep track of their prescriptions.
Let me give you a specific example. I have spent about nine years in
providing hospice care on an on call basis. I take call for hospice
patients on weekends. We happen to have the luxury of a situation where
we have an electronic health record for each of our clients. We can do
that, because we have well over 100 clients at any one time that are in the
care system. It is not cheap, it is complicated, however, I have at my
disposal and laptop, where I can pull up the health record of that individual
when I get called. I can look at what has been prescribed for them.
I can look at what other kinds of situations, I can look at all the visits that
have been made, all the telephone calls that have come to and from the family
on behalf of the patient. However, if I am on a weekend and it is 2
o’clock on a Sunday morning and I need to call a physician and get an order,
call a pharmacy and get that medication filled, my options are pretty
limited. I can put that into the system, but how does that actually get
to the pharmacy? That is one question. The other question is, how
does it get confirmed by the physician or whoever is providing the order in the
end? It can take many days before that whole loop has been completed.
In the past, we always had a multi-carbon copy. It got sent around,
it came back, et cetera. With an electronics system, how do we do that in
a timely fashion? I have heard from the joint commission as they have
done surveys in a hospice and home care environment that they are seeing
exactly that kind of problem; they are seeing hospices and home care agencies
with patient records on a laptop, in a system. They get the verbal
orders, but when do the verbal orders get confirmed. There are gaps in
that process that need to be considered as you look for the big solutions in
e-prescribing.
It gets bigger than that. As you know, with hospice and palliative
care patient populations, we are talking about a lot of scheduled drugs.
We are talking about things that have the potential of being abused, although
with most hospice with patients and their families, we don’t see that, but the
potential is there. But how do we manage those scheduled drugs?
That makes this more complex.
We also end up with situations where pharmacists are being asked to
compound special preparation of medications. You have a 44-year-old
patient who has a brain tumor who is no longer able to swallow. However,
she is getting large doses of Decadron, seizure meds and so forth; how are we
going to get those medications in? We are often asking for pharmacists to
compound special preparations as suppositories or whatever in order to get
those drugs in. How do we manage that kind of situation in the
e-prescribing you talked about?
Other issues that are out there, coding for complementary therapies,
samples and other components that are not officially dispensed. We know
that goes on everywhere, physicians’ offices, other kinds of practices and
providers. How do we track that? Do we care? It is possible
that several months of some particular medication, and at this point I don’t
think there is any way to know that is occurring.
Compounded pharmaceuticals and those engineered and tailored for genetic
therapies have yet to be addressed. We have e-prescribing methodologies
needed as a format for ordering services and procuring other resources, such as
diet, nourishments, appointments, lab and radiology studies, PT, OT, other
consults, BME. It is not just a matter of prescribing medications.
When you get right down to it, it is a matter of e-prescribing for all sorts of
special needs that a patient population has. The home care population,
the palliative care, the hospice care populations, they often need supplies
such as PCA pumps, OT saturation meters. Those patients are being cared
for in an environment where those kinds of supplies are not there, they are not
ready at hand, they have to be ordered. There is a process that has to go
on to make that happen.
Other kinds of issues that have already been talked about.
Authentication and credentialling of a prescription author. If there
needs to be a co-signer for students, PA, PRNs and others, how does that
process happen?
We also know that there are medication orders that are obtained that may be
sent to the pharmacy, that may be given to the patient, they may be filled, but
maybe they are never filled, or maybe they are filled but are never picked
up. What happens to that? How does a pharmacist know that that is
what to do with that? How does the physician or other care provider know
that the patient has not followed through and picked up that
prescription? How do we know what the interaction is or what the dosage
is if it has never been completed? We also know that patients get
multiple prescription dosages. They are tried on 2.5 milligrams of
something, then they go back to the physician and they order a higher
dose. How does the rest of the system know that that has occurred, and
how are we going to manage those kinds of issues?
Other kinds of issues. Linking with diagnoses, outcomes and other
interventions. I think we want to be able to look at our e-prescribing
and talk about the fact that this relates to this particular problem, this
particular diagnosis or set of interventions. It is not just a blank
slate where we pop in all of the things that are being ordered. We would
like the system to relate to something besides a list of prescriptions.
Compile a consolidated drug list from many sources. Again, when
patients get medications from a variety of sources, how do we compile that so
that is all happening in one place? Record retention, disposition and
discovery, down time and emergency recovery procedures. That has been
alluded to here this morning. Let me tell you that if a system goes down
and you have medications and other things that have been ordered and you want
to get that into the central system from the laptop, the system is down, you
may not be able to do that. If you need to be able to get from the
central system to your laptop particular patient information and the system is
down, you may not be able to get that, either. Both things happen to
nurses in the field and others who have the electronic medical record.
How do we create backup systems for that?
I thank you for your time. These are a number of issues that I think
broaden the perspective. They come from a variety of provider settings
and situations. I would urge you to look beyond just the physician
practice or an acute care setting or your traditional ambulatory care settings
when you talk about this issue, because I think it is broader than that.
Thank you.
DR. COHN: Thank you, a very interesting set of discussions. We
will have questions and discussion now for the next while.
David, I wanted to start out with a question for you. We have heard
discussions yesterday and for the last couple of sessions about this issue of
standardizing SIGs on prescriptions. I was reading through your
testimony, I know you didn’t address it verbally, but we have heard from at
least those who have been trying to deal with the standards or whatever that
the complexity of coming up with a standard for SIG seems to be either daunting
or overwhelming or something. I’m not quite sure the right terms to use,
but it has been a major barrier.
I notice in your testimony that was representing a large physician provider
group that you seem to feel that this was a necessary activity. I am
curious about your concepts as well as others here about whether A, it is
necessary, and also whether it is obtainable.
DR. KIBBE: I think it is both necessary and obtainable. We may
have to lock a few people in a room for 48 hours and not let them out until
they come out with something that is reasonable, but yes, I certainly feel that
it is necessary, and we can do it.
There is a very large number of people who understand that that has to be
done if we are to do this right. So I don’t think there are too many
people who would argue — I wasn’t here in the last day or so, but I haven’t
heard people argue that we don’t need to do that. Most everybody I have
talked to believes that that is very important. It doesn’t have to be
perfect, but it has to be done.
DR. COHN: Do you think it is a low-hanging fruit, something that can
be done in the near term?
DR. KIBBE: I would guess so, yes. I’m not sure it is a
low-hanging fruit because I think it is going to be difficult. I think it
is going to take collaboration. But I think it is the kind of thing that
you all could facilitate to happen, and by getting the right people together, I
think it could happen fairly quickly.
DR. COHN: Other comments?
DR. HALE: I’d like to make a comment. I think it is very
important. I don’t think it needs to be done all at once. You can
start building it with all the standard ones. I have seen some testimony
suggesting where to start.
I think it is really important, because it is going to be really hard to do
any kind of dose checking and checking against diseases and lab and other
things if you don’t know exactly how much the medication, or how the patient is
taking the medication. That is what the SIG is about. So I think it
is necessary to have all the basic parts put in. The fine detail, unusual
medications that are ordered, no; we will need to have free text in every area.
But getting the basics done I think is very critical.
DR. NARCISSI: I just wanted to comment that I think it is possible,
too, but I think the most important thing to consider would be the testing
before it would be required for use. I think that was lacking in many of
the HIPAA standards, and we are paying for it now.
DR. KIBBE: Let me just add one thing. What makes this possible
and doable in part is that there are actually so few people doing e-prescribing
now, of the kind that we are talking about. If you already had 60 percent
and we were trying to change systems, I think that would be very daunting,
because you would have very established interests. But if we are
combining this with testing, as Pat says, I think it can work.
DR. COHN: Patricia, this is a question for you. You had
commented early on that you had had hands-on experience a couple of times with
trying to get e-prescribing up. You didn’t really reference exactly what
were the major roadblocks, or what caused things to not work. I wasn’t
sure whether the issue was connectivity or standards, or the fact that the
systems were so burdened with required data entry fields. Can you comment
on this?
DR. HALE: I’d be glad to. There were two different situations.
One situation was trying to build it out of a home-grown electronic medical
record. The second situation was trying to do it through one of the
companies that came and went back in the dot-com era.
There were several problems. Some were technology, but most were
interactivity. The technology part was that a lot of the technology that
was used at that time, and some of it still is, was difficult. It just
doesn’t work in your hands as well as it is supposed to. Trying to use
radio frequency to print something out, trying to set up a network that can
transfer the information through your office, those kind of things are not as
easy to set up as people thought when they were trying to do it.
But the biggest problem was the interoperability with our patient
management system, getting information across, patient names and demographic
information. Even though we did that, it wasn’t continuously
updated. Then we didn’t have updated formulary information. Even
though it was updated on a every so many months scale, it was completely
outdated frequently, so we were still getting all the phone calls. We
didn’t have the ability to have the process go all the way through to justify
to authorize the prescriptions when there was a problem with whether it was a
certain number of doses and things patients could have. So we didn’t have
that.
Then when we actually tried to connect with the pharmacies, the local
pharmacies couldn’t connect. The ones that even had fax machines did not
leave them on, and didn’t want to have to police them and didn’t want the
responsibility of making sure that they got the prescription. So all the
way through, every single step, there was a roadblock. Many of those
roadblocks still exist.
We also have a number of — our biggest health plans in my particular area
are very small relative to nationwide, and are not nationwide players. So
we have multiple health plans to try to get formularies from. They are
small health plans which don’t necessarily have the resources to do this sort
of thing easily.
DR. COHN: Thank you for reminding us of the reality of the situation,
especially the rural practice. Other questions, comments?
DR. STEINDEL: Thank you, very nice presentations from all of you,
thank you very much. You did a very nice job of covering the issue of
standards, so unlike the last two days where I have asked everybody standards
questions, I am not going to do that this time.
But I do have a question that started with Jean’s talk from the AMA, and
was a theme picked up by everybody, where she used the words undue financial
burden, about putting this system into place. Obviously, doing this is
going to cost something, and where the cost goes and how it goes is an issue
that needs to be resolved.
I would like to ask Jean and then the rest of the panel to pick up on this
and discuss it further. Could we also use the words undue increased
financial burden? Meaning that if we could put this in as a revenue
neutral proposition, is that palatable?
DR. NARCISSI: Last fall in AMA News, there was an article after the
MMA was signed. There were some estimates pulled together by AMA. I
couldn’t track down all the resources, so I didn’t include it in my
statement. But they were estimating that it would cost a physician’s
practice $140,000 to implement just e-prescribing, the e-prescribing end of
it. But I don’t know if that would be dependent on the electronic record
system also.
So there has to be some kind of financial incentive that will help the
physician do this, or even non-physician providers.
DR. HALE: I’d like to comment on that, too. I think that the
actual cost of buying the system is important, and you can do calculations on
return of investment and perhaps upfront buying and payback. But the part
that is going to be really hard that you need to cover is the actual
implementation of change in work flow, teaching and education, the time that
the practice is going to be down by about 25 percent of productivity
implementing the things, and trying to change things in the work flow.
So I don’t know how you quantitate those and make it budget neutral without
a lot of work.
DR. SCHUMAN: It seems to me also that we are not just talking a
physician group or a provider group purchasing this system and having the
development of it, but everybody that it touches as well. It is the
pharmacist, it is the DME, it is everyone else who gets connected to that
system in some fashion that is going to have the expense. It does change
the work flow. There are always questions with these kinds of systems of
user friendliness, and have they been instructed in such a way that real people
can use them, and do they do so in a way that is relatively efficient.
If we go in and say — I think you are right, we are certainly not going to
go in and say we are going to save money. We probably aren’t even going
to save efficiency in the short run. It usually takes more and creates
more time and effort than it does reducing it, at least from the systems we
have implemented in other kinds of settings. That is my perspective.
DR. KIBBE: I would like to take a slightly different cut at that,
because I don’t think you can answer that question, anyone can, until we have
much, much better data about this industry.
We are all going forward towards what is becoming a very large scale
societal transformation around health information technology. We are
seeing considerable alignment among many of the parties, and significant
collaboration of parties who previously have not been as collaborative.
However, what strikes me is that we have almost no data about this
industry, meaning that we — I’m sorry.
DR. STEINDEL: Can I ask a quick clarifying question? When you
say industry, do you mean the health care industry as a whole or the
e-prescribing industry?
DR. KIBBE: I was going to define that. I am really talking
about health care providers, hospitals, doctors. I am particularly
interested in the small and medium sized medical practices. So let me
just be very clear, in terms of the use of information technology in the small
or medium sized medical space, market, the $200,000 or so small and medium
sized medical practices in this country, there is virtually no baseline
information regarding their use of information technology, the kind of
information technology they use, the cost of that information technology that
they currently use, what they are going to need or replace in the next six, 12,
18 or 24 months, et cetera.
The government and the industry in tires know exactly how many sized tire
comes out, what tread size they are, can make projections on the basis of the
next week, month, year, with regards to sales and so forth. We can’t do
even the most elemental projections around the adoption of the technology in
this marketplace, because we don’t have any information. We need to start
immediately gathering that information.
It was interesting that — I heard David Baylor speak at Tepper last week,
and had a chance to talk with him. This was one of the points that he
made as well. As we approach this very large scale conversion in most
small and medium sized medical practices in many other organizations in this
country, from a paper based system to electronic systems — and I have every
reason to believe that the majority of practitioners in those spaces want to do
that — we really need to understand that without these data, we may end up
thinking about an e-prescribing industry off on its own.
Indeed, there may be an e-prescribing industry formulated, but I would
suggest that that is a very piecemeal way to look at this emerging
market. It is much more — it is not about putting faxes into doctors’
offices. It is about putting complicated decision support tools into
doctors’ offices over time. If we don’t think about it in a sophisticated
enough way, we will end up with enormously increased costs of care, without the
payoffs we want in terms of quality, safety and efficiency.
So I think we can all speculate about what the costs are or what doctors
are willing to spend and so forth. But I think it is really important to
understand that they are struggling right now to acquire affordable standards
based systems for their offices, both for practice management and for
electronic health records. Our members tend to see electronic health
records as part of that. To ask them to go off and purchase systems that
don’t fit in anywhere, or are simply another sliver application is defeating
the purpose.
Additionally, without the raw data that we need and I think we could get if
we had a concerted effort to do so, it is going to be hard to know if we are
making progress. How will we know two years from now whether or not there
are more physicians using electronic health records and benefitting from them,
and the patients are getting better quality care and safety is improved, if we
don’t start to measure these things fairly accurately right now? In my
mind, we won’t know. We will still be speculating.
DR. COHN: I think Patricia is going to comment on that.
DR. HALE: Yes, just one very small comment. There have been
suggestions that upfront loans and things that can be paid back once there was
a return on investment might be a model. One of the problems with that
is, at least in my area, the general physician and ACP’s physicians, we lack
such confidence in our ability to be able to meet our economic demands on a
year to year basis. We don’t have any confidence that we can ever pay a
loan back.
We also feel like all these things are so much more complex than ever
predicted, because every time something gets implemented, it always costs more
and it is much more difficult to implement. I think it would be very hard
for people to take that step and take a loan to put something in and feel like
they could pay it back.
DR. STEINDEL: This is a followup question for Jean. You
mentioned during your talk that the AMA was gathering some statistics in this
area. How much of the list that was described by David might you
have? And as a corollary to that, will you be ready and able to present
that at the NHII conference in July? I think it would be important to do
so.
DR. NARCISSI: I’m not sure it will be ready by that time, but I’ll
check when I get back. The past few years, we have done a few surveys
regarding use of technology in physicians’ offices. We have done them
like every two years, so I think we have done two of them so far. This
will be our third. Although they are pretty high level, what we have
found in the past, what is the percentage of physicians with access to the
Internet. It is pretty high, although not within their office. They
may be the one that has access to the Internet on their private desk, but their
staff may not have access. So those are the kinds of things we were
finding.
But hopefully we will find out who is currently using e-prescribing.
That could be a different definition among the people that are responding to
the survey. So we will see. We also added a few questions regarding
basic billing types of questions on practice management systems and
attachments, claims attachments.
DR. COHN: I’d just follow up on that. If you can’t do it for
the NHII, certainly we are very interested in that information.
I was just going to follow up about your comments, David, in terms of the
lack of data around what is happening in the industry. I think of it in a
slightly different way. The IOM patient safety data standards study which
recently concluded — I understand the book is out, although I haven’t seen it
yet. As part of that, I did a review of all the information. I
think there were some very large number of studies that had reported to do
exactly what you are describing, an evaluation of the status of the information
infrastructure out in the health care industry.
What you saw was that because of varying techniques, varying definitions,
varying samples, that depending on the state you looked at, I think we are
certain that there are somewhere between three and 20 percent of implementation
of the EHRs out there in the industry, exactly the conclusion you reached.
So I don’t think it is for lack of people out there doing the
studies. It is just that people seem to be — I think there is an issue
about methodology, perhaps.
DR. KIBBE: May I differ with you respectfully, Dr. Cohn? I
think there are many, many studies of the nature that you have spoken of.
They have tended to focus on physician adoption of electronic health
records. They have been very — two things. One, they have been
focused on physicians’ opinions about things, not what is really happening in
the industry.
For example, no one has really surveyed the 267 electronic health record
companies that we have identified at the Academy that are used by our members,
their numbers, in the same way that many other industries in this country have
data gathering efforts in which the suppliers of the products or services
report their information.
So it isn’t simply going into the doctor’s office and asking them, do you
want to use e-prescribing, or do you use an electronic health record. We
need to think about this in ways that are non-traditional. This is an
emerging market. This is a huge investment that hundreds of thousands of
people are going to be making over the next few years. Some of them are
family physicians, some of them are internists, some of them are pediatricians.
A second point that I would like to make is that we can’t do this just on
the basis of a single medical specialist society. We have a fairly large
scale study that we did in early 2003 or 5500 family physicians, and we found
that there were 267 different vendors, there were 1300 members, 24
percent. That information is all very piecemeal.
So what I am suggesting is really to rethink how we gather data from what I
will suggest is an emerging market of small and medium sized retail units in
health care — I would include small hospitals in that — and their use of
information technology now, detailed information, which can only be done
through a rather comprehensive sampling methodology. It would give us a
real sense of where we are now.
You are right, those studies vary from three to 23 percent, who
knows? I have heard people who are in the so-called e-prescribing —
replicating the same mistake; we have got people in the e-prescribing industry
of products and services who will tell you that 50 percent of the pharmacies
are hooked up, or 12 percent of the doctors are already using it in this
state. I don’t believe any of those data, because we really have no way
of assuring the quality of those data. So I am suggesting this is very
important.
DR. COHN: I should just jump in and apologize to one of our members,
who I think does one of those studies, in case I have in any way made any crass
comments. My apologies.
MR. BLAIR: Well, MRI should be involved, no question, in that whole
thing.
MR. REYNOLDS: I have a question for each of you on each of our
testimony. Jean, — could you give me a few more words on that?
DR. NARCISSI: What I have heard from practicing physicians is that
some of the detail isn’t there. For instance, an example that was shared
with me was the flavorings for pediatric prescriptions are not included.
If you have dealt with children that have to have a specific kind of flavor,
who wake up screaming for their medicine at three in the morning, you will
understand that that is a very important thing. So it was detail like
that.
What I think needs to happen, not just with the Rx Norm, but with all the
standards, is that we have to have all of the potential users at the table when
we develop these. We found from experience with some of the HIPAA
standards that that wasn’t maybe the case. So somehow we need to get the
providers to participate in standards development efforts a little bit more.
DR. COHN: Can I ask a followup question on that? I am an
emergency physician, I am not a pediatrician, and I was wracking my brain as
you used that example of flavorings. I am well aware of many over the
counter medications that have multiple choices of flavorings, and I am just
trying to wrack my brain about pediatric prescription medications available in
four or five different flavors. Is my brain just not functioning well
this morning?
(Simultaneous discussion.)
DR. HUFF: That particular issue is one that Rx Norm is well aware
of. But there are two solutions. One is to make precoordinated
cherry amoxycillin versus strawberry amoxycillin, or to have another field on
the pad that specifies the flavor. So it is not part of our Rx Norm, but
it is part of the prescribing process. So that is the usual discussion of
where do we put that. That is the same case for when you want to say
sugar free or alcohol free or all of those other kind of things, which are also
important. It can be done as a pre-coordination in the code system, or it
can be done as a separate field. Right now, people are leaning towards
having that be a separate field that is specified separately, because it
happens infrequently and it is really hard to keep up with.
DR. HALE: I just want to briefly say, from an internal medicine
standpoint, the one thing that bothers me the most is that the other
ingredients that are in there — and gluten is a major one. I have a
number of patients that have gluten enteropathy. It is a big problem,
trying to find out what medications they can take, because that information is
not readily available.
So those kinds of ingredients, allergies, there are people with allergies
that die and things like that, that is critically important, and that is not
included, and that could be.
MR. REYNOLDS: I’ll go to my next doctor visit and ask him for a
flavor chart. It should be fun to watch his face. He is very
electronic, but it will be interesting to watch what he does.
Patricia, you mentioned the no commercial basis. We heard some
opinion on that yesterday, and I would like to see if yours relates to that, or
if you present it in some other way.
DR. HALE: What we have been saying from an ACP standpoint in terms of
commercial bias is that we don’t think that there should be direct influence on
the prescribing practice. There should be information provided to a
physician when they are making a decision, so that they can make the best
decision, but they shouldn’t be pushed into some particular direction.
For example, you could build a prescription program — and it is not like
this has been abused so far in the past, but it is a risk — you could build a
prescription program where I would have to go through three different clicks to
click a medication that was less desired, and then I would only have to do one
to get them one that was desired, or the list could have the ones that were the
most wanted, and they may not even be listed, ones that I would have to look up
through an alphabet or something to get to other ones.
Those are the kinds of bias that we are really concerned about.
MR. REYNOLDS: David, you mentioned —
DR. KIBBE: Do you want me to comment on the commercial issue?
MR. REYNOLDS: You didn’t put your hand up for that, but I have
another question. The question I had is, you mentioned that e-prescribing
must be free to the doctor, and we have heard that before, and you said the
patient. What are you envisioning? What takes you to that point,
and how would you see that e-prescribing would have some charge to the
patient? I didn’t quite understand the relationship there.
DR. KIBBE: I don’t mean that e-prescribing technology per se needs to
be entirely free to the physicians. I am talking about primarily the
issue of how the information is transmitted and transacted. We don’t want
to see transactions costs, either direct or indirect, in that exchange. I
think that would really put an undue burden on particularly the primary care
physicians.
I also think that there are hidden ways of doing this that may involve
asking physicians to voluntarily surrender certain information that they
supply, either as part of the e-prescribing process or along with that, and
then people monetize those.
This is a very tough issue, but the principle that the Academy has outlined
is one of data stewardship. That is, as physicians are using electronic
health records and other electronic means to collect, store, manage and
transfer information that comes from their practices and involves their
clinical decisions, we want to put a stake in the ground, and we are doing so,
saying that the way these data — and I am talking primarily about
non-identified or de-identified and aggregatable data, so HIPAA has a floor on
this, and we are talking about issues that are above that floor — we need to
assure that those data are used ethically, that the physicians who provide
those data have some say in how those data are used, that we enter into a
dialogue with those commercial parties prior to the use of that information.
MR. REYNOLDS: This is my last question. As we have been
listening for the last few days, up until your comments it was a lot more
linear than you made it seem, which is good, that is a good perspective to
have. A little more on the second signatures. In other words, as
you think of the physician and then the pharmacy and then the PBM and
everything that we are looking at, when you throw in a second signature, that
obviously deals in a different way. So if you could help me understand a
little bit about what second signature means.
DR. SCHUMAN: For a number of providers out there, depending upon the
state in which you work and the circumstances and the classification of the
drugs that you might be prescribing — for instance, nurse practitioners may
have to have a collaborating practice agreement, or they may actually need to
have a second signature on a medication that they might prescribe.
PAs would always work under the auspices of a physician, so there may be a
second signature required in those circumstances. In an electronic record
system, such as I have seen with hospice work, for instance, you take a verbal
order or you put in a verbal order. Even with a scheduled drug in the
middle of the night, I can call the pharmacist that I have a contract with as
the organization and say, I need to order oxycontin X, Y or Z, or liquid
morphine, five to 20 milligrams Q one to four hours, and the physician that has
agreed to this order is such-and-so, and give them that information.
There needs to be a followup then. Even if I put that into the system
electronically, somewhere there has to be a completion of the link, that that
physician has to then also electronically or some other way confirm that yes,
that was a verbal order, that was the appropriate order, et cetera.
The pharmacist may or may not choose to call the physician in the middle of
the night. If they are comfortable with the physician and their typical
ordering practice for hospice type patients, they would probably accept
that. Or they have been around this route with this patient over the past
week, and they know that is where it is going. In other cases, they are
going to have to call, but even that is still a verbal order.
So how do you complete the link when you have an electronic system?
MR. REYNOLDS: One of the things that we got into yesterday, we got
into the idea of, a patient goes to a facility, and if the doctor leaves, the
facility can still do refills and everything. You are now taking us back
— at least, I think I hear — taking us back down to the actual person doing
the actual thing and their actual involvement.
DR. SCHUMAN: Right.
DR. COHN: I think for narcotics, which is a particular issue, clearly
that is not something that just goes to a facility. That physician is
very critical.
DR. NARCISSI: I just wanted to comment. Working with the AMA
Internet idea a little bit, that is an authentic patient product for a
physician. We have been trying for the last several years to figure out
how we could either develop a product for the physician’s staff or delegate the
use of that product.
We found that it has been extremely difficult, because although for
authorization for the physician, we can use the AMA master file, which is a
huge database of all physicians, but there isn’t anything similar to that for
the other staff in the office. So this whole area of authenticating that
I wrote the prescription, or I am the second signature, or I have permission to
use my physician’s digital certificate, has not been figured out yet.
DR. WARREN: My question is to follow up with a comment that David
made. I am trying to understand. You had mentioned that one of the
requisites that you had wanted to see in e-prescribing was for the physician or
the provider writing the prescription to have access to a medication list, and
that was especially important, or if you had multiple physicians or providers
caring for that patient.
But you also made the comment that you didn’t want to have any kind of
requirements for physicians to have to do e-prescribing. So I just wanted
to hear how you — at one point you were saying we want to have this, and at
another point you are saying, but we don’t want to mandate this. So how
are you going to deal with physicians who want this information when they are
prescribing, but they are not sure they will get it because other of their
colleagues may not be participating.
DR. KIBBE: Good question. I’m not certain that in all cases,
those are completely resolvable. However, I would suggest to you that
most of the prescribing information about physicians who write paper
prescriptions now, which is of course 98 percent of them or more, is currently
available — could be available to them electronically, because those
prescriptions get into the pharmacies’ prescription information systems, into
the PBM’s prescription information systems, into the health plans’ prescription
information systems.
Suppose I was a physician who didn’t do e-prescribing. It is still
conceivable that a third party could fax to my office a list of all the
medications that had been prescribed by a physician, let’s say, in the last six
months from those other databases, even if I weren’t participating, not that
that is necessarily a good thing to do.
Understand, we are in favor of e-prescribing if it is done right, and we
will certainly work very hard, and we are doing so now — in our partners with
patients program, we have EHR vendors and we have RxHub and SureScripts as part
of that group, so we are really trying to make those connections.
DR. WARREN: That is why I asked. As we have been listening over
the last two days, everybody has been talking about, we need to close the loop
on all of these prescriptions. When I was listening to your testimony,
here was this hole for me. I was trying to figure out how we can close
that hole, because that will probably be a reality. It will be a long
time if ever that everybody is going to be totally online. So thank you.
DR. HALE: Can I just make one quick comment on that? There are
levels of communication and there are levels of knowledge needed to enhance
what you are prescribing. The simplest one is, all you need to know is
what is the list of medications the patient was on. That can be derived
from other sources, what has been prescribed. A health plan may know, and
various places may know.
But in order to make the best decision is when you can include decision
support. Once you get to the decision support part, I need to know what
diagnosis that physician chose. It may or may not have been listed in
billing information. Oftentimes that is extremely erroneous data.
So to get the really good prescription writing and to have the enhanced
knowledge support, that is when the communication between the various
physicians is really needed.
DR. WARREN: We have heard a lot about that, about needing to have a
clinical diagnosis, not a billing diagnosis. I am trying to pull all this
together. I have been trying to keep notes, what are all the pieces here,
because we are just beginning to find out that there are pieces in this process
that probably no one person has identified until they start hearing from
everybody else.
DR. HALE: Then people like me don’t find out until you try to do it.
DR. WARREN: Before we started this, I thought I knew what a
prescription was.
DR. HALE: I know the feeling.
DR. COHN: I was just going to observe that there is a very slippery
slope between fully being able to prescribe well and the functionality of an
electronic health record, a very slippery slope on this one. Jeff, you
had a question?
MR. BLAIR: Yes. Could all of you help me understand a little
bit better the overlap between prescribing within an ambulatory setting and the
acute care setting, both of which may go to retail pharmacies? What is
behind the question is, clearly we have some ambulatory settings that are
implementing EHRs, and then as part of that, there will be an integrated
e-prescribing capability. Others are implementing separate e-prescribing
capabilities on networks, with the idea that at some later point they will have
an EHR. Other folks are winding up doing both, but they are separate.
All of those things are the way the market is emerging. It is
something that the subcommittee, NCVHS, has to consider from the standpoint of
whether standards could remove barriers to the later integration of these
systems.
So my question is getting beyond the ambulatory piece to trying to better
understand when physicians are placing prescriptions, and to what degree we
need to consider the integration between the ambulatory and acute care
environments with respect to standards for e-prescribing. Could you help
me understand that environment a little bit better?
DR. NARCISSI: I think what we are seeing right now with the current
systems are use of a lot of proprietary types of standards, if you want to call
them that. So I really believe that no matter what the setting, no matter
what the specialty, you need to try to figure that out in the standards that
are currently identified as being possible e-prescribing standards. We
need to have more outreach maybe to include those people that are at the table.
I like the way HL-7 handled the EHR ballot, when we were trying to look at
what pieces were necessary for the electronic health record standard.
They went to different regions of the country. I think medicine is
somewhat regionalized. I think you do need to go to maybe different areas
and figure out what our the needs. But as far as the overlap, I’m not
sure I could comment on that.
DR. HALE: I can comment on it, because I have been in that situation,
too. I work in and out of a hospital in the outpatient setting.
What happens is, when you admit a patient to the hospital, there is a separate
formulary for the hospital, and usually a different dictionary database as
well. So my patient may be on a specific beta blocker when they go to the
hospital, and once they get there, the formulary of the hospital is using a
different vocabulary or dictionary as well, and switches them to a different
beta blocker. When they are ready to come home, that hospital may or may
not remember that they were on a different one when they came back to the
office. So they can come back to the hospital with a different
medication.
This is a real problem in terms of errors, when there are new medications
started, when you are missing that link between the inpatient and the
outpatient. I had an error a couple of months ago, when a patient was
discharged. It was a handwriting issue, and it was misread in the
pharmacy, and the patient was started on a psychotropic instead of a blood
pressure medication. The nurse came back in, and the family was saying,
she is acting wacky and her blood pressure is still up, and they were trying to
figure out why she was on that medication.
But those kind of errors are going to be greatly increased and those risks
remain when you don’t have things synchronized from the inpatient to the
outpatient. They are not using the same dictionaries. You don’t
have SIG all the way detailed out so you know you are going to be doing the
same dosing. That continuity of handing a patient on and off really can
have a lot of errors associated with it if that is not taken care of.
DR. SCHUMAN: Let me add the human element to this as well. We
likewise see that. They are at home, they are being managed at
home. They go into the hospital, they come back out, the instruction to
that patient or that family is also not clear. So they come home with a
handful of different medications than what they went in with, all different
names. It is not clear what they are for in some cases. They have not
been given very thorough discharge instructions regarding meds. Then they
had a whole host of different medications at home, and they are going, should I
be taking all of the things I already had at home and these new ones? Do
I take just the new ones? I always took two of these at night, and how
you are telling me I should only take one.
DR. HALE: They will double up.
DR. SCHUMAN: Yes, that is exactly right.
DR. HALE: Two different beta blockers.
DR. SCHUMAN: So it is a nightmare. So when we come in and do an
assessment to readmit them to hospice or home care or a palliative care
program, we are sitting there with a handful of stuff that we are trying to
figure out. Unless there is someone like a nurse or someone else who can
advocate to the physician and say, I have 40 medications here, what ones are
they really supposed to be on, the family may not be able to do that, or at
least speak enough of the same language to be able to accomplish that
successfully.
DR. KIBBE: You always have a way of getting to the heart of these
problems. It is amazing. I don’t know the answer to your question
precisely, that is, how can standards help these divides, but I can make some
comments on what those divides are.
Historically, there has been a divide between information technology in
large institutions like hospitals and integrated delivery systems and academic
medical centers, and the relative lack of information technology penetration in
the small and medium sized medical practices, where most of the care is
delivered and most of the prescriptions are written.
So one issue is that, as a result of this whole push of e-prescribing, we
are running right up against that divide. That divide has a number of
different dimensions, which are cultural, economic, maybe spiritual, but one of
them clearly is the vendors. In the 267 vendors who were in our study,
they are almost all very, very small companies, who do no business in the
hospital space. And in the hospital space, you have got another group of
vendors who generally speaking don’t understand ambulatory care electronic
health records, although we are starting to see that change. Many of them
have entered this space and failed miserably in the past. They have
bought up products and services that didn’t work in the ambulatory care
world.
So one of the issues is to begin to address those divides. This is
another area where data is called for, and to encourage those bigger companies
that are moving into the ambulatory care space, where there are products and
services to do that well, so that they don’t fumble and drop the ball again,
because it would just set us back with the whole e-prescribing process if we do
that. Clearly, one of the ways they can do this well is by working with
the medical specialty societies, who are here today, and others.
MS. CRONIN: Two questions, one related to drug information and one to
knowledge representation or decision support.
ACP had a specific recommendation in their written testimony regarding
universal vocabulary and classification system for drug information, and how
that should be maintained and available in the public sector, as a public
good. While we do have efforts under way to get a structured product
label out there, we have yet to really seriously consider how it might be done
for off label information, which everyone realizes is very important, given the
amount of prescribing decisions that are made for indications that are not part
of the label.
So I am wondering if the panel could comment on whether or not commercially
available systems in your perspective are adequate, and what the government’s
role should be in perhaps providing that information.
DR. HALE: I don’t think that they are adequate. I think all the
standards have major gaps, and that we do need something that unifies them, and
that there is a standard amount of information.
You are never going to be able to build something that has every detail
that makes it a perfect system, that is understood. But you can have the
major 70 percent or 80 percent of the prescribing information that is most
commonly used. That you can have in an agreement across.
It is important, because as we move forward and want to bring more things
into the decision making process, unless there is confusion and unless there is
variability in those parts, and the more standardized that is, the more
information I can bring in to be part of that decision making process, and the
less likely there are going to be errors, because of divergent pieces of
information that didn’t get lined up and didn’t get properly matched.
DR. NARCISSI: I just want to emphasize that whole testing phase
again, whenever applications are out there, to really test those.
This whole area of the controlled substances, that has not been figured out
yet by DEA, although I do understand that that does represent a small
percentage of the prescriptions that are being written with the old 80-20
rule. So I don’t know if that could be factored into any type of testing
or not.
DR. KIBBE: My only comment would be to echo that we really need good
pilots and demonstrations under very controlled circumstances, so that we don’t
make assumptions about things that are going to work, that really aren’t going
to. It is a perfect opportunity to do so, because I think we have got a
lot of willing parties to do that.
A lot of the stuff needs to be tested, and I think where things don’t work
and where the gaps are can be tracked very carefully, and we will find out how
to do that.
MS. CRONIN: It seems as though from a qualitative perspective, AFP
has already made the assessment that there are some gaps out there, in terms of
commercially available sources, given your written testimony.
DR. KIBBE: Yes.
MS. CRONIN: But what you are saying is that there has never been a
written evaluation of the adequacy of that information.
DR. KIBBE: Correct.
MS. CRONIN: My second question has to do with decision
supports. Since this is an area that is very immature in the scope of the
standards that we are considering, I am wondering if you all have some specific
comments regarding the scope of standards for decision support, and the need
for flexibility or anything from a user’s perspective that is important for us
to consider.
DR. KIBBE: I’ll start off. One area that I would suggest to you
that may not seem obvious, but I think has been very significant and will lead
to better decision support is the AMA’s physicians consortium standardization
of specific quality and performance measurements has now found its way into the
docket project, and may help in the development of a pay for performance
program that has standard measures, standard specifications for those measures,
so that the vendors can — that piece is very important, because much of that
information is what would then go back into any algorithms about what has been
done or what hasn’t been done or what should be done.
So although it doesn’t appear at first blush to a lot of people in terms of
decision support, I think the collection of those data are a very, very
important founding brick if you will in that edifice.
DR. HALE: In terms of decision support, that is a very broad
topic. In the EHI document we had, we were going all the way from
decision support to drug-drug interactions, all the way up to pulling in
something like Peer. It is a very, very broad topic.
Certainly the basic drug-drug interactions and straightforward alerts and
things are reasonable. However, the biggest problems of the decision
support are going to be choosing what kind of alerts are mandatory, important,
and making sure that they are not used so often that they get turned off,
having limitations on how much you can turn off things without using it if it
is not usable.
Then going all the way up to the other side is, when you get to the more
complex decision support, it has to be billed in such a way that it is not
pushed at the physician, and the physician can go and drive down into the
information. There are pilot projects going on now, just being started,
to look at how you can provide that kind of information.
There is a peer project with one of the vendors under way, where when I am
prescribing, I can click a little icon and go in and drill down to further
information. I can get CME credit even while I am doing it, but I can
drill down in the prescribing process to have that information help me.
But it is not pushed at me.
I think the biggest challenge is being able to decide where those lines
are, what you want to push at the physician and what you want to make available
to make a better decision.
MS. CRONIN: Can I just ask which pilot is that?
DR. HALE: It is with AllScripts.
DR. NARCISSI: I just wanted to comment on some of the stories that I
have heard from physicians. Depending on what type of vendor they were
working with, sometimes it would take them longer to an get electronic
prescription, which means time with the patient, than it would to write out a
paper prescription and follow the paper process. So I think that has to
be considered, because there aren’t really standards out there for different
vendors.
The other comment that I just wanted to follow up on is that the AMA has a
consortium pulled together of all the specialty groups, and has been developing
performance measurement sets, but there still is a lot of work to be done.
DR. SCHUMAN: I would like to also comment that from a broader
provider perspective, I think at least in nursing, we are trying to work on
protocols and decision support models that help us get drilled down to, if a
patient is restless, is this terminal agitation, is this something else, are
they hypoxic, what is going on, and then make a decision about what is the
appropriate treatment for that, and/or how do you coach the family or anyone
else to provide that treatment.
There is relatively little of that out there. Even when it comes to
pain management or whatever, there are some standards about, here is the best
range to give and so on and so forth, and how to titrate that. But the
protocols that are established are still pretty vague, so there is a lot of
work that needs to be done around that as well in order to be able to kick it
into a more complex decision support process. There is a lot of work to
be done in clinical practice that needs to feed into this.
MS. FRIEDMAN: My question has to do with the minimum data elements
that would be required in a prescription. We have heard about it in two
ways. One is the content, for example, Mary Jean brought up about the
second signature. That might be something to consider.
The second is the format, would you need free text to deal with the
flavoring choices, for example, or the things that are infrequent. We
have also heard that there is a lot of variation among states on this issue.
So my question is, given all that, do you have any advice for us as to
where we should look for the minimum data elements and the standards for those?
DR. HALE: I have a slight bias, since I worked on the HI
project. I actually was the one that worked on that table that is in
there. There wasn’t total agreement. There were gray zones, but
there are a few things that everybody pretty much agrees on. You need to
have a single format, not differing state to state, on the drug, the dose, the
SIG, how you say DAW 19 different ways. There were some things that are
generally agreed upon. As soon as you get out of those into the next step
of trying to do something more with the prescription process, I think that is
where it gets a little bit more vague.
MS. FRIEDMAN: Refresh my memory real quick; that is in the EHI table?
DR. HALE: Yes. There is a table in that 101-page report, yes.
DR. KIBBE: One quick place you could look is the XML tagging for the
continuity of care record.
DR. HALE: The difference is, in that report we did classify them as
required and not required. We did try to do that. I couldn’t recite
the whole list because I just don’t have the detail, but they are classified
that way.
DR. KIBBE: That may not be sufficient to answer your question, but
that is an exercise that has been recently done with regards to the content you
are looking for.
MS. FRIEDMAN: (Comments off mike.)
DR. KIBBE: It is a committee under ASTM International. It is a
new data content standard.
DR. NARCISSI: I just wanted to comment on the difference in the state
laws. You might have to look at it like you did with privacy, and have a
floor instead of a ceiling.
MR. BLAIR: This question is difficult to ask, because it is
ambiguous. Patricia, you would up indicating you worked on the health
initiative on e-prescribing. So maybe you could give us a feeling for
this.
What I am struggling with is trying to find out where the balance is.
It is very impressive on our first day, and even on the morning of our second
day, seeing how much progress has been made by the e-prescribing networks and
vendors. There are a lot of systems out there that are providing value
today with the standards that are out there. They may be imperfect.
I think you wind up saying there are a lot of standards gaps.
They fill them in with proprietary solutions. The proprietary
solutions tend to work for the folks they have focused on, and they focus on
for the most part getting the pharmacies connected, making sure that the PBM
formularies are in place, and getting the connection to the drug knowledge base
vendors.
It appears to me, and tell me if this assumption or perception is correct,
because if it is, then I will have my question, I have the perception that the
major standards gaps are the lack of the standards that provide information to
the physician, for the physician to make decisions that would respond to a
formulary limitation, that would respond to a drug to drug interaction, drug to
allergy or whatever, if they don’t have NEH already, just an e-prescribing
system. But those set of standards are the ones that are the primary ones
where the gap is.
Patricia, is that observation correct?
DR. HALE: I think in general, that is true. However, there are
some other things that I think we have major concerns over. I think
having the multiple different proprietary dictionaries is an issue.
I think what a lot of us are concerned about is that, because of the
limited situations where these have been tested, they have been tested in the
perfect setting, where all the pieces are in place and then they are
tested. But some of these problems are just sitting barely in the
background, and if you try to reach out and use it in any other kind of
setting, they are going to be a big problem. When you try to do any kind
of communication between networks, then they are going to be a big problem.
So I would add to that that having different proprietary dictionaries, and
some of the things where the vendors have gone off on their own ways and have
multiple versions of things, that is a big concern.
MR. BLAIR: Then here is my question then. It is a tough
question. None of us may be able to answer it now, but nevertheless there
may be a feeling of perception. I got the impression that a number of
those standard steps in terms of messages could be standardized. RxHub
indicated it would wind up turning over its messages on formularies and on
access to medication history, for example, so that they could become
standardized. Maybe that could be done fairly quickly, within the next 12
or 18 months or something like that.
The other piece is that we already know the National Library of Medicine is
working on the enhancements and the mappings between Rx Norm and the drug
knowledge base of interest. So that addresses another major impediment.
We have got a few others. I won’t go down the list, the prescriber
identifiers and all the others. But on some of those that would at least
start to get the physician more enabled with the information, to be able to
improve patient safety. That seems to be one of the major criteria for
standardization, is patient safety. It seems to me that that would rise
to a very high priority in terms of what we address in terms of standards gaps.
My groping is, I don’t want to go to something where we wind up going for
the perfect, which may take three, four, five years. We still may want to
go down that direction and work on those things, but if there is a
demonstration project that begins in January 2006, is it your perception that
if we focus on addressing the standard gaps that enable the physician between
now and then, so that those standards could be ready, plus — and I think in a
lot of ways, those same standards between the Rx Norm stuff and the messages
that Express Scripts would provide, hopefully they will, they said they plan
to, that that might be the low-hanging fruit, so that we could have
demonstration projects in 2006 which would significantly increase
interoperability and the sharing among a lot of facilities, and enablement of
physician involvement to improve patient safety.
This turned into a little bit of a speech here, but the question is, do you
think that this thinking is on target for low-hanging fruit near-term
improvements?
DR. HALE: I think it was a good speech. It tied in a lot of the
things that we have been emphasizing, so that is great. It reinforces
that we are all on the same page.
I think two things. What is your goal for low-hanging fruit? Is
your goal for low-hanging fruit projects in specific areas? One way is to
pick these perfect areas all the way across the country, where all the pieces
are in place, get them going, get the benefits in those areas. They are
there first; that is low-hanging fruit.
The next thing though is, if you are talking about trying to capture the
best situations in the most diverse practices, that is a much bigger
challenge. As soon as you put something in my office that I use in part
of my patients part of the time, it is not really usable. I have been
there, it is not usable.
So if you are going to start looking in the future for that last mile,
which is me out there in the Adirondacks, you are going to have to do test
systems and demonstration projects that not only are out in these kinds of
areas, but also have mobile physicians and mobile patients that are moving from
place to place and system to system, because you are going to have to deal with
that. I think that kind of interoperability is the one that is really
going to be difficult.
MR. BLAIR: Any other comments?
DR. SCHUMAN: I guess the only thing I would say that ties into what
Patricia said, is that the latest statistics that I have seen indicate that 46
percent of the hospitals in this country are hospitals under 100 beds, which to
me connotes that they are largely rural environments. While they are not
taking care of 46 percent of the patients in this country, they are 46 percent
of the hospitals. So if you pick the low-hanging fruit, you are picking
no more than 54 percent of the fruit, in a sense.
DR. KIBBE: Jeff, I think you are right on. I think we are
closer than we think. I think we are getting better at all this, so we
are doing a lot of nitpicking, appropriately so. We understand the stakes
are very high, and we don’t want to fail.
But I think you are right. I would like to acknowledge the work that
has been done by a lot of people, RxHub, SureScripts and the folks that they
represent, to make progress. I think this is very doable.
DR. HALE: One of the things that made the EHI initiative so great was
how well everybody worked together. We had real diverse groups, and that
is exactly the environment that we saw, that everybody was trying to say, what
can I do on my side to get this going. So I am very optimistic,
too. I may sound not optimistic, but I am.
DR. HUFF: Just a comment. I would like to second Dave in
saying, doing structured coded SIGs is really doable. I say that because
a lot of systems are doing it. A lot of the inpatient systems have done
structured SIGs for 20 years.
What is needed is a standardization. It is getting the people
together and standardizing how it is done. That mostly is just saying
exactly what abbreviations we are going to use here, take into considerations
the JCHO’s things about ones you shouldn’t use, and some stuff like that.
But I think it is very doable.
My question though would be to Mary. You made a lot of great
points. I was thinking how do we apply that to the standards that we
adopt. The two things that I came out with were, one, the code sets need
to accommodate alternative care, more devices than we might have thought of,
that sort of thing. We also need to think about the second signatures and
co-signing and that sort of stuff.
Is there anything else you can think of that are implications for the
standards that are — those are the two things that I could think of
specifically. Are there other things that specifically need to be
accommodated in the standards to take care of your needs?
DR. SCHUMAN: No, I think that is a good start. I think the
coding has been a huge issue. I’m sure you have heard representatives
from AMA talk about that before, and the second signature. I think that
gets us a long way down the road.
I think the other challenge is the multiplicity of settings and the
vehicles that are used to provide care, when you step outside of your typical
ambulatory practice or physician practice, whether it is big or small.
But that is less of a standards issue and more of an implementation issue, I
think.
DR. NARCISSI: I just want to make one little comment. I think
one piece that I mentioned in my statement was this whole area of prior
authorization. No matter what the setting, no matter what the specialty,
I think that needs to be taken care of.
It seems like there is a disconnect on the end of the PBM and payor side of
the standards. So perhaps that could be focused on.
DR. HUFF: Yes, we have heard that one a bunch.
MR. REYNOLDS: When you think back to HIPAA, and some of the people
mentioned some of their struggles with HIPAA, one of the things that you
mentioned needs to be tested and needs to be clear. One of the comments
that was made a minute ago answering a question was, I can do it faster with
the pad than I can with the electronic.
Any time you implement something, you need service level agreements.
You need the idea, what is the test, what is the gold seal that it
worked. I think as an industry, we sometimes wait for the end to decide
the success factors, or decide that the service level agreements are the things
that have to be in it.
So when it is talked about — we got from the panel, here are the things
that need to be involved in the test, and here are the things that make it be
reasonably successful. For example, only taking that particular answer in
a particular setting, if you want drug history, formularies, some of the other
things, and also beat the pad.
For example, Judy can’t type in faster than I can write. That doesn’t mean
hers aren’t formatted better, it doesn’t mean she has got better quality
notes. But I can beat her, and I can write all over the page.
So I’m not picking on your answer, so please don’t take it that way.
I am trying to understand that when we talk about piloting and we talk about
testing, if we can get a precursor from this panel at some point of what that
test is, and what success equals — because if you do it a little bit slower
than the pad, but you reduce all the callbacks because of others — but we
never did that with HIPAA. We said it was going to save money, and then
we all went off and did it, and now we are all trying to figure out what is
success.
I can tell you right now, success is still not clear. We are all
working on it and we are all going to get there and we are going to get done,
but do you check it off and go, nice job. I think we have an opportunity,
with this much of a cross section of the people that are going to use it,
involved. Because at some point, they are going to come back and present
to us the test result, and we are going to get some sense of whether or not
that is good or bad. Right now, from hearing all the testimony, I
wouldn’t even begin to list what I think —
DR. COHN: We heard some of that yesterday. If one reflects on
the results that were being brought forward, but I am curious to hear what this
panel says.
DR. NARCISSI: Maybe you could look at some of the applications that
are currently in place, for example, Wellpoint. I heard last week in a
presentation that they are giving computer systems for billing applications,
but also for e-prescribing, to several thousand physicians around the
country.
So that might be something to look at. Maybe they would have the data
on those physicians that have the capability of e-prescribing and those that
don’t.
Also, the Florida Medicaid project that they are doing, that might be a
good thing to look at, and then gather from that what is missing from the
testing.
MR. REYNOLDS: The way I was going with the question was, we all found
out yesterday, if we don’t satisfy the subscriber, it is not going to
happen. Even though these things are going on, if you would also look at
those things and begin evaluating what you think success is or isn’t, because
we can look at them and think what is success, but if you who represent the
people that are actually going to do it don’t agree that that is success, that
would be helpful. It is not to say that if you come back and testify, we
can’t change it. But it is a struggle.
DR. HALE: It isn’t simply that it has to be as fast as how I
prescribe something else. That isn’t necessarily going to be a
success. For me, if I find that it is more helpful to use the system
because it may take a little longer, but I have better information to make a
better decision, that can be just as valuable, or it is safer for my
patient. Those are the things I value.
The bottom line is, I will use it most of the time for most of my patients,
and tomorrow I use it and the day after, that is success. That is
criteria one, but it is criteria one in multiple settings, too.
My biggest problem with the projects that have been out there is that they
are just not the real world. And the statistics don’t tell you how many
times that physician used for how many of their patients, did they use it all
day every day for all of their patients, for all of the prescriptions.
You don’t get that information. That is what I need. I need to know
that, and I need to know in multiple settings and moving patients to
physicians.
DR. KIBBE: Harry, I think if you looked at the Future of Family
Medicine report that was published several weeks ago in the Annals of Family
Medicine, that is really a road map about what we think success in family
medicine in the future depends upon. I think e-prescribing will be seen
by the AFP and our members as the success of it being part of that success.
That is not a very concrete answer in a way, but in another way, it is that
we will look at these kinds of solutions for quality and safety enhancement and
efficiency enhancement within the context of that Future of Family Medicine
report.
MR. REYNOLDS: Agreed, but do the other three people on the panel
agree that that is the picture we are trying to draw? That is my point.
DR. COHN: Maybe you can’t come to a conclusion right this
moment. Is this something you want to have them send us a note on?
MR. REYNOLDS: I think this is a great opportunity, like we talked
about, for NCDPD to get together and other people to get together. We
have got four people on the panel. If they drew us a picture, or drew us
a set of what the basic requirements would be, philosophically agreed upon
across all the environments we heard, that would be at least a nice — because
we are not trying to implement it for any one segment, we are trying to come up
with standards with every group.
MS. FRIEDMAN: If you all would like to do that and e-mail it to me,
if you feel compelled to write it down on your way back or whatever, I would be
happy to get it out.
DR. HALE: You’re going to need the elephant pulled down into a little
bit more of a shoebox. I want it all.
MR. REYNOLDS: You get to draw the animal.
DR. HALE: We addressed that in EHI as well, in our format of what we
thought that was, too, as AFP did in their report. So I guess we need a
little bit more detail of what you are asking for, how practical are you
talking about.
DR. COHN: I think we probably need to form our question a little more
precisely, I think is what you are saying. But I think it is a question
we need to start asking people as we go forward.
DR. SCHUMAN: It seems to me that the tradeoff is between convenience
for the prescriber and payback to the patient. You asked before about
balance, and if that can be in balance, that may be one of the recipes for
being successful. It seems that if it increases the continuity of care to
the patient, the access of the prescriptions and/or whatever the services are
that the patient needs, and at the same time increases the knowledge base to
the physician in a way that improves their practice, that that is where the
tradeoff may need to be.
I agree, I don’t believe that we have ever thought that it was related
necessarily to speed or to decreased costs, although unfortunately many of the
organizations we are familiar with may disagree with that. But from a
purist perspective, it is the tradeoff between the patient and the ability to
get compliance from the large community.
DR. WARREN: Hopefully mine is a fairly concrete question, and it is
directed at Jean. When you were answering the question about decision
support and the need for standards in that, one of the comments you made was
the number of clicks it would take a physician to drill down.
From my perspective, the number of clicks is more software specific and
developer specific. So what kind of standards are you talking about that
are gaps within the decision support arena that need to be there for you?
DR. NARCISSI: What I was trying to express is that I think that
during this whole standard development process, we also need to consider the
vendors, which were not a part of the HIPAA standards. They waited, and
when everything was finalized, they implemented whatever they could. So I
think they have to also be at the table, just as much as everybody else.
DR. WARREN: So your question was more to make sure that we hear from
the vendors, not to request standards for the numbers of clicks or the way
decision support was presented to you?
DR. NARCISSI: Exactly.
DR. HALE: But the other point too is that from the ACP point of view,
we are talking about clicks. You do need to have some sort of — I don’t
know if standard is the right thing to say, but there should be some kind of
rule of some sort saying that you can’t put unnecessary amounts of steps in
order to do something that isn’t based on the necessity to get the information
and ease of use, but is based on some other need, whether we are pushing to use
a certain kind of formulary, or some kind of other push from the outside.
DR. WARREN: Just so I understand what you are talking about, are you
talking about — when I am talking about standards, I am talking about someone
like ASTM or HL-7 or some ANSE group coming up with standards. What I
hear you talking about are more guidelines that are out there.
DR. HALE: Yes, this would be more of a guideline rather than a
standard, yes.
MS. CRONIN: Can I ask a followup question there? So with that
in mind, could that be part of a certification process of products, if that
were to happen?
DR. HALE: Absolutely, yes.
DR. COHN: I really want to thank our panel for what has been a
fascinating discussion. We really appreciate your coming and joining
us. I have a sense that the conversation doesn’t quite feel
finished. As you know, we are having more hearings in July and in
August. We will talk about it after the break, but I’m sure you will be
hearing from us additional questions, and we obviously would welcome additional
thoughts from you all as we go along with this.
Now we will take a 15-minute break. We will get together at 11:25.
(Brief recess.)
DR. COHN: Welcome back to our three-day marathon. What we are
going to do with this session is, first of all, starting with an open
microphone, just to allow anyone in the audience who wants to make any other
comments about e-prescribing or any of the testimony or issues we have been
considering over the last two and a half days, the opportunity to make brief
comments. After that we will go to subcommittee discussions and a variety
of issues.
Questions or anyone interested in a comment?
DR. MARTIN: My name is Ross Martin. I am a medical
informatician with Pfizer. I really appreciated hearing from the
providers. I want to emphasize the importance of hearing how their
interest can be represented by the committee in ways that they may not be able
to defend themselves, if you will.
If you look at what happens in standards organizations typically, the way
things happen in the standards organizations and the people who are typically
at those organizations doing the detail work, such as NCPDP and HL-7, are
typically the vendors, the technical people, not so much the advocates of
general interest, because they don’t have typically the technical depth to get
involved at that level.
So what happens is, we quickly arrive at solutions to move the engines
forward for business transactions, and don’t always take into account some of
the needs that were articulated here. So the examples of being able to
get what the doctor ordered, and clearly articulate that in a standard format
is a physician need specifically and a patient advocacy need; the
prioritization question that was brought up a couple of times also.
So that is the kind of thing that you may not be able to find a standard in
existence for those today, and there likely won’t be one in the future without
somebody saying to those bodies, you have to do this at some point in
time. That was one point.
Then just the thing about Rx Norm in particular. I think Stan brought
up a great point about the difference between structuring the Rx Norm so that
it can capture what I have been calling differentiating features. There
are two ways to approach that. One is to try to codify all of them into
what an individual product is, and the other is to do a pick list method, where
the doctor will say I want bubble gum flavored that is gluten free, and send
those little notes, if you will, along with the Rx Norm code for the drug that
they want and the dose and the dose form.
Now, in my view, what is missing in that is the standardization of that
actual list. So it could still be a list, but if you standardize that
list so that everybody can pick from the same things, and then the
manufacturers like Pfizer can say that their products that they put on the
market, they assert that they have these qualities that are in that pick list,
the Rx Norm function can serve to lump together all the manufacturers and all
the manufacturers’ products that exist that can fulfill the request that the
doctor is looking for.
So today, a doctor can say I know that Brand X of amoxycillin is banana
flavored, so I will prescribe that and say dispense as written, but that is not
what you want. You want to be able to say there are 20 different
manufacturers out there, five of them fit the specifications that they are
looking for, and you want those manufacturers to be able to articulate that in
a structured way. By asserting that they have those qualities, what they
do is, they are asserting that yes, we are gluten free and we will always be
gluten free, and we will go about the manufacturing process to be sure that
that remains true for that product, versus, we happen to be gluten free today,
but it is not an active ingredient, and I’m not even sure that would change
that that would change the NDC code, for example, if that manufacturing process
changed for inactive ingredients. I’m not sure what all the new criteria
are going to be for that. But these are the things that we can capture
and identify and even become part of the structured product label, potentially,
if it is included in that manner.
So I just wanted to bring up that concept of something in between those
either-or extremes of coding the Rx Norm to include a thousand potential
different initiations, versus having a standardized pick list that is also a
vocabulary that you can choose from, that are recognized issues for
patients. You don’t have to list every inactive ingredient; you have to
list the ones that somebody is going to have to make a decision clinically
based upon that, or clinically a personal preference.
Things like animal products; an animal product doesn’t have a clinical
function if it exists or not in a compound, but if somebody has a religious
preference, there have literally been wars fought in history because of the
application of animal products to — I think the Boer Wars were about putting
pig grease on musket bullets, and you had to rip this off with your mouth, and
people who don’t want to be associated with things porcine got pretty upset.
So those preferences need to be able to be identified, articulated by the
doctor, and prescribed and understood at the other end. I think that is
why it is important to be able to codify those things in a structured fashion.
Thank you.
DR. COHN: Steve, did you have a comment?
DR. STEINDEL: Yes, if I could comment. Ross, if you could
comment back on my comment and maybe question. There has been a lot of
discussion on Rx Norm and some of the details, et cetera. One thing we
heard the first day was the perception that there was an inability of Rx Norm
to define regimented steroid products, birth control pills, et cetera.
The comment came back that they are there as a drug delivery system. So
there are various places some of this can be expressed in Rx Norm. The
idea of flavored vehicles to deliver drugs, maybe Rx Norm can handle that in a
systematic fashion by just having a set of flavored ingredients and say it is
part of the drug delivery system. So there are ways that they can handle
that issue.
The other side of it is the gluten free, the allergen ingredient. One
of the goals of Rx Norm is to link the drugs to lists of NDC codes. It
would seem to me that the inert ingredient there is important for the
successful link to be made. I don’t know how to handle that within Rx
Norm. I think that is something that NLM needs to think about and the
community needs to think about from the informatics point of view, because
without that being embedded in the clinical drug, I don’t know where the link
can occur.
DR. MARTIN: The first point about the packaged drug, that was a
recent change that was implemented in Rx Norm from Stuart Nelson for that
need. That was great, because that covers a lot of these issues.
This was one of the things that brought it to Pfizer’s attention to begin
with, was the idea of the packaged product and the importance of that for
compliance, et cetera. So we are very glad that that got covered.
As we did that exercise, looking deeper at the issues, we came up with the
thing about, at 2 o’clock in the morning, when your kid needs amoxycillin for
an ear infection — I’m not going to speak about whether that is clinically
valid or not, I know that is a controversy — what is therapeutically
equivalent in that situation at 2 o’clock in the morning. Cherry is not
therapeutically equivalent to bubble gum to my son, to my three-year-old.
He will take one and not take the other. You can offer it all you want,
but it is not a therapeutic equivalent.
It is an issue that — there is no price involved in that. You have
got to be able to state that preference. This is why I wanted to
emphasize why it is important in e-prescribing versus what is going on
today. It is because in e-prescribing, the goal is that that drug will be
ready for the patient when they get to the pharmacy, instead of handing it off,
running errands for an hour, and coming back. That hour already happens
because the prescription precedes the patient. If you cannot articulate
that need — and I think it should be done in a structured way, then you miss
out. Then you get it home and you find out what flavor it is, and we know
we asked for cherry and it didn’t show up, and there you go.
Now, in terms of how you structure it, I would think that in the Rx Norm
environment, one of the things that you could do is have this hierarchy if you
will of the Rx Norm in terms of those particular needs.
What I mean by that is this. Let’s use amoxycillin as an example,
continue to use that. If you have a product that you have on the market
that is gluten free and it comes in these three flavors, most of the time
somebody is not going to care about it. A doctor and a patient is not
going to care about any of those things. So they will ask for the top
tier amoxycillin Rx Norm code. But if there is a need for a preference,
you can drill down. You can imagine this in almost like a Windows folder
structure, the acknowledgement that a plus sign says there are differential
features for this if you need them, and then you can drill down and say, I need
gluten free. Now it reduces it down to, there are five products on the
market that do that. There are five generic manufacturers, there is one
brand, or whatever it might be. That is associated with that code, and
you keep drilling down.
That is very easy to map once you have those articulated in a structured
fashion.
Does that answer your question?
DR. STEINDEL: I think so, thank you.
DR. COHN: It sounds like we will need to hear from the NLM sooner
rather than later. Any other public testimony? Please.
PARTICIPANT: A few observations from today I wanted to share.
Harry asked the question about how to measure success. I think there are
a couple of things to think about here. In my mind, the simple answer is
lower cost and higher quality.
With Medicare in particular, measuring the lower cost may be difficult,
because to a large extent those people today are invisible, because there is no
benefit. The easy way to measure a cost savings is a pre and post study,
where you look at peoples’ expenses before a period and then after a
period. With electronic prescribing in particular, the after is
relatively easy, particularly when you have got a managed benefit, because you
can see what people get and you can see the costs.
In the managed benefit world, the pre and post studies have been very
difficult, because the adoption has been very sporadic. So even when you
have what you might think is a perfect scenario, where you have a managed
group, you could say July 1 and then plan your calendar year and say what
happened before July 1 and what happened after.
What you need to do that well is, you need a large group of physicians
covering a large portion of that group’s population to adopt on July 1, or have
been adopted on July 1, so that you can get that true measure. What we
have seen to date has been sporadic adoption and even sporadic use, so
physicians aren’t necessarily using it for every patient, for every drug.
So the real point I am making is, there may be some planning that needs to
take place now, thinking about how do you identify populations, and is there a
way to start gathering pre data with respect to both quality and cost, so that
when you implement a managed benefit with electronic prescribing, you have got
something to compare to.
The other comment I wanted to make dealt with this universality
question. Pat Hale made a comment about, electronic prescribing doesn’t
work for me unless it works for every patient. There has been a lot of
discussion today about decision support as a component of electronic
prescribing and so on and so forth. I think as you add those components,
the formulary piece becomes secondary.
One of the things that early on in electronic prescribing, one of the
reasons we built RxHub, is to try and build that momentum, to put the three
large PBMs together so that as a percent of any one physician’s practice, we
had a better chance of getting the lion’s share or at least some majority of
their practice.
But again, the thought there was, formulary was one of the primary drivers,
cost savings was one of the primary drivers. As electronic prescribing
has evolved, there are a lot more benefits that come from it. Formulary
is one component of it. When you add the Medicare population to any
physician’s practice, plus the big three PBMs, that question becomes even
smaller, because in a lot of primary care practices, Medicare alone can be 40
percent of a practice. So if the players in Medicare are connected, we
have a better chance of getting wider formulary coverage, but to the extent
that electronic prescribing is doing even more, that is just a value add.
Another point I think is worth making here that hasn’t been made is, one of
the big issues that we have seen and was identified yesterday by a lot of the
vendors, is that one of the primary barriers to getting adoption of electronic
prescribing is accessing the member data in the physician’s practice. If
a physician can go in and find me quickly, the more likely to keep using
it. One of the big obstacles that vendors have had has been getting to
that data.
So there is a constituency that hasn’t been represented here, and that is
the practice management vendors. One of the things to think about might
be standardizing in some way an extract methodology for electronic prescribing
vendors to get at the database of patients in a physician’s practice, so that
they don’t have to build a hundred different extracts and hack their way into
some of these systems, because some of them are harder to get to than others,
and there are literally hundreds of these systems, and many versions of each of
those systems.
So if you really want universal adoption, you need to get to an easier way
to access patient information at the practice level, so that physicians can
adopt easier, and it is easier for vendors to get to more physicians.
The last point I wanted to make dealt with some comments this morning about
multiple clicks and so forth, and trying to think about certifying
vendors. Most of that kind of problem I think has worked its way through
in the electronic prescribing industry over the past few years. Early on
there were lots of different ways to drive behavior, and people were trying to
monetize their success rates in doing that. So you might have popup boxes
to say, are you really sure you want that drug, and there is another
drug. People have come to realize that those often just create obstacles
to adoption, because physicians get frustrated.
I think when you are looking at electronic prescribing from a Medicare cost
savings perspective, you need to be careful in saying what you can or can’t
do. Let the market work a little bit there. Some of those tactics,
people have found, actually may help, and they certainly will help in a cost
savings context, and people could do it in ways that physicians could live
with. To say that you can’t do those things really undercuts some of the
value.
We talked about message fatigue for physicians with respect to drug-drug
alerts. This could happen in this area, too. At the same time,
physicians tend to ignore some of this stuff. So if you give a physician
a list of drugs and you don’t create an application that helps drive them to
the most cost effective choice, they very likely are going to use these systems
to continue the habits they have today, which is, they like one stat and they
like one SSRI, they like one antibiotic, and that is what they use, and people
get to the pharmacy and the antibiotic, when there is a ten dollar choice.
So there is some benefit to reminding a physician. I want to make
sure that we don’t restrict the capabilities of these systems so much that we
lose a lot of the cost savings possibilities.
DR. COHN: I’d just make a comment. I was struck yesterday that
when one heard of successful implementations of all of these things, that
nothing worked very well when things were done or dumped on desktops, be it
information systems or use or otherwise. I think we heard — and maybe
this is also my background with successful implementation to change — that all
of the things you are talking about are probably best done in some sort of
relationship that involves face to face interactions, training on the use of a
system, explanations of why things are done with information systems.
I think that is probably a whole other aspect. From what you were
saying, it sounded to me like people were going to be dumping decision support
on physicians’ desktops, somehow anticipating that this would be a cost
benefit, or something would happen, as opposed to, without an understanding of
why they might be there, they would be generally probably ignored.
PARTICIPANT: I think you are right. I do think the market is
going to work itself out. I have seen decision support systems that no
one would use, and I have seen some that are really slick. Most of them
haven’t gone very far. I think it is just recently coming more on
peoples’ radar screens. I think some of these applications will go
places.
In addition to the value of the applications, the usability varies a
lot. Some of them do require a lot of coaching and a lot of hand holding,
some of them are very intuitive. But you are right, I think people have
to understand all the benefits they are getting. In particular on the
cost savings side, physicians have to understand that driving from one drug to
another is actually helping the patient. It is not the big bad PBM that
is forcing the practice.
We fully recognize, and we have always said, that physicians have to be
free to choose the most therapeutically relevant drug, but we also recognize
that in a lot of cases, there are more than one choice. If a plan selects
one as preferred, or they are trying to drive to a generic in a particular
class because there is a lot of things, all else being equal, it is worth
reminding physicians of that, and it is worth making them aware that this is
going to save your patient $90, as opposed to, this benefits the plan or the
PBM.
Particularly in the Medicare context, I think you have to keep in mind that
very often, if people don’t get the most cost effective choice, they may not
get anything, because people a lot of times are inclined to say, no, thank you,
when they get to the pharmacy and the cost is $150.
MR. BLAIR: I would agree with everything that you were saying.
I think that some of the anxiety that we are hearing is that there be
transparency in the presentation of the formulary, because some folks are just
— some physicians can’t see the choices that — don’t have quite enough
information to be able to see the information that would support exactly what
you are saying. Does that make sense? In other words, it is not
just a code, but it is actually a price. You have a list that is clear,
that is organized the way that they need it, and they could wind up also having
the ability to electronically, real time, if they have prior authorization, to
be able to do that without delays.
PARTICIPANT: I think that is right, although I think a lot of the
concern about transparency — we have always driven toward relative cost,
because it is very difficult to get the actual price. I think I made a
comment about that the other day.
I think a lot of the concern is driven by things that happened years
ago. Early on in e-prescribing, Parkstone was selling banner ads, and
there were vendors who were trying to sell commercial formularies, where they
wanted to get pharma companies to pay them to show something as preferred when
nothing else was preferred. So if you have a cash payment patient, they
were going to try to drive people toward a formulary that really wasn’t a
formulary.
Those vendors by and large have gone away, because people recognized those
aren’t viable models. I think in the physician community, there is still
a little bit of a perception that there is something being hidden. I
don’t think it is really there today. There really isn’t, at least from
what I have seen, things like that out there today, where there is some sort of
hidden agency, other than what the plan has actually designated as the
preferred. That may again be another education issue, as to what we are
thinking about today.
DR. COHN: Thank you. Tom?
MR. BIZZARRO: Tom Bizzarro with First Data Bank. I’ve got a
couple of comments about the testimony we heard this morning from the ultimate
end users. I think it is good to get their perception. I also think
there is some confusion in that end user group about what is the responsibility
of the content provider, the application provider, the network and the
standards. We heard some comments about the inadequacies that they
see. I think they are not sure where to assign the blame, or where the
problem is. That is very common in an end user, because you don’t see it,
you’re not technical, you don’t know.
Now, certainly there are some issues in there that need to be addressed,
but I think that those things have to be addressed through all four of those
different entities.
The second comment I have is, I am having my 30th anniversary of having
become a pharmacist this month. Thirty years ago, there were no
electronic systems in pharmacy. Twenty years ago they started to come in,
and they were terrible. If your experience with those systems is in the
beginning when they first came out, and the gentleman before me just mentioned
about some of those issues that we had with the original e-prescribing vendors,
about some of those problems that we had; those things have changed.
In this world, two years is a lifetime in the development of
applications. So I really hope that our physicians and end users take a
look at the current systems that are out there. I also think that the
marketplace is going to drive the best systems to the forefront, and that that
is what is going to keep this moving forward. Some of the applications
that we saw three and four years ago really were terrible. They were not
driven from a physician’s point of view, they didn’t fit within the physician’s
work flow. If that was your experience, and you are afraid that that is
what we are still going to have to go forward with, I think that you would
never want to. This would scare you to death as an initiative for
e-prescribing.
DR. COHN: Thank you.
DR. MARTIN: This is Ross Martin from Pfizer again, just to comment on
the observation that the market is changing in terms of what vendors are
offering about things like banner ads for e-prescribing.
I agree that the banner ads are going away, but we as a company are still
being approached by vendors to be able to talk to doctors at the point of care,
based on when they are making prescribing decisions, saying we will make money
when we get them to switch.
Now, I know that that is happening, but they are getting pretty smart about
how to do it. It is much more subtle. It is not a popup ad that
makes five clicks. It is the quieter, gentler kind of nudging. We
know that these things are effective. We know that they do that.
One of the principles that we have come up with is, there should be no
commercial messaging, like when you walk up to the ATM. I think Kevin
Hutchinson from the first session made a comment about the banking
industry. Once you put in your card, you can’t be doing advertising at
the ATM. If I am putting in my card in somebody else’s ATM, that other
bank can’t start telling me you can switch over and we will take care of your
account. They have worked those issues out. That is one principle
that we think is important.
A correlate to that is the idea that clinical information that is provided
to the clinician at the point of care for prescribing cannot be created and
managed by someone with a financial interest in the clinical decision that they
make. That includes the manufacturer, that includes the payor, that
includes anybody that is going to make money, depending on how they
choose. If that entity is also the ones making the clinical rules, they
are inherently biased in how they are going to make those rules, whether or not
something is considered to be therapeutically equivalent, whether or not — if
somebody makes a differential income on choosing a generic brand, or whatever
those decisions might be, or choosing my brand over somebody else’s brand.
In the case of the pharmaceutical industry, every word that we say to a
doctor is highly regulated by people from the FDA. We cannot talk to the
doctor about off-label uses and things such as that. Our conversations
with that doctor have to be cleared, go through all sorts of hurdles.
There are no similar hurdles for anybody else in this kind of
conversation. So payors don’t have to qualify what they say. They
can say, studies show that Brand X is the same as Brand Y. That doesn’t
have to be a defensible statement, really. Those things aren’t currently
happening. So that is one of our concerns.
The same thing with the editorial boards with the compendia. They are
a little better in this position, because they make their money off of selling
their compendia. They don’t make their money off the decision necessarily
that the doctor makes.
However, they can be influenced by that. If their editorial board has
shares in their company and their company makes a lot of money off the payors,
for example, using the product or let’s say the manufacturers are able to buy
these tools for doctors, then they would have an interest in pleasing that
customer.
So this is a sticky issue. I think it is not a standard in terms of a
technical standard. This is where we get back to that idea of policy
standards and best practice standards, how is that going to be done. One
principle that we think should be applied is that there should be editorial
independence of the people making the clinical statements about a drug, and
obviously cost statements we think should look at the total cost of care.
From a Medicare perspective, if they save the patient ten dollars or 50 dollars
or whatever, but if it results in hospitalization, that is going to cost us all
a lot of money. So it really has to look at the total cost of care.
So a ten dollar drug may not be as effective as a $100 drug. It may not
be as cheap as a $100 drug in the long run. Those kinds of things I think
need to be brought to the front as well. That is the issue.
MS. CRONIN: I’d just like to make a followup comment. I think
all the things you said are very important, but health economic information, or
more specifically, pharmaco economic information is regulated under Section 114
of ADAMHA, while industry now gets away with providing a lot of that
information to payors for reimbursement or policy decisions, because it is
considered an unsolicited request.
That information often serves as the basis for decision making, and they
can use that to inform decision rules. So there is a circular effect here
in the real world. A lot of evidence that is generated is funded by
industry.
But I think you do have a really important point, in that we need to make
sure we are very aware of the biases that are inherent. If there is
widespread adoption, and eventually we are functioning with decision support
across the board at a high level, there should be very credible evidence that
it is available. Perhaps it is in the form of a clearinghouse or some
other mechanism, but that it is done in a way such that the biases are well
understood.
DR. STEINDEL: Ross, I have a followup request. I won’t put this
in the form of a question, because I really think I would be putting you on the
spot to answer that. I asked the question a couple of times about
off-label use, and you brought that up in your comments. I think it would
be very helpful to this subcommittee if we have some thoughts from the
pharmaceutical industry about how off label use should be handled in the
e-prescribing world.
DR. MARTIN: Okay.
DR. COHN: And I think that is a general question. We can ask
other pharmaceutical companies also later on in testimony.
PARTICIPANT: I don’t necessarily disagree with Ross, but I wanted to
clarify one thing. The kinds of information that we are talking about
providing is really formulary information. And formularies are created by
P&T committees, which are — I don’t know if you are familiar with that,
but that is a process unto itself.
So we are not advocating that at the point of care, we or the point of care
vendors engage in a debate as to the merits of Lipitor versus Ocor versus
Lovistatin, but we want the ability to make clear to the physician that
Lovistatin is cheaper than Lipitor, and the Pfizer rep very ably will be able
to let the physician know that Lipitor is more effective in lowering
cholesterol than Lovistatin, and the physician can make a decision with respect
to this patient, which is more appropriate.
So I don’t disagree that we should be using electronic prescribing to
disseminate information in some sort of unregulated manner, but formularies are
formularies, and they are facts of life. So we just want to make sure
that physicians have the whole picture as to the cost of what they are
choosing, and then they can make clinical decisions based on what they have
received from other sources.
DR. COBUZZI: Lou Cobuzzi from Pharmacy Benefits Management at the
Veterans Health Administration. The Veterans Health Administration has
been using e-prescribing for over 20 years now. We are probably
responsible for about 95 percent of the prescribing that is electronic in the
United States.
You have to forgive me; this is the first time I have been involved in
this, so if my insights seem superficial to you, please understand that it is
just me getting involved in the process.
One of the things that — over the last three days that I have been going
through those 20 years, and every time somebody would bring up a different
point, something would ring; that is reasonable, I would discount that, that is
something everybody should consider. I would suggest that any
e-prescribing standards should be as broad as possible, because nobody knows
where this is going to go.
So from a physician’s standpoint, if it appears that there is going to be
too much information, extraneous information, that may not be the case ten or
15 or 20 years from now. We have seen it evolve to where the efficiencies
of the process allow you to do so much more in the same time frame than it used
to take to just write a prescription. There is more data that is
available to providers, there is more capability. There is the ability to
electronically trigger labs that heretofore we never used before. When I
hear physicians say that they use the process, and that process is filled with
all kinds of concerns and everything and they actually abandoned it, if they
actually saw the concerns that are associated with the paper process today, and
that was being constantly flipped up into their face and thrown, I’m sure they
would want to abandon that even more.
One of the things that hasn’t been discussed is the functionality of
collaborative practice. There are about 46 states in the United States
now that have collaborative practice agreements between pharmacists and
physicians. In some states it works very well and it is comprehensive, in
others it is in the neonatal stage.
I think that many of the concerns we have heard about the flavors, the
textures, the additives, when they say I don’t need to know the color, you do
need to know the color. The pharmacist knows that. If they want to
know what the flavor is, you do need to know the flavor. But that doesn’t
need to be in all instances something that the physician is going to number one
pick up, because a lot of times in these pediatric environments, those people
are out the door before they even ask the question what flavor does the child
like. The pharmacists are supposed to pick this up at the end. So
that collaborative practice needs to work.
The extraneous question that physicians get all the time doesn’t need to be
there if there is understanding between the pharmacist and the physician about
what is important to that physician and what confidence they have in the
pharmacist.
So I think a lot of the dialogue that is taking place has nothing to do
necessarily with e-prescribing, but with the practice that we are dealing with,
which is health care. So if you could keep that in mind, I think it is
not so much what we do see as what we don’t see that we are concerned about
today, and just keep those standards as broad as possible.
Thank you.
DR. COHN: Thank you for also making me not feel quite so silly not
knowing all the flavors of amoxycillin.
DR. ROBINSON: Just a quick comment, more in the area of decision
support. As I have been hearing the hearings, it is occurring to me that
one thing we can’t do through standards is fix the problem of information being
stored in disparate locations, as we have all these different providers.
We are not proposing a centralized patient record anywhere in all this.
It lives in the ether among all these different providers.
So in the area of decision support, if you are doing it in a computerized
model, decision support is only as good as the information that you have in the
clinical record. So if you are depending on your medication history on
claims data, at that point in time you don’t know if the patient is actively
taking all these medications, or again, are there medications that didn’t go
through the claims process and that aren’t there. Similarly, drug-disease
contraindications are only as good as the patient problem list that you
have.
The comment I have, and maybe it would be directed more toward CMS, in all
this I am not seeing the patient centric approach, where perhaps someone is
compensated to take a patient holistic view from the case management
perspective, and let’s pool all this electronic information from all the
disparate parties and let’s sit down with the patient and say, Mrs. Jones, are
these all the medications you are taking. Maybe someone is checking with
the physicians involved, and do we really have an accurate picture of the
clinical side of the patient, are we really capturing all the labs.
So you could have the best decision support database in the world, but if
the information that you have collected is fundamentally flawed, you are going
to get bad results, bad decisions on drug-drug interactions, bad decisions or
errors of omission on drug-disease contraindications. A lot of this is
what is driving adverse events in the patients, is the information not known,
or decisions made on information that is just not up to date.
So I don’t know if this would be a CMS initiative. We have talked
about what type of incentives we can have for physicians. Later maybe it
will be pharmacists. Maybe the area of case management fees in a
structure that everybody is comfortable with would be a positive way to perhaps
have somebody be an agent of the patient to try to pull everything together.
DR. STEINDEL: This is just a very quick comment. After we have
heard about not making commercial announcements, I would like to make a
commercial announcement. A lot of what you have mentioned and a lot of
what other people have been mentioning will be a topic of discussion at the
NHII conference around July 20. The commercial part of it is that if you
register before Monday, you will save $100 on your registration fee.
DR. ROBINSON: Stand by, operators are waiting for your call.
MS. CRONIN: I think we can actually say something more specific to
address your comment. There is a lot of active consideration going on in
terms of pay for performance that incorporates electronic prescribing. I
think the details that will be discussed at the NHII meeting will probably be
of great interest to you.
DR. COHN: I think it is time for us to wrap
up. I want to thank everyone for the open microphone. We will
continue through all of our many upcoming sessions to have open microphone
sessions. They have actually been quite useful, and very interesting.
Now, let me just get very concrete here for a minute and talk about a
couple of the next steps. I don’t think we are at this point going to
start going through all of the notes that everyone has taken, but we will
reflect on how we are going to put them together.
I think as you all know, we have an upcoming full committee meeting on June
16 and 17, on the 17th in the morning from 8 o’clock to 9:50 we have a meeting
of the Subcommittee on Standards and Security. It is a breakout, and I am
sure we will be spending much of that time trying to put together some of our
thoughts on this hearing and the last hearing in preparation for the next
hearing. By that time, plans will be well under way for that next
hearing.
Now, for your information, at the end of June, on the 24th and 25th, there
is a meeting of the quality work group. This is not an e-prescribing
discussion, but I just wanted to let subcommittee members know. I know I
have discussed this with a couple of you privately. One of the topics has
to do with reviewing possible changes and modifications to either billing
practices, data collection practices that relate to the HIPAA administrative
and financial transactions. These are recommendations coming out of the
quality work group that are going to have a broader hearing.
This is not per se a joint meeting. I don’t think we are going to
have a quorum of everyone, although if we have three maybe we will make it a
joint meeting. But I do want to have you note those dates. If you
are available, everyone from the subcommittee are more than welcome to sit in
on this. I think this has particular interest for those of us who are
involved with the administrative and financial transactions. So I just
want to point that out to you. If you are available and want to come, let
me know, and rather than just sitting in on the session, we may cause it to be
a joint hearing.
Now, going back to e-prescribing, we do currently have hearings scheduled
for July 29 and July 30. I have had conversations with some of you.
We are going to hold July 28 as a possible expansion back to another three-day
hearing. We will only do that if we think by doing that, that we can
accelerate progress with some sort of deliverable available as early as
September for the full committee. That would mean that we would be able
to hopefully cancel September and/or October as hearing dates. We are not
going to do it if it is just more days of being in Washington for more
hearings. So I will be working closely with Maria, our lead staff, and
Jeff, our Vice Chair, to make that determination over the next week to ten
days, I would expect. But I am asking you to hold the date while we make
that determination.
Now, we will also have hearings on August 17 and 18. My understanding
of the next several hearings — and once again, we are figuring out exactly
what is going to be when, but the next steps involve hearing from a variety of
other users.
Once again, I was just going through our work plan which, Jeff, I want to
thank you for putting together here. We will be hearing in July from
pharmacists and pharmacies, PBMs, state boards of pharmacies, state boards of
medicine, state Medicaid agencies. If we consolidate July and August, we
will also be hearing from private sector payors, government payors; we already
have them at the table, so we may not need to hear too much more from them,
consumer advocacy groups, and a variety of other stakeholders in relationship
to all of this. At some point either in August or September, beginning to
hear back from the SDOs as we get a better idea of what the requirements are
from the SDOs what is there now versus what is in the near term as they improve
functionality versus what is long term, to help us advise the Secretary on what
is doable in the near term.
So anyway, that is what is going on. We have people who have been
taking notes, Stan and Margaret and Maria, and Judy. What I think we are
going to be doing over the next several weeks is a combination of putting
together peoples’ notes in a way that hopefully will help inform the
subcommittee. Part of this is also looking back at our previous hearings
and beginning to incorporate that information into it, so we begin to develop a
body of knowledge and a model of where we are that we can reference.
There are no conclusions associated with them, but it is more notes about where
we are, the outstanding questions and issues, so that we can keep a ready
reference, and it may help frame some of the future conversations.
MS. FRIEDMAN: And also some of the major themes that have emerged
already. Just sitting here the last couple of days, some things are
becoming a lot more clear than others.
DR. COHN: So anyway, the intent will be that this will be a living
document as we move forward and begin to either understand or further resolve
issues. I’m sure that we will be identifying other issues as we move
forward. But we will be looking forward to a compilation of that into a
document that is a near term deliverable, which hopefully we will have a chance
to review and talk about at the June breakout.
Am I missing anything?
DR. HUFF: What I am wondering, Simon, is, certainly we don’t want to
make decisions prematurely, but there are things that we are hearing again and
again. The thing that occurs to me is, if we need to address — it seems
to me that there are certain standards that we will choose. It is also
clear that those standards have deficiencies. I am wondering if there is
a way we could basically make note of that early, so that work could start on
it now rather than either six months or a year from now.
DR. COHN: I don’t think you should take any of this process to be six
months or a year from now. However, I do think it is premature until we
have heard from all the stakeholders to start getting too specific on what
needs to be done. I am just concerned that we have heard from some; we
need to hear from the stakeholders to be sure that at the end of that day, that
we come up with the same understanding of what the gaps and the issues are.
Now, that may be done as early as sometime in July or as late as early to
mid August, so we are not talking about a six-month lag before any motion
occurs. That is why I am — rather than start pointing out issues and
gaps, I am trying at this point to consolidate notes and identify issues that
we need to be further exploring.
I don’t know that I have satisfied the issue. Like you, I would like
to have this problem solved tomorrow, but — what you would like to do is
publish the gaps that we are already seeing?
DR. HUFF: I would like to even say, let’s apply resources to fill the
gaps. Just to be specific, it seems to me that it is hard for me to
imagine — we want to get all of the input from everybody, but e-prescribing is
in fact sending a prescription from a physician’s office to somebody who is
going to fill a prescription. It is hard for me to imagine that we
wouldn’t say the script standard is going to be the standard that does
that. At the same time, we have heard testimony of people who have said
these things need to be addressed, and there might be additions that we get
from other folks.
We also heard pretty much universally, Rx Norm ought to do this, but there
are problems with Rx Norm. I’m just looking for opportunities where we
could from this committee say, could we have resources assigned to start
addressing the issues with script, with Rx Norm, some formal support for HL-7
script collaboration for the structured SIG.
There just seem to be some things that are non-controversial. We want
to learn more and we want to get the other opinions and the list might grow,
but it seems like there are some things that we could actually say very quickly
that are things that need to be done.
I’ve said enough.
DR. STEINDEL: Simon, I agree in principle with what Stan is
saying. I think there are a whole lot of common themes we have seen
identified as gaps. What I would like to see is a consolidated set of
notes, and just to point out that our next session when we get together is a
little over two weeks away, and maybe we can have some consolidated set of gaps
to discuss at that session, and then make some decisions as to whether we can
move forward, and make some recommendations on those gaps very quickly.
We are not prepared to do it right now.
MS. CRONIN: I think that is a great idea. I would also like to
add that the more time we have to think specifically how we are going to act on
the gaps and to address some of the standard development needs before January
’06, the better off we will be in planning a meaningful pilot and implementing
a meaningful pilot.
MR. BLAIR: I think there are certain things I need to go on record
with here.
I feel uncomfortable setting forth recommendations before we have heard
from the folks — we are going to try to see if we can hear from all the folks
in July, that the law indicated we should listen to before we make
recommendations. So that is one piece.
Then if we listen to — after we have gathered in July the input from all
the stakeholders that use these standards in August, we will be in a position
to get an assessment from the STOs and terminology developers and NLM on their
ability to respond to those, so that we can make some meaningful interim
recommendations by September, when our full committee meets. Even though
it is an accelerated schedule, I think we will have met the intent of the law.
I think it is more than just meeting the intent of the law. I think
that there is a lot of folks that have been coming up to me — they may be
coming up to others here, from pharmacy associations, from PBMs, from the
pharmaceutical industry, that we haven’t received testimony from, and I think
they are going to be pretty upset if we start making preliminary
recommendations before they have had an opportunity to be heard, and state
agencies.
So I feel uncomfortable with us making preliminary recommendations before
September 1, which is a two month acceleration from November for the interim
recommendations.
The the thing that I believe is that to just go forward and saying the
obvious, that we know NCPDP script will be used, and we know that Rx Norm has a
role to play, I feel as if the real value in what this committee has to offer,
especially in this case, is identifying those gaps that can be addressed
quickly and marshaling the industry.
So here is where I tend to agree with you, Stan, only my date is a little
bit different. I feel like if by September 1 and 2, when the full
committee meets, we could wind up saying we want accelerated development within
NCPDP to address certain specific message gaps, and certain specific things we
want to try to either accelerate or have additional government support to be
able to make sure that the mappings from Rx Norm to the drug knowledge base,
and a few other things in terms of preauthorization, and in terms of the
identifiers and other things, I think that way, we will be making meaningful
things to move the ball forward.
Now, that is September of 2004. That is 14 months before the
demonstration projects, and the industry could prepare. They would have
14 months to try to fill in those gaps and address those needs, so that when
those demonstration projects occur, the demonstration projects are really
meeting the goals that the law prescribed, and that our testifiers have told us
are so important in terms of interoperability and in terms of patient safety.
MR. REYNOLDS: We have heard a lot over the last three days, and we
haven’t really had a chance as a committee to talk about it. I think the
exciting part about the next full committee meeting is that we have time.
We have no testimony, we have time.
I think Stan the other day put up a straw structure to at least begin
discussing it. We all heard what we heard. There is a difference
between — and I’m not sure I am ready to decide yet whether we should move
forward fast, slow or what, because once we put up a straw structure of what we
think we heard, what we think we believe in, minimally it allows us to listen
different going forward. It allows us to ask questions going forward,
because we have at least come together in some kind of way that we think we may
have a direction.
Now, we may at the end of that two hours go, wow, that one is clear, and it
doesn’t matter, that one is clear. The point is, that structure is going
to allow us to take what we have been flooded with for a couple, three days,
plus what we are going to get flooded with on the rest of it, and at least put
us in some kind of order, rather than waiting until the end of the whole time
and then deciding that we all agree philosophically to disagree on what we
think we heard.
Sometimes you can let the whole journey go, and then you don’t really like
the vacation, rather than stopping halfway through and saying, I don’t want to
go anyplace. I am tired of getting in the car. So I think we have
got a long trip going here, and we might want to sit down and we get a chance
to talk, and we get a chance to decide. Then once we do that, then
whatever is decided will be a little clearer at that point, rather than rushing
forward.
DR. COHN: I think regardless, our next step is probably putting
things together. Even though, Stan, you see things very clearly, I think
the real question is, does everybody have the same level of understanding that
you came up with out of these three days. I think that is going to be
part of the next steps. That is why we are going to collect everybody’s
notes, try to put them together, analyze our last session and try to create a
framework that is not just yours, but is everybody’s thoughts to date.
And we will see where we are with that.
Of course, these notes are available to HHS, so if there is something
obvious out of there, they will do whatever they do. But that is how
things work in an open federal process.
With that, I think you have all heard what is happening. Are we
missing anything else? Well, the good news is that it is 12:41. We
said we were going to adjourn at 12:45, and with that, we will be adjourned.
(Whereupon, the meeting adjourned at 12:41 p.m.)