[This Transcript is Unedited]

Department of Health and Human Services

National Committee on Vital and Health Statistics

Subcommittee on Standards and Security

February 1, 2005

Hubert H. Humphrey Building
200 Independence Avenue, SW
Washington, DC 20201

Proceedings by:
CASET Associates, Ltd.
10201 Lee Highway
Fairfax, Virginia 22030
(703)352-0091

TABLE OF CONTENTS

P R O C E E D I N G S (9:10 a.m.)

Agenda Item: Call to Order – Welcome and Introductions – Simon Cohn, Chair

DR. COHN: Good morning.

I want to call this meeting to order.

This is the first day of two 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.
Department of Health and Human Services on national health information policy.

I am Simon Cohn, chairman of the subcommittee and the newly-appointed
chairman of the full committee.

I am also the Associate Executive Director of Health Information Policy for
Kaiser Permanente.

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.

And, of course, as always, I want to remind everyone to speak clearly and
into the microphone.

As always, we have a lot to cover over the next two days.

This morning, we begin with a CMS update.

I want to, first of all, congratulate Karen Trudel and Maria Friedman and
your staff on the publication of the recent notice of proposed rule making for
e-prescribing. Obviously, you have our congratulations on that effort.

Karen, I know, will be providing an update and overview of the proposed
rules as soon as we are done with introductions and welcomes this morning.

Then we continue our discussions and testimony on issues related to our
next set of e-prescribing recommendations. This begins with an update by Lynne
Gilbertson and industry work groups on progress to improve the e-prescribing
standards.

As many of you will recall, the previous NCVHS recommendations called on
the industry to work together on a number of areas to improve the e-prescribing
standards. These include things like codified SIG, the development of standards
for transmitting formulary and benefits information, and, finally, about the
harmonization between the HL7 e-prescribing standards and the NCPDP SCRIPT.

Obviously, I want to thank Lynne for providing leadership and support to
these efforts, and we’ll be looking forward to hearing from her.

Additionally, I believe, Lynne will also be – we know that Lynne has been
leading discussions within the industry to identify best practices related to
e-prescribing security and authentication, and we are, hopefully, going to be
hearing some about that from her later on this morning.

Following the morning break, we have asked Eleni – and I will call her A.
Is that okay?

MS. ANAGNOSTIADIS: Yes, that’s –

DR. COHN: Okay. Thank you. (Laughter). To review with us state approaches
and issues related specifically to e-prescribing of controlled substances.

As all of you remember, this is really to follow up on discussions we had
at the last set of hearings with the DEA around the issues of prescribing for
controlled substances.

After lunch, we have an update on RxNorm and use in the e-prescribing
context.

And, then, after the afternoon break, we will have an open-microphone
session and then move into subcommittee discussions.

I want to emphasize, of course, this is an open session. Those in
attendance are welcome to make brief remarks if you have information pertinent
to the issues which we will be discussing today.

We will also have time during the open-microphone session for brief remarks
and comments by those in attendance.

Finally, for those on the internet, if you have any comments about any of
the issues being discussed, we do welcome email and letters on any of the
issues coming before the subcommittee.

Now, just a quick look on to tomorrow. Tomorrow, we do start at 8:30 a.m.
At that point, the Designated Standards Maintenance Organizations, the DSMO’s,
will be giving their annual report. Then, we’ll spend the remainder of the
morning on subcommittee discussions of our next set of e-prescribing
recommendations and next steps.

Now, with that, let’s have introductions around the table and then around
the room.

For those on the national committee, I would ask if you have any conflicts
of interest related to any of the issues coming before us today would you
please so publicly indicate during your introduction?

Jeff.

MR. BLAIR: Jeff Blair, Vice President, Medical Records Institute; Vice
Chair, Subcommittee on Standards – oh, well, yes. I’m Vice Chair – (laughter) –
and there is nothing that I am aware of in terms of the testimony today that I
have a conflict-of-interest with.

DR. STEINDEL: Steve Steindel, Centers for Disease Control and Prevention,
staff to the subcommittee and liaison to the full committee.

DR. HUFF: Stan Huff with Intermountain Healthcare in the University of Utah
in Salt Lake City, and I don’t have any conflicts, as far as I know for – any
potential conflicts for testimony today.

DR. FITZMAURICE: Michael Fitzmaurice, Agency for Healthcare Research and
Quality, liaison to the full committee and staff to the subcommittee.

And congratulations, Simon, on your appointment for chair.

DR. FERRER: Jorge Ferrer, Veterans Health Administration, staff to the
subcommittee.

And kudos to Simon –

MS. TRUDEL: Karen Trudel, Centers for Medicare and Medicaid Services. Staff
to the subcommittee.

MS. AMATAYAKUL: Margret Amatayakul, independent contractor to the
subcommittee.

MS. PICKETT: Donna Pickett, National Center for Health Statistics, Centers
for Disease Control and Prevention, and staff to the subcommittee.

MS. GREENBERG: Marjorie Greenberg, National Center for Health Statistics,
CDC, and executive secretary to the committee.

MR. REYNOLDS: Harry Reynolds, Blue Cross Blue Shield of North Carolina,
member of the subcommittee. No conflicts.

DR. WARREN: Judy Warren, University of Kansas School of Nursing. Member of
the committee. I am not aware of any conflicts I have today.

MS. FRIEDMAN: Maria Friedman, CMS, lead staff to the subcommittee.

MS. SQUIRE: Marietta Squire, CDC, NCHS and staff to the subcommittee.

MS. LAVIN: Kelly Lavin, American Osteopathic Association.

MR. BRUCKLE: Mark Bruckle(?), Academy of Managed Care Pharmacy.

MR. BIZZARO: Tom Bizzaro, First DataBank.

MR. ROBINSON: George Robinson, First DataBank.

MS. GILBERTSON: Lynne Gilbertson, National Council for Prescription Drug
Programs.

MR. ROTERHMICK: Phil ROTERHMICK(?), Express Groups.

MS. SHERWOOD: Dana Sherwood with PCM.

MR. MARTIN: Ross Martin with Pfizer.

MR. VLOWAY: Roger Vloway(?) with the RXPC.

MR. POLLARD: Michael Pollard, Medco Health Solutions.

MS. VARTANE: Laura Vartane(?) with Wexer(?) and Walker for IMS Health.

MS. BLODGETT: Holly Blodgett(?), Amisis Center Tech(?).

MS. ANAGNOSTIADIS: Eleni Anagnostiadis with the National Association of
Board os Pharmacy.

MR. WHITTEMORE: Ken Whittemore, SureScripts.

MS. FORQUET: Lori Reed-Forquet, ASTM.

MS. ZIGMAN-LUKE: Marilyn Zigman-Luke(?). America’s Health Insurance Plans.

MR. SIMKO: Mike Simko, Walgreens.

MR. BROOK: Richard Brook, ProxyMed, an active participant with NCPDP and
the workgroups.

MR. SCHUETH: Tony Schueth, Point of Care Partners.

MR. LEVINE: Gary Levine, Medco Health Solutions.

MR. MILLICAN: Mark Millican(?), Medi-Span, a part of Walters-Cooler(?)
Health.

MS. ECKERT: Karen Eckert, Medi-Span.

MR. DE CARLO: Michael De Carlo with the Blue Cross Blue Shield Association.

MS. RIPIN: Helga Ripin(?), Office of the Assistant Secretary for Planning
and Evaluation.

DR. COHN: Okay. Well, welcome, everyone.

I think before we start into our first session, Jeff, do you have any
comments?

Obviously, want to thank you and Maria for your leadership in putting these
sessions together and moving forward the work plan.

MR. BLAIR: No.

DR. COHN: No?

Agenda Item: Proposed E-Prescribing Standards
Regulation Overview – Karen Trudel

DR. COHN: Well, Karen, why don’t you lead off, and, again, congratulations
to you and your team.

MS. TRUDEL: Well, I think congratulations are in order for the subcommittee
as well, and I was very happy to see that that participation was prominently
mentioned in the administration’s press release.

Other thing I would like to do, it takes a village to build a regulation
and there are an awful lot of people who were involved in this behind the
scenes. Maria and I are the public faces, and I am not going to go through and
mention every name like the Academy Awards do, but I just do want to recognize
Gladys Wheeler, because she was the principal author of the regulation, and I
can’t begin to tell you how difficult a job that is.

So, with that, we are very, very happy to talk to you today about the
regulation and what it does; and Steve Steindel, in a magnanimous email to me
the other day, classified it as boring, because it was – its content was
somewhat to be expected, based on the discussions that the subcommittee has had
before, but I do want to talk about the dates and some of the content.

The proposed rule was announced by the President last Thursday, and it did
go on display at the Office of the Federal Register, which means there is a
paper document there that people can come and see. It becomes a public
document.

However, it will not be actually published in the Federal Register until
this Friday, February 4th. That is what starts the 60-day clock for
the comment period, which will, therefore, end on April 5th.

This particular NPRM has to be looked at in the context of other Part D
Regulations that are either published or underway, and I would like to just
point out some of those linkages to you before I go on.

The Medicare Prescription Drug Benefit Final Rules, which we colloquially
call Title I and Title II, were published on January 28th. They
don’t say an awful lot about e-prescribing, but they do require Part D plan
sponsors to support and comply with electronic prescribing standards.

In addition, I know we are all aware that there is a Stark exemption for
e-prescribing and also a requirement for an anti-kickback, safe harbor. There’s
a separate proposed rule in process now on the Stark exemption, and it is
moving very, very quickly. It will clarify how physician self-referral will
work in the context of e-prescribing, and there is also an anti-kickback
regulation that is being worked on by the Office of the Inspector General.

So what is in the rule? And I am sure most of you have actually read it in
great detail, but we do take a long time to describe the NCVHS process,
summarize the recommendations. Why did we go to the effort of putting that in
there? Because it provides so much background in terms of why we feel so
confident about these foundation standards, the number of people we have
listened to, the depth of the discussion, the amount of time was spent. I think
all is very valuable in showing the kind of very broad base that we discussed
in order to come to this place.

We do propose foundation standards.

We explain that we are talking about an incremental strategy here, and this
is one of the things where we specifically asked for comments from the
industry. We were very clear that we are talking about a set of foundation
standards, that we will build on these foundation standards based on the
pilots, and that the ultimate place that we want to be is not to build an
e-prescribing stovepipe or a Medicare Part D stovepipe, but we want everything
that we are doing to roll into a consolidated electronic health record, and we
clarify that this is consistent with many of the other initiatives that are
going on in the department, but we, again, request comment on how all that can
work together.

It discusses state preemption, and here are the proposed foundation
standards. Obviously, the SCRIPT standard, Version 5, Release 0, except for the
Prescription Fill Status Notification Transaction, because that was the one
that the subcommittee mentioned that was really not in use, and we could not
say that there was adequate industry experience.

One of the things that I would suggest to consider is that when we began to
look at the entire transaction, the entire SCRIPT standard, there were some
ancillary-messaging and administrative transactions – like receipt
notifications, password changes, et cetera – and after some discussion, we
concluded that leaving those pieces of the standard out would make it more
difficult for the parts of the standard that we were keeping to work. So the
ancillary-messaging and administrative transactions we included as part of the
standard that we are incorporating by reference in the Federal Register.

So, in addition to those new prescriptions, fill request and response,
change request and response, cancellation request and response are all
suggested as foundation-standard usage for the SCRIPT. Also, the X12 270/271,
Version 4010 and Addenda, for eligibility and benefits inquiries and responses
between prescribers and Part D sponsors.

We also – although this was not discussed by the NCPDP – felt that we
needed to adopt the Telecommunications Standard Version 5 and the associated
Batch Standard for eligibility and benefits, inquiries and responses between
dispensers and Part B sponsors.

Now, that may seem like a somewhat redundant approach, because that NCPDP
telecommunication standard, which is, essentially, the retail-pharmacy drug
claim and related transactions, is already a HIPAA requirement. It is already a
HIPAA standard, and anyone who is a covered entity would have to use that
transaction in any case. We decided to play it safe and adopt it officially as
well under Part D.

So this is simply the eligibility-and-benefit portion of those transactions
that we are saying we would use as a foundation standard.

Formulary and benefit standards are a little bit different, and that is
because, as you recall, there was not a clear recommendation that those
standards were ready for prime time. The sense was that if NCPDP could get
through the process of having those standards be accepted as accredited SDO
standards and that they had not been changed enough to say that there is not
adequate industry experience, then they could be adopted.

What we did to address some of those concerns was to set up what some of
the – well, there’s one typo – ANSI.

We set out – rather than adopting those standards specifically, we set out
a number of criteria that said this is what an appropriate formulary standard
would look like. This is what an appropriate medication history standard would
look like, and, in this particular case, we said they needed to be accredited,
that the standard would permit interface with multiple products, point-of-care
vendors, that it is not a single application standard, that the standard would
provide a uniform means for PBM’s to communicate, arrange formulary and benefit
information to prescribers – and what that means is that no matter what your
formulary structure looks like, this formulary standard has to be able to
accommodate it. So, in essence, the format is not dictating the formulary
makeup and composition. It accommodates – it is flexible enough to accommodate
any formulary that could be designed, that the standards would cover a range of
formulary and benefit data, medication-history standards would be adopted –
again, if accredited – again, point of interface – and that they provide a
uniform means for a prescriber, dispenser or payer to list the drugs that have
been prescribed or claimed for a patient within a certain time frame.

Foundation standards, obviously, recommended by NCVHS, and, in the
regulation, we actually discussed the notion of adequate industry experience
and attempt to put some parameters around that, too, that adequate industry
experience means a standard needs to be ANSI accredited. It needs to be
implemented by entities to which the final standard will be applied in multiple
programs with more than one external partner, so that we have some sense of
transportability here and interoperability, and it needs to be recognized by
key stakeholders as an industry standard.

The first set of standards doesn’t represent the full set. The additional
standards will be identified, pilot tested and proposed through a separate rule
making, and if you think back to the timetable, we had the pilots in calendar
year 2006. We are required to do a report to Congress early in 2007, publish
regulations by 2008. So that is the timetable for this second set of standards.

Some of our proposed definitions. For e-prescribing, transmission using
electronic media of prescription or prescription-related information between a
prescriber, dispenser, PBM, health plan directly or through an intermediary,
including an e-prescribing network. That is fairly broad.

An electronic prescription-drug program is a program that provides for that
for covered Part D drugs prescribed for Part D-eligible individuals who are
enrolled in Part D plans, and so, basically, we take the notion of
e-prescribing, which is this big and we ratchet it down, so that it fits within
Medicare Part D.

We found that we didn’t have definitions for prescriber or dispenser, so we
have proposed those.

Prescriber is a physician, dentist, other person, licensed, registered or
otherwise permitted by the U.S., or the jurisdiction in which he or she
practices, to issue prescriptions for drugs for human use.

And a dispenser is a person or an other legal entity, licensed, registered
or otherwise permitted, which would include a pharmacy, by the jurisdiction in
which the person practices or the entity is located to provide drug products
for human use.

So, again, those are basic definitions of prescriber and dispenser, and
we’ll be interested to hear what comments there might be on those.

State preemption was clearly a big issue that the subcommittee heard about.
There were two essential points of view here. One was that preemption would
only apply to transactions and entities that are part of an electronic
prescription drug program under Part D of Medicare, or a broader read would
apply it to a broader set of transactions and entities and apparently apply it
across the board in the sense that any state law that appears to be a barrier
to e-prescribing could be preempted under these requirements.

Our proposed rule takes a narrow interpretation, and, again, we have
specifically invited public comment on that issue. That particular portion of
the regulation has an extremely legal flavor to it.

Talk a little bit about the use of the National Provider Identifier. We
invite public comment. We do not make a specific statement about it. We ask for
information on the possible use of NPI, the earliest time when NPI should be
required, and we ask about the effect on industry of accelerating the use of
NPI for e-prescribing ahead of the HIPAA compliance date, which would be in
2007.

We talk a little bit about alternatives to the NPI, especially in the event
that we have not been able to enumerate sufficient providers to make this work
in 2006 and talk about options for phasing in the use of NPI.

One other thing that these slides don’t cover that I do want to make sure
that you are aware of and that is the issue of whether the foundation standards
or the standards are required in completely internal transactions.

There was a discussion about whether existing entities that were doing
totally internal transactions – in other words, the prescriber, the dispenser
and the plan were all within the same organization needed to use these
standards or if they had adopted HL7 standards and were using those – whether
those would be sufficient.

The subcommittee had suggested that that would be appropriate. We don’t go
that far in the regulation, partially because we came up with some potential
legal concerns that that is an approach that is contrary to the one that we
adopted under HIPAA, where, under HIPAA, we said if you are doing the
transaction, if it meets the definition of a claim, whether it is internal to
an organization or external is irrelevant, we want you to use the standards.
So, then, again, that is an issue where we basically raise and discuss both
parts of the issue, both perceptions, and ask for comments, and we’ll go
forward, then, in the final rule and make a final judgment on that.

Let’s talk a little bit about pilot testing. The MMA requires pilot testing
in calendar year 2006 for the standards not in widespread use. We will soon
start to solicit applications for organizations that wish to voluntarily be
part of the pilots. We will be doing that by a public procurement. We’ll also
be obtaining a contractor to evaluate the results of those pilots, and we will
be looking at the new or emerging standards. We have a checklist that Margret
prepared for us based on the NCVHS recommendations and we’ll be working to get
those pilots going as quickly as possible.

We want to make sure that the criterion and the specifics about the pilots
are on the street as soon as possible, so that people who may wish to become
participants will have ample opportunity to understand what is involved.

And that concludes my report remarks.

Can I take questions?

DR. COHN: Well, Karen, thank you very much.

Again, congratulations.

I just do want to remind the subcommittee – or, actually, not remind you,
but explain – that the publication of, obviously, the notice of the proposed
rule does afford the subcommittee, and, actually, the full committee, an option
to also respond and provide comments to the rules. So one of the issues that we
will have to discuss, when we get into discussion later on today and tomorrow
morning, was whether or not – in case you hadn’t thought we had enough to do
for the next month – (laughter) – to develop a letter for the full committee in
March relating to the proposed rule, so just in case you thought that things
were moving ahead well and we wouldn’t need all the time that we thought we
might have.

Anyway, questions or comments from the committee?

Jeff.

MR. BLAIR: My congratulations to you and Gladys and Maria for an
outstanding job in pulling together the NPRM so quickly.

And one of the areas, when I looked through the NPRM this weekend – and I
am assuming that the reason that a number of our recommendations, in terms of
what should be pilot tested, were simply not appropriate for the NPRM. Maybe
that is going to be coming out in further guidance, you know, later, or maybe
you could help us understand what those are –

MS. TRUDEL: Yes, you are absolutely right, Jeff. What is in the regulation
is the issues that are necessary to bring up a basic e-prescribing program with
foundation standards in January of 2006, and that is it.

So the recommendations that were made about pilot testing and things that
need to be pilot tested you will see when the request for proposals or rollout
materials on the pilot become public.

MR. BLAIR: Okay. Given that we’ll be having additional recommendations in
March in the subcommittee – or NCVHS will – was it – how does that play – would
you need a separate – I hope you don’t need a separate NPRM for anything we
recommend, you know, at the end of March.

MS. TRUDEL: Unless there should happen to be an additional foundation
standard that no one was aware of before, I would say that the NPRM that came
out is the only one we are going to see until we get to the end of the pilots
and that whatever recommendations come from the subcommittee or the full
committee between now and that time would play into the pilot process, rather
than the regulation process. So, yes, I have no wish to write another NPRM
right at the moment. Thank you.

MR. BLAIR: Thank you.

DR. COHN: Other comments?

Harry, did you have a question or comment?

MR. REYNOLDS: Yes, and I would also like to thank Steve. He had sent out
the email that Karen mentioned, at least helped some of us also look at the
NPRM.

But since, you know, as a subcommittee, we discussed the HL7-NCPDP
interaction, and the fact that a lot of hospitals use HL7, does the reg
preclude them from – if a person is in the hospital and they do e-prescribing
with their own pharmacy, does it preclude them from using HL7, and how does
that actually play?

I am trying to understand – you know, we had talked about what – kind of
what we thought it might look like, and then it doesn’t mention HL7 in the reg.
So how do you – if a hospital wanted to be involved in a pilot test or wanted
to be involved in a pilot test of sending things back and forth with other –
how would you see that work? Or if you were – how do you see them approaching
that?

MS. TRUDEL: I think the first thing is that a participant in the pilots –
and while we haven’t specifically written down what all the criteria are, we
would be expecting that that participant would have some sort of – that their
operation would have some applicability to the Part D benefit.

I would have some trouble, on the face of it, figuring out how a – for
instance, an inpatient hospital e-prescribing program – would be able to inform
Part D, and so I am not sure that that is one of the constituencies that we
would be reaching to for pilot participants.

DR. COHN: Jeff, is this a follow on or is this a –

MR. BLAIR: Yes. One of the areas – at least this is my understanding of our
own recommendations. I realize that sometimes you don’t remember entirely
accurately what you have recommended, but in terms of if a prescription is
generated from within an acute-care institution, a hospital, that we were
looking at that environment, that scenario as one which would be driven by the
standards requirement of the dispenser, the pharmacy, which we had previously
identified as saying it must – any prescriptions that the dispenser receives –
retail commercial pharmacies – has to be received in NCPDP SCRIPT format.
Therefore, if the original prescribers is in a hospital or acute-care situation
and if that original prescription is in HL7 mode, it has to be translated,
which the networks seem to be able to provide that service, which – and we
declared that prescriptions within acute-care environments were out of scope.

Now, when I read the NPRM, I really got confused, because I just didn’t see
these concepts. So did I misunderstand them or is this something which is just
too detailed for us to clarify easily here?

MS. TRUDEL: We didn’t address the acute-care setting at all for the exact
reason that you mentioned, in that it is out of scope for Part D.

We did talk about the concept of transactions within an organization – for
instance, a staff model HMO – and you are absolutely correct. The
recommendation there was that if you had, for instance, a staff-model HMO,
where the prescriber, the dispenser and the plan were all part of the same
organization, the NCVHS recommendation was that they be permitted to continue
to use HL7, except in a circumstance where they are transmitting out of
network, and in that particular instance it is the receivers’ – the dispensers’
requirements that rule, and my point, a few moments ago, was that we didn’t
even go that far in this regulation. We basically set out the options and
requested comments, and, again, the reason that we did that was because the
concept of inside and outside of an organization was somewhat counter to the
way we have handled these same concepts within HIPAA. So that entire argument
is – discussion is in there.

MR. BLAIR: Okay. Okay.

DR. COHN: Yes, so, Jeff, you’ll have the option to decide if you want to
comment on that in a letter –

MS. TRUDEL: Right.

DR. COHN: – or reference our previous recommendations.

Karen, thank you. I think you articulated that very well.

MS. TRUDEL: Thanks.

DR. COHN: Michael.

DR. FITZMAURICE: My congratulations, too, Karen, to you and your staff for
a fine job in this regulation.

Jeff picked up on exactly the point that – which I had some confusion.

If I go into the hospital as an inpatient, moved into the outpatient –
discharged and moved to the outpatient, and, then, at the outpatient side, I
say I would like to have a prescription sent to my local pharmacy, well, in the
hospital, it uses HL7, and I want to send it to a local pharmacy that is not
part of the organization. Must that hospital transmit it to the local pharmacy
in SCRIPT or may it use HL7, if they have a trading-partner agreement – or is
that simply not addressed?

MS. TRUDEL: I think that is not addressed. I would even wonder whether that
is a Part D issue.

DR. FITZMAURICE: Well –

MS. TRUDEL: In other words, if the drug is dispensed and ordered in the
hospital, is that even something that is covered under Part D? I think we would
need to talk about that some more.

DR. STEINDEL: Yes, that is ambiguous.

DR. COHN: Judy.

DR. WARREN: I think, Karen, some of the confusion we have is that we are
thinking of examples to where you have kind of a multi-service organization,
such as Kaiser or some of the academic medical centers, where you have both
acute care and ambulatory care and even home care all as part of one
organization. At which point – if they are using HL7 internally, at which point
do they need to switch over to an NCPDP standard? And my memory and
understanding is is if they are sending a message to a pharmacy that is not
part of that organization, that message needs to be the NCPDP. So is my
understanding –

DR. COHN: What happens to our recommendation?

MS. TRUDEL: Well, again, that is what –

DR. COHN: That is not –

MS. TRUDEL: That is what the committee’s recommendation was –

DR. WARREN: That is not what –

MS. TRUDEL: – and the regulation did not take a position. The regulation
did not take a position.

DR. WARREN: Okay. Okay. I need to go back and reread.

MS. TRUDEL: It simply stated the issues, discussed and requested comment.

So I think if you have a place where you think that line should be drawn,
that is a comment that you should make, and that may be something that the
subcommittee wants to weigh in on in communicating back to the Secretary.

DR. WARREN: Thank you.

DR. COHN: Yes.

Steve.

DR. STEINDEL: My comment was just what Karen made, you know, that this was
an area where the regulation laid out two alternatives, did not take a
position. NCVHS has taken a position in its deliberations, and I think it is up
to the subcommittee and the full committee to decide whether they want to send
specific comments on what they feel the response should be to that request for
comment in a letter.

MS. TRUDEL: Yes. Any other questions?

Thank you.

DR. COHN: Well, Karen, thank you very much. Great work.

Agenda Item: Report on Industry E-Rx Work Groups
and Best Practices – Lynne Gilbertson

DR. COHN: Okay. With that, we move on to getting back to our next set of
e-prescribing recommendations, and happy to have Lynne Gilbertson coming and
joining us.

MS. GILBERTSON: Good morning. I am Lynne Gilbertson from the National
Council for Prescription Drug Programs giving a report to NCVHS on the status
of some of the e-prescribing items we have taken on or put on our plates.

You can follow along in the testimony. This is basically the same setup as
we testified in – was it just January or December, whichever one it was – and I
followed the same format, so that you could compare the progression for
different items.

Related to Observation 3, which is the Prescription Messages, I am just
going to go through the actual highlighted ones and not read all the different
parts of the recommendation.

This was related to the fill-status notification function of the NCPDP
SCRIPT standard in the 2006 pilot tests.

The status is that NCPDP Work Group 11 Prescriber/Pharmacist Interface,
their RXFILL Task Group has created implementation and operational guidance to
pharmacy and prescriber system participants for the consistent utilization of
the Fill Status Notification transactions. The Task Group isled by Teresa
Stickland of Healthcare Computer Corp.

The task group has met one to two times per week.

The guidance recommendation has been submitted to NCPDP for approval at the
Work Group meetings, which will take place the first week in March.

If the guidance is approved, the updates will be made to the next
version/release of the SCRIPT Implementation Guide and given to the NCPDP Board
of Trustees for approval. It would be assumed the approval would be asked for
in the April-May time frame.

The guidance includes operational challenges, such as automatic triggering
of fill status notifications, triggering on return to stock, inferring pick up,
privacy, liability, coordination in medication history, a patient changing
physicians and other guidance.

Observation 4, Coordination of Prescription Message Standards, this is
related to the NCPDP-HL7 collaboration.

DR. COHN: Lynne, I’m sorry. I don’t mean to break in. I’m just confused.
Are you referencing this document or – Okay. I apologize.

MS. GILBERTSON: That’ll be in a minute. I just don’t want to have to switch
–

DR. COHN: Okay. Okay.

MS. GILBERTSON: – AV materials.

DR. COHN: Thank you. I – thank you. Okay.

MS. GILBERTSON: Okay. Sorry. Oh, I’m sorry. I didn’t realize I was pushing
buttons.

MR. REYNOLDS: We do have a copy of what she is – It is in the book.

MS. GILBERTSON: Let me just close this out, and we’ll just deal with it
that way.

Okay. All right. This is related to the NCPDP-HL7 collaboration.

The collaboration is actively underway. Attendees met last week at HL7 on a
Connectithon(?) to demonstrate the interoperability of new prescriptions and
medication history transactions.

The demonstration went very well. There was quite a bit of interest in the
transactions that were flying around the room, and this will be shown at the
HIMSS demonstration coming up in February.

The mapping team meets, two calls per week, to discuss the mapping of the
standards. There is also a business meeting.

Mapping continues on the refills and on the code sets, which are some of
the action items for the team.

They are determining right now face-to-face meetings having to do with
mapping and educational sessions. Those still are in the works right now.

And, let’s see, the mapping working documents will eventually be turned
into a guidance document for the industry.

And other work items have been identified, such as change management of
taking this guidance document forward throughout the years, process flow and
education in HL7, Version 3.

Regarding Observation 5, Formulary Messages, the status on the NCPDP
standard for formulary and benefit information file transfer using RxHub
protocol as a basis, the status on this is that Work Group 11’s Formulary and
Benefit Task Group is led by Teri Byrne of RxHub and consists of many, many
industry representatives.

The task group will be submitting a standard for approval to NCPDP at the
March 2005 work group meetings.

The draft standard includes:

The sharing of formulary status lists, which are codes to explain how to
treat non-listed brand, generic, over the counter; whether the drug is on
formulary or preferred status; its relative value limit and other fields.

Formulary alternatives lists, which are alternatives for specific drugs –
the source and the alternative.

Benefit coverage lists, which are conditions under which the patient’s
pharmacy benefit covers a medication.

Benefit copay lists, the extent to which a patient is responsible for the
cost of a prescription. The specification supports multiple ways to state this
cost, including flat dollar amounts, percentages and tier levels and it
contains a cross-reference file of user-recognizable health-plan product names
to the identifiers used for the Formulary, Alternative, Coverage and copay
lists.

Discussion is continuing on the use of RxNorm. At this point, a placeholder
is included in the implementation guide, but no detail has been determined. The
implementation currently supports multiple drug identifiers. Further analysis
will be done to understand how and if RxNorm may be used and what level RxNorm
will be qualified.

Observation 6, Eligibility and Benefit Messages, this is related to the
270/271 of the ASC X12N, Health Care Eligibility Inquiry and Response Standard,
with the action item of NCPDP’s efforts to create a guidance document to map
the pharmacy information on the Part D Pharmacy ID Card to the appropriate
fields in the 270/271.

This task group, under Work Group 3 Standard Identifiers, is led by Todd
Walbert of Walgreens. They are collaborating with X12N Work Group 1 Health Care
Eligibility, as needed, and they are forming their working goals and their
deliverables as we speak.

Observation 7, Prior Authorization Messages. This is related to prior
authorization workflow scenario recommendations for the designing of 2006 pilot
tests and automating prior authorization communication between dispensers and
prescribers and between payers and prescribers in the 2006 pilot tests.

The status: NCPDP Work Group 11, Prescriber/Pharmacist Interface Prior
Authorization Workflow to Transactions Task Group is led by Tony Schueth of
Point of Care Partners and consists of X12N Work Group 10 Health Care Services
Review Co-Chairs and a lot of other interested stakeholders.

The task group has created draft flows for the medication prior
authorization process and has identified where standards exist and where gaps
exist.

The task group is working to understand the dialogue – the questions with
answers – that is necessary for prior authorization processing.

Now, we’ll move to the presentation.

The task group actually put together some thoughts in a presentation format
for your benefit.

The task group, as I mentioned, is part of Work Group 11. At a task-group
level of NCPDP, anyone is welcome. You do not have to be a member to
participate in the task group, which is our lowest working group, and they do a
lot of work via conference calls.

Participating organizations represent standards organizations, professional
organizations, pharmacy and physician vendor systems, long-term care, health
plans, formulary aggregators, PBM’s and network switches. I hope we haven’t
forgotten anybody.

The task group goals are to:

Understand the prior auth workflow in the physician the pharmacy and the
plan environment; to identify additional standards needed to support
prescription drug prior auth; to work to develop the standards with the
appropriate SDO; and to make recommendations to the committee on whether and
how to include prior authorization in demonstration projects.

This chart, I am not going to go over in detail, but it is a framework the
task group put together to put everybody on the same page, basically, to try to
define what all happens in the prior auth environment from the – basically, the
time the patient visits and a drug is going to be prescribed, through to the
end, where, hopefully, the drug is prescribed because the prior authorization
requirements have been met at whatever step of the way they are required.

It was interesting when the task group met to discuss this, there were a
lot of, move this box. Let’s put this box here. Oh, we need a duplicate box
there. Oh, yeah, but it could happen if you go around here, too. So I’m
surprised we got it to one page.

They discussed some of the things that are concerns with this process,
basically, known to the subcommittee, as we heard testimony in 2004.

Basically, there is patient, pharmacy, prescriber hassles and
inefficiencies throughout the process with when the prior authorization is
finally flagged, and it may be after the patient has left the prescriber’s
office. That might delay treatment, while all the paperwork or the interactive
communication is taking place.

Sometimes, it shows up at the pharmacy time frame, where the pharmacy has
to call back and check around, get things done before the patient can get the
actual prescription fulfilled. There’s health-plan inefficiencies, obviously,
dealing with labor-intensive processes.

Other considerations the task group is looking into is that there are time
limits placed on prior auth drugs, that these need to be considered as part of
the environment when you are talking about the workflow.

There is the notion of temporary authorization that must be included in the
workflow.

There is the action item of denials, which can result in appeals or
grievances that must be taken into account as well, as well as a denial being
reversed.

And then long-term care has their own business needs that must be included
in this workflow consideration.

Prior authorization components. The prior auth criteria can vary from plan
to plan, even for the same drug. Some prior auths are very simple in their
requirements of what they are looking for. Some prior auths are very complex,
might require lab results, might require attachments.

These next few slides show examples of a particular PA form.

For growth hormone, just kind of lists the different types of information
that may be requested before the prescription can be fulfilled, and, as an
example, there may be lab results that are included.

The task group has spent some time and is continuing to spend time on doing
some analysis, and this is one page of the analysis to show you. It is for the
growth hormone, once again, and what they have done is gone through the PA
forms, the PA requests, the PA processes that – and identify the different data
elements or needs that the forms or the interactive systems might require, and,
then, obviously, blinded health plans, but to show the differences and the
similarities in the kind of information that, for growth hormone, for example,
these six, eight different health plans, what kind of requirements they may
need. So that is some of the analysis that is going on.

The task group also identified what current standards are. Moving from the
workflow, now, into the standards, and this is a prescriber-initiated prior
authorization, and many of these are known, as we discussed during the
testimony, having to do with information from the payer into the prescriber
system showing that this drug may require a prior authorization. This would be
done via the formulary-and-benefit standard that is under development or is
also being used in existing protocols today.

When the patient visits a physician, the prescriber has some information
available about the drug requiring a PA for that health plan. The prescriber
writes the prescription and submits the PA request, which can be done via the
X12N 278 in its form today.

The prescriber can then submit the electronic prescription to the pharmacy
of the patient’s choice, using the NCPDP SCRIPT.

The pharmacy has the – then dispenses the drug and can file the drug claim
using the telecommunications standard, which updates the payer system as well.

The formulary and benefit standard, as we have talked about, it is under
development. Its purpose is for transmitting formulary and benefit information
from payers to the e-prescribing systems. Drugs requiring PA will be flagged.
Requirements for prior authorization fulfilment will be requested for
incorporation when determined.

The task group discussed that wouldn’t it be neat if, when the formulary
and benefit information is being sent down from the payer or health plan to the
e-prescribing system, that, at some point, we can get to the – that the
standard information could be sent down to the e-prescribing system of what are
the questions necessary to fulfill this prior authorization, so that that is
loaded in, so when the prescriber writes the prescription, not only do they
know it has a PA, here’s the things that must be fulfilled. The vendor systems
have the ability to pull any of the related information off the system,
wherever it might be, that would help fulfill the questions that need to be
responded to, and then that could be sent to the payer or the PBM as part of
the fulfilling of the prior authorization request. So that is in the scope of
the work.

The X12N 278 health care services review. It is a standard for sending and
receiving prior auth communication between physicians and insurance review
boards for procedures and services. It is a HIPAA-mandated transaction. The 278
supports the ability to request additional information from the provider. It
supports LOINC codes to request that additional information. It also does not
limit the additional information being provided via an HL7 CDA. It supports
many means to supply the additional information, including fax, mail, phone
call, et cetera.

The scope needs to be expanded, which – as they are part of the task group
– has been identified and will be worked on to support the more specific PA
information of the drug products, perhaps PA attachment; alignment to SCRIPT,
telecom, formulary and benefit standards; and integrate drug prescription
terminology and identifier standards, where necessary; and attachments that
were developed for claims may be leveraged and used for PA, and additional
attachments may need to be developed.

We spoke with the HL7 A SIG – the Attachments SIG – last week and proposed
the idea – what we are working on, seeing if we could find some synergies and
some assistance, and it was very responsive. We have a couple of volunteers who
are going to help us. They did, obviously, a lot of analysis when they built
the attachment booklets, and so we can leverage off that, and they have also
made the offer that when the PA Task Group has the data elements determined
that they think they need, they will be very glad to help write an attachment
booklet for this business use.

NCPDP SCRIPT, we have talked about, the e-prescribing between prescribers
and dispensers, and with the addition of medication history that is under
ballot, also payers, and the telecom standard, which is the electronic
communication between the dispenser and the payer.

The task group has built a straw model which looks quite a bit like the
original current model, because the players are basically the same, but it
discusses some of the original draft thoughts of what could happen, and,
basically, some of the information between the prescriber and the payer is
where the work would be developed, which is the prescriber – having the
information ahead of time from the health planner, the payer, having to do with
what the PA clinical rules are that can be shipped down to the e-prescribing
system, the prescriber, then knowing what – the system popping into for the
prescriber what needs to be completed for a structured question-and-answer
prior authorization session, and then the completion of the prior auth, and the
submission to the pharmacy of the completed prior auth or at least that the
prior auth is in flight, and, basically, we have the same standards in place
with the thought added about the X12 278 and with an HL7 PA attachment, if
needed, for lab results or whatever the other information is.

Additional gaps have been identified related to a structured
question-and-answer process within the clinical system. You have all this
information of what is required for a PA, but what does it take to get all the
information collected in the e-prescribing system? The ability to extract
supporting data from the clinical system or wherever that data mine exists –
the lab results, whatever else is required – and then the, obviously, the
aggregation of the prior-authorization rules.

The initial recommendations: To work with the HL7 Attachments SIG, as I
mentioned, to use as much synergy as possible.

Members of the task group are also conducting additional research on the
structured PA dialogue, possibly leveraging some work being done at HL7 as
well, which has to do with consideration of standardizing the structure and the
content of what the question and answer needs to look like, but, obviously,
leave the choice of the content to the payers, but it standardizes how you do
the question and answer, so that we don’t have 50 different forms going back
and forth with the same questions, but in different orders, things like that,
trying to standardize that aspect, and there is some good effort going on in
HL7, from what I understand, that members are really investigating.

And, as a note, that it is possible that, at some point, this task group
may come back to the subcommittee to request funding and support for
face-to-face meetings or webcasts or developers to work on structured clinical
dialogue or for the pilot process itself.

That’s it. Back to page 4 of the testimony. (Laughter).

Okay. Observation 8, which has to do with medication history messages from
the payer to the prescriber. The recommendation was for an NCPDP standard for a
medication history message for communication from a payer/PBM to a prescriber,
using the RxHub protocol as a basis.

The status on this is that RxHub did submit a Data Element Request Form at
the November work-group meeting for the protocol, which was based on SCRIPT.
The request is being balloted now and we will adjudicate the ballot in March.

Observation 9, Clinical Drug Terminology. This is related to the RxNorm
terminology.

The status: In August, NCPDP requested NLM map examples to show the flow of
a prescribed clinical drug using RxNorm through to the pharmacy dispensing of
an NDC.

We ended up that examples were created by NCPDP members and presented
during the November Work Group meetings.

Work Group 2, Product Identification created an RxNorm Task Group. They
have met and begun discussions of how it would be used in electronic
prescribing, in medication history, in prior authorization, they have
identified some gaps in usage, and they are building recommendations.

Task group is led by George Robinson of First DataBank and Karen Eckert of
Medi-Span.

And we’ll present more information on this during the RxNorm session this
afternoon.

And I understand that during the NCPDP Annual Conference in March, the NLM
has accepted an invitation to speak on RxNorm.

Related to the SPL, the Structured Product Label, the status: NCPDP Work
Group 2 Product Identification created an SPL Task Group to collaborate with
HL7. The task group is led by Tom Bizzaro of First DataBank.

The task group has identified their goals, which are:

Review the HL7 SPL Implementation Guide and Structured Product Labeling
Release 2 document. They will review and suggest improvements and changes to
Release 2.

They will review areas in the SPL that would benefit from codification. If
there is consensus on what code systems should be used in specific areas, the
group will suggest the use of those code systems, and ensure that, if the
Medicare Part D Model Guidelines and NDF-RT differ, an accurate mapping exists
so that they both can be used successfully.

On Observation 10, Codified and Structured SIG – this is related to
addressing SIG components.

The Work Group 10, Professional Pharmacy Services, has the Industry SIG
Task Group, led by Laura Toper of Allina Hospitals and Clinics, and Keith
Fisher of SXC Health Solutions, Inc. with over 50 members, now, actively
participating, including doctors, pharmacies, representatives from SNOMED,
ISMP, HL7, CMS, the VA, NCPDP, pharmacies, health plans, vendors, processors,
clearinghouses, et cetera, et cetera.

They meet biweekly via conference call, and they are holding another set of
face-to-face meetings during the NCPDP March work-group meetings, and they have
built a draft implementation guide. It is being revised as we speak.

They have combined the work of NCPDP, HL7, continuity-of-care record, Dr.
First and anyone else who has brought forth some suggestions on how to codify
SIG or standardize SIG or whatever term you want to use.

They are, again, looking for potential consistencies and efficiencies
between in- and outpatient settings, so if you are going to use a SIG, here is
how you would use it, regardless of the setting.

The focus is based on U.S. activities. They are keeping an eye on
international work, but there was a long discussion, for example, about what
languages SIG might be written in, and they decided, for the purposes of trying
to get stuff done, they would stick with English to start with, and, obviously,
English has Latin in it, but they would stick with that as a start.

They are going to try to maximize the use of the standard for, as I said,
in-patient and outpatient. They accept that some SIG’s might not fit the first
cut. They are going to try to do an 80/20 or a 90/10 rule and deal with the
exceptions, hopefully, second phase, just to make sure that they are getting
their goals accomplished and not getting bogged down with the oddities, and
they want to optimize technology by using computable fields, where possible.

Their assumptions have been there will be no abbreviations used within SIG.
They are going to leverage and maximize existing standard vocabularies – for
example, SNOMED efforts – external code lists and data dictionaries.

Defaults may be overwritten. For example, the verb take or apply. The user
must be able to change that to something else, such as, dissolve under tongue,
or, if a system defaults a common set of instructions – one po two times daily
– the user must be allowed to change to two po one time daily, if desired, and
still have the same results.

The textual representation will accompany any codification.

The standard is about interoperability transfer, not interface creation,
and the standard would be able for usage to be incorporated into various
standards, such as how HL7 might represent it, how NCPDP SCRIPT might represent
it, how medication history – whatever transaction you might be thinking about
or whatever syntax could represent it – the continuity-of-care record, for
example – but the intent would all be the same by the time you got to the end.
It is just the framework.

They are taking common SIG’s and SCRIPTS and putting them into the proposed
format to make sure things are working. The final reviews are underway of that.

As I mentioned, the implementation guide is being written, and they are
having face-to-face and – calls, as I mentioned, scheduled for the March
work-group meetings.

They intend to – in the summer of 2005 – have this document ready. They
recognize that it will not be an approved standard. It will be draft for people
to make comments on, but it could be prepared for the January 2006
implementation of pilots, if people so choose.

Observation 13, which is Test Pilot Objectives related to the whole pilot.
Now, this action may be a collaboration or whatever, based on what Karen just
said this morning.

I wanted to bring this to your attention, and if there is an action we can
do, based on that, great. If not, well, we made the offer.

Jill Helm of Allscripts suggested to me that an approach might be to
develop a matrix with the NCVHS recommendations and standards as a benchmark.
That is what is in use today, and then you apply some best-practice information
to determine what features and functions for the pilot seem to fit together and
what estimates are, in terms of the number of physicians, pharmacies, health
plans that would be needed for that pilot of that case, and determine what the
interdependencies are between the attributes, suggest the metrics that might
need to be collected and define the success criteria.

Jill is beginning this work. Once there is a rough draft, NCPDP will ask
electronic-prescribing organizations, the whole cast of characters, if they
would like to join this effort. It is important that CMS have active
participation in this effort. This document would provide a guidance and a
framework, but there would still be a lot to be done by CMS to actually
coordinate and execute the pilots.

So, based on what Karen mentioned this morning, if this is something that
perhaps we can assist – not reinvent the wheel – or if there is something you
wish us to do, Jill has made the offer to start. I don’t know. We can probably
talk off line, if you would like. It was just a thought that perhaps the
industry could help, since the last time we testified on this, it wasn’t quite
sure what pilots would look like or what metrics would be taken, things like
that. So that was the suggestion.

Under the NCVHS recommendations for next steps, these two items appeared.

The electronic signature for use in e-prescribing. The task group, led by
Dan Staniec of Caremark and Ken Whittemore of SureScripts, has been working on
an authentication-practices document since our January session – yes, this is
only February – and they have some real questions that they would like to ask
the committee, which Ken is here and has made himself available to ask some
questions of the subcommittee, maybe during breaks or lunch or something like
that, however, just to get a little bit more framework and feel on some of the
questions that the task group has brought back dealing with trying to make sure
they are meeting what the subcommittee was asking for.

And the last status is on the Provider Broadcast. This is a task group led
by Allan Smith of ProxyMed and they are completing an implementation guide
based on SCRIPT.

The implementation guide is planned to be brought forward at the May work
groups for balloting and approval, and, if approved, the ballot would then
occur in the summertime.

That’s it. Thank you very much.

DR. COHN: Lynne, thank you very much.

I think we’ll maybe discuss this now. Given that you have some time left,
you may want to cede some of your time to Dan Staniec to – is it Ken Whittemore
or is Dan here?

MS. GILBERTSON: Dan was going to come as well –

DR. COHN: Oh, okay, but just Ken –

MS. GILBERTSON: – but, at the last minute, he can’t. Ken is here.

DR. COHN: Okay. Well, okay. To Ken to discuss his issues or concerns.

But why don’t we talk about what you have here first, and then we’ll move
to that other issue, if that is okay.

Comments, questions from the subcommittee?

Harry.

MR. REYNOLDS: Yes, Lynne, on – well, first, excellent work, and, second, I
learned a new way to present. If you keep moving between documents, they’ll
stay with you. (Laughter). You all gotta pick up new tricks – (laughter). I
like that. You had everybody really jumping on it.

On the first page of this document – (laughter) – not the presentation on
prior authorization, you mentioned – at the end of the first status, you
mentioned inferring pickup. What does that mean?

MS. GILBERTSON: Oh, I knew you would ask. I’m trying to think. I haven’t
been as active in this task group as I have in others. If I am not mistaken,
this has to do with – as you go through the workflow of – let’s see. I’m
thinking it has to do with the fact that because I have refill environments
that I can – no, wait. I’m drawing a blank.

MR. REYNOLDS: Yes, well, that’s fine.

MS. GILBERTSON: I’ll have to look that one up. I don’t remember.

MR. REYNOLDS: The reason I am asking the question is, as you think of our –
the whole process of e-prescribing and the medication download – the
medication-history download –

MS. GILBERTSON: Um-hum.

MR. REYNOLDS: – things like inferring pickup, knowing that they picked it
up and might be taking it is one thing. Inferring that they picked it up and
are taking it, so that the next doctor they go to downloads a medication – a
set of medications and whether or not they actually did or didn’t – so
inferring was –

MS. GILBERTSON: Right. It is not guidance to infer. It is guidance to be
careful that you don’t infer.

MR. REYNOLDS: Oh, okay. That’s –

MS. GILBERTSON: It is all the different aspects of what if I didn’t get –
you know – if I didn’t get a message or if I didn’t look in the right place for
my message or things like that. They tried to really present all different
aspects of, well, what happens if your partners are only enabled in certain
functions or what if they only want to see based on – they want to know fills
for certain drugs or they only want to know fill notifications for the first
fill, but they don’t want to know for every refill. So they look through a lot
of the different business cases of what people could support at different times
within the growth cycle and making sure that they gave you caveats and things
to think about as you determine.

I spent a lot of time thinking in terms of what if you only chose to
support some of the functionality first, what could be the repercussions of
those choices, and here’s some of the things you really have to think about
before you decide I am only going to implement this aspect of it first.

So the only other thing –

MR. REYNOLDS: That’s fine.

MS. GILBERTSON: Richard is here. If you remember the inferring discussion.
I’ll have to look it up.

MR. BROOK: That was pretty much – you described it correctly.

The biggest concern we had, again, is if the patient picks it up are they
taking the medication, and we can’t guarantee that.

The other thing was what we tried to trigger on the fill status was that
some of the pharmacies might be enabled to do this at one point. Others will be
at a different point. So depending on when this goes into play, they might get
the fill status back for patients when they have picked up the medication or
they would get a negative fill status back if the patient didn’t pick up the
medication. So, again, depending on how they would implement this is where we
were trying to figure this out.

MR. REYNOLDS: One other – thank you. One other comment. On page 3 of this
same document, at the end of the first – right before Observation 6, you talk
about the – being able to – it says the implementation guide currently supports
multiple drug identifiers, but – and then you are going to see how RxNorm – so
you are – basically, you are saying that with what you have in place, multiple
drug identifiers can be handled. So if RxNorm does or doesn’t

work out – and we’ll hear more about that this afternoon – would you
recommend continuing to support multiple drug identifiers or go to RxNorm? And
if you want to wait ‘til this afternoon and chat about that –

MS. GILBERTSON: Yes.

MR. REYNOLDS: – (laughter) – that’d be fine, but –

MS. GILBERTSON: Well, if I had to put a point in time right now, it would
be we support the multiple-drug identifiers, because that is what we have and
that is what we know and love and we can feel confident in, and that will
continue to exist, because it is assumed that because of all the knowledge that
exists in the drug database – you know, the files that e-prescribing systems of
all makes – the prescriber, the dispenser, the health plan – will continue to
have those drug files because they need the drug utilization review
information, the interactions, the things that RxNorm is not meant to and
doesn’t intend to cover. So you will still need those identifiers and you will
still haul those around, because if you and I are talking the same identifiers,
then we might as well use those for the clear concepts of what we are trying to
do.

RxNorm would be added, obviously, as a good function for mapping when you
and I aren’t talking the same drug databases, but until we know at what level –
and you’ll hear more in the testimony, because there’s so many different use
cases, you have to make sure what the intent of the sender is, and you get the
right RxNorm concept and that you feel confident you have got a one-to-one
match that we don’t know exactly what that means in the formulary and benefit,
where there are different concepts, and it just – really, it just boils down to
saying, okay, in this use case, in the formulary-and-benefit environment, when
you are transmitting information about alternative drugs, what level of drug do
you want to relay to a prescriber system or e-prescribing system? That
corresponds to what level of concept in RxNorm? And getting that very clearly
defined, so that people know what it is they are using at that level, and then
taking that same analysis and saying, okay, when we are talking about a new
prescription, what is the level of what the prescriber’s intent is for the
drug, and then turning it into the appropriate concept level at RxNorm and
making sure those instructions are clear, so that people who are using it get
the one to one.

MR. REYNOLDS: Thank you.

DR. STEINDEL: I have a followup to Harry’s question. What does RxNorm add?

MS. GILBERTSON: Perhaps that should be discussed later? (Laughter). I mean
–

DR. STEINDEL: That’s a fine answer. Thank you.

MS. GILBERTSON: Yes, I mean, a part of it is, as the e-prescribing
testifiers said – gosh, early in 2004, I think it was – right now, they are
using the text – the label name or the – you know, the drug name when they are
– in new prescriptions, for example, from a prescriber to a pharmacist, the
best thing to use, at this point, is the free text. Well, you can’t do much
computationally with the free text. Pharmacies can – once they have dispensed a
drug – turn it into an NDC or a – you know – a nine-digit NDC, a shortened
version of it just to get the intent. So, now, you have a computational. Your
clinical systems don’t talk the same language is what the testifiers mentioned,
so that you can link back if you are using one drug database and I am using
another and you want to – if you want to get to codes, and we are using two
different ones, we need some kind of linkage. So in that particular intent –
and from what I understand, that is what RxNorm’s intent is. It is not to be a
drug database with all the bells and whistles that are out there. So as far as
a mapping, you know, I would think that is the function.

DR. COHN: Yes.

MR. BLAIR: Now, you knew that, so what was the thrust of your question?

DR. STEINDEL: I wanted a clarification, Jeff, on exactly where RxNorm fits
in between the use in clinical systems for one purpose and then the use in the
external prescription world, and even the internal prescription world within
the hospital, for other purposes, and I think one question that still is in my
mind is do we have enough of a mapping system or relation system between RxNorm
and the actual order codes being developed yet to instantiate that relationship
and make it useful?

DR. COHN: Okay. And, Steve, since we are going to be talking about that
this afternoon, let’s – why don’t we hold that question?

DR. STEINDEL: That’s –

DR. COHN: Thank you.

Michael.

DR. FITZMAURICE: Just like we congratulated Karen on her fine work, this is
a lot of fine work by you, your organization and by the team leaders and the
people who work underneath them.

As I went through your slides, I was thinking, gee, now, the hassle is just
moving from the pharmacy to the physician, but my daughter worked in our family
physician’s office for about a year, and one of the things she had to do was
call up insurance companies about drugs and devices and coverage, and I
realized the physician has always had the hassle, that the pharmacy might be
relieved of some of the hassle if the physician is dealing directly with the
benefits – the payers – but it is going to take more time, and there probably
will have to be something where the physician him or herself isn’t spending a
lot of time, but somebody else is delegated to spend all that time working on
this.

I liked what you said about it would be nice if the prior authorization
questions were known up front. It is remarkable to me that they are not, but I
know, from a fact, that they are not.

The other thing that you mentioned in your presentation and I think should
be a really strong goal for all of us is interoperability transfers the goal,
not a creation of interface, that it is interoperability that we are after.

Secondly – or lastly – your final offer to develop a standards benchmark
and roadmap for pilot testing, it is very perceptive about what is needed, I
think, and it is a very generous effort to work with CMS to get the job done
openly and expeditiously, and I hope CMS takes you up on that offer.

DR. COHN: Michael, thank you.

Karen.

MS. TRUDEL: Yes, I would echo what Michael said. I can’t believe the amount
of work that you’ve gotten done in just a short month or two, and really think
you should – I’ll be very proud of that.

In the portion of the proposed rule that talks about the formulary and
medication history standards, we say that the – we acknowledge the fact that
the RxHub protocols are out there and that there is an activity underway to
bring them within the SDO realm, and we say that if they become accredited
standards and they do not change significantly, so that we can still say there
is adequate industry experience, that could be something that we would do.

I am kind of interested in getting a sense of how much they have changed in
the course of these deliberations. Hardly at all? A lot?

MS. GILBERTSON: Medication history is under ballot, so when we review the
ballot comments, we’ll know what modifications might come in. I do not know at
this point. The presented material did not significantly change at all. In
fact, all we did was really incorporate it into – and I think I SCRIPTized one
or two things, rather than RxHub, you know.

Medication history. Formulary and benefit. There has been addition, but
most of it has been clarification or further information related to – for
example, you know, they were only supporting one set of drug qualifiers, and,
now, there are multiple sets of drug qualifiers or things like that.

I don’t think – I know content-wise, there hasn’t been a lot of addition.
There was some cross referencing that was added, based on MedaMedia’s(?)
experience, but that was additional information that was added and would be
used by partners that need it, but, basically, structurally, it is the same. It
has just been a little tweaking, you know, add a hold for an RxNorm code and
things like that that have been added. So it is not a lot.

MS. TRUDEL: Thank you.

DR. COHN: And maybe I’ll follow up, because I think Karen went about
three-quarters of the way with the question I was going to ask, which was, of
course, knowing in the back of my mind that we had, in our recommendations,
said, you know, this will be a foundation standard or something that would be
piloted and we will advise the Secretary later on in the year, and I don’t know
that we have to do that for March 2nd. We could do it for April
whatever or June whatever.

The question is is where do you think it is going to end up? Do you think
it is going to be something that the industry will be behind as something that
has two aspects that should be part of the foundation standards?

MS. GILBERTSON: I don’t want to presume to speak for the industry –

DR. COHN: Sure.

MS. GILBERTSON: – but we’ve got an awful lot of participants that are
putting an awful lot of time into it, and I would think that is a show
immediately that they are willing to buy into it, and some of the participants
are already using the protocol that is out there and are just monitoring the
changes and helping to understand.

Other ones are maybe new payers, for example, or new health plans to the
environment, and they all seem to be extremely supportive, and they have run it
through their systems folks to see can we do it and come back with minor
suggestions. So I can’t presume, but I really think it is going to have a very
positive impact, yes.

DR. COHN: Okay. Then maybe I want further clarification just for my own
understanding.

Both of these are apparently going to be – let’s see. I guess one of them
is being balloted now with adjudication in March.

MS. GILBERTSON: That is medication history, correct.

DR. COHN: And the other one is being –

MS. GILBERTSON: Will be brought up for approval in March and would be
balloted after that.

DR. COHN: I see.

MS. GILBERTSON: And are just – you know, one came in first, and the next
one is coming in.

DR. COHN: I see.

MS. GILBERTSON: The medication history is the first ahead, but it is only
basically one quarter ahead, because, now, formulary and benefit will come in
and be discussed next month and we’ll prepare a ballot within about six weeks
and get that ballot ready to go, will be out there.

DR. COHN: I see. So –

MS. GILBERTSON: And we ballot at least four times a year. So things happen
constantly.

DR. COHN: Okay. So that would be on the ballot after this, in other words.
So that would be the June ballot probably.

MS. GILBERTSON: Let’s see. March – May.

DR. COHN: May ballot. Okay. So it sounds like we would have – the
subcommittee could choose to hear more – I mean, obviously, it is premature to
ask until after you finish balloting –

MS. GILBERTSON: Right.

DR. COHN: – and the adjudication of the ballot, but it sounds like as early
as April we would know about one and maybe by June the other. Is that –

MS. GILBERTSON: Um-hum.

DR. COHN: Okay. And, at that point, the industry could comment about
whether they are feeling comfortable enough about these things to be foundation
standards –

MS. GILBERTSON: Right.

DR. COHN: – or pilots or whatever.

Okay. I mean, thank you.

Harry.

MR. REYNOLDS: Yes, along the same track, all of your standards have to go
to your Board of Trustees for final approval, is that correct?

MS. GILBERTSON: That is correct.

MR. REYNOLDS: And how often do they meet – so back to Simon’s point, and
Karen’s – so if you – something starts showing up in March and then it gets
balloted and then – so how many times do they meet a year, so how many more
shots this year – (laughter) – are there to actually have, you know, something
be completed?

MS. GILBERTSON: You have a shot each month.

MR. REYNOLDS: Each month?

MS. GILBERTSON: Right. When the board does not meet, they do things via
email.

MR. REYNOLDS: Oh, okay.

MS. GILBERTSON: And they approve documents via that. So if something is
ready at a particular time, then it can go into the next board packet for that
month.

MR. REYNOLDS: Oh, that’s good. Okay.

MS. GILBERTSON: But, now, recognize that – I mean, per the NC procedures,
every ballot goes through the initial ballot, and if there are negative
comments, those must be adjudicated, and then there would be a recirculation
that addresses all those comments. So if there are negative comments, we go
through a recirculation, but that – in NCPDP terms, that is quarters. So you
might have a new ballot in one quarter, a recirculation in the second quarter,
with, hopefully, approval right after that.

MR. REYNOLDS: Again, shooting towards the ‘06 pilots and shooting
towards some of our goals of what to do – I just want to understand – in other
words, if they met twice a year –

MS. GILBERTSON: No. No, no, no.

MR. REYNOLDS: – and the first one was in April and the second one was in
July and game over, a little different –

MS. GILBERTSON: That is why I always cringe when someone suggests SDO’s
move slowly. I’d say, how much faster would you like us to go? Because then –
you know – you approve standards, I mean, for example, in the rule it names
SCRIPT 50, well, we are balloting 80 already. So we have moved that far
already, so – you know, and the other factor is the industry has to decide they
want to move to that version, too.

MR. REYNOLDS: I would commend you. That is a well-oiled process. I mean, it
is that available that often, that is pretty cool.

DR. COHN: Jeff is next.

MR. BLAIR: You know, I am going to get on a long list here of folks that
are complimenting the tremendous amount of work that has been done, and,
specifically, the willingness of RxHub to submit its protocol, so that they
could become recognized as standards by going through the ANSI accredited
standards process.

The other thing that – there’s two other areas.

You indicated Jill Helm has volunteered to kind of take the lead on trying
to pull together a matrix of the information that would be necessary to begin
to plan the pilot tests and a roadmap for the pilot tests, is that correct?

MS. GILBERTSON: She is shaking her head yes as well.

MR. BLAIR: Yes? Okay. Thank you, Jill.

And one of the thoughts that I have that you might include in the roadmap –
and I understand you have to do it step by step by step – but I have a feeling
that the roadmap that you begin to create is going to lead to the requirement
for CMS to support and sponsor an array of pilot tests, because I don’t think
you could craft or design one single pilot test that will test all of the
things that need to be tested. So – you know – as you go through that process,
you probably wind up identifying the requirements and the metrics, and then
maybe there is a Stage Two where you begin to say, okay, which of these can be
pulled together into Pilot Test A and which will need to be grouped into B and
C and all of the rest, and I just wanted to make that observation, because I
think that that has sort of been an unsaid assumption that a lot of us on the
subcommittee have had for a long time, and now that you are doing that, you
could probably document where those multiple pilot tests are needed.

MS. GILBERTSON: That is very correct, Jeff, because we have identified
early on in the process that this is multiple pilot tests. It is not just one
big glump of people all doing the same thing altogether, and we need different
viewpoints of people to say what are the metrics and the deliverables from
different viewpoints of what we can glean out of this, so that we are not just
– you know. I mean, if you had me do it, I would have, probably, a couple of
single focuses, because that would be my perspective, but we’ve got the
opportunity to grab a whole bunch of stuff, and let’s grab it and see what we
can use, whatever it is.

MR. BLAIR: One last observation on the requirements for the pilot test and
roadmap for the pilot test. I imagine that a lot of those pilot tests can be
run in parallel. However, there might be a few where the results from a certain
set of pilot tests would determine the design or construction or the metric for
a second set, and there’s a year there, so there may be the opportunity to test
certain things which would then feed into a subsequent set of pilot tests still
within the year. So that is just a thought.

MS. GILBERTSON: I noted that, yes.

MR. BLAIR: The other thing is, Simon, you just mentioned the fact that it
might be possible for CMS to recognize some of the messages that have been
converted into NCPDP SCRIPT standards that RxHub has contributed faster than
before the pilot test is completed, and I hadn’t been thinking that way, but if
I start to think that that might be possible to accelerate it, then, I would
ask the question – well, Lynne, the industry that makes up the constituency for
NCPDP is somewhat different than for many other standards organizations in
healthcare, and it may be possible for you to actually get a documentation of
the breadth of usage of the standards. It has been harder, for example, the HL7
constituency to come up with a number or ASTM or a number of the others, but
you might actually be able to document widespread marketing – that has been
kind of the criteria NCVHS used in the past to get a comfort level on
recommending a standard. So you might think about that, because that could be
helpful.

MS. GILBERTSON: Okay.

DR. COHN: And, Jeff, just to clarify, the reason I was bringing the issue
up was more that our recommendation letter of September had sort of stated that
we were going to provide additional guidance on those two areas. I have no
particular opinion whether it is a pilot or a foundation standard. I was just
curious of what their perspective was on that.

Okay. Other questions before we – I actually would like to have us have a
little time to have Ken come up and talk about his concerns or questions – not
concerns, but questions around best practices.

But anybody have any final questions on this one?

Well, Lynne, hopefully, you are comfortable having Ken take some of your
time here.

MS. GILBERTSON: Definitely, especially on this topic.

DR. COHN: And, again, Lynne, thank you for some very good work. We really
appreciate it.

MS. GILBERTSON: Thank you.

MR. WHITTEMORE: Good morning. Ken Whittemore from SureScripts, and before
we start, I would like to thank Lynne for putting to everybody that I was in
the audience today. (Laughter).

I think probably the little statement that Lynne made at the beginning
didn’t quite capture what we have encountered with the task group that has been
looking at electronic signatures, and I think it would probably be instructive
to the committee if I just kind of gave a little bit of history of how that
task group came to be and what were the driving elements and, again, how we
wound up where we are.

As I recall, the task group was formed probably last May in St. Louis. Is
that – that is what I seem to recall. I know you go to a lot of meetings – and
it was primarily at the impetus of the concern that some of the members of
NCPDP had in terms of what they were running into with respect to
interpretations of the meaning of the term electronic signature among the
primarily state agencies, state boards of pharmacy and the like, and we had a
fairly lengthy discussion at that first meeting and shared experiences and what
not, and we decided it would probably be a good idea if NCPDP put together an
effort that would communicate to the world at large, primarily the state
regulatory agencies, exactly what was being done in terms of electronic
signatures for e-prescribing, and so it was primarily created as an educational
effort. It wasn’t so much designed to create a standard, per se, because I
don’t think anybody really thought that was necessary. There are a lot of
entities out there and you have heard from most of them during your testimony
back during December and again in January that there are a lot of different
types of processes that are being used, and everybody, within their own
business environment seems to be happy with the way they are working.

I think where my anxiety level got raised a little bit was in the January
sessions where I guess I really hadn’t fully appreciated what the subcommittee
needed, but Dr. Cohn turned to Lynne and said, we are really anxious to hear
what you are going to tell us about best practices in electronic signatures,
and the next phone call that we had with the task group was kind of like one of
those situations where, yes, we can do that. Yes, we can do that, and then you
get together – How are we going to do that? (Laughter).

The people who are participating in the task group are folks, again, who
have run into some of these issues at the state level. There are a few people
who have a great amount of expertise with respect to authentication and those
types of things. I am not one of them, and I made this known up front when I
was, shall we say, volunteered to be one of the task group leaders, because I
had run into some issues, and I think what we have run into is we have made
several attempts, since coming to understand that the subcommittee wanted more
out of the task group, to put something together that might be useful, and the
document that we kind of – or got distilled down, if you will, last week, the
three of us, Dan Staniec and Lynne and I looked at it, and we just did not feel
that it was something that the subcommittee would find useful.

So, in terms of questions, I guess the question is – the primary question
is how urgent is your time frame in terms of getting some kind of guidance?

We can probably go back to our various organizations that are doing
electronic prescribing and ferret out those people who do have the technical
expertise to put together something that would point to some best practices,
but we would have to do that and then get those people focused on working on
the goal that has been put before us.

So it is more a question – today, I think it is a status report and our
concern that, above and beyond the guidance that NCPDP has already given the
subcommittee, which I think was back in December, we don’t have a lot in the
way to offer from our current task group, and so we would have to basically
reconstitute or I guess beef up a task group before we could give you what we
think you are looking for.

DR. COHN: Well, I don’t know if that is so much a question as a statement.

MR. WHITTEMORE: Well, that’s why I said I wasn’t – (laughter). I think this
kind of mischaracterized a little bit what our discussion was last week, but
that’s –

DR. COHN: No, and I do understand. There is nothing like dealing with sort
of deadlines that other people are putting on you.

Mike, do you have a comment, and I’ll ask others –

DR. FITZMAURICE: Just that it might be useful for the committee to learn
what the industry thinks is the state of the art and what they could do, but I
wouldn’t want the industry to feel pushed to do something that they are not
prepared and not already implementing.

We have known for quite a while that the electronic signatures – we are
waiting for a standard. We are waiting for a solution that’s adopted across the
industry, and we haven’t seen it so far, and so, to the extent that you can
tell us, you know, here is how close we are, that would be good, but you
shouldn’t feel forced to do something that might not be implementable and might
not be compatible with what the industry wants to do.

MR. WHITTEMORE: When we were having discussions in the task groups, one of
the things that we came back to quite regularly was that diagram that NCPDP
presented, because we rather felt that captured a lot of what everybody was
doing. I don’t guess you could really look at it as defining the best
practices, per se, but every time – I know every time we had a conference call,
sooner or later, we were kind of coming back around to it and the processes
that are represented in that document.

DR. COHN: Do others have comments?

DR. FITZMAURICE: Just –

DR. COHN: Okay – please.

DR. FITZMAURICE: Put yourself into the Secretary’s shoes or the Secretary’s
chair and the Secretary wants to be helpful to consumers, to physicians,
pharmacies, to the industry, and what would you advise the Secretary to
consider in the terms of electronic signatures? Watchful waiting? Take one of
these and see if it works in the morning? What would the industry have the
Secretary do at this point? I mean, you don’t have to answer that now –

MR. WHITTEMORE: Okay.

DR. FITZMAURICE: – but that is the kind of thing that if you put yourself –
(laughter). I would love to have your answer, though. It is the kind of thing
that you probably wouldn’t want somebody to jump right in, but, on the other
hand, you don’t want things to just languish around, and we are looking for
what is the state –

MR. WHITTEMORE: I could probably give a little bit of feedback. I think
there is a sense among the task group that something like watchful waiting
might not be a bad idea, because, again, there are a lot of different
processes, a lot of different people using which seem to work well for them.
Maybe, over time, these processes will coalesce and kind of head in the same
direction, and then that is the time to say, okay, here’s the best practice.
Here’s the standard.

MS. GILBERTSON: Well, then you get back to – I was kind of surprised when
you threw the term e-signature in there, because I wasn’t expecting – I was
expecting the best practices for e-prescribing, not for e-signature, which is a
different twist, and, once again, we go –

DR. FITZMAURICE: – too much.

MS. GILBERTSON: Yes, we go back to my intent to sign may be the fact that I
have enrolled in a program and I send a user password or something, you know,
or I use an ID to that network may be my electronic signing at that point.
Obviously, there is the digital signature and all that that could come even
above that. So that’s a little bit of a question mark as well as to – because
the other thing the task group was focused on is the fact that there is the
security regulations, which say you must analyze your risk and take action
based on that analysis, and that really is what everyone is using, and so it
doesn’t – you don’t really need to have a paper that says see the security
rule. You know, play with that or you got problems.

So they are still – you know – it’s two things, the security rule is out
there and that is what they are operating under, and the diagram explains how,
at a high level, authentication and things like that are occurring today. So
they are trying to figure out, okay, past that, what could we provide?

DR. COHN: Okay. Steve has a –

DR. STEINDEL: I think Jeff is probably going to make it.

DR. COHN: Oh, Jeff. Okay. Why don’t you make it and then I will –

MR. BLAIR: Well –

DR. COHN: – and then I will also try to sort of –

MR. BLAIR: I’m bridging off a little bit – but I don’t know how to do this,
and I just am going to suggest a need, in case the industry – in large part,
represented by NCPDP – might be able to construct something, so that – clearly,
the pilot tests are going to be pilot testing a lot of the work that you have
reported on, Lynne, to enhance the capability of e-prescribing over
e-prescribing networks, and if there’s some way that maybe – I don’t know,
maybe Jill can begin to include it in the pilot tests or you can all work
together. I don’t know how to do this, but we really don’t know the level of
security that we need when we run over an e-prescribing network. I mean, I
don’t think – at least I haven’t heard anything, and if there is some way to
include in the pilot tests some type of a measure of what is sufficient, what
is adequate, what is secure, that would be a helpful result out of the pilot
test.

So, Steve, did you have –

MS. GILBERTSON: You’re grimacing right now.

MR. BLAIR: He’s grimacing.

DR. COHN: Yes, and I guess I am grimacing a little bit, too, only because I
don’t think I have heard anything in any of the testimony that tells me the
things going on the networks aren’t secure –

MR. BLAIR: But we don’t know, and that is what I am saying.

DR. COHN: Well, it isn’t that we don’t know. It’s just that – I think that
the best practices have not been well described, which is very different than
we don’t know.

You know, Steve – I wanted Steve to make a comment, but I guess my general
view on this one is is that – and maybe I’ll jump in just to make a general
comment or two – is is that I want to sort of pull you all out of this one a
little bit in terms of – I feel like we are getting deeper and deeper into a –
either it is a marshmallow field or a morass or some – something in on that
way.

If there had been something that you had been able to put together, it
would have been very helpful, just because we could have referenced it and sort
of looked at it, you know, something perhaps a little beyond what we received
in the testimony, though, certainly, what we received in the testimony, I
think, was a very important piece of sort of best practice and architecture we
can reference.

I think not having something today is not terrible. It changes the nature
of what recommendations we may have going forward, since we can’t say, look at
this. This really looks good or not, but it is not – you know, I mean, it’s not
a terrible situation.

I do think that what we would ask for your help is as we go through and
begin to talk about recommendations and how to frame things and all that as we
move forward over the next day and a half, as well as, potentially, in other
public conference calls we may have between now and the end of February, we
might ask your participation and involvement to sort of provide perspective on
all of this, and that, I think, would be primarily what I would ask.

I mean, certainly, there may be aspects – and, once again, this is going to
be a conversation we’ll have today and tomorrow – that there may be certain
things about either authentication or e-signature, and I guess I tend to think
of this as authentication or non-repudiation – let me be very exact in the
terms here – that may need to be evaluated, and that is something I think we’ll
be talking about, but I think that is something that we still need to get our
arms around, and, once, again, you may have some perspective on those specific
areas, which I really think is the areas that we really have been talking about
over the last couple of months.

Now, Steve, did you have a comment? Am I getting closer?

DR. STEINDEL: I think the only comment that I have to make is you said
exactly what I wanted to say, probably much better than I could.

DR. COHN: Okay. Other comments?

Harry.

MR. REYNOLDS: Yes, I just had one – I think that Ken’s testimony has just
reiterated that there is not – the industry has come forward with a number of
standard recommendations. There is no standard recommendation for this
particular situation. There is an agreement upon the security concept and the
fact that people are securing what they are doing, but something for us to just
pick and say that’s it is not available, and I think some of the information –
the way Margret and Jeff have captured it in some of our draft work that we’ll
see tomorrow, I think – I actually feel better about the letter –
(laughter) – the proposed letter than I did, because it just reiterated we
didn’t miss something, and it wasn’t something out there that was easy to pick
up, and I think Ken’s testimony says it.

DR. COHN: Other comments from the subcommittee?

I guess should ask Ken are you feeling better now or worse?

MR. WHITTEMORE: Oh, much, much better.

DR. COHN: Much better. Okay. (Laughter). Okay.

And, Lynne, I, obviously, want to thank you both. I mean, Lynne, do you
have any final comments on all this stuff? I mean, this is going to be,
obviously, an ongoing conversation, but we are really very impressed with the
amount of work that has gone on and we really want to thank you for the amount
of energy from – and your leadership as well as, obviously, all of NCPDP and
the hard work that has gone forward to move us this far. So thank you.

Now, with that, I think we have a well-deserved 15-minute break, and we’ll
take that.

(11:02 a.m.)

* * *

(11:25 a.m.)

DR. COHN: I think we are going to start – Karen Trudel, you had a comment
or two on pilots that you wanted to share, please.

MS. TRUDEL: Yes, I was a little bit surprised at the NCPDP offer with
respect to setting up some of the criteria for the pilots and metrics and
critical-success factors. I was not personally aware that there was any
consideration to doing that and wanted to just explain where we are coming from
and the fact that we have a different set of rules here. We are talking about a
procurement, because we will be signing contracts or cooperative agreements
with these sites, and, yes, it will be more than one, and, yes, we do
anticipate that the standards being tested at the various pilots will not all
be the same, but considering that we are doing a procurement and there are
federal rules that cover such things, I think we are not able to work
proactively with groups to develop some of these kind of roadmaps.

I would encourage anyone who has any thoughts about what should be in the
pilots, what standards should be tested by the pilots, how they should be put
together, how they could be evaluated, the critical-success factors, et cetera,
I would encourage anyone to let us know what their ideas are. We’ll take
everything into account, but the ability that we might have to work with one
particular group or another to come up with a model is very limited.

So any thoughts about pilots should be directed to Maria, and her email
address and contact information is already widely available at this point, and
I can – (laughter) – and I can assure you that we’ll take any input into
consideration.

DR. COHN: Maria, do you have a comment on this?

MS. FRIEDMAN: I would just like to say that we are working on the RFP. Our
goal is to get it out quickly, and so any input that you have we would
appreciate it quickly. I can’t give you a specific time frame, because you know
how long these things take, but, still, we have to get this thing out on the
street pretty quickly in order to do the open competition and to make the
awards and get – you know, have the lead time that we need to make sure
everything gets up and running by January 1st.

DR. COHN: Okay. Maria, I just wanted to thank you.

I actually have a comment also, and then Mike does.

And I guess I would also observe, as I was listening to, I think, some
people’s hopes for pilots, just the observation that standards are a
constantly-evolving process, as well as improvements to software, networks and
everything else is always a constantly-improving process, and these pilots are
not the last time that there will be ongoing improvements in all of this. So
pilots are going to be part of that ongoing activity. We’ll hopefully
accelerate and identify for the department directions around for the final
standards, but, certainly, not going to be the last time or the only time that
there will be need for new or improved standards. So this will be, obviously,
an ongoing process.

Mike, you had a comment.

DR. FITZMAURICE: I didn’t want the offer by NCPDP to go by the wayside, and
Karen left the door open by saying that we would be willing to take any input
from the industry that came in to help formulate thoughts on the RFP.

I wondered if the input from the industry came in by – as late as March
30th that that would still be useful in terms of here’s the
situation with the standards and what is ready, what still needs to be tested,
from their point of view, and how things might work to be ready by the time of
the pilot. Would that be useful information if received by March
30th?

MS. FRIEDMAN: I would say sooner.

DR. FITZMAURICE: Sooner? So we are talking by March 1st it might
be useful then.

MS. FRIEDMAN: In order for us to write an RFP and take into account
everything we need to take into account, and, especially, you know, be very
helpful for things like metrics, those kinds of things, in order for us to get
it out on the street and one last look, we really need it sooner than – I would
say March 1st would be better or even – hate to say sooner, but we
are really trying to fast track this.

DR. COHN: Okay. Karen, Maria, thank you, and, Maria, congratulations on
your new assignment.

Agenda Item: State Issues Related to
E-Prescribing of Controlled Substances – Eleni Anagnostiadis

DR. COHN: With that, we turn back to our discussion for the day, and,
Eleni, I want to thank you for joining us. Please go forward.

MS. ANAGNOSTIADIS: Thank you.

Dr. Cohn and Members of the subcommittee, on behalf of NABP, I would like
to personally thank you for inviting us here today to provide you with
information regarding state-related issues that relate to the electronic
transmission –

DR. COHN: Eleni, you’re going to have to get closer to that microphone.
Yes.

MS. ANAGNOSTIADIS: I’m sorry – of controlled substance prescriptions.

The state boards of pharmacy and NABP recognize the importance of creating
a regulatory environment that facilitates and regulates the electronic
transmission of prescriptions for both controlled and non-controlled substances
in the interest of patient safety.

We have had an opportunity to testify before the subcommittee a couple of
times, so I am not going to repeat myself on the information that we shared
with you in the past, but pursuant to the current request of the subcommittee,
today’s testimony will focus specifically on state issues related to the
electronic transmission and prescribing of controlled substances.

It is important to note that the state board of pharmacy authority
encompasses both controlled and non-controlled substances. However, regulations
regarding controlled substances, including the transmission and prescribing of
prescriptions for controlled substances, bear significantly on the regulations
and policies of the Drug Enforcement Administration.

The primary concern NABP has regarding the electronic transmission and
prescribing is that we want to be sure that the prescriptions are
authenticated, legitimate and have integrity for all prescriptions, both
controlled and non-controlled.

I am going to split the testimony up into two separate sections today. The
first part, I am going to talk a little bit about the DEA regulations for
controlled substances, and then I am going to shift over to the state-specific
controlled substance act.

So, you know, we not only have the state boards of pharmacy. We also have
the Drug Enforcement Administration that I will refer to as the DEA, and then
we also have another group that sometimes works with the board of pharmacy. It
is under the purview of the board of pharmacy, and, sometimes, it is not. They
have to work with another agency within the state that have authority over the
state-specific controlled substance act.

I’ll begin by giving you some background on the DEA regulations. State
regulations pertaining to controlled substances intersect with the federal
Controlled Substances Act, which I will refer to from here on out as CSA and
DEA authority.

The CSA, which focuses on the ordering, handling, distribution and
dispensing of controlled substances, is enforced by the DEA. Although states
have no explicit authority to enforce federal laws, states have enacted
state-specific controlled substance acts, which incorporate the federal CSA, at
a minimum, and, in a limited number of states, include more stringent
provisions.

The complimentary and sometimes conflicting relationship of state and
federal laws regarding controlled substances highlights the importance of
defining federal requirements and DEA policy for the electronic transmission
and prescribing of controlled substances.

NABP understands that the DEA presented information to NCVHS outlining
their position on the electronic transmission of controlled substances. Beyond
the information presented to NCVHS, NABP is not aware of any proposed changes
in the CSA or regulations from the DEA in regards to the electronic
transmission of controlled substances.

We are aware that the DEA has authorized a pilot project at the Hines VA
Hospital to evaluate the effectiveness and security of transmitting controlled
substances – and I believe we heard testimony – you heard testimony a couple of
weeks ago for the transmission of controlled substances from CII’s all the way
to CV’s.

NABP is hopeful that the pilot-project results will provide the DEA with
the information needed to develop and release requirements for the electronic
transmission of controlled substances.

The release of these requirements is critical to any recommendation of the
NCVHS or action by the states because the states typically defer to the DEA and
federal CSA for guidance.

And I would just like to share with you – and I have it in the testimony as
well – as recently as January 2005 – this month – or last month – in the Nevada
State Board of Pharmacy Newsletter – and I quote – they say, “The near
future will reveal a federally-approved Drug Enforcement Administration
electronic prescription prescribing system. The board office has been hesitant
to establish one mechanism, soon to be superceded by another. Regardless, any
electronic signature transmission system needs Board of Pharmacy approval and
none have been given.”

And I guess the point that I want to make – and it is not just Nevada, but
a lot of the other states that I have been talking to – they are very hesitant
about even coming up with regulations regarding electronic transmission of
prescriptions in general, if they don’t have any in place, and specifically for
controlled substances, for the reason that I stated here.

In any event, the more stringent laws, whether they be federal or state,
will take precedent, whether the regulation comes from the DEA or the state
board of pharmacy.

NABP strongly recommends that the NCVHS exert whatever influence it may
have to foster the release of regulations and requirements from the DEA.
Recommendations regarding the electronic transmission and prescribing of
controlled substances must encompass the actions or anticipated actions of the
DEA. To do otherwise would create conflicting regulations at both the federal
and state levels.

As mentioned earlier, the states will follow the lead of the DEA and
incorporate additional security measures and limitations placed on the
electronic transmission and prescribing of controlled substances into their
state requirements.

NABP is hopeful that the electronic transmission and prescribing
requirements for controlled and non-controlled substances will be consistent in
order to minimize fragmentation and potential barriers to electronic
prescribing and transmission.

Standards for electronic transmission and prescribing should incorporate
the necessary security, accountability and privacy domains whether the
substances are controlled or non-controlled.

What I have tried to do – if you look at Appendix A – is I have tried to
give you an analysis of the current DEA regulations. I went through them one by
one and listed them, and then I – on the far right-hand column, you’ll notice,
I mentioned whether or not these regulations would or would not be impacted by
the electronic transmission of prescriptions or would have different
requirements from the paper requirements, and, then, in the center, I mentioned
whether or not the state regulations differ from the DEA regulations.

There are really only three that I want to point out to you here. On the
first page, we have one, and on the second page, we have two. Only three of the
regulations out of the 15 listed here show that they will be impacted by the
electronic transmission and the DEA will have to address.

On the first page, about halfway down, under Prescription Requirements,
Where an oral prescription is not permitted – and that is in the case of a
controlled – a CII – a prescription must be written in ink or indelible pencil
or typewritten and must be manually signed by the practitioner.

Today, the state regulations – the states defer to the DEA for that
particular regulation. However, that will be impacted by the electronic
transmission.

On the next page, under Schedule II Substances, today, all Schedule II
substances must be handwritten by a physician and must be signed by the
practitioner. Again, the states defer to the DEA.

And, finally, the Schedule III through V Substances may be communicated
orally, in writing or faxed. Again, the states defer to the DEA on that
particular requirement, and, as you can see, the states, right now, are looking
to guidance from the DEA for their proposed regulations in the area of
electronic transmission of controlled substances.

Also absent from the current DEA regulations and requirements, but a
significant consideration, is the use of electronic or digital signatures, and
I know they mentioned that during their testimony.

NABP anticipates that this area of interest will be a primary focus of any
DEA regulation and will be developed specifically to meet the concerns of the
DEA, regardless of the relation to overall standards or requirements for
electronic transmission or to creating a regulatory environment that
facilitates the electronic transmission or prescribing of prescriptions.

As mentioned earlier, the states do not have the ability to override the
more stringent DEA requirements. Under current DEA regulations and
requirements, prescriptions for controlled substances cannot be transmitted or
prescribed electronically until the DEA updates their regulations to allow for
the electronic transmission of those prescriptions.

If the standards or requirements for electronic transmission do not satisfy
the DEA’s needs and result in a separate set of standards for electronic
transmission of controlled substances, then a cumbersome and fragmented system
will result, imposing burdens on practitioners and patients throughout the
system.

I am going to switch gears now and give you a little bit of background on
the state controlled substance acts.

In all states, the regulation of the electronic transmission and
prescribing of controlled substances will be impacted by any regulation or
requirement issued by the DEA. Absent any federal regulations or requirements,
the states will be forced to develop requirements for the electronic
transmission and prescribing of controlled substances.

In the past, the states have treated controlled substances differently than
non-controlled substance and imposed additional requirements for controlled
substances because of the highly-addictive and susceptibility to diversion that
characterize these substances.

Upon conducting a brief search of state regulations regarding the
electronic transmission of controlled substances, NABP identified New Jersey
and Wisconsin as states that have language in their regulations that would
potentially allow for the electronic transmission of controlled-substance
prescriptions pursuant to the DEA establishing federal regulations allowing for
it.

Wisconsin’s regulations do not specifically include language, but they do
say that they will not allow for the electronic transmission of CII
prescriptions.

And I share with you, and I noted the New Jersey regulations, because I
think they have some really, really good language in here that addresses
controlled substances and the electronic transmission. However, before I get to
that, I just want to say that most states do not specifically address
controlled substances in their regulations regarding the electronic
transmission. Again, they are waiting to see what the DEA comes out with before
they update their particular regulations.

New Jersey just recently, in June of 2004, updated theirs, and I think that
their language is very progressive.

I’ll skip down to, “A pharmacist may fill a prescription for a
Schedule II controlled substance transmitted electronically, provided that the
original signed prescription is presented to the pharmacist prior to the
dispensing of the controlled substance.”

That is what occurs today. For example, if a doctor wants to fax in a CII
prescription and the patient can potentially bring in the written prescription.

However, the language that is of great interest is, “If permitted by
federal law, and in accordance with federal requirements, an electronic
prescription shall serve as the original signed prescription.”

So what New Jersey is saying is if the DEA allows for it, they, in the
State of New Jersey, will allow the electronic prescription as the originally
signed prescription.

According to the NABP 2005 Survey of Pharmacy Law, all 50 states, as well
as the District of Columbia, have a state-specific controlled substance act.
While the CSA in nearly half the states falls under the purview of the board of
pharmacy, in the remaining states, some aspects of the regulation of controlled
substances may fall to another state agency, such as the dangerous drugs
bureau, drugs and narcotics agency or the bureau of drug control. It is through
state-specific CSA’s that state regulatory agencies define their authority for
the regulation of controlled substances.

Ultimately, authority for the electronic transmission and prescribing of
prescriptions for controlled substances will rest with the individual state
board of pharmacy or similarly charged agency, because this practice is an
integral component of the practice of pharmacy.

Recommendations from the NCVHS must account for the regulations and
requirements of the DEA, as well as existing state regulations or requirements.

NABP, again, urges caution in any preemption of state laws and regulations,
particularly those governing the dispensing of controlled substances, because
of the highly-addictive nature of these substances and problems with diversion
and trafficking the states have experienced.

In closing, NABP recognizes the benefits of the electronic transmission of
prescriptions and understands the positive impact this technology can have on
patient safety and the prescribing of all prescription orders.

Electronic or digital signature considerations and qualifications will be
critical to the entire validation process and extremely dependent on the
technology and standards used to ensure the authenticity, legitimacy and
integrity of the electronically-transmitted prescription.

While many arguments can be made to support the rapid adoption of
electronic prescribing, consideration should be given to the development of a
national standard that is focused on patient safety, public protection and the
provision of quality health care.

NABP is committed to assisting the NCVHS, CMS and other interested
stakeholders in developing standards, laws and regulations for the electronic
transmission which ensure appropriate regulation and safeguards that enhance
public safety and engender public trust.

Thank you again for your time, and I’ll be happy to address any questions
you may have.

DR. COHN: Thank you very much.

Questions?

Michael.

DR. FITZMAURICE: Well, this kind of, to me, blows things out of the water.
(Laughter). What I see is that the Department of HHS cannot go ahead with pilot
testing of real electronic prescriptions without a waiver from DEA of those
regulations for the pilots or without a change in DEA regulations. Is that a
fair interpretation?

MS. ANAGNOSTIADIS: Well, I think, for the pilots, it would be fine, and –
if that is the decision that HHS – the direction that they wanted to take.

I think our concern is if you take faxed prescriptions, years back, the DEA
did not take a position on faxed prescriptions. So the states came out, one by
one, and started saying, We allow for faxed prescriptions. All of a sudden, the
DEA came out and said, No faxed prescriptions will be allowed. So they had to
halt the activity that was going on.

So I think the point that we are trying to make is any final standards that
are put in place should incorporate the requirements of the DEA regarding the
electronic transmission of controlled substances.

DR. FITZMAURICE: So maybe I am overreacting. Maybe the fact that they are
silent on electronic prescriptions means that we can go ahead with the pilots
until they say no. Is that – (laughter).

MS. ANAGNOSTIADIS: You know, I can’t speak for the DEA –

DR. FITZMAURICE: It’s a legal question. So I’m not a lawyer and you are
probably not a lawyer.

MS. ANAGNOSTIADIS: I’m not an attorney either.

I think that if you look at the example from the Hines VA that the Hines VA
had to go to the DEA to get authorization to be allowed to do that pilot
project.

I can’t speak on behalf of the DEA, and maybe there is somebody in the room
that could address that issue a little bit better than I can.

DR. COHN: I think Karen may be trying to – I don’t think she is speaking
for the DEA –

MS. TRUDEL: No, I’m not speaking for the DEA, but I would like to point out
that we are only talking about a restriction for Schedule II substances, which
is a very small percent of the total prescriptions, so that while we could, if
we were going to test some sort of authentication or electronic signature,
possibly go to the DEA and pursue that, the other alternative is to simply run
the pilots on the –

DR. FITZMAURICE: Catch me if you can?

MS. TRUDEL: No. No, no. No. In other words, what people have been telling
us is that where they have electronic prescription programs, the Schedule II
substances just get reduced to paper and signed perfectly legally.

MR. WHITTEMORE: Mr. Chairman, one point of clarification?

DR. COHN: Sure. Why don’t you come up to the microphone and introduce
yourself and make a comment.

MR. BLAIR: Introduce yourself.

MR. WHITTEMORE: Ken Whittemore from SureScripts again.

I just want to point out, pursuant to what Karen was saying, that
controlled substances, on the whole, are only about 15 percent of all
prescriptions. So, as things are concerned today, 85 percent of the
prescriptions are not in any way encumbered by what the DEA does or what the
state does specifically related to controlled substances.

And, then, in addition, as Karen said, a small portion of overall
controlled substances are Schedule II controlled substances.

DR. COHN: Okay. I think there are people who have questions. Does Steve
want to –

DR. STEINDEL: Further point of clarification.

DR. COHN: Okay. And then Jeff – and then Jeff. Okay.

MR. BLAIR: Mine is a clarification also.

DR. STEINDEL: Karen, I don’t know if I am totally certain that we are clear
from an e-prescribing point of view from DEA on III to V that we can do it
electronically. I mean, just because they authorize it orally and by fax – and
from what I have heard with the testimony that we have today, there may be some
issues concerning doing it electronically, and we might want to just approach
DEA and get a clarification on that.

MS. TRUDEL: I am hearing in a sidebar a conversation from Lynne that the
DEA testimony that we heard last time was pretty clear on that point, that III
through V was not restricted.

MR. WHITTEMORE: I was one the people who was squirming in their seat at the
last testimony, because I don’t think the question was posed properly to Mike
Mapes, and so, after that, I talked to Annie McFall at the DEA to clarify,
because I think the impression was left that digital signatures were only going
to apply to CII’s and III’s through V’s could be done electronically without
digital signatures. That is what I thought the impression that a lot of people
in the room last time might have been left with.

So I asked him the specific question, Do you plan digital signatures to be
applied just to CII’s or to all II’s through V’s? And he said, We have always
intended that it would apply to all controlled substances, II’s, III’s, IV’s
and V’s, and so if the – I would encourage the committee to have another
interaction with the DEA to really get the final word on this.

DR. STEINDEL: To expand on that, that was – my impression was that they
were planning on digital signature for all controlled substances, but there was
a predisposition that they could be influenced on III to V, but that needed to
take place.

MS. TRUDEL: Well, obviously, we need to go back and clarify that –

DR. STEINDEL: Um-hum.

MS. TRUDEL: – but I was responding to Mike’s point, which was we can’t do
pilots at all –

DR. STEINDEL: That’s wrong.

MS. TRUDEL: – until we have something worked out, and I think my answer to
that is I think that is a bit of an overreaction.

DR. STEINDEL: Yes, I agree with that.

DR. COHN: Please introduce yourself and –

MR. BROOK: Yes, Richard Brook from ProxyMed.

Karen made the point before that – as we testified and some of the other
e-prescribing vendors – as you are currently aware, the CII’s, when they go to
do an electronic prescription, it automatically prints out.

As far as III’s, IV’s and V’s, in certain states, as we stated before, even
if it is sent electronically, there is an oral confirmation. So I think, again
– Ken and I had this conversation as well – there really needs to be some
clarification from the DEA, because all of us, for the last several years, have
been lost when it comes to certain regulations, and we never get specific
answers or we get different answers from each person we speak to at the DEA.

MR. WHITTEMORE: Actually, if I could follow up, usually, the answer is,
Wait for the NPRM. (Laughter).

MR. BLAIR: May I?

DR. COHN: Yes, Jeff, did you have a question, please?

MR. BLAIR: A question, a clarification here.

One of the problems that I think has continued to foster this ambiguity and
confusion is that not everybody is using the same definition of electronic
signature, and there also continues to be an overlap. Somebody says digital
signature, some people think of PKI as digital signatures and other people see
a distinction between PKI and digital signatures.

So when we get this clarified, at least in my mind, here is where I think
the lines are drawn, and if this is not right, then let’s get it to find some
other way, so that when we speak to DEA we really get this clear.

My thought was that electronic signatures, from the standpoint of our
committee work, was not focusing on facts. It was focusing on computer to
computer. It might include a digital signature, but, for III, IV and V,
probably wouldn’t include PKI. That is an important distinction, at least at
this time. It may later, but not at this time.

And so I think we gotta get down to the right terms, and in our
recommendation letter there, Margret made an attempt to define each of these,
and I think we need to review those definitions of terms, so that when we
discuss these things we are all talking about the same thing.

DR. COHN: Okay. I actually have a question for – since we have our two
network carriers here to assist us in all this stuff.

So let me just make sure I understand all of this, and this is actually why
I wanted Eleni to come and join us, only because I came away a little confused.

Now, first of all, obviously, I worry when our directions from the DEA come
through trying to understand testimony at NCVHS hearings. I personally would
like to see something written or some guidance or whatever, but just to make
sure I understand what you are doing in the real world versus what Eleni is
talking about, obviously – and I tend to think of the world in three buckets –
non-controlled, Schedule III through V and then Schedule II, just for
simplicity’s sake, because I think it makes some sense.

Now, non-controlled, you send things via electronic prescribing, generally
permitted by state laws and federal laws and all of this stuff. You may have
some problems with some states, but, generally, things sort of work.

Then, Schedule III through V, I think what I am hearing from you – and I
guess maybe I am reflecting my own confusion – is that there seems to be a fair
amount of confusion relating to how III through V’s are handled. So, for most
states, unless otherwise precluded, the work process is is that something gets
sent electronically, but, then, there is a follow-on call by the physician or
his designated associate or something is additionally faxed or something else,
and then II’s basically get out in paper and basically handed to the patient
and away they go. Is that sort of how it works?

MR. BROOK: Yes. This is Richard Brook. A hundred percent CII’s from the
e-prescribing systems that connect to our network automatically printed out
with the doctor’s DEA and they require a signature.

As far as III’s, IV’s and V’s are concerned, when it is sent
electronically, they get confirmation from the pharmacist calling back the
physician as far as orally accepting the prescription, even though it was sent
electronically.

MR. BLAIR: Now, when you say electronically, are you talking about a fax?

MR. BROOK: No, I am talking about both, conceivably fax, but computer to
computer as well.

MR. BLAIR: Okay.

DR. COHN: Ken.

MR. WHITTEMORE: And this is where it really starts to get confusing. I
mean, basically, for our network right now, we do not electronically transmit
controlled substances – II’s, III’s, IV’s or V’s. We force the printing out of
that. The physician can then sign it and either hand it to the patient or they
can fax that through a traditional fax to the pharmacy.

And here is one of the other areas where it has gotten confusing, because
there are really two types of faxes that exist today, and the regulations that
the boards of pharmacy adopted, for the most part, are silent as to whether
they are talking about – I think we can presume they are talking about the
traditional kind that people think of. You’ve got a machine sitting on your
desk. You put a piece of paper into it and it pops out the other end. So I call
that traditional fax, but there are computer-generated faxes. So a physician
could use a hand-held device or whatever, create a prescription. It gets sent
off to a fax server and that pops out of the fax machine at the pharmacy
without having been signed by the physician, except for electronic signature,
and that is when you get back into this grey area where the DEA hasn’t issued
guidance on what is acceptable in terms of electronic signatures.

So some types of faxes the DEA has said are okay and other types they have
said that has to wait for our NPRM.

MR. BROOK: As Ken was discussing about computer to computer, one of the
things that we do, and have done for several years, is when the prescription is
sent electronically from the physician’s office, it goes from the physician’s
office to our network and we do the same validation procedures as we would do
as a true EDI prescription, and the places where the pharmacies can’t accept it
truly electronically, we do the validation procedures as far as – not computer
from a physician’s office directly to a pharmacy, but computer to our network
to do the validation procedures and so forth.

So there’s actually a third type of fax that takes place. Again, the
handwritten fax that goes done. The second one is it goes from the physician’s
office directly to a pharmacy, and the third one is it goes through the same
procedures and the validation part, that takes place at our network, just as if
it was an EDI prescription where the individual pharmacies can’t receive
electronically.

DR. COHN: Let me just ask a followup question on this one. I think,
hopefully, this is helpful to Karen, too, because this is the confusion I had
at the last – when the DEA came, just because – trying to figure out how to
handle the III through V’s.

Now, Mike actually asked a question about getting clarification for the
pilots. I guess I am trying to think in my own mind of, given that we are
talking about foundation standards and really the industry moving forward with
certain things being piloted, I think, really, what we are talking about is
general clarification or clarification for the Part D program or something like
that, because I don’t – and I guess I am trying to think of what – of how
waivers specifically related to Schedule III through V might work, just for the
pilots. Would that be useful even? Would that change – would that be helpful to
people or is it really more the general clarification?

MR. BROOK: Again – this is Richard. I think it would be good clarification.
I mean, the fact that we would be comfortable doing – not just for Medicare
Part D, but for the other patients.

Again, we have to be careful about one of the things. We talk about
Medicare Part D. However, we need to talk about the general population that was
in the physician’s office when it comes to the standards across the board.

So to answer your question, the answer is yes, it would be helpful. I think
that it would clear up that grey area, at least even for the pilot program,
until we got – quote/unquote – the final ruling from the DEA or had some more
clarification from them.

DR. COHN: Okay. Mind if I ask one final question – and that is, I guess, a
question about the New Jersey wording in their recently approved state regs.
Does this type of wording from the state help in terms of moving us forward on
e-prescribing, and is that something that states could do as an independent act
to help move things forward?

MR. WHITTEMORE: I think the language is – Eleni has said is good. Some of
the states have taken a very hands-off approach in terms of controlled
substances, and deferring to the DEA. At least, their intent is clearly stated
here that, in theory, you can do electronic prescribing, but you have to have
the DEA’s blessing first. So as soon as the DEA issues its rules, you may
proceed. That is language that is – you know – conditionally permissive, I
guess is what you’d – you know. So I think it is a good thing.

MR. BROOK: Yes. I agree with Ken. We look at model language from different
state boards of pharmacies, because a lot of state boards say to us, Do you
have some recommendations, and we have them look at other boards of pharmacy as
well, and that is part of the way that I operate myself, as well as
SureScripts. We look for model language, like, potentially, New Jersey, that we
would suggest to the other boards of pharmacy.

DR. COHN: I apologize, but I am still just trying to get down to this.

From – I mean, both of you are obviously involved in trying to promote
e-prescribing to practitioners and pharmacies nationally. From your
perspective, is the uncertainty and the issues relating to controlled substance
prescribing is that a barrier to adoption of e-prescribing? And I would have
you ask differently for Schedule II versus III through V. I mean, is this
turning out to be a barrier or is it just something that we are having concerns
about?

MR. WHITTEMORE: I don’t know I would characterize it as a barrier, maybe
more of an irritant, because whenever the physician companies want to move
forward, they say – you know, we tell them you can go ahead and you can
transmit everything except for controlled substances, and then they were like –
Well, you know, our prescribers, as they are writing those things out, I mean,
I guess they have a general notion of what is controlled and what is not, but a
lot of them don’t think about it, and so they are going to be putting something
into their handheld device or whatever and then they are going to have to
follow – kind of divert that from the workflow and follow a different procedure
in order to make sure the patient gets that medication, and so – like I said,
more of an irritant or a headache than – I don’t really think it is – again,
because we are talking about five out of six prescriptions don’t have this
restriction, and so all of that volume can go without being encumbered.

MR. BROOK: And, again, what Ken is saying is, well, it becomes – it is just
an issue that takes place, but, again, when you go to do this and when you work
with partners, you explain to them that we don’t use this as a barrier, as Ken
said, because it is not – you know, if it was 80 percent of prescriptions, we
would have a real issue. Obviously, when it comes to pain-management
physicians, they could be a real issue there, because the majority of their
prescriptions are doing that. So, again, to what you said, it hasn’t been a
barrier, just a nuisance.

DR. COHN: Well, thank you, Judy. I’m sorry for monopolizing here.

DR. WARREN: No, actually, I wanted to follow up on the barrier notion,
because I am still struggling with state laws and what they are saying, and
when you have the code in there from Nevada’s thing, it said regardless of any
electronic signature transmission system, it needs the board of pharmacy
approval and none has been given. Does that mean for any electronic
prescribing?

MS. ANAGNOSTIADIS: Yes, that’s what they have, and the way that this
portion of the newsletter – it is talking about faxed prescription reminders
and how some fax orders come in with a rubber stamp or say signature on file,
so that the pharmacist would have to call and follow up and treat that as an
oral prescription, much like Richard just said, and then they go on to say – I
guess they were talking about, you know, the broader electronic prescribing –
that they are waiting, basically, for the DEA before they implement, I guess,
you know, broader regulations, and then I think as a separate piece, the
electronic signature transmission, I believe – and I would have to go back and
look at their entire regulation, but I believe it refers to all – transmission
of all electronic – all non-controlled – and I guess controlled they can’t
do at this point – substances.

DR. WARREN: Yes, I guess my question – let me just clarify my question and
then answer it, because I am trying to get this straight in my head.

It would appear to me that what the Nevada newsletter is saying is we are
not going to do anything on e-prescribing until DEA makes up their mind, which,
then, to me, becomes a barrier, like Simon was talking about. So, I mean, are
you seeing that in state laws, that they are waiting for this to do
e-prescribing for other than scheduled drugs?

MR. WHITTEMORE: Ken Whittemore, SureScripts again.

Not in general. I have had an interaction with Nevada and I think what they
are speaking to there is there are a few states that want to have a demo done
of an e-prescribing system before they will approve it in the state. The
feedback that I’ve got from Nevada is any physician application they want to
see it and make sure that they are comfortable with how it handles electronic
signatures, and there are a few other states that do that. So I think that is
more what they are –

MR. BROOK: And that is exactly it. Richard Brook again. Ohio, for example,
each application has to be approved. As I mentioned in my testimony once
before, Washington State has entrusted ProxyMed or SureScripts to make sure
that the e-prescribing vendors on our networks have met whatever the criteria,
because they didn’t want to speak to each individual e-prescribing application,
but Ohio still strongly about that. They actually go out and look at physically
the application that sends the prescriptions back and forth.

DR. COHN: Yes, could you introduce yourself?

MR. SCHUETH: Tony Schueth, Point of Care Partners.

A couple of comments.

First of all, what these gentlemen are saying, I guess I – I sort of do see
different state board of pharmacy rules and regulations as a huge barrier. New
Jersey was one that was pointed out today.

In my previous employment, I worked for four years working with the State
Board of Pharmacy in New Jersey before the rule was actually approved, and so,
therefore, you know, things are really slowed down in the State of New Jersey
as a result of that.

The other comment that I want to make is to what Richard – the question you
had asked before about the barrier that Richard – I’m sorry. I’m losing my
train of thought here. That you guys had asked – (laughter). I’m sorry. Lack of
sleep last night. You know what, I’ll come back. (Laughter).

DR. COHN: That’s fine –

DR. WARREN: Can I just summarize? So you see that, from a state law
perspective, for e-prescribing we are in pretty good shape. The DEA is an
irritant, in that states are holding out to see what they are going to come out
whenever they file their NPRM, but it shouldn’t hold back them working with
e-prescribing.

MR. WHITTEMORE: That is correct, and I think Tony is right that the various
state regulations have been somewhat of a barrier, but I think both of our
organizations have gotten far enough down the road working state-by-state
through those processes – I mean, we now consider 37 states good for
SureScripts’ partners to move forward in, and so if you had talked to us two
years ago, we’d have said, Yes, this is a huge barrier, but, over time, we have
dealt with it.

DR. COHN: Harry.

MR. REYNOLDS: Yes, a couple of questions, and any of you can answer this.

Not being a doctor, what percent of the medications are non-controlled?

MS. ANAGNOSTIADIS: The number I have heard is 85 percent.

MR. REYNOLDS: Eighty-five percent – Let me go the other way. Just let me go
the other way. CII’s.

MR. WHITTEMORE: Fifteen percent of all prescriptions are controlled. Of
that 15 percent, another two to three percent – well, I am not stating that
rightly. Two to three percent of the total are CII’s – okay? – of all
prescriptions. CII’s is a subset of all controlled substances, and that is 15
percent of the total of prescriptions.

MR. BLAIR: Oh. Oh. That clarifies a lot.

MR. REYNOLDS: Okay. That’s – Okay. That’s getting clear now, and so
non-controlled –

MS. ANAGNOSTIADIS: Is 85.

MR. REYNOLDS: – is 85.

DR. COHN: Right.

MR. REYNOLDS: So – so –

DR. COHN: Yes.

MR. REYNOLDS: Okay. That’s different than I thought.

DR. COHN: Yes, so you got 85, 13 for Schedule V, IV’s and III’s and
Schedule II is around two percent, which is consistent.

MR. BLAIR: So when you kept referring to the 15 percent before, you weren’t
saying 15 percent of the total. You were saying 15 percent of the controlled
substances were Schedule II.

DR. COHN: No. No.

MR. WHITTEMORE: Well, actually, if you do the math, it works out to about
that.

MR. BLAIR: Well, just a minute.

MR. REYNOLDS: Yes, it’s pretty close, yes.

MR. BLAIR: What percent of the total is Schedule II?

MR. WHITTEMORE: Two to three percent.

MR. BLAIR: Two to three percent. Okay. And what percent of the total is II
through V?

MR. WHITTEMORE: Fifteen percent.

DR. COHN: Which is why I have been trying to put them in different buckets.
Remember, I have practiced for many years and written many triplicates and
other things over the years. So – I mean, I would sort of comment that, at
least in the state that I practice, somehow things didn’t work exactly the way
that the DEA always prescribed them, but –

MR. REYNOLDS: Okay. Thank you for the clarification – (laughter) –

MR. SCHUETH: I’d like to go back to –

DR. COHN: You had a question?

MR. REYNOLDS: Yes, a couple of questions.

So, Eleni, when you talk about the state laws, it appears that, obviously,
everybody follows the DEA, and there’s a few states who – there are other
states who say no e-prescribing unless the DEA says so.

MS. ANAGNOSTIADIS: Right.

MR. REYNOLDS: Okay. Then you have only a couple of other – this is a great
chart – only a couple of other things where somebody else kinda – well,
actually, one – where somebody said, keep them a little longer.

MS. ANAGNOSTIADIS: Right.

MR. REYNOLDS: Okay. So with that as a framework, and I guess what I – so,
minimally, for Part D, we are talking about 85 percent of all medications can
be dealt with in pilots – minimally. If the DEA only stays with CII’s, we are
talking about two-to-three percent impact. If they went as far as three to
five, you are still talking about only a 15-percent overall impact.

So the reason for – so when people talk about preemption, one of the pluses
of preemption, I would assume, is that once everything is set up it is set up
in a way and states can’t necessarily come in and then keep messing with it,
philosophically. In other words, you got a structure. So if DEA says this and
you’ve got a structure and then it continues to be that way, so everybody kind
of marches after the DEA, rather than states going out and saying, Well, I want
this a little different and this a little different. So is that a fair
statement?

MS. ANAGNOSTIADIS: Right. The DEA would create a federal regulation that
all the states would follow. If there is not preemption in place, the states
could make regulations that are more stringent. Most likely they won’t, but
they usually defer to what the DEA has to say or what the DEA regulation is.

MR. REYNOLDS: And in my final comments, I agree with you. I think that any
communication we do should clearly state – in other words, controlled substance
here kept flipping back and forth. I couldn’t tell which – so I like your
buckets a lot, and I think any communication we make we may want to make sure
that we use those clear buckets on what we are saying or not saying or what is
or isn’t effected. Kind of back to Michael’s point. I think it is a lot more
helpful, because controlled substance, everybody in this room, if we took them
in one at a time, it means something a little bit different, and I know I just
got mine cleared up, so I thank you for that. Thanks.

DR. COHN: Did you have a comment? And then we’ll try to begin to wrap up,
so we can give people some lunch, but –

MR. SCHUETH: My comment and the reason that I came up here was because I
have a slightly different experience or perspective than Ken and Richard on the
subject of does the controlled substance – a Schedule II, is that a barrier to
prescribing, and, you know, they both just testified it is a nuisance, it is an
irritant, and I would argue that there is a certain percentage of doctors that
start using electronic prescribing solutions and then stop for one reason or
another, and one of the reasons that I have heard from doctors in the field, in
the trenches, one of the reasons that I have heard is because – you know, I’m
going and I’m writing all these scripts, and then when I get to a Schedule II,
I have to set it down and I have to go and I have to – you know – use another
process, a paper process.

I guess it depends on how you define barrier, but from my definition of
barrier, I would argue that it is – I would argue that it is stronger than an
irritant or a nuisance. I would argue that it certainly is a barrier to some
doctors who expect that the prescription – that the electronic prescribing
solution is going to be just as easy as the prescription pad, and there’s a
percentage of doctors – and those folks – guess what – you know, they are not
part of ProxyMed today, and they are not working with the SureScripts folks or
the companies that are connected to SureScripts because when they started using
Dr. First and when they started using ZIX(?) and they came to that one barrier,
then they set it down, and, now, it is an irritant for those folks that are
using SureScripts and Dr. First and Web MD. They are past that, because they
have already dropped off the plane, because when they encountered that barrier,
they said, I don’t want to use this anymore. So I would argue that it is a
barrier.

MS. ANAGNOSTIADIS: I think one area you would see that in the dental
profession, because a lot of dentists tend to write for antibiotics and
controlled substances, you know, for – and so you may see that, and I have
heard similar comments from some dentists that I know when – just having
discussions about electronic prescribing where that issue has come up.

DR. COHN: Yes, well, certainly, every healthcare specialty is different in
its prescribing patterns, and I think that is what you are reflecting, and if
you are a pain-medicine specialist or an oncologist, obviously, you have a
different perspective on these things.

You know, first, I want to thank our – I mean, our sort of spontaneously
created panel of experts here. (Laughter). Actually, I think it is one of the
great strengths of these hearings is that we can really pull on experts to sort
of get clarification and verification as we go forward.

You know, I don’t want to come to a conclusion on this one, but this was a
very important piece here. As I said, I am hearing more and more that we are
all comfortable with this sort of putting things in separate buckets – you
know, non-controlled, Schedule V through III’s and Schedule II’s – and it may
be that we come up with different recommendations relating to them.

I mean, certainly – and I don’t know what exactly it is, I mean,
recognizing that for anything that is controlled it is the DEA that is, at the
end of the day, doing a lot here, and, clearly, part of what I wanted Eleni to
talk about was to make sure that there weren’t additional significant state
barriers that even if the DEA got everything clarified, things still couldn’t
work, and I think what we are hearing is a report that we’ll know that, really,
people are pretty much deferring to the DEA, so that is helpful, but, clearly,
there’s probably opportunities for collaborative work between HHS and the DEA
in a number of these items to get further clarification either in pilots or for
Part D or whatever, and it is something we need to think about in our
recommendations sort of going forward. So I want to thank you all for helping
to frame it, which I think was very helpful for us. So thank you.

MS. ANAGNOSTIADIS: Could I just make two comments?

DR. COHN: Please –

MS. ANAGNOSTIADIS: If I can just leave you with these two thoughts. If you
enter into a dialogue with the DEA is if – they did – they had the Hines VA
pilot project, and it appeared from the presentation that we heard that they
had some very positive responses and results to that. If you could talk to the
DEA, if there is any way they can utilize the information that they learned
from that pilot project to maybe come up with some written regulations or
proposed regulations, because they have done a pilot and I think that they said
that the pilot has ended, but the Hines VA is continuing to use the electronic
transmission of prescriptions for, I believe, all controlled substances, if I
remember correctly. So if you enter into a dialogue with them, that would be
very helpful to help, you know, move this process forward.

DR. COHN: Was there a second one –

MS. ANAGNOSTIADIS: About the Hines VA, and then, also, trying to get the VA
to put some of the regulations in writing and, you know, have some proposed
regulations for public comment.

DR. COHN: Okay. Well, thank you, Eleni, and, hopefully, you’ll be sticking
around, so as we start talking about things we can get some more of your advice
and guidance.

MS. ANAGNOSTIADIS: Yes, I will be here. Thank you.

DR. COHN: Thank you.

MR. REYNOLDS: Can I make one quick comment –

DR. COHN: Please.

MR. REYNOLDS: It is my understanding, though, that the pilot that they are
doing is in a closed environment, point to point, which is not what the
outpatient industry and the commercial pharmacy industry is. Is that correct? I
believe that is what I heard.

So that will be – you know, non-repudiation works perfectly pretty much in
a point-to-point – in the land of the point-to-points. Whatever you call them,
but as we overlay it on the rest – that is where I think we are struggling with
it breaking down.

MS. ANAGNOSTIADIS: Or maybe even advising of a pilot project that isn’t
point to point, that could be –

MR. REYNOLDS: Okay. I got you. I just wanted to make sure – I thought
that’s what I understood, but I wanted to clarify –

DR. COHN: Yes.

DR. STEINDEL: Harry, you picked on some of my points that I would like –
you know, would have made concerning the pilot is that it is very closed. It
still has – it has very specific certification requirements that are handled
internally within the VA, and it is not even widely used within Hines.

DR. COHN: Sounds like we have some things to ponder over lunch. With that,
it is 12:20. Why don’t we take an hour break, and we’ll come back at 1:20, and
thank you.

(Whereupon, a luncheon recess was taken at 12:20 p.m.).

A F T E R N O O N S E S S I O N (1:16 P.M.)

Agenda Item: Use of RxNorm in the E-Prescribing
Context -Stuart Nelson, M.D.

DR. COHN: Okay. This afternoon’s session is on RxNorm and its use in the
e-prescribing context. Stuart, I think we are going to start with you, and
please go forward.

DR. NELSON: Okay. Is this on? Yes, I definitely – How is that? Is this
close enough to me? Okay. Fine.

DR. COHN: Stuart, now, you’ve gotten too far away from it.

DR. NELSON: How about if I just hold it?

You are going to get me from both sides?

I think it has been about six months –

DR. COHN: Closer.

DR. NELSON: He is triangulating me. (Laughter).

I think it’s been about six months since I talked to you about RxNorm, and
I just thought I would take the opportunity to give you a little bit of an
update and see – talk to you about where we are and where we are going with
RxNorm.

Well, first of all, one thing that became quite clear is that we needed to
develop a maintenance environment for RxNorm outside of the UMLS. There were a
number of problems that were happening inside. UMLS was never built – its
maintenance environment was never really built to create vocabularies, and
there were certain kinds of problems that were happening, some of which have
been picked up on and commented on already – and I think George may comment on
some of the other ones, including the fact that, unfortunately, codes for
RxNorm concepts were constantly being regenerated. So the codes were being
churned, instead of having permanent codes, and I am terribly embarrassed about
that, but I think what I can tell you is that with a new maintenance
environment, such will no longer be the case.

The challenge is to have a maintenance environment outside the UMLS and
still maintain currency with the UMLS and the links to the UMLS concepts that
all of the other vocabularies in UMLS provided, and that is an interesting
technical challenge, but I think we have a model for doing it.

We are starting to issue updates outside of the UMLS. We had a first
release in November of 2004, and our plan is that every time that there is a
new release of the UMLS – there has just been a 2005 AA release of the UMLS.
There will be a complete release of all of the RxNorm data, including the
vocabularies that are linked directly, the source vocabularies for RxNorm will
be in that full resynchronized release. So it will come out right after the
release of the UMLS, a full release of the RxNorm data and those source
vocabularies with links to the current – in the UMLS will be released.

Following that, until the next release, there will be interim updates that
will be issued both on a monthly basis, which is a cumulative since that last
full resynchronization release, as well as a weekly interval update will be
released, and I think that that is going to achieve an important goal towards
trying to maintain currency with what is currently available on the market.

I think that the documentation for RxNorm is available now on the RxNorm
website. The documentation does address such things as what our current
editorial policy is, our business rules for how we compare strengths of
medications and has some suggestions about how to go about implementing RxNorm
in systems. It is hard to write something that is generic enough to be useful
for everybody and specific enough to be helpful at the same time.

We are currently revising the document to reflect some of the newer things
that have happened.

The other thing that I wanted to point out is that – for those who really
want to see the model and kind of explore what is currently available in
RxNorm, there is a browser available on the RxNorm website, called RxNav, that
was built by some folks in Lister Hill Center, which allows you to see the
RxNorm names and codes, but not the proprietary drug codes that are source
vocabularies within RxNorm, so that you could actually see a lot of the things
that are relationships and look to see coverage, because I think it has been
hard for individuals to talk about what is in RxNorm without having some way of
looking at it and exploring it other than – through large files and similar
types of things.

We had a study done by Point of Care Partners and – to talk about how
compatible the standard using RxNorm would be with an NCPDP standard.

My experience with medical subject headings and Pub Med has been that,
while it is very easy to talk about as a first order of proximation, you have a
terminology model and an information model that – that they can be developed
independently.

Fundamentally, when it comes down to it, when you are really trying to make
something work and work on an efficient, regular basis, those things really are
not independent and really have to – there’s a lot of careful consideration
that has to go into making sure that they do work.

What we learned from that report, two observations, I think, were very
important. One is that the current NCPDP standard does have a character limit
which is pretty short and requires a fairly short field for expressing the
meaning of a concept, and I think that that perhaps explains why some of the
source drug vocabularies have had – in the past – have had somewhat peculiar
abbreviations at times in order to fit into this character limit.

I think – in modern systems, I think character limits like that are
probably a little antiquated, but it’s something where both the terminology
standard and the messaging standard have to work together and resolve that.

The other thing that becomes clear to me, as I have looked at it and looked
at the report from Point of Care Partners, has been that in any system that you
are going to develop there has to be some ability to handle exceptions. We were
hearing testimony this morning about the exceptions about controlled
substances, but there are other kinds of things that become exceptional that
need to be handled.

I was sitting with a group of fellow internists and we were talking about
the numbers of prescriptions that we had written over the years that had
certain kinds of very stringent requirements, the lactose-free, the
cartrazene(?)-free types of prescriptions and so forth, and most of us could
remember a handful out of – I’d say, conservatively, I wrote a
quarter-of-a-million prescriptions in my career as – when I was practicing –
but I think things like that which are beyond the scope of RxNorm to address at
the present time can be quite easily handled by exceptions, by some
exception-handling mechanism, and I suspect that no matter what is done that
there will always be some need for exception handling in the system. So I just
wanted to point that out.

As far as content of RxNorm is concerned, we have 5,000 ingredients, 9,000
brand names, 27,000 clinical drugs and 14,000 branded drugs currently
represented. So it is a pretty broad coverage at this point.

I have talked before – I mentioned before that completeness is a hard thing
to measure. It is a universe which is constantly changing.

What we are doing now with the content is now that we have a fairly secure
environment for maintenance, we are adding more and more updates from sources.
We expect to have a resynchronized release, a full release of RxNorm in
synchroneity with the 2005 AA UMLS release by the end of this month, and
continue on to do more sources.

NDC codes, I think originally, when we were including NDC codes in the
RxNorm, we didn’t think it was necessarily important to try to make sure that
we had the NDC code for the brand name, as opposed to just linking the NDC code
to the clinical drug. I think that we have been convinced by the comments that
we have received from NCPDP and the report and other things that it will be
more important to link them, where we can, to the branded drug, if at all
possible.

And, you know, the NDC codes remains a problem, in terms of what
information is available and how complete it is, and I think that that will be
a problem until the FDA resolves that situation. It is not something that the
library can resolve.

We are focusing largely on prescription drugs at present. Another thing
that – and I think we will continue to for the foreseeable future, over the
counters, and – I should say we are focusing largely on U.S. prescription
drugs. We have – in UMLS and other sources, we have lots of names of
medications, brand names from outside the United States, as well as clinical
drugs from outside the United States. We are trying to stick largely to U.S.
prescription drugs at present. I think it is going to be important, at some
point, to be able to identify which are the U.S. drugs and which are the drugs
from other countries, and I suspect that we will end up with some sort of – no
pun intended – flag associated with it.

Also, comes a question of what do we do about obsoletes. What do we do
about a drug that is withdrawn from the market or that is no longer listed by
one of our source vocabularies? We certainly do not want to lose that
information over time, and the relationships between the drug and the
ingredients and so forth. We need to know that when phenopropanalomene(?) is
withdrawn from the market, when somebody took phenopropanalomene – took a
medication containing phenopropanalomene in the past, we need to be able to
identify that.

Now, the sources have various ways of dealing with names of drugs that are
withdrawn from the market, but what we will do is we will label them – where we
know them as being obsolete, we will label them as being obsolete and identify,
as near as we can, the date that it became an obsolete, in that terminology, as
well as when we have an RxNorm-branded drug or clinical drug which no longer
has a link to anything that is active on the market, we will put up an obsolete
flag on it, so that – and use in the clinical system for transmission of
prescription data or something like that, you would not want to use it.
However, if you were doing an epidemiologic investigation, you might still want
to be able to pull that up. So I think that that is a reasonable way of
representing the senescence of some of this material.

Where are we going in the future? One of the areas that we need to address
is consistent name for the devices containing medications. The oral
contraceptive packs, the kits that have several medications involved in them
and so forth.

I did indicate the last time I testified that the library does intend to
approach this problem of providing coverage for reimbursable supplies, since
that was one of the areas that – things that are prescribed, but are not –
don’t quite fit into a drug model, but we are going to try and do coverage for
it, reimbursable supplies.

The other thing I want to point to is that we are expecting later on this
year to actually start receiving from the Food and Drug Administration the
structured product labels as they are approved for medications, as they are
released. I think it’ll be – the FDA estimates it is going to be a seven-year
process before every label, every package insert is available electronically,
but as soon as they become electronically available, they will be feeding them
to the library for distribution. As we do that, this will enable us to be more
current on newly-available drugs and we’ll also be able to link RxNorm names to
all of these structured product labels.

On a personal note, I have been doing three jobs for the past two years,
getting a little worn out from it all. I am adding a – we are adding to the NLM
staff an RxNorm specialist whose specific functions include being a liaison to
the standards development organizations, particularly HL7 and NCPDP, and also
to provide additional support to users and to help users understand and
understand what users needs are. So I am looking forward to that. I expect him
to join us in April.

We are also going to add another individual, kind of lower-level
individual, who’ll help keep track of the sources that we are using for RxNorm
and UMLS and make sure that we stay very current.

The other thing is I think we are approaching a fairly stable staff for our
systems work, which is important, because the learning curve on how to build
and maintain a vocabulary maintenance environment is considerable, and building
up a staff that is secure and stable and capable of doing the job does take a
while, but we are happy that we are able to do that, and, certainly, the amount
of funding that we have received from ARK(?) has been a great benefit in doing
a lot of this work, and I don’t think we could have done it without it.

And one last thought, Donald Lindbergh’s favorite saying about
vocabularies, they get better as they are used, and the more you use them, the
better they get. You know, I have the privilege of being responsible for the
medical subject – which has now been around for 45 years, and – yes, 45 years,
the first vocabulary developed for use in a computerized system – in medicine,
developed in 1960. So it gets better every year. Perfection, I think, is a far
way off, but things do get better as they are used over time, and I’m sure that
this will happen with RxNorm as well.

So that is all I have to say and open for questions.

DR. COHN: Well, Stuart, we’ll let everyone comment and then we’ll have sort
of a general discussion.

Now, George, are you on – George, are you talking or is it Lynne or both of
you?

MR. ROBINSON: Tag team.

DR. COHN: Tag team. Okay. So we have Lynne Gilbertson from NCPDP and George
Robinson from First DataBank. Anyone else?

MS. GILBERTSON: She’s hiding in the back. I guess not.

SPEAKER: I’ll be there for questions, if there are any.

DR. COHN: Okay.

MS. GILBERTSON: This is in your testimony, what is on the screen, but just
in case you don’t want to read the writing, it’s on the large screen.

I am Lynne Gilbertson from the National Council for Prescription Drug
Programs, once again, this time reporting on the NCPDP Task Group which was
formed under Work Group 2, Product Identification, to discuss working with the
RxNorm; and the task group prepared this testimony which I am going to go over
some parts, and then George will chime in in the real important stuff.

The purpose of the document, as presented to the committee, is to describe
the functional requirements of a drug concept interoperability vocabulary,
detail the steps required for implementation of a drug concept interoperability
vocabulary within the health information systems, assess the current
acceptability of RxNorm in supporting drug interoperability requirements,
provide recommendations to this committee regarding the adoption of RxNorm
within applicable messaging standards.

The assumptions: One of the primary assumptions is that existing local and
proprietary drug vocabularies will continue to be used within healthcare
provider and payer systems. This assumption is based on the current widespread
adoption and implementation of preexisting medication vocabularies within the
point-of-care health information system applications.

And as we also talked about earlier, that there are still functions that
the drug vocabularies will support. That is not the intent of RxNorm and should
not be considered part of that.

As drug concepts are exchanged as part of messaging transactions between
healthcare providers and/or payers, it is proposed that an RxNorm concept would
be included in addition to the drug concept code and text description native to
the center’s application. Thus, the primary purpose of the RxNorm concept is to
provide the automated translation of drug concepts between health information
systems that use disparate drug vocabularies.

The proposed MMA drug concept interoperability requirements. The ideal drug
concept interoperability vocabulary has the following characteristics:

Drug concepts are assigned permanent identifiers.

The meaning of a unique drug concept does not change over time.

Drug concepts are made publicly available on a timely basis. Publication
should mirror the market release date of the related packaged product.

Editorial policy for the naming of drug ingredients, brand name, dosage
form, strength and strength unit of measure is published and is consistently
applied.

Change history is published. For example, the obsolescence or replacements.

The comprehensiveness of the interoperable vocabulary should cover all
medications and medical supplies likely to be prescribed or ordered within the
U.S. market.

A liaison is provided to the industry to handle editorial policy questions,
provide implementation support on a timely basis and participate in related
industry workgroups.

The entity responsible for authoring interoperable vocabulary content
provides a responsive customer support service.

Cross-references between the drug concept interoperability vocabulary and
external source contributing vocabularies are maintained and published on a
timely basis, used within licensing constraints specified by source
vocabularies.

The meaning of the linked interoperability vocabulary term must be
equivalent to the meaning of the external source vocabulary term.

A governing body is established to approve editorial policy, establish use
cases, review content authoring processes and review ongoing quality
improvement procedures.

The governing body also deems the interoperable vocabulary as ready for
production release and use within health information systems.

This is a product. The industry will expect to support a product, as you
would if you were doing it from your own company. So those kind of pieces of
information are very important going forward as we trust the use of these
products.

Messaging transaction candidates. The following list of ambulatory
service-based message types are candidates for the inclusion of
interoperability drug concepts. Reference is made to the possible terminology
type that would be applicable from RxNorm or administrative code sets.

I am not going to go over the extreme list, but the task group looked at
different kinds of use cases of where you might fit an RxNorm into the business
case, which would then go into the messaging standard. Some of these – for
example, prescriber inquiry into the prescription drug plan for the formulary
to determine eligibility, and the recommendation sub-bullet under there is it
should support the brand name, the ingredient, the semantic drug brand and the
semantic clinical drug or NDC.

So there is a list of use cases having to do with prior authorization, with
formulary, with drug history, with prescription fills, the Rx fills, with
discharge medications, all the different kinds of use cases that it is possible
to make sure that we have the correct use of the RxNorm concept built into that
standard or the knowledge of how to use it.

The task group recommended some steps required for industry implementation
of RxNorm. A governing body deems the interoperability vocabulary as ready for
production release and use within the systems.

The industry becomes convinced that a demonstrable business need is present
for the inclusion of RxNorm within that applicable messaging standard.
Widespread implementation will be stymied unless vendors and self-programming
entities perceive that the integration of the interoperable drug vocabulary
will provide value or an adequate return on investment in exchange for the
associated development costs.

RxNorm is published on a regular basis using permanent identifiers.
Incremental updates are provided at an interval no greater than weekly.
Medications likely to be prescribed must be coded to semantic clinical drug
concepts; associated semantic brand drug concepts are required for innovator
products.

RxNorm cross-references to UMLS contributing vocabularies are published on
timely basis.

An RxNorm liaison is made available to the industry to assist with
implementation activities and to resolve outstanding content issues.

Messaging standards are updated as needed to accommodate the inclusion of
RxNorm concepts. Related messaging standard implementation guides are written
and provided to the industry.

Drug knowledge base vendors and entities using in-house drug vocabularies
import RxNorm into the database files. As needed, additional cross-reference
work occurs to ensure that RxNorm ingredient, brand name, semantic clinical
drug and semantic brand drug concepts are linked to analogous concepts.
Integrated RxNorm files are then made available to their customers.

The industry becomes comfortable with the use of RxNorm cross-reference
files and concerns regarding potential liability are eased.

Prescriber, pharmacy, prescription drug plan, long-term care systems and
other ancillary systems are modified to import the RxNorm concepts. Systems are
further modified to send and interpret RxNorm values within applicable
electronic transactions.

Note that, typically, a significant time lag between the development of
updated health information systems and the subsequent implementation within
end-user systems occurs. This time lag can range from months to years,
depending upon the number of production versions supported by the vendor or
self-programming entities.

I am now going to turn it over to George, who is going to do some detailed
discussion about some of the effort that has been done in the concepts and
mapping and cross referencing and all that.

MR. ROBINSON: I am George Robinson with First DataBank. I have been working
with an NCPDP work group that has been discussing issues related to RxNorm,
and, you know, the time is good to have Dr. Nelson here, in that some of the
comments and concerns that we had, I think, are ameliorated with what I see
happening.

From our perspective, one issue that we see is no clear owner of RxNorm as
far as stepping forward to clearly state what the business requirements are
intended for support of RxNorm.

What we have learned through experience is there are subtle differences
between the way drugs are ordered from a prescriber perspective in an
in-patient hospital scenario or work environment versus the way drugs are
ordered by prescribers in an out-patient environment and that the level of
detail – at least in dosage form – is made much clearer in an ambulatory
environment.

Secondly, it is one thing to describe a drug in the context of how a
prescriber perhaps is ordering it. It is another to represent how a pharmacy is
representing, in terms other than an NDC, a drug that they actually fulfilled
or gave to a patient. So we’ll go into a little bit of detail on that.

Without some clear business requirements from some entity that is
credentialed or empowered to really say what RxNorm needs to be, and a plan for
acceptance testing, it is really difficult for the industry to say, with
conviction, that the design of RxNorm is adequate to meet all the range of
needs that we are discussing.

The business need for mandatory inclusion of RxNorm concepts within
applicable messaging standards is really yet to be established or yet to be
convinced to the industry that it is needed.

In the absence of having an acceptable interoperability standard in many
applications, of course, the NDC, the Universal Product Code and health-related
identifiers are used, at least as a work-around, certainly in billing
transactions of which a known packaged product was used was actually dispensed
to a patient, that makes sense.

Certainly, some of the early testimony that I saw where we are talking
about an exchange of patient drug history, where it is coming from a
pharmacy-benefit organization back to the prescriber again, that is using the
NDC. So – now, this was written before the standards were actually released in
electronic prescribing, and I think there was a similar approach taken in the
authors of that document that we should proceed with electronic prescribing
without having RxNorm be a dependency at this point. So looks like – the point
of that paragraph was that we are already making inroads and progress in
electronic prescribing without an interoperability standard present. Let’s
continue with that momentum and then bring RxNorm in at the appropriate time.

As Dr. Nelson related earlier, there have been some issues with RxNorm
concepts being assigned permanent identifiers. The relevance of that to the
industry is that concept-unique identifiers, fondly known as CUI’s, are
assigned to concepts only through the quarterly release of UMLS. Well, because
we have a business need to describe drug concepts on a more frequent basis, as
an interim, what we will be cross referencing to is the RxNorm code itself,
which eventually is linked up to a – identifier which assigns a level of
permanence within UMLS.

What was frustrating is that some of the early cross-reference work that
the industry was doing had to be repeated every time a new release of RxNorm
occurred, because those numbers are regenerated. So it is nice to hear that
that is going to be addressed and taken care of.

Timely publication of RxNorm. What we have heard is a process that really
looks like it will be effective and responsive. I think it is in the early
stages of implementation. So I think we are in a wait-and-see perspective to
see if that actually occurs. What we have heard is that, ultimately, it will be
published on a weekly basis for update between cycles, and that is very good
news.

Also very good news is the appointment of an RxNorm liaison. As more and
more people hear of RxNorm, there are a lot of questions with it, and having a
knowledgeable source at the Library of Medicine, other than just Stuart, would
be – will be nice, so we can get the word out and have proper implementations.

One critical issue that we have is that, on the whole, drug-knowledge-based
vendors describe drugs in terms of pharmaceutical equivalents, and what I mean
by that it’s a distinct combination of ingredients, plus there are salts, a
dispense-dosage form, strengths, and a primary route of administration.

The functional purpose, historically, for drug-knowledge-based vendors in
publishing these types of identifiers has been to group like drugs more from a
– in some cases an Orange-Book substitutability, as far as candidates or
substitution.

The roots of RxNorm and its creation, really, I believe, go back to input
from HL7, where, from a hospital environment, when you are talking about a drug
from a clinical perspective, many times, the salt is not included when the
doctor is ordering a drug entity, unless, of course, the salt is clinically
significant.

The subtle difference is that with that approach taken in creating RxNorm,
there are times when a single RxNorm concept will span multiple pharmaceutical
equivalence concepts.

The significance is if I am trying to represent with a code something I
dispensed or a very specific representation of ingredients related to a branded
formulation, I am losing some of that knowledge when I go to some of the RxNorm
codes that are presented today, and we’ve got a few examples here, just to talk
about some of the subtle differences.

For example, one thing RxNorm does is it attempts to normalize the strength
and strength unit of measure to the administratable or the administrated form
of a drug.

In Amoxicillin, 50 milligrams per ML, what we see is most of the
drug-knowledge-based vendors would differentiate between a formulation in which
it is 250 milligrams for Fidenal(?), which is related to a larger bulk
container of 150 ML’s. For example, this would be intended as a suspension of
Amoxicillin more for pediatric use versus the true 50 milligram per ML is
normally related – well, it is related to products that – it is dispensed as a
dropper. So that is intended more for infant use.

You know, when you look at the drug, it is a subtle difference, but it can
differentiate between physician intent that, Hey, I want this drug dispensed,
and I want it in the dropper form, because it is a very young child and I get
greater accuracy in the dosing delivery, versus, No, this is for an older child
and it’s okay for them to have a teaspoonful, for example, per dose.

Another subtle difference that we have picked up deals with the salts, and
this is where I think if we were able to work with a liaison and work through
case-by-case issues that perhaps the right thing to do with RxNorm would be add
salts where we need to.

The relevance of Erythromycin is the fact that there is estolate, stearate
and base that is defined. At least with the estolate, I believe, there are some
hepatic side effects that, if you are trying to aggregate clinical knowledge
against one concept or another, you would want to differentiate between
Erythromycin estolate and Erythromycin stearate or the base.

So in this table, what it is illustrating is that, in comparison to the
drug-knowledge-based vendors, including the VA, is typically these formulations
of Erythromycin are described at a greater level of specificity that includes
the salt. So the relevance, again, is if we look at RxNorm as a translation
vehicle, what would happen is if I wrote an order specifically for Erythromycin
base to 50 milligrams, that would be translated to Erythromycin 250 milligrams
without a salt specified. Presented back to the pharmacy as candidates for
dispensing would the base, the stearate and the estolate.

So what we need to work through, again, is how interoperability
application, the intent of the prescriber is not lost through the translation.

Now, one way that is expressed with greater specificity is if the
prescriber is writing the order using a brand. So if they write – I don’t even
know if Ilastone(?) is still made, but if it was Ilastone 250-milligram oral
capsules, we know that that resolves itself to a more specific formulation, and
if we are able to do the interoperability, exchange it more a brand name, then
I think less is lost of the prescriber’s intent.

But, again, this is an example of a situation where I am very comfortable
that if we had a working relationship with a liaison from RxNorm that we could
work through issues like this, and the right thing to do would be, perhaps, add
greater specificity in formulation descriptions within RxNorm, as needed.

The last example was going through an example where if you look at the
difference between Diprazone(?) and Dipralene(?), Dipralene is delivered in a
formulation in which what is considered an inactive ingredient, being the
vehicle, which I think is white petrolatum, it actually increases the potency
of the drug. So, although Dipralene and Diprazone span the same RxNorm concept,
they really have very different levels of potency. So, again, if an order was
expressed as Dipralene and then was converted to the RxNorm concept of
Betamethasone, the way it is presented is that when it is – again, when it is
presented to the pharmacy for dispensing, there’s the loss of knowledge of the
specificity of that order, of knowing what exactly should be filled.

Another issue which Stuart addressed which would be very welcome to those
of us that have the task of publishing cross references would be having
semantic brand drug links to the NDC. As he mentioned today, the only links we
have are at this generic clinical drug level. So, in the example I mentioned,
all forms – well, Filoxatine(?) 20 milligram oral capsules would link to all
indices, no matter if they were branded Prozac or not, versus what would be
most useful to us would – the knowledge of knowing that if Prozac 20 milligram,
all capsules, is actually a semantic brand drug concept in UMLS, that it would
have a direct link to indices that are actually, indeed, labeled Prozac.

The importance of that to the drug-knowledge-based vendors is, with that
solid name base linked to indices, we can also do joins to our name-based
identifiers, and we can bring to the market a cross reference at a much faster
pace.

So, with that, Lynne, has some closing thoughts, and then we can entertain
some questions.

MS. GILBERTSON: Recommendations. Once again, a governing body should be
established to define RxNorm requirements, certify acceptance testing,
production processes and editorial policy.

Once industry participants have completed the analysis necessary to
establish proper usage requirements from the drug databases to the RxNorm
concepts for the appropriate use cases, it is expected that a submission for
the appropriate RxNorm concepts will be included with the appropriate NCPDP
messaging standards.

We recommend that RxNorm continue to be funded, developed and tested. As
RxNorm matures and interoperability business needs become more apparent, we
believe that business partners will eventually call for the addition of RxNorm
concepts within the messaging standards on a development time-line independent
of MMA.

Thank you.

DR. COHN: Thank you.

Questions or comments?

Jeff, you ready?

MR. BLAIR: Yes.

DR. COHN: Okay.

MR. BLAIR: Stuart, Lynne, George, thank you.

I agreed with the articulation of the outline, Lynne, that you presented.

When it got to the business cases, I started to get a little nervous. Not
that – I mean, I very much agree that there is a need to establish the business
cases, but I worry a little bit, because when we deal with business case for
something like a clinically-specific terminology like RxNorm and we know that a
lot of the benefits are in clinical-decision support and patient safety, where
avoiding an adverse-drug event, the benefits there may accrue to avoiding a
hospital stay. So I got nervous that when the vendors are looking at it for a
business-case standing, that benefits like that and the cost savings of adverse
drug events might fall outside the scope of a business-case evaluation of the
benefits of RxNorm.

MR. ROBINSON: This is George Robinson. Can I comment on that?

What RxNorm will provide is instead of a pair of eyes reading a piece of
text, which, at the very least, would be not handwritten, that pair of eyes is
translating that text into a codified term that is compatible with their
information system. So I don’t see that all those objectives that you describe
are not contingent on RxNorm or an interoperability standard.

What is contingent is that, within the application that is used in
different roles of healthcare, that there is timely access to decision support,
relevant information related to the drug. That is available today, based on
several providers or homegrown information that is actually maintained in
several institutions.

So that is why, when we mentioned the business case, what we have to
convince the industry is is the convenience of actually having an automated
translation between that term that is coming in the prescription to my system
worth the investment that I have to do to make it happen, and, certainly, if we
know that it is trusted, there’s all kinds of automation that we could
anticipate that could be supported with that. So that would really make a nice
business case.

MR. BLAIR: Is it a business case that you feel can be established readily
within this next year or two?

MR. ROBINSON: Personally, what I am looking for is this governing body.
Now, is the governing body CMS?

MR. BLAIR: Um-hum.

MR. ROBINSON: And then the context of MMA comes forward and says, RxNorm
will be successful if it meets these conditions in a) the transmission of a
prescription to a pharmacy, b) the request of formulary information from a PDP
to the prescriber.

It is just that what we have are several different application. Each have
subtle differences in the level of drug extraction that needs to be exchanged,
and also a difference in the information that is expected to come back with it,
and so you can’t look at just one narrow application, I believe, within MMA and
say, If we satisfy this one need, then we know that RxNorm is acceptable and
that life is good. We have to look at it from each individual message that I
believe that we need to work on, as far as defining the requirements and say,
Does RxNorm meet this need? And, ultimately, at the end of the day, what is the
measure of its acceptability?

DR. COHN: Okay. Stan and then Mike.

DR. HUFF: I think the point about an editorial board or somebody who is the
owner of this is very good, and I guess the – I mean, another alternative is,
in fact, to have the owner, if you will, be, you know, one of the standards
bodies.

And, at this point, I disclose, again, that I have – you know, I am tied to
HL7 as the Vocabulary Co-Chair, and so we are not voting on anything, but be
aware of that potential conflict of interest, since I have some – don’t have
any financial interest in that, but, obviously, I am a participant.

So, in terms of the use case, I mean, I think the – I had some of the same
thoughts – maybe I would express them a little different than Jeff.
Representing the provider organization, you know, our use case extends beyond
what you said. We are trying to enable the possibility for us to interoperate
with lots of different drug vendors, drug-knowledge-based vendors and make that
an easy transition, so that there is open competition, and that may work
directly against people who have big market share now, and so there are
different motivations, basically, for this, and I think, in NCPDP, in
particular, and even in HL7, the providers are under represented. You have
vendors and you have pharmacists and you have the drug-knowledge-based vendors
and you have the pharmaceutical company, all of which are good folks, but they
have different motivation than the provider organizations like IHC and other
people like that, and so, I mean, as I came to this meeting, Keith Larson, who
is over our drug stuff, said, you know, Tell them to do what they need to do to
get this done, because I want it. He doesn’t need to be convinced of the use
case for this. He wants it today.

The other kind of thing that we are trying to enable is, in fact, that
there would be open collaboration in development of knowledge bases around this
standard coding system. I mean, there could be all kinds of neat whiz-bang
things that graduate students and others in informatics would develop that
could be shared, you know, openly, if there were this standard open
terminology.

So I think – but all of those things, I think, do come back to who is the
owner and how do we establish that those are the use cases, and I don’t know
how that works with Stuart. I mean, in a sense, we are sort of like trying to
vote the new boss for the NLM or for this particular activity, and – you know.
I mean, it is one thing – we can say what we want. If they are unwilling to be
owned – (laughter). I’m not sure – you know.

So I guess I would ask maybe the question of Stuart is there – are there
other kinds of patterns for this, for instance, in the maintenance of MESH(?)
and other things, where there are citizen panels or editorial boards or
something that get involved in your work and how would you see this happening?

DR. COHN: And, Stan, maybe I can even enlarge that before you answer,
Stuart, that, actually, a number of other models exist for public vocabularies.
We have one at the table. The National Center for Health Statistics maintains
ICD, for example, and CMS maintains other terminologies that are owned by the –
you know, in the public sector. So, Stuart, I just wanted to elevate the
conversation a little bit.

DR. NELSON: Right.

I wanted to observe that I think it is one thing to talk about using a
consensus-based approach to developing standards and policy, but I don’t think
it is actually feasible for day-to-day operations about making decisions about
things to follow that, and so I think if you talk about, well, are we willing
to have a group of individuals – now, before I go too far, you know that NIH
Advisory Committees require all sorts of levels of approval and so forth. So I
don’t think that we necessarily can say, well, we can constitute a board of
individuals to serve to advise us, but, you know, it is quite possible to
develop some voluntary organization which can contribute to helping formulate
the policy and carrying it through. So I think that that is quite possible, but
whether to go – we could go through a process of formally constituting an
outside advisory board would be something that is beyond my level to discuss
and would have to go to the director or the deputy director to discuss, and I
certainly will discuss it with her tomorrow morning.

Yes, I wanted to make – so I would say that we are open to hearing about
how to do that and try to find a mechanism, certainly.

You know, I have expressed to NCPDP that, you know, I am certainly very
willing, and I certainly want to – I am looking forward to the arrival of some
help to listen to any and all suggestions about our policies, given that what
we are trying to do is trying to address this problem of interoperability, and
the only way we can address it is to understand what people’s needs really are.

The whole RxNorm project came out of meetings at HL7 where people started
discussing some of the difficulties that they had, and, specifically, what they
were discussing was the difficulties that people had in comparison of First
DataBank names to Medispan names to Multim(?) names, and the fact that they
could not interoperate between them, because each of them does take a slightly
different cut in terms of this whole idea of pharmaceutical equivalence about
representation of salts, and so forth.

So when you recognize that each of them has done it a little differently,
you know, one of the things that we were trying to do was to try to find that
level of interoperability between the different ones.

One other comment that I wanted to make, George, was about the
Erythromycin, where our model in RxNorm actually does call for the expression
of this precise ingredient, which includes the salt, where it is clinically
significant, and I would say – speaking frankly – looking at the Erythromycin,
that somebody dropped the ball on that one and it will be corrected, because I
also believe that Erythromycin, it really does make a difference what the salt
is; but, on the other hand, I don’t think with Amoxicillin it is terribly
important what the salt is.

So we are concerned about that, and I think it is very important that we
try to address those on a regular basis. There’s kind of this low-level quality
assurance. Yes, we made a mistake on Erythromycin, and then there’s the higher
level of do we go out and try to express every drug according to the particular
salt that is associated with it, you know, and how do we deal with things like
the augmented delivery of Dipralene versus Diprazone?

Interestingly enough, I am just back from Vietnam, and before I went, my
internist prescribed me some Dipralene to take with me. So I took it to my
local pharmacy and got generic Betamethasone ointment. (Laughter). So so much
for a pharmaceutical equivalent. I think George would argue that I didn’t get
the pharmaceutical equivalent from the local drugstore, but pharmaceutical
equivalence is a difficult thing to do.

I have taken a look at the Orange Book, where they publish the material
about it. Currently, I don’t see a way of incorporating the information in it
electronically without a great deal of manual labor. So, Mike, if you want to
send me some more money – (laughter).

DR. HUFF: Still my turn?

DR. COHN: I think, Stan, it is still your turn.

DR. HUFF: A couple of other questions.

One of the significant questions, I think, has been – and, again, this
would come back to how we could do this with a liaison or some input – but
whether you see it within the scope of your current work activity, in fact, to
be pharmacy supplies, if you will, as opposed to purely drugs. You got a work
plan for including band-aides and gauze and tape and –

DR. NELSON: We are planning it right now, how to include – the question is
where do you draw the line there, and one thing might be a list of written
Medicare-reimbursable supplies, because I don’t necessarily want to have
band-aides, but I do think that you see a fair number of prescriptions for
insulin syringes, for glucose test strips, and there are an infinite variety of
manifestations these days for those types of supplies.

One of the interesting supplies at the VA is a pill cutter. It rates high
up there in their prescription list.

DR. HUFF: My final question, you know, implied by the recommendations from
NCPDP and the testimony, may, in fact, be some things that are only solved by
more people looking at things, you know, and you mentioned that it might – Are
you sufficiently staffed and funded to do all of the things that are being
asked?

DR. NELSON: I think, at the present time, we are sufficiently staffed and
funded. I would like to fill those – you know, I am looking forward to the
arrival of the RxNorm specialist. I think that having somebody else along on
board tracking sources, I think we are doing reasonably well, in terms of
staffing, yes.

Some of this is just a matter of time to make it all work, but I think
where we are right now is we are looking at a system that will avoid the
churning of codes that can still maintain synchronization with the UMLS – but I
am going to recommend that for people who are going to be using RxNorm that
they use the RxNorm codes because they will be current and available on a
regular basis and not dependent upon the quarterly updates of UMLS.

DR. COHN: Okay. Michael.

DR. FITZMAURICE: As Stuart alluded, in ARK’s budget is a patient safety
health and information technology component. We have spent – oh, over $50
million on more than 100 grants and five state contracts, and a prospect of
potentially more.

As part of that program, there’s $10 million for a data-standards program,
which I was tapped to head up, and I have spent – this was in ‘04, and the
money will be continued in ‘05, and I spent about 70 percent of it on
reducing adverse drug events, specifically to FDA and to National Library of
Medicine, to help supply the structured product labeling, a list of all the
products that have been approved by FDA – drugs, non-drugs, and so forth – and
to NLM for RxNorm and for conducting a bunch of mapping, which I think will be
extremely useful.

I am pleased with the progress that NLM has made with RxNorm and it is
pulling together.

I am a little troubled that it is not meeting all of the needs of the
industry, and so I think there needs to be some meeting of the minds, some
adjustment, and I hear Stuart saying, Well, we’ll change the Erythromycin and
we’ll get a liaison and we’ll interact more with the industry. I think that is
a good and a positive approach.

I want to continue supporting NLM in development of RxNorm and the use of
RxNorm.

So I guess some questions I have are, Stuart – maybe also from Lynne and
George – what changes do you see are needed in RxNorm? And I think you have
laid them out very well in the paper. Is it hard to make these changes, and can
RxNorm be piloted to find out its strengths and actual use? That is, in the
pilots that would start January 1st, I realize this would probably
need a private-sector partner or two, someone who is saying, I’ll submit a
proposal to do this, and I’ll propose to use RxNorm as my vocabulary, and we’ll
see how far we get. Are these things possible? I’m looking to get something
that is useful clinically and also useful for the business use.

I think clinically is essential, because we want to reduce adverse drug
events.

The business uses, we want it to be used, and often this information will
flow through the same kinds of networks that claims information flows right
now.

DR. NELSON: I expect by this summer that you will be seeing the RxNorm
updates coming out weekly, and I think it will be essentially with all the
current source vocabularies, so I would say, in terms of the clinical drugs, by
January 1, there should be a pretty complete representation.

Whether it fulfills every need in representing what is dispensed, as
opposed to what is represented and what is prescribed, I cannot – I’ll offer no
guarantees on that, because I don’t think I know enough about dispensing, but I
think, in terms of at the clinical level of a provider, I think it will be very
expressive.

So I think that it would be – and, you know, one of the things is when
people use things, then you find out where you need to make the changes, and so
part of it is that process of getting out there and doing pilots and finding
out what needs to be changed, but I think the processes will be – as far as the
other recommendations, I think, essentially, other than the editorial board,
which is a policy thing that I can’t really speak to, I think most of the other
things we are already working to address, so I think that they should be there.

DR. FITZMAURICE: I want to say that I also applaud your getting a liaison,
hiring a liaison. I talked with Betsy about getting another person – and I am
convinced now that it is essential – that is the liaison with the community
that would be using RxNorm that needs to provide that feedback, and it is
difficult for us to do it in the government without these sufficient ties, and
I think getting a person is going to help with those ties.

DR. NELSON: Yes.

DR. FITZMAURICE: So I applaud the work of NLM in this.

DR. NELSON: Thank you.

DR. COHN: Harry.

MR. REYNOLDS: Yes, I have started my first novel ever, Called RxNorm for
Dummies. So – (laughter) – you are in the medical profession, so you are going
to have to help me for a minute here – (laughter). Got a couple of questions,
and I may have to recuse myself, because I don’t understand if I don’t ask a
couple – (laughter).

MS. FRIEDMAN(?): That is not allowed.

MR. REYNOLDS: I know.

What is a concept?

SPEAKER: Whoa.

MR. REYNOLDS: No, I mean, just a quick definition –

DR. NELSON: Without going into a deep philosophical –

MR. REYNOLDS: Yes, if you go deep, you’ll lose me again. Keep –

DR. NELSON: You know, I would love to do that, but maybe afterwards over a
beer. (Laughter).

Essentially, a concept is representation of when more than one thing means
the same thing at a given level of abstraction. So you can name it in a whole
variety of ways and say, yes, at this level of abstraction, these things all
mean the same, and so all those names constitute the concept.

Now, what we have found in doing UMLS work is that meanings of words change
over time. So, ordinarily, we say that the meaning resides in the identifier
that we give that concept, and we put the names that mean that in that concept
class.

MS. FRIEDMAN: Give an example.

DR. NELSON: Yes, can you go back and put up your Amoxicillin example?

DR. FITZMAURICE: Stuart, I wonder if I could try a simple example, because
I started off at the kindergarten level and maybe I have moved into the first
grade.

DR. NELSON: Okay.

DR. FITZMAURICE: If you think of the concept of –

DR. NELSON: Now, if you scroll over to the right, so you can see the
generic descriptions here, we have all of these different names. Okay.
Amoxicillin trihydrate, 250 milligram per five ML oral suspension
reconstituted. Oral, parens, milliliter. Amoxicillin trihydrate, 50 milligram
per ML, oral drops, reconstituted, oral. All of these things, at the level of
abstraction of an active ingredient, strength is expressed in concentration,
and form mean the same thing.

MR. REYNOLDS: That’s good. So it’s active ingredient, strength and format.
That’s good enough for me for right now.

DR. NELSON: Okay.

MR. REYNOLDS: This gives me some kind of sense.

Now, my next question. So if you take the numbers that – you take NDC,
which I am familiar with, then you take what First DataBank calls something and
Medispan and Multim and the VA, and we are saying that RxNorm is the one that
links those together at that concept level, correct or not?

DR. NELSON: RxNorm tries to express them all at a concept of that, which I
just described to you, which is a clinical drug, or the branded drug, at that
level of abstraction, and NDC code represents a much finer level of granularity
of a product, a manufacturer, a labeler and a packager, size of package in that
code.

MR. REYNOLDS: Okay.

DR. NELSON: Okay. So that an NDC code, there might be hundreds of NDC codes
associated with a given drug, which makes it hard to interoperate, because
people don’t necessarily all know what those are.

MR. REYNOLDS: Okay. Thank you.

DR. COHN: Oh, I thought we had a follow-on – no.

MR. REYNOLDS: No, I’m just working on chapter one right now. (Laughter).

DR. COHN: Okay.

MR. REYNOLDS: I don’t want to get too many extra notes –

DR. COHN: Okay. I guess I would like to sort of try to put this together
and maybe – bring it to a close, unless there are others that have comments.

I think – first of all, I am sort of reflecting that, obviously, our
recommendation – and I think the plan for 2006 would hopefully be to pilot
RxNorm. Nobody is talking about implementing it, you know, industry wide as a
required foundation standard for January 1, 2006. So a lot of the things that I
think it was very useful hearing from Lynne and others – hearing from both of
you – about the issues with RxNorm – I am actually delighted to see Stuart
attempting to be responsive and I think beginning – I think we are hearing
things sort of getting closer as the process occurs, but, clearly, I don’t
think anybody is saying that this is – has been around 30 years or 45 years or
whatever, and I think that is the purpose of the pilots.

Now, hopefully, some of them may turn out to be CMS pilots. One would hope
that maybe ARK might have funding for pilots themselves to identify the – how
well this all works.

Now, Lynne, I have one question. I just couldn’t tell from all the
testimony. I am presuming that there will be from NCPDP draft standards for
trial use that enable the testing of RxNorm for the 2006 pilots, is that
correct, or is it that already you can use RxNorm in the SCRIPT standard
without any modification currently? Can you clarify that for me?

MS. GILBERTSON: As of right now, the SCRIPT standard does not support
RxNorm.

DR. COHN: Okay.

MS. GILBERTSON: It doesn’t know what it needs to do to make sure it clearly
states what concept level – what level of anything RxNorm should be used. So
that discussion will have to take place during this year to determine what is
it that clearly needs to be probably just a qualifier and a placeholder for a
code, but of what? Is it just all the standard needs to have in it is a
qualifier that says RxNorm and a field of X number of bytes that whoever fills
in however they want to or do we need to provide some level of guidance of when
you are doing a new Rx, this is the level of Rx norm concept you should use. If
it is being used for refills, this would be the level, because you might be
linking it to NDC’s. I don’t know. That whole discussion has to take place
before I can say we are going to throw something into the standard, even at a
draft, for people to use.

DR. COHN: Okay. Well, I’m glad –

MS. GILBERTSON: I mean, people can use the standard and try a pilot without
having an approved – quote/unquote – NCPDP standard. That is no problem. As we
are trying to play through things, that is just fine. Then we’ll bring it
forth, once we have an answer, could be part of how we do it, too.

DR. COHN: Okay –

MS. GILBERTSON: There’s nothing in the pilot says it has to be an absolute
approved – blah, blah, blah – standard. If two parties agree that this is how
we are going to try to do it to try to prove it works, that’s fine with me, and
then we’ll bring it forward as to what we find is the answer.

DR. COHN: Okay.

MS. GILBERTSON: Do you see where I’m –

DR. COHN: Yes. No –

MS. GILBERTSON: I’m just real confused how to give guidance other than to
say just throw in an RxNorm code field without giving any kind of guidance of,
What are you going to do with that? Where do you get it from? What level are
you trying to relate? Or maybe the standard should be completely silent. I
don’t know ‘til we have done a lot more work with, hopefully, this liaison
position to really hammer those kind of things through.

DR. COHN: Okay. Well –

MS. GILBERTSON: Does that make sense?

DR. COHN: I think Maria has a comment, then I’ll probably follow up.

Sure, but thank you for clarifying that, because I was wondering – wasn’t
sure as I was reading this if that was sort of the intent of some of the
testimony.

Maria.

MS. FRIEDMAN: Thank you both.

Lynne took me part of where I needed to go, and that’s as somebody who is
going to be involved in pilots and – you know, we are putting out this RFP for
pilots, and my presumption had been, up until about two seconds ago, that we
would all go merrily along; and, now, I hear that there’s a lot of work that
needs to be done to pilot RxNorm, and I have timing issues and I have a lot of
other issues. So do you think we are going to be able to pilot RxNorm in 2006
at all, given the fact that there’s a lot of work, obviously, on your end that
needs to be done?

MS. GILBERTSON: Well, the standards perspective, I think, is – I mean, the
analysis and all that is part of the work. It is probably the easier part of
the work, compared to the – you know, the getting of the database, munging(?)
it with all the different database companies and them spitting out the –
whatever they are going to do to their customers, so that their customers can
pull it off the file and put it into the SCRIPT message and transmit it and
then translate it. That is probably more of the difficult work of the actual
implementation, but there is definitely that analysis work that – you know, and
part of it could be my ignorance, but it also is until we’ve got a good idea of
what this morphs into, that analysis just has to take place.

DR. HUFF: But, I mean, another view of this is that – I mean, I don’t know
exactly how you are constituting these pilots, but, I mean, you could let the
pilot decide what they think the use case is and all that sort of stuff; and,
as Lynne said, you can put it in the message and it wouldn’t be an
approved-standard message, but it would be a message that would work just fine
with everybody’s software to go ahead and pilot that.

So, I mean, I don’t – You know, to get it into the standard, all of the
work that Lynne described has to be there. To do a pilot, you can let the
people who are piloting create the use case and do all of the work that would
say how to use it, and that would – actually, we would learn a lot from that,
and that could inform the process of how it is put into the standard.

MS. FRIEDMAN: Well, I obviously misunderstood or had a critical brain
freeze in the middle of all this, because I agree with what you say, Stan. That
was my idea of the pilot. I just was concerned from the discussion that was
going on, and not being a technical person, that we wouldn’t even be able to do
this much at this point, and I just want to make sure that we can.

MS. GILBERTSON: Right. And one of the concerns – I mean, this is a mapping
exercise. It really is. We each are talking in a different language and we need
some translator in the middle.

There is a part of me that says we need to be very concerned that both
parties know what the rules are for translating, so that that – whatever it
gets translated to is the same every single time, and, at some point, that
information is on the burden of the drug database companies, but I am also
wondering if there’s some responsibility that has to be on the vendors that –
the e-prescribing vendor systems that are using this.

And, you know, for example, when I was doing my RxNorm for Dummies –
because I am still struggling with this myself – what if you take a concept or
you go from one drug database to another database and you arrive at an RxNorm,
but when you translated it back out, the text names are wildly different? Well,
there is nothing in this process that says humans are going to relay and review
that the RxNorm translation on both ends worked, because nobody is reviewing
data names to see if the drug names turned out the same, but we have to have
the security and the comfort that that will happen behind the scenes. So it is
just kind of working through those kind of things, and those kind of RxNorm for
Dummies is probably a real good way of thinking about it.

DR. COHN: Yes, but we are not hearing anything that gets in the way,
necessarily, of doing some pilots.

But, Jeff, go ahead with your questions, and then we’ll, hopefully, at that
point, finish up.

MR. BLAIR: Yes, I have the same concern that Maria had and I think Stan
alluded to, and so my question deals with a little bit of reassurance here.

I think what I have heard is that it is entirely possible to have RxNorm
codes included in NCPDP SCRIPT messages for the pilot test, but I just – I was
hearing so many issues being raised that I wasn’t entirely sure that either
there was a commitment or – I thought that the RxNorm test group that you have
– that is the group that has the responsibility for seeing to it that it
happens. Is there another missing element, other than the capability of doing
this? Is there a lack of –

MR. ROBINSON(?): Karen Eckert is coming up.

MS. ECKERT: No, keep going.

MR. BLAIR: Keep going? Am I digging myself deeper? I mean, if you were
about to answer the question, then please jump in. I am just trying to make
sure that this won’t default unnecessarily.

MS. GILBERTSON: Okay. Let me see if they want to – Do you want to answer
that from the task-group perspective of what you think you want to take on?

MR. ROBINSON: Well, I have a couple of comments.

Part of the recommendation to have a governing body is that I don’t
necessarily know if it is NCPDP’s role in life to validate that RxNorm is
adequate to meet the business cases expressed in the MMA. In my opinion –
George Robinson speaking – that would be CMS to actually look at the particular
statutory requirements specified, look at the needs of interoperability
vocabulary, and if they are comfortable with the content and the mappings
that’s there, then perhaps suggest and go forward.

What we have here is a chicken-and-egg thing. It’s – we talk to the
industry, especially, perhaps, pharmacy, and, you know, I don’t want to rely on
an automated link until I know that it’s bulletproof. So, you know, in lieu of
that, what they are forced to do is continue to look at the text that is
associated to the drug concept that is selected in the prescriber system and
compare it to the concept that has been translated to an RxNorm, you know,
until at least we reach that comfort level. So it is really – a lot of this is
a confidence thing, and, also, the reason why we went through and detailed the
steps required for industry implementation is that some of these are
dependencies that have a large amount of time related to them.

So one point that Karen had mentioned to me that we need to get across is
in the absence of brand-drug links to NDC, the drug-knowledge-based vendors or
self-programmers will need to build cross-reference links themselves between a
concept such as Prozac 200 milligram oral capsules to Prozac oral milligram
capsules within RxNorm. So that adds to the implementation time line.

MR. BLAIR: If I hear you correctly, I think what I am hearing you say is
you are worried that you are going to be stuck with making the financial
investment to develop those links for the pilot test and that that expense is
not going to be recoverable to you. You don’t have a business case to do that
for the pilot test. Is that what I am hearing?

MR. ROBINSON: We have had that expressed from – not necessarily the
drug-knowledge-based vendors, from the industry themselves.

And it looks like Tom Bizzaro is stepping up to the microphone. He may have
some background from providers that he has talked to.

MR. BIZZARO: This is Tom Bizzaro from First DataBank, and I want to make a
couple of comments.

I think there are two things that have to happen for RxNorm to be part of
the standards then to be used within the pilot.

To be used within the pilot, really nothing has to happen within the
standards. To make them part of the standards, there has to be a sponsor or an
advocate for RxNorm that comes in – CPDP – looks at all the different standards
and looks at the proper data elements in areas within the standards that now
have to have RxNorm as a qualifier so that it can be included into those
standards.

So someone who is that advocate – whether it is from NLM or whoever it is –
has to come in, examine the standards and say, Okay. I want – this concept
belongs in the standard in this particular place. So that is some work that has
to be done, and I think that is outside of the pilot, because the pilot should
go forward, no matter what.

The other thing about NCPDP is it is a pragmatic organization, and,
generally, they want to see a business need for changes to the standard and
inclusions of new data elements or fields. I think that is a good thing, and I
think that RxNorm will prove itself to be valuable to pharmacy, both within
NCPDP and then to institutional pharmacy and healthcare, which is a little bit
less represented in NCPDP, as was mentioned earlier.

So from my point of view, I think that there needs to be an advocate for
RxNorm. You need to come to NCPDP, and someone has to learn the standards and
see where RxNorm should be placed into those standards, and I actually think it
is inappropriate for the task group at NCPDP to move that forward. I think it
is someone who has that business need and desires to see it as a part of the
standards that should come forward and do that.

MR. BLAIR: Now, you are saying nobody has volunteered to be that champion
or advocate.

MR. BIZZARO: Not that I am aware of right now, no, they have not.

And I do want to comment a little bit about will the database organizations
spend the time and money to do the linking to RxNorm? And I can certainly only
speak for my own organization. We are a market-driven company, and if the
market shows us that need, we will take that initiative and move forward to do
those links. It is, right now, I can tell you, certainly, our intention to do
that, but, again, being a for-profit organization, we are going to have to see
where does that marketability of that product come forward, and we do think,
internally, that it is going to be a market need. We just don’t know when.

DR. COHN: Thank you.

MS. FRIEDMAN: I have a follow-on.

DR. COHN: You have a follow-on. Okay. Please.

MS. FRIEDMAN: I’m still confused. Are we still going to be able to do
pilots or not?

MS. BIZZARO: In my opinion, you absolutely will be able to do –

MS. FRIEDMAN: If you still have this significant investment to make to do
the crosswalk, is somebody going to be able to step up to the plate for the
pilots?

MR. BIZZARO: I’m not sure you need to do the crosswalk within the pilot. It
depends what you want to prove within that pilot. Do you want to prove the fact
that in an e-prescribing system you can send RxNorm – you can send a
description of RxNorm and from that be able to fill a prescription in a
dispensing pharmacy? If that is your proof of concept, then you can do that. If
you are expecting there to be a full-blown cross reference between the database
companies and RxNorm concepts, I don’t think you’ll see that.

You’ll certainly have, within NLM, some of those cross references already
in place, because NLM is doing that cross referencing.

So I don’t know what you’ll see from the database companies, because I
think we do have to see something of a marketable product and – you know,
because we are not sure yet either what the market is going to want to see from
us as it concerns RxNorm. I think with NLM now and within the cross-referencing
they are doing, there’s enough there right now for you to prove the concept.

MS. FRIEDMAN: Thank you.

DR. COHN: Okay.

MS. ECKERT: Can I add on to that?

DR. COHN: Sure.

MS. ECKERT: Karen Eckert from Medispan.

I was trying to keep my – to myself. That’s why I was in the back of the
room. I apologize.

I think to build on Tom’s case, you know, from the Medispan side, we are
also very market driven as to what we need to do, but if it turns out some of
our partners are wanting to be part of the pilot tests and they bring to us
what business need and which levels they are wanting us to map to, I mean, we
will work with them as what we can.

I don’t think we could produce a product and get it out the door and make
it widely available to everybody in time for the pilot tests, but we will make
something available to those who want to partner with us to make the testing
possible, but, again, it helps if we can go down and define which levels of
RxNorm we mean, and so that is where the use case with your vendor partner that
you would need to have that.

It’s just there’s another level on top of their development, then, that we
need to be doing that coding for as well as them writing their code to use that
data before the test could start, and just realize it does add some more time
to the development.

DR. COHN: Karen, thank you.

Steve, I think you have the final question, comment.

DR. STEINDEL: Yes, and I am trying to rephrase it, because my initial –
it’s nice to be at the end, because as I had my initial comments, they have all
been addressed, and, Tom, you did a very nice job of addressing a lot of what I
was going to bring to the table.

I think Stan and Mike and Stuart presented a very good case for the need
for RxNorm in the drug process – ordering, dispensing, et cetera. It serves as
a very good tool for linking a lot of things together and has a lot of
important purposes and needs to go forward.

And I think what we have also heard is that there is, at best, a mixed case
for the use of RxNorm in the actual e-prescribing process, sending the
prescription, because, at this time, it is being identified by other means, and
from what I have heard, for the near future, which may extend more years than
we want to talk about, that is going to continue to occur. So, at best, in the
e-prescribing process, we are going to look at a duel-coding system, if RxNorm
is included, and if the mappings between your existing coding systems exist
someplace else and you can do that using something else, which we have already
noted is being developed within the UMLS and readily available to anybody, I am
not convinced that we need to send the RxNorm codes with the prescription, if
people have availability of the set of mapping tables.

And the reason why I am saying this is because we have two things going on
here is – you know, first of all, the development of each one of the drug
terminologies, RxNorm and the various ones that are being used, and second is
the maintenance, and the maintenance of those mapping tables, and it would be
best if they were left centralized and people didn’t have their own problems,
and we have brought up the whole question of mapping at NCVHS in the past, and
it is not just RxNorm to drug terminologies. There’s the clinical terminologies
to the ICD terminologies. There’s the adverse-drug-event terminologies, et
cetera, and, you know, we have said that NLM should take ownership of these
mappings and develop processes to make them available and keep them adjudicated
for clinical use, and I think we should just still look at it that way.

Stuart, you look like you had comments.

DR. NELSON: Yes, I want to say something about your comment about mapping.
There is no –

DR. STEINDEL: It is probably the same as concept.

DR. NELSON: There is no mapping between First DataBank’s drug vocabulary
and Medispan’s. There is no mapping between First DataBank and Multim. The fact
is is that none of them can speak to one another, and the only mapping that
there is of them is in RxNorm, which is the only mapping between them in the
UMLS. Okay?

DR. STEINDEL: Thank you for your clarification, and, actually, that is what
I meant.

DR. NELSON: All right.

DR. STEINDEL: Yes, I appreciate the clarification there.

DR. NELSON: I wanted to comment a little bit about Tom and the mapping and
the flagging work.

I always viewed the use of RxNorm in interoperation between systems as
being something of a preprocessing step, that it doesn’t necessarily have to
get into the heart of how you do drug-interaction checking and so forth, but
just to go to say, Okay. Here is an RxNorm ID for an SBD. This is what our
identifier within the FDB system is, and this is what it will do, and this is
what our identifier is in Medispan and this is what it will do, and I think
that that – when you look at it at that level, just saying that there’s a
preprocessing step, I think doing a preprocessing of an RxNorm code between
other codes is a lot easier than trying to make it inherent in all the
software.

MR. BIZZARO: And I would agree with Stuart. That is exactly how we see it
used, and it is why we don’t feel threatened by RxNorm, because we think that
it is an opportunity for us to get our information out there in different ways
for multiple people, and, as – I don’t know if it was Michael or Stan talked
about does it generate competition. Well, yes it will, and it should, and that
is good thing.

DR. COHN: Well, I want to thank everyone. We are running about a half-hour
late, but that’s fine. Obviously, Lynne, thank you very much for spending the
morning and afternoon in front of us. George, thank you; and, Karen, thank you
also for coming up, as well as Tom.

We are going to take a 15-minute break, and then we’ll come back and just
sort of open session, open mike, and then we’ll start talking about the letter
after that.

(2:48 p.m.)

* * *

(3:10 p.m.)

Agenda Item: Open Microphone

DR. COHN: Okay. Our first order of business is basically an open-microphone
session, and what we want to do is sort of invite anyone who would like to
comment about any of the issues we have discussed today, or any other issues,
to come forward for brief comments, if they would like.

Please. And, obviously, you need to introduce yourself and state your name
in the introductions.

MR. SIMKO: Thank you. Mike Simko with Walgreens.

I just wanted to make a brief comment on the earlier discussion today
talking about the DEA and controlled substances and the need for PKI or the
confusion about the need for PKI.

Schedule III through V, controlled substances, I think it was discussed,
account for – you know, the discussion was around 15 percent of prescriptions
filled, and that number is about accurate, and Schedule II’s do account for
about three percent of total filled prescriptions.

However, when you look at the most effected group of patients that the
Medicare Act may cover, they really use a higher amount of controlled
substances for a lot of variety of reasons. When you start factoring away
pediatricians and everybody else who writes a lot of prescriptions, but very
little controlled substances, that number is a little bit more significant, and
it does become a barrier to physicians in using electronic prescribing,
especially for that population of patients, because much of it is given in
combination therapy with other medication, especially patients either on
chemotherapy or a variety of other reasons that they may be taking
prescriptions that, if he has to dispense controlled substances one way – by
either a phone call, fax or writing a prescription – and if you think he is
going to send the other medication electronically, it probably won’t happen.
It’ll be more of a barrier, especially in this time of year. We are seeing a
lot of cough and cold medication and antibiotics, and they can write the – or
send the antibiotic electronically, but they have to either phone in or have
the cough syrup, if it is with codeine or a controlled substance, verified by
the pharmacy, and, again, that verification process, if we get – a controlled
substance is sent electronically, we have to make a phone call back to the
physician’s office for verification. Most times, we never get past whoever
answers the phone is just verifying that, yes, the doctor sent it. So, again,
you know, they are allowing controlled substances to be sent verbally to a
pharmacy, but we can’t accept it electronically, and to many of the physicians,
that is confusing and rather nonsensical, and it really does become a barrier
to their using electronic prescribing, because then it forces them to support
two different methods and really doesn’t help the patients as far as getting
the medication to the pharmacy quicker. It doesn’t help – you know – in all the
other benefits of electronic prescribing with transcription errors and getting
the accurate information, and to that group of people that that may effect the
most, it is the most important functions of electronic prescribing, and then we
seem to be shortchanging that group that can be most beneficially-effected.

So I would urge the subcommittee to reapproach the DEA regarding the
requirements on Schedule III through V’s, that they should be allowed to be
sent electronically. We don’t seem to have as great of an issue with Schedule
II’s and the use of PKI or some more stringent types of identification, but at
least with III through V’s, not being allowed to be sent electronically will be
a barrier and will hinder the adoption of e-prescribing.

MS. FRIEDMAN: Mike, you said that the elderly, the beneficiary population
for Medicare Part D generally tends to have a higher use of controlled
substances, for a variety of reasons, like for oncology and that kind of thing.

MR. SIMKO: Sure.

MS. FRIEDMAN: Could you take a guess at percentage-wise?

MR. SIMKO: If you take – 15 percent has been the number that has been
talked about as the total amount of controlled substances that are being
dispensed by pharmacies, and if you slice away pediatrics and everything else,
which are very high prescribers, generally, in our pharmacies, I would probably
say then you are in the 30-percent range, and that number, then, becomes a lot
more significant.

Thank you.

DR. COHN: Yes, Mike, maybe I can ask – can I ask you just one other
question?

So you said slice away pediatrics. So are you referring to the adult
population or are you referring to the Medicare population?

MR. SIMKO: No, I was just talking about the prescription numbers as a
whole, that controlled substances are always represented as being 15 percent of
prescriptions, but if you take pediatricians and prescriptions written for
children, the population that gets controlled substances, that percentage is
much higher than the total amount.

DR. COHN: Sure, but I just wanted to clarify whether that population was
the adult population –

MR. SIMKO: Yes, it is.

DR. COHN: – which is what you are describing –

MR. SIMKO: Right.

DR. COHN: – as opposed to the Medicare population. So I just wanted to make
sure that I understood that.

MR. SIMKO: The older the patient, the higher the percentage.

DR. COHN: Great. Okay. Thank you.

Other open microphone? Other testimony or comments?

Okay. Now, you all have your chance to – oh, is there another one? Please.

MR. BROOK: Richard Brook from ProxyMed.

I think I was going to wait until you started to craft the letter. However,
as we talk about some of the transactions that have been recommended with CMS
and MMA and everything else, one of the transactions we talk about is RXFILL.
We still haven’t come up with a true business case. Obviously, we know the
benefit of RxFill, but the question is when a network, like a ProxyMed or a
SureScripts, whoever is in the middle, is going to be doing transactions
between the physician’s office and the pharmacy, who is going to pay for that
transaction?

So we have had several discussions and we have had some – you know –
discussions. One of the former companies said they would be willing to pay for
that. So as we talk about the RxFill transaction, we are still coming up with –
and we speak about this, and I was on the RxFill task group, and if you have
received the letter about RxFill, the task group, our recommendations is we
still were trying to figure out, when it came to RxFill, who was paying for
that transaction. Is it the payer that is at risk for the patient making sure
the patient necessarily got the medication, didn’t necessarily take it?

So one of the things we need to consider is the fact that we have been very
involved with this discussion for quite some time, but we are still trying to
draw a real business-case decision: Who is going to pay for the transaction? I
guess that is more a comment than a question.

So I guess as we address – and I don’t know – as we talk about that, since
RxFill isn’t in wide use yet, we are always trying to figure out, as Tom
Bizzaro mentioned before, you know, we still have to figure out a company like
First Data to go and do something where there is a business decision to be made
where it becomes cost effective for a company like ProxyMed.

Thank you.

DR. COHN: Can I ask you a question then?

MR. BROOK: Sure.

DR. COHN: Well, the other parts of the SCRIPT transaction, who pays for
those and is it divided by –

MR. BROOK: Generally speaking, when it comes to – prescriptions generally
pay a subscription fee for using an application, and pharmacies pay for the
transaction that generally results in a fill. So, for example, a new
prescription is where pharmacies pay for, as well as if a pharmacy generates a
refill request, we get paid on the response, and, generally speaking, the
response rate is high into the 90-percent range. So that is how we get paid on
the revenue model.

As you talk about a change, a request and response, it really becomes the
same thing. So where it results in a benefit to whomever is receiving it is
generally the one that pays.

Again, physicians pay for subscription fees to use an application, or part
of their EMR, their practice-management system, and pharmacies pay a
transaction fee; and the way ProxyMed has come up with a model, it is it just
about equally shared between the pharmacy and the physician, when you look at
subscription fees and you look at transaction fees.

DR. COHN: Okay.

MR. BROOK: Thank you.

DR. COHN: That is helpful. Thank you.

Any other comments?

Okay. Everybody’ll have their chance as we get into the letter of
recommendations.

Agenda Item: Subcommittee
Discussion

Now, first, you know, obviously, we sort of wanted to begin to talk through
a letter, but I just wanted to sort of open it up for subcommittee members. Is
there anything that – from today that you want to talk about, at this point,
that we should be recording and saving, lest it gets forgotten?

MR. BLAIR: There was a number of issues that were raised today, especially
about the business-case issues, and I think when we get through the letter and
we start to think of our recommendations, I think that the things we heard
today will probably help us to think whether we are crafting recommendations
that are appropriate and effective. So I think that is going to help us.

DR. COHN: Okay. Harry.

MR. REYNOLDS: Yes, what we may want to do – back to Jeff’s point. What we
may want to do when we get the letter finished is step back from it and say,
Okay. With what we have sent forward before and what we are sending forward
now, it is kind of an e-prescribing package. Are there any things – almost a
matrix. Which things help the patient? Which things help the process? Which
things – you know, and so on. In other words, it’s a nice little matrix
run-through just to make sure – you know, some things are business-case
related, some things protect the patient, some things are just part of the
process, and maybe that is another cut. We could take a look at it, just to
make sure if it brings anything else out that we want to say. Just a good –

DR. COHN: Okay.

MR. REYNOLDS: – step-back look from it, from a –

DR. COHN: Yes –

MR. REYNOLDS: – maybe a different view than just – you know, linearly going
through a letter.

DR. COHN: And I think Margret can maybe help us with that also –

MR. REYNOLDS: Yes.

DR. COHN: – as we move through.

MR. BLAIR: There is one thing I might say as we start to get into this
letter.

For those of us that – for all of you who are in the room who know and love
us here on the NCVHS – (laughter) – and have followed through this last year,
you realized that we were on accelerated schedules, and we did the best to
balance off trying to anticipate and do things in preparation and still do
things in an open manner where there was plenty of time for public comment and
reaction to the way we went through things.

So, in the case of this letter, Margret has helped us to start to draft
this letter, and she has accepted suggestions and inputs from all of us. This
does happen to be the first meeting when, as a subcommittee, we will be all
looking and commenting at the letter together.

DR. COHN: Okay. Phil, looked like you wanted to make a comment.

MR. BLAIR: Draft letter. Preliminary letter.

MR. MICH: I was struggling with whether now is the right time.

You mentioned earlier today the possibility of the subcommittee commenting
on the NPRM, and I was just, I guess, curious how we may have an input into
that process and how that may be together or separate from the process of
creating the next letter.

And the reason I raise it is, I, among others, have made a lot of comments
about the preemption issue, and there is a lot of conversation today about the
controlled-substance issue and how that sort of confusion could create
obstacles to adoption, and I would submit that the issue of having a separate
standard for Medicare prescriptions, separate and apart from 50 state standards
for other prescriptions, and introducing into the prescribing process a
question the physicians may have to ask as to what drug coverage a person has
is going to have a lot more to do with adoption than the existing issues around
controlled substances, because people have found a workaround.

And I am just concerned that if the end goal is to foster more electronic
prescribing that the issue with preemption could accelerate that or
significantly decelerate any adoption that is going on, and I think it would be
useful for the subcommittee to weigh in on that issue, understanding that there
is a legal issue and that HHS has to do what they have to do, but to the extent
the subcommittee is talking about adoption issues, I just wanted to put that on
the table as well, because I think that has the capacity to create real
confusion out there, and I’m not sure – looking at it as purely a legal issue –
if there’s a way to find a solution – you know, a different reading or whatever
– and it is not up to the subcommittee to solve that, but I think to the extent
you recognize that issue it could be important.

DR. COHN: Let me try to answer – I’m not sure I’m answering any of your
questions, but let me, of course, observe that the issues that you are bringing
up, of course, were just at the beginning of the public-comment period. So it
isn’t just the NCVHS. Your organization and many others, I’m sure, will be
providing input to CMS about all of that.

Now, in terms of our process, in terms of what we are going to be doing, I
guess I should just sort of be clear with everybody sort of how we are trying
to structure this.

There is something called the second set of recommendations to the
Secretary on e-prescribing, which we are going to work on. There is going to be
a separate letter that will be sort of our response and comments on the NPRM.

I was sort of, I think, joking earlier today that this would be yet another
thing that we hadn’t really – you know, unplanned mandate for us to come up
with a letter like that.

I suspect some time towards the mid-to-late morning tomorrow will be the
time that we have to sort of reflect on that, at least briefly, with the idea
being that we will have a number of open conference calls between now and
probably the second or third week in February. I guess we are almost the first
week of February now, aren’t we? So the second or third week of February, we
will publish the information on when those are going to be, and, obviously, our
first priority is to get this letter done.

Second priority is to develop a draft letter for the full committee on
recommendations relating to the NCVHS comments on the NPRM. So just to give you
a flavor.

Yes, Karen.

MS. TRUDEL: If I could just add a little bit of procedural minutia –

DR. COHN: Oh, sure, please.

MS. TRUDEL: The process for commenting on an NPRM, in terms of where the
comments go and the time period in which the comments are provided, is quite
different than the process for the NCVHS to submit recommendations to the
Secretary. So in order for us to manage this workload, I would suggest keeping
those processes very separate in all of our minds.

In other words, any problems with the NPRM – and I would say that that
would be either a situation where the committee felt that HHS did not follow
their recommendation and that we didn’t have an adequate reason for it or if
there is a situation where the committee revises its original recommendations,
for some reason – should be included in the comment letter that would go to a
particular address in the Federal Register, et cetera, et cetera, by April
5th, which is the closing date. If that could be kept separate from
the recommendation letter, that would make things much easier for us, and it
would just simplify everyone’s lives considerably.

So with respect to Phil’s issue, if there is a sense that HHS did not –
that there’s a disagreement with our position or that if NCVHS wishes to revise
its original recommendation, that would be appropriately in that comment letter
on the NPRM, and if it is not, if it somehow managed to get into the other
letter, we wouldn’t be able to consider it, because it is not part of the
comments.

DR. COHN: Yes, Karen, hopefully, I was sort of trying to say what you are
saying, and I agree.

The only issue, of course, is is that the next full meeting of the NCVHS
occurs right at the beginning of March, and we don’t have another meeting of
the full committee, which is the group that would have to, obviously, consider
–

MS. TRUDEL: Right.

DR. COHN: – any letter after that. So we would either have to have it ready
at the same time, though clearly a separate –

MS. TRUDEL: Right. Right.

DR. COHN: – letter, or we have to invoke some sort of an emergency process
to get that letter completed.

MS. TRUDEL: Right. And that said, I just want to follow up on one thing,
and that is that we are now in a comment period, which puts everybody in a
completely different place. It essentially says that for this period of 60
days, the NCVHS is no longer the funnel for industry comment and input.
Everyone has a direct link to HHS during the comment process, and I encourage
people to take advantage of it.

DR. COHN: Very well said.

And, Marjorie, did you have a comment?

MS. GREENBERG: That is what I was going to say, that, although it will be
an open process, the NCVHS comment letter, as well as your other
recommendations, people should not see that as a substitute for commenting
directly.

DR. COHN: Absolutely.

Now, the other piece, of course, -I would just remind everybody, lest I
forget – is that tomorrow we will be scheduling those three open conference
calls, and so we ask everybody to come with their calendars, so when Maria sort
of sends things around to the subcommittee and staff we can identify when we
are going to be available, so that she can do that very quickly. Okay?

Okay. Now, any other – Are we okay at the moment here? Any other
housekeeping issues? Any other stuff?

So basically, just to reframe this, what we are doing is obviously going to
be talking about a draft letter related to our sort of second set of
recommendations.

I do want to remind everyone that this letter includes, obviously, our
recommendations relating to hearings we have had about e-prescribing – I’m
sorry, e-signature – but also will include information being brought forward by
the Subcommittee on Privacy and Confidentiality related to any privacy and
confidentiality issues, and my memory is there’s a couple of paragraphs that
that subcommittee had developed which I haven’t seen for while, but we’ll put
that in there, as well as, I think, other additional comments we may make,
knowing that, in our first letter, there were a series of things, many of which
we were updated on today, related to whatever additional advice, at this point,
we may want to make on those sort of outstanding items, and, of course, for
those outstanding items, we have a choice of saying – making a definite
statement or continuing to defer them, only because the information is not
complete at this point. So just – and there may be yet other issues, but I
haven’t figured out what they are yet. Okay?

Now, with that, are we comfortable with having Margret beginning to lead us
through a – as I said, a draft letter where – I think, primarily, the focus of
today will be beginning to look at some of the introductory materials
preparatory to any sort of recommendations.

MR. BLAIR: Could I suggest if Margret could take us through the framework
of the letter first, let everybody know how far we have gotten, because we
don’t have everything in the letter yet, but, at least, if she shows us the
major headings and the subheadings, before we start to go into the detail, then
I think everyone in the audience will sort of know what to expect, what is
coming.

MS. AMATAYAKUL: Am I on?

Okay. We have a couple of opening paragraphs, just setting the stage that
this is the second letter on e-prescribing.

Then we have e-signature background that includes some information on a
prescribing overview, e-prescribing networks, use of PKI in e-prescribing, and
industry experience.

And then we have the observations and recommended actions, and they are in
the same form as – or they will be in the same form as we had for the other.
There’ll be observations and recommended action observation, recommended
action.

MR. BLAIR: And an appendix?

MS. AMATAYAKUL: Oh, yes, and then there’ll be the other items and the
privacy items, and then there’s an appendix – two appendices – that we’ll see,
and then a glossary of terms, if you wish.

MR. GREENBERG: The privacy items will be in the body of the letter.

DR. COHN: Yes.

MS. AMATAYAKUL: Yes. I don’t have them, though, yet.

So shall I read the first paragraph here?

DR. COHN: Sure. Why don’t you actually read the first two paragraphs, if
you would.

MS. AMATAYAKUL: Okay. The National Committee on Vital and Health
Statistics, NCVHS, has been called upon by the Medicare Prescription Drug
Improvement and Modernization Act of 2003, MMA, to develop recommendations for
uniform standards to enable electronic prescribing, e-prescribing, in
ambulatory care.

This letter is the second set of recommendations on e-prescribing and sets
forth recommendations relating to electronic signature and other important
issues relative to pilot testing for e-prescribing.

The first set of recommendations since September 2, 2004, addressed
message-format standards that provide communication protocols and data-content
requirements, terminologies to ensure data comparability and interoperability,
and identifiers for all relevant entities within the e-prescribing process.

The recommendations were sequenced so that any foundation standards that in
themselves did not need to be pilot tested were identified first. Then, actions
were identified to address additions to the foundation standards, development
of new standards and pilot testing of standards necessary to meet the MMA
requirements.

The September letter also provided a description of the current prescribing
environment, current e-prescribing capabilities and the process NCVHS used to
evaluate standards.

DR. COHN: Okay. I’m not sure that we want to do terrific wordsmithing, but
I would just suggest that maybe at the end of the first paragraph you have – it
says relative to pilot testing for e-prescribing, and I guess I would question
whether that is really the case or whether it is just other important issues,
period.

MS. AMATAYAKUL: There are some recommendations for pilot testing, but I
think that could be assumed.

MS. FRIEDMAN: The way it is written, it looks like all of the
recommendations refer to pilot testing.

MR. BLAIR: Could we have, other important issues, and then, in parentheses,
such as pilot testing?

DR. COHN: Well, I didn’t know that pilot testing was necessarily the focus
of these – I mean, I guess if it is, maybe we can hold that one pending what
our recommendations are, because I didn’t think that pilot testing was
necessarily a focus of these recommendations, but I don’t know. So why don’t
you put that as something for us to revisit, but maybe removed.

MS. AMATAYAKUL: I have another version that is the original. So –

DR. COHN: Oh, so when you remove things, we don’t have to see track
changes, huh?

MS. AMATAYAKUL: Well, if you want, I can.

DR. COHN: Your choice.

MS. AMATAYAKUL: Just reduces the size to look at.

DR. COHN: Okay. Any other major comments? Are we directionally correct?

Okay.

MS. AMATAYAKUL: Electronic signature background, prescribing overview.
Prescription writing is a critical factor in patient care and patient safety.
The National Association of Chain Drug Stores, NACDS, estimated that 1.4
billion new prescriptions were written in 2003 with another 2.0 billion refills
and renewals processed.

The Center for Information Technology Leadership, CITL, estimated that $154
billion was spent on prescription drugs. CITL also estimated that, as a result
of adverse drug events, ADE’s, approximately $2 billion was spent in
ADE-related hospitalizations and visits.

DR. COHN: All right.

MS. AMATAYAKUL: The U.S. – next paragraph. The U.S. Department of Justice’s
Drug Enforcement Administration, DEA, has regulatory authority over prescribing
and dispensing of controlled substances. These are substances that have
addiction and abuse potential. They are divided into five schedules. Schedule I
substances are illegal and may not be prescribed. They are not applicable to
these recommendations. Schedule II substances are highly addictive, and their
prescriptions must be handwritten and delivered to a dispenser. They may not be
phoned or faxed to a dispenser.

Not sure that statement is quite correct, after what we heard this morning,
but we can talk about it.

Schedule III through V substances are controlled, but may be phoned or
faxed to a dispenser.

Prescribers must be authorized to prescribe controlled substances by the
DEA and receive a DEA number for this purpose.

Substances which are not controlled are usually handled by prescribers and
dispensers in the same way as Schedule III through V substances.

Seems to me that we heard that you could send a Schedule II, but it had to
be followed up with a call.

DR. COHN: No, it actually had to be followed up by handwriting.

MS. AMATAYAKUL: Handwritten – that the patient brings in that handwritten –
and that Schedule III through V had to be followed up with a call, in some
states. Is that –

DR. COHN: Actually, I thought it was all states.

MS. AMATAYAKUL: Somebody –

DR. COHN: And either had to be faxed, written or oral.

MR. BROOK: It depends on the type of fax, again.

DR. COHN: That’s right. It depends on the type of fax.

MR. REYNOLDS: Yes, Simon, I had a question on that word fax, because I
think Ken made it clear today that most people would read that letter that fax
means I have a piece of paper and I fax it in, but there are, clearly, maybe
even more situations where it comes from a computer to a fax machine, which I
think is a quite different –

MR. BLAIR: Somebody –

MR. REYNOLDS: – whether we parenthesis – yes, or something – because I –

MR. BLAIR: Someone behind me suggested that we use the word, in general.
That way, we could keep it as generalities, without getting into all the
exceptions different states might have.

DR. WARREN: I think we need to keep this distinction on faxes, because I
think that is critical to all this.

DR. COHN: Okay. Judy, you need to get to a microphone.

DR. WARREN: I think we need to keep in mind the distinction between the two
kinds of faxes, since we do have a fax that is computer-based and one that is a
traditional fax, instead of just talking about this in generalities.

Maria suggested maybe a footnote or something by fax could handle – we
could put a footnote and say, Faxes that are paper-to-paper or faxes that are
not computer generated, and that could handle – it would read better.

DR. COHN: Okay.

MS. AMATAYAKUL: Let me wordsmith that later, okay?

DR. COHN: Okay.

Now, help me with the last sentence of this second paragraph.

MS. AMATAYAKUL: Substances which are not controlled –

DR. COHN: Yes.

MS. AMATAYAKUL: – are usually handled by prescribers and dispensers in the
same way as Schedule III through V substances.

DR. COHN: What does that mean, exactly? Because I don’t think you are
right, but I’m just – I’m not sure that I understand that well enough to know
whether I agree or disagree. I mean, my understanding is that Schedule III
through V require – can either be oral, written or faxed with a footnote.
(Laughter).

Basically, things, however, substances that are not controlled can just be
– you know – sent by a computer, at least in some states. So it isn’t –

MS. AMATAYAKUL: Can we just take this sentence off?

DR. COHN: Okay. That’s good. Okay.

MS. AMATAYAKUL: Thank you.

Onwards.

MS. GILBERTSON: What is the title of this section?

MS. AMATAYAKUL: Prescribing overview.

MS. GILBERTSON: It discusses controlled substances in the overview.

MS. AMATAYAKUL: The intent here was to describe the scope of the number of
transactions we are talking about, because later we are going to talk about the
percentage that are Schedule II, and, then, to identify the different schedules
and the DEA’s responsibility, and also just to alert readers that there is a
DEA number, et cetera.

DR. COHN: Yes, and, actually, maybe this is almost better. Rather than
prescribing an overview, maybe this is e-signature and e-prescribing.

MR. BLAIR: Oh, okay.

DR. WARREN: Simply don’t talk about signature.

MR. BLAIR: Well, yes, we start –

DR. COHN: Or maybe we don’t. Maybe you are right.

Make a note that we need to change the title here a little bit.

SPEAKER: (Off mike).

MR. BLAIR: No, well, it will be. It is the foundation upon which the
e-signature discussion will rest, but, nevertheless, I think you are right.
Let’s not have e-signature quite yet here.

MS. AMATAYAKUL: Okay.

DR. COHN: Yes, just a comment here, and I’m trying to think of whether it
is said later on and all that, because, obviously, we dip very quickly into the
world of DEA. You may want to put – we may want to put something maybe even –
maybe almost at the beginning of the second paragraph which talks about, you
know, the percentages of prescriptions. I mean – something that talks about –
not controlled, but, you know – obviously – you have essentially removed
basically non-controlled substances, and if you can figure out a way to say
that in a positive fashion, rather than a non, we should probably make a
comment about that generally and then dipping into DEA after that.

MS. AMATAYAKUL: So I could say something about the overall percent of – 85
percent is not controlled, et cetera.

DR. COHN: Yes, something like that.

MS. AMATAYAKUL: Okay.

DR. COHN: That seemed like it might be a place to – I mean, just generally
introduce and then move down into that, because it seems like sort of an abrupt
beginning here.

MS. AMATAYAKUL: Okay.

MR. BLAIR: Do you think it is worthwhile adding something to the fact –
this is Medicare prescription drug program – that the percentage of
prescriptions that will involve controlled substance are higher in the adult
population, in the Medicare population?

MS. FRIEDMAN: The elderly and the disabled.

MR. BLAIR: The elderly, um-hum, and disabled, and we may not have a number,
but just a statement to that effect.

SPEAKER: (Off mike).

MR. BLAIR: I don’t think we have – I wasn’t – I don’t think we have a
statistic, but I am just saying, just a general statement saying that it would
be higher – the percentage would be higher.

MS. GREENBERG: If you don’t actually have a reference, I think you could
say testimony indicated – Testimony from the national committee –

MS. AMATAYAKUL: I’ll work on that.

DR. COHN: Okay.

MS. AMATAYAKUL: Next?

Okay. E-prescribing –

MS. GREENBERG: Actually – excuse me. I was thinking that it is possible
that the National Ambulatory Care Survey at NCHS might have – they collect drug
data. So it is possible that you could get some statistics from them. I’m not
sure.

MS. AMATAYAKUL: From where, Marjorie.

MS. GREENBERG: Hum?

MS. AMATAYAKUL: Where did you say?

MS. GREENBERG: It is the National Ambulatory Medical Care Survey conducted
by the National Center for Health Statistics – and I am not sure the level of
detail, but they do collect some drug data.

MS. FRIEDMAN: The current beneficiary – out of Medicare Data may have
something, too. I haven’t looked at it lately to remember whether they have –
what kind of utilization data they have.

MS. AMATAYAKUL: Okay. E-prescribing networks. Today, most prescriptions are
handwritten by prescribers on paper. Prescribers may fax or phone these to a
dispenser or give them to the patient. The patient may then – may take them to
a dispenser or use an online or mail-order service.

Prescribers are starting to use their computers to send faxes to
dispensers, and still others are starting to send prescription transactions via
e-prescribing networks. E-prescribing networks are value-added networks – that
receive prescriptions from prescribers in the form of a transaction using the
NCPDP SCRIPT standard. These are routed to the designated dispenser.

Testimony to NCVHS indicated that the electronic transaction form of
e-prescribing offers the greatest potential to improve patient safety and
achieve quality benefits as outlined in MMA.

Electronic transaction conducted over e-prescribing networks is the
definition NCVHS uses for e-prescribing in these recommendations.

And this was where I was concerned that we saw in the NPRM that
e-prescribing could be direct or through a network. So I think we should
clarify this.

DR. COHN: Well, Margret, let me ask on this one, because I am actually – is
it that the recommendations herein refer to e-prescribing that is conducted
over networks or is there another assertion here?

MS. AMATAYAKUL: I think it is primarily over the networks, but I think
there’s sufficient reference to other forms would require other measures. So we
might want to just make a note and see whether we want to limit it or not.

DR. HUFF: Yes, you know, this is one of the places where I noted. It just
seemed like we were being too constraining here to just say vans(?). I mean, it
seems like our recommendations, in fact, could be e-prescribing where it was
over the internet via VPN, and there’s not really a van involved in that, per
se.

MS. AMATAYAKUL: Exactly.

MR. REYNOLDS: Can we say that e-prescribing networks can take many forms,
many routes, many –

MS. AMATAYAKUL: No, I think the point that I was concerned about and that
Stan said is that, basically, you’ve got a network that you pay to transmit and
it could be a van, it could be just the switch. It could be whatever, but you
pay to make that, and they add that security value for you, or, alternatively,
you could create your own VPN or you could send it as an email message. Who
knows? It doesn’t say an NPRM, that it is limited to a secure network.

MR. REYNOLDS: And that’s why I said many forms. (Laughter).

MS. FORQUET: Lori Forquet.

You might simply recognize that the vast experience that is out there today
utilizes the vans, but the considerations are not restricted to the vans.

DR. COHN: Thank you.

MR. REYNOLDS: I have another – are you going on to the next paragraph? I
have another question there.

Help me – where it says testimony to NCVHS indicated that the electronic
transaction form of e-prescribing offers the greatest potential. So does
e–prescribing offer the greatest potential or the electronic form of
e-prescribing?

MR. BLAIR: Could I chime in on that a little bit?

I think the intent of the sentence, which we may want to change, was to
somehow communicate that the MMA wanted e-prescribing because it was going to
improve patient safety and quality of care and reduce costs, and patient safety
was first, and in order to do that, there was a whole list of functions that
would be needed for e-prescribing to achieve patient safety and cost reduction,
and, at this point, as we look at the marketplace right now, it looks as if the
e-prescribing networks – and I am not going to be cryptic about that right now,
but it is the value added networks that are bringing e-prescribing to the
marketplace. They are driving the market – that they have the greatest
potential to get us to the goals that MMA wants. Now, maybe there is a better
way to say that, but that was the intent of the sentence.

MS. AMATAYAKUL: I actually thought that the intent was to distinguish
between an electronic transaction using SCRIPT versus what some people call
e-prescribing that is only electronic facts.

MR. REYNOLDS: Okay. I can buy that.

DR. COHN: Yes, and I agree with that, though we need to figure out some way
to say that a little plainer.

MS. AMATAYAKUL: Let me make a note and I’ll work on it.

DR. COHN: Yes, it probably has to do with the electronic transaction form
of e-prescribing –

MR. REYNOLDS: Yes, because if you look at the definition –

DR. COHN: – is probably the – is probably a – it’s something we just made
up. (Laughter).

MR. BLAIR: The other thing is – in the future, in the letter, a lot of the
recommendations relate to e-signatures used in e-prescribing networks, and so
we sort of wanted to make sure the definition of that and the context of
e-prescribing networks was well understood, which is why we have this little
section on e-prescribing networks.

MR. REYNOLDS: I’m fine with the section. I just thought that was – that
sentence needed looked at. (Laughter).

MS. AMATAYAKUL: Okay. Next paragraph is, E-prescribing networks use a
combination of the following as a means to provide proof of prescriber and
dispenser identity and security: a minimum of a user ID and password for access
to e-prescribing software, credentialing upon enrollment in a value-added
network van, use of a network-assigned electronic signature process and
transmission of the prescription message through a private lease line or
through the internet using a virtual private network VPN connection or the
secure socket layer SSL protocol.

The next paragraph is a very long sentence, but it is because I couldn’t
help it. (Laughter).

SPEAKER: Is that the insanity defense?

MR. REYNOLDS: So for the record, get over it, right?

MS. AMATAYAKUL: The electronic signature process used by e-prescribing
networks varies with the network. In general, the process includes
identification of the source system – that is, prescribers, e-prescribing
system or dispenser’s pharmacy system – date and time stamp, sending system
identifier, prescriber’s name, DEA number, internal sender ID, name of
prescriber’s agent, if applicable, destination, dispenser name, address and
phone number and destination dispenser internal receiver ID.

Simon, could we take a comment on that first?

DR. COHN: Oh, I’m sorry. Sure.

MR. BROOK: Richard Brook.

One of the things I want to be pretty clear about, the name or prescriber’s
agent, if applicable, all of the time the prescriber’s agent is sent. Ken,
would you confirm that?

MR. WHITTEMORE: If there is an agent.

MR. BROOK: If there is an agent, but, generally, it is – generally
speaking, I guess it is applicable may not be the term, because most of the
time or I would say the majority of the time the designated agent is sent. So
how do we do that?

MS. AMATAYAKUL: If required?

DR. COHN: It’s probably if indicated is probably what it is.

MR. BROOK: That’s better.

Thank you.

DR. COHN: Thank you.

SPEAKER: DEA is not required, is it?

MR. BROOK: Currently, it is. Currently, it is. That is the only way –
Currently, the DEA is required, because it is the only way to route
prescriptions.

DR. COHN: Interesting.

Harry.

MR. REYNOLDS: When you think of the whole network process and it is passed
through multiple places and so on, I struggle that it is the
electronic-signature process. I do agree that it is the security process, but
an electronic-signature process seems to say to me I signed it and that is
exactly what gets there, but it gets opened, it gets translated, it gets
touched. So I am not sure my electronic signature makes anything other than an
identifier that I started it and others pick it up. So – Because, you know,
e-signature is kind of a term of art –

DR. COHN: Yes –

MR. REYNOLDS: – and I’m not sure it plays through this process and still
comes out the other end looking like an e-signature. I think there is a
security process that goes step by step, but –

DR. COHN: Okay.

MR. REYNOLDS: – I don’t know. Somebody might read that –

DR. COHN: Well, Harry, let me get back – that gets back to words of art
here. Are we talking about security? Is it security and authentication? Is that
what we are really talking about here?

MR. REYNOLDS: That is really what we are talking about. There’s some kind
of a –

MR. BLAIR: Authentication is part of security.

MR. REYNOLDS: Yes.

DR. COHN: Well, I guess the question is is what are we – Well, I guess we
need to decide what we are talking about. We decided maybe e-signature was not
the right term here. Security may be the right term or is it authentication? I
don’t know.

MS. AMATAYAKUL: Actually, Simon –

DR. COHN: Yes.

MS. AMATAYAKUL: – if you looked on line 18 –

DR. COHN: Yes.

MS. AMATAYAKUL: – this whole thing really is related here. So we are
talking about proof of prescriber and dispenser identity and security.

DR. COHN: So authentication and security works then.

MS. AMATAYAKUL: So we could say this process even to eliminate confusion.

MR. REYNOLDS: I’d be good with that.

DR. COHN: Okay.

Please.

MS. FORQUET: I just want to reiterate that according to our ASDM
electronic-signature standard, those would be signature attributes which for
health care we determine need to be part of the healthcare electronic
signature, not the digital signature. Just the healthcare electronic signature
needs to be able to carry those other signature attributes such as time stamps
and credentials.

DR. COHN: Great.

MS. AMATAYAKUL: Well, let me put in parentheses we can think about it some
more, okay?

DR. COHN: Okay.

MS. AMATAYAKUL: Dispenser – this is a continuation of the paragraph.
Dispensers rely upon the network to verify that the sender and receiver are
authorized users of the network, that none of the signature components are
missing and that the message is in the NCPDP standard format and version.

Parentheses, Some e-prescribing networks translate the message format or
version of messages that they receive into the appropriate format or version
for subsequent transmission. There may also be pass offs from one van to
another or to a gateway to route the prescription to its final destination.

See Appendix A for an illustration of electronic-prescribing security and
authentication standards – which I have to change the title of – being used by
e-prescribing networks. And I was going to change the title to Current Security
and Authentication Practices in E-Prescribing Networks.

MS. FRIEDMAN: In case people don’t remember what that was, that was the
diagram that Phil – and Terry Swanson presented in December on security.

MR. BLAIR: Now, is that the one where it has the word standards in the
title?

MS. AMATAYAKUL: Yes, I am going to change that, Jeff.

DR. COHN: Okay. I think this may just be wordsmithing, but it says some
e-prescribing networks translate. Would it be more appropriate to say
e-prescribing networks may translate or something like that?

MS. AMATAYAKUL: I think we wanted to get across that some do, but some
don’t, but I mean we could start with saying e-prescribing networks may.

DR. COHN: Why don’t we leave it alone for the moment, and we’ll – if we
decide we need to change it, we can do it later.

Okay. So, basically, we are – seem to be settling on security and
authentication as another way of describing this, so that is probably useful.

So anything else on this paragraph?

Okay. Let’s keep going.

MS. AMATAYAKUL: The current e-prescribing transaction communication process
uses a signature that is consistent with the electronic signatures in Global
and National Commerce Act e-sign definition of electronic signature that is,
quote, an electronic sound, symbol or process attached to or logically
associated with a record and executed or adopted by a person with the intent to
sign their record, unquote.

Although e-prescribing transactions are reportedly occurring in between 39
and 41 states, only nine state-pharmacy statutes or regulations address
electronic signature directly, and only three reflect the e-sign definition.

It is important to note that e-prescribing is not used for writing
prescriptions for Schedule II controlled substances. It is estimated that
approximately 15 percent of all prescriptions written are for Schedule II
through V controlled substances and that less than two to three percent are for
Schedule II controlled substances.

DR. COHN: Yes. Question. I think that you sort of probably have the right
numbers there. I guess I am wondering whether those last couple of sentences
maybe need to be – Well, I think it actually needs to be in that front area
when we were sort of introducing controlled and non-controlled prescriptions.
I’m just not sure what –

SPEAKER: Plus the numbers are wrong.

DR. COHN: Yes, probably the numbers are wrong, too, but –

MS. AMATAYAKUL: I thought all controlled substances were 15, and two to
three percent are two –

DR. COHN: What?

MS. AMATAYAKUL: – and I’ll cite testimony today.

MR. BLAIR: Yes.

MS. FRIEDMAN: And could you also provide a citation for the previous
sentence about the states?

MS. AMATAYAKUL: Sure. That was testimony earlier.

MR. REYNOLDS: Simon, I have a question on –

DR. COHN: Sure.

MR. REYNOLDS: – where it says it is important to note that e-prescribing is
not used for writing prescriptions for Schedule II, would it be good to – it is
important to note that because of DEA regulations, e-prescribing is not used
for those? In other words, rather than just somebody wonder why it’s not being
done, go ahead and cite that that is really the thing that’s tending to drive
it.

DR. COHN: Yes.

MS. HELM: Hi, this is Jill Helm from Allscripts.

I am just wondering if maybe we should make a distinction here between
electronic prescribing and electronic transmission. There’s no states that I am
aware of today that prohibit the creation of a prescription for a Schedule II
using an electronic system. There are signature requirements and there are
requirements in terms of the transmission of that prescription.

DR. COHN: I guess you can write anything you want on a computer system.

SPEAKER: It’s what happens to it after that that’s the problem.

DR. COHN: Yes. Good point.

Now, Margret, I guess I would ask, and I guess I am still asking is this
the right place to put that information or do you want to take a look and see –

MS. AMATAYAKUL: Yes.

DR. COHN: – whether it really – I mean, those last two sentences should
really be –

MS. AMATAYAKUL: Up here –

DR. COHN: – in the prescribing overview area?

MS. AMATAYAKUL: I’ll put it up here.

DR. COHN: Okay.

MS. AMATAYAKUL: Yes.

I am probably making you all dizzy. I’m sorry, but I –

DR. COHN: Yes, Tom, do you want to –

MR. BLAIR: I’m doing fine. (Laughter).

MR. ZORTOV(?): Simon, I’m not sure what the intent is here, but when I read
this it seems to me that if I am reading it I’m saying, Okay, so that is not a
big problem, because it is a small percentage of the SCRIPT. I don’t know if
the committee wants to indicate that we do see this as a possible – If the
committee sees it as a possible barrier to the implementation of e-prescribing,
is there something that you would want to say here that can reinforce the fact
that this is an issue and needs to be addressed?

DR. COHN: Good point. Let’s put that in parentheses as a note.

Yes.

DR. WARREN: Do we want to put in here – we said up there on line 45 it is
important to note because of DEA regulations. Do we want to put in there that
those regulations require a wet signature, so that we really make the
distinction between something that is transmitted and something that is a
physical entity?

MS. AMATAYAKUL: Done.

Okay. The next section is Use of PKI in E-Prescribing. NCVHS heard
testimony regarding the current status of digital signature, and, in
particular, public key infrastructure, PKI, because the federal government and
other initiatives seeking to strengthen authentication and provide
non-repudiation from messages transmitted over the open internet.

DR. COHN: Open as opposed to?

MS. AMATAYAKUL: Private network.

SPEAKER: As opposed to VPN –

MS. AMATAYAKUL: VPN.

DR. COHN: Maybe you want to put quotation marks around open.

MS. AMATAYAKUL: Wide open was the word used in – (laughter) – Can we use
wide open? (Laughter).

MS. FRIEDMAN: Back up in sentence seven, do we want to say the current
status of use of digital signature? I’m not – it seems like there is something
missing there. I’m not sure –

MR. BLAIR: Could you help me –

MS. GREENBERG: To respond to, instead of because of? I’m hanging up over
that because of. In response to – I don’t know.

MS. FRIEDMAN: I would also split those sentences up.

MS. AMATAYAKUL: Okay. I’ll work on that. I’ll make a note, work on it,
okay?

MR. BLAIR: The only thing that I would – I’m sorry. The only thing I just
want to verify is, Maria, you just used the word digital signature. The title
is PKI, and I’m just trying to make sure we’re keeping them straight.

MS. FRIEDMAN: Well, I am just using the words that were on the screen. We
haven’t gotten to PKI.

The sentence reads, NCVHS heard testimony regarding the current status of
the use of digital signature and in particular public key infrastructure, PKI.

MR. BLAIR: Oh, okay. Okay. It had them both. Okay.

MR. WHITTEMORE: Ken Whittemore, SureScripts.

I would just add the word proposed use, because I think the testimony was
that it really isn’t being used very much at all right now. Something that
conveys that.

MS. FRIEDMAN: In fact, that helps the sentence out, because you could take
away – if you say proposed use, you could say regarding the proposed use of
digital signatures and take out current status, and that helps your –

DR. COHN: You know, I guess I am having some trouble, and, Margret, I think
I may have to let you wordsmith a little bit, but I think I would not
characterize digital signature as not being used and only being proposed. I
think that what we heard was there are certain specific uses that occur in the
federal government and elsewhere for this, and I don’t think we want to
understate that.

Would you like to comment?

MR. FORQUET: Yes, I am wondering whether or not the only reason is because
of the federal government and other initiatives or if it is also because of the
trend toward health informatic standards and it is also toward the trend of
proposed DEA initiatives.

MS. AMATAYAKUL: Well, we’ve got the initiatives coming we haven’t read yet.
So –

Let me take a look at this.

DR. COHN: Yes, you can play around with –

MS. AMATAYAKUL: Okay. One initiative was the attempt to include digital
signature in the proposed HIPAA regulations. Parentheses, However, standards
for digital signatures were not retained in the final HIPAA regulations.

DR. COHN: Yes, actually, can I ask a question there? So that was – I mean,
as I remember, the HIPAA regs called for standards for electronic signature.

MS. AMATAYAKUL: Digital signature.

DR. COHN: Was it digital signature? Okay.

MS. AMATAYAKUL: I checked that out.

DR. COHN: I stand corrected, but – (laughter). Okay.

MS. FRIEDMAN: I don’t think that needs to be a parenthetical, though.

MS. AMATAYAKUL: I don’t either.

Another initiative is the Federal E-Authentication Initiative, promoted
through the Office of Management and Budget, OMB, Authentication Guidance for
Federal Agencies, M0404. This is based on the National Institute of Standards
and Technology, NIST, Electronic Authentication Guideline, SP 863.

NCVHS sought testimony from OMB and NIST to thoroughly understand the
E-Authentication Initiative and how it sets forth standards based on risks
associated with information being exchanged over the open internet.

NCVHS’ analysis of the OMB guidance is provided in Appendix B. I could show
you that later. This analysis highlights the level of security risk associated
with transmissions sent over the open internet. This was helpful, but does not
address the risks when secure protocols at other practices are used to
authenticate prescribers and dispensers as currently employed by e-prescribing
networks.

MR. REYNOLDS: What are you trying to say in that last sentence?

DR. WARREN: Do we need to add why it wasn’t helpful? Because, I mean, it
wasn’t helpful, because we are already talking about some security that is
being added there.

MS. AMATAYAKUL: It was basically helpful because we finally understood it
applied directly to the open internet.

MS. FRIEDMAN: And not to the other use cases.

DR. WARREN: Right, but you are left hanging with completing that thought at
the end of the sentence.

DR. COHN: Yes, I guess the question I have is is that it isn’t the risks
that are the issue. I mean, it was basically that – I mean, I think it did a
very good job of describing the risks. It was more the – what needed to happen
as a result of that risk assessment that we thought was – at least that I
thought was not applicable here. Am I –

DR. WARREN: My understanding from that, when we finally got clarity on
this, was that that whole OMB guideline was for, if you were conducting
interactions over the open internet, these were the things you would need to
consider. However, we are already looking at things that are done by VPN and
SXL. So there’s already a level of security that is there. So, now, we have to
consider what the risks are. So I mean, there’s –

DR. COHN: Okay. Well, let Karen first, since she has been patient.

MS. TRUDEL: Yes, I just want to follow up on that. I think the thing that’s
left out is that there’s already been mitigation of the risk.

DR. COHN: Yes, maybe that’s what I’m trying to say is is that it is more
like what happens as a result of those risks.

Jeff, I’m sorry.

MR. BLAIR: Yes, I was going to weigh in supporting Simon’s thinking on
this. I think that the discussion or the matrix that OMB/NIST offered, in terms
of what are the risks, based on the applications, you know, the type of
information you had, it was helpful to go ahead and see the way they viewed the
risk in terms of potential privacy violation, personal healthcare information,
those types of things, but then it got to the area of what do you do to address
the different levels of risk, and, there, Simon is saying that is where we felt
that, since it is not an open network, that their solutions were less relevant.

MS. FRIEDMAN: You are saying that the risk assessment is a good thing to do
in any case, but the applicability to the environment –

MR. BLAIR: Right. The methods to address the level of risks wasn’t relevant
to the e-prescribing environment we face, but the analysis of what the risks
are, that was, I think, useful information for us.

DR. COHN: Yes.

Margret, do you want to read what you have or are you intending to work on
that?

MS. AMATAYAKUL: Well, I’d like to work on it some more. What I’ve got so
far is it was helpful to understand that this guidance addressed risks for open
internet, but does not address that there has been mitigation of the risks when
secure protocols and other practices are used to authenticate prescribers and
dispensers as currently employed by e-prescribing networks.

I don’t think that still says it quite right, so I’ll work on it.

MS. FRIEDMAN: It’s not just authentication. It’s the transmission over the
secure networks, the secure connections.

MR. BLAIR: I kind of felt like Simon and I were thinking a little
differently, Margret. It was helpful because it identified the levels of risk,
based on the information that needed to be protected. That was the part that
was helpful.

However, its value was limited because it didn’t apply to the business
environment that we are looking at, because most of that business environment
is over the e-prescribing networks, not the open internet.

MS. AMATAYAKUL: All right. I’ll work on it some more.

NCVHS also heard testimony and studied guidance and standards developed by
the Electronic Financial Services Council and from ASTM International –
Committee on Health Informatics.

DR. WARREN: Do we want to say what they gave testimony about?

MS. AMATAYAKUL: I’m sorry?

DR. WARREN: Well, it just said, we heard guidance – we heard testimony for
guidance and standard development, but it didn’t say about what topic. I mean,
do we need to say that about what we heard or –

MS. AMATAYAKUL: On PKI, digital signature. Also heard testimony – guidance
and standards on digital signature? Okay.

MS. AULD: But even go a step further. Does it support what you said in the
previous paragraph or say something different or what?

MS. AMATAYAKUL: I think it was just to be complete that we heard testimony
from these groups. I could add something, if you want.

MS. FRIEDMAN: Also, in the appendices are we going to list the testifiers?
Didn’t we list them all –

MS. AMATAYAKUL: Yes –

MS. FRIEDMAN: – at some point in the other letter?

MS. AMATAYAKUL: Right. Yes.

So do we need this? Are you saying we don’t need it or –

DR. COHN: I don’t know. Actually, I was trying to remember back to their
testimony, and I thought – now, I can’t speak for the Electronic Financial
Services Council, because it was a couple of hearings ago, but I know that ASTM
talked about e-signature as well as digital signature, and so I’m – Did the
Electronic Services Council talk about both or was it just –

MS. AMATAYAKUL: E-sign.

DR. COHN: So that was electronic signature, then.

So I guess I’m trying to figure out why we are referencing them talking to
us about digital signature and PKI.

SPEAKER: Because they did.

DR. COHN: No, I think they only talked about e-sign and electronic
signature.

Well, Lori, can you help us with that one?

Well, no, I know ASTM talked about digital signature.

MS. FORQUET: Yes.

DR. COHN: I was actually referencing the Electronics Financial Services
Council, if they were talking about e-sign.

MS. FORQUET: Yes, but – Lori Forquet with ASTM – I would also like to
reiterate that ASTM gave testimony on electronic signatures, digital signatures
and PKI.

DR. COHN: And PKI. Okay.

MS. FORQUET: And network, internet, intranet security.

DR. COHN: Okay.

MS. FRIEDMAN: Take Electronic Financial Services Council out of there and
put them in a separate sentence.

I would take out the Electronic Financial Services Council from that
sentence and let that sentence reflect the range of testimony we heard from
ASTM, and then give the Electronic Financial Services Council some sentence.

DR. COHN: That’s a great solution.

MS. AMATAYAKUL: I didn’t hear the last part.

MS. FRIEDMAN: Move – eliminate the Electronic Financial Services Council
from that sentence and give them their own sentence.

MS. AMATAYAKUL: Okay.

DR. COHN: Harry, it sounds like you have a –

MR. REYNOLDS: Yes, when you think of this section, aren’t we really just
basically trying to say we heard from X amount of testifiers about the possible
use of PKI and digital signature and whatever else it says, and, basically,
isn’t that all we are trying to say here? Because, later on, we have
recommendations and whatever we heard or didn’t hear is going to be distilled
into that. So I’m not sure we couldn’t say this a whole lot –

MS. FRIEDMAN: I think it is important because of OMB and NIST, you know,
who they are and where they sit, that we explain what they said and why we
thought it was helpful, but not 100 percent on point.

MR. REYNOLDS: But in the rest – well, just one more, but in the rest of the
letter – this is the first one where we kind of take a stand. I thought –
usually, we would take our stand in the recommendation. So this would say we
heard from everyone and how it related, didn’t relate, because, otherwise, we
have made a recommendation that none of these are –

MS. AMATAYAKUL: All we are saying here is that we also heard testimony from
these groups that helped our thinking, period.

MR. REYNOLDS: That’s – and period. That’s –

MS. AMATAYAKUL: And maybe we don’t even need to reference them here, if we
have them in the list.

MR. REYNOLDS: I understand what Maria is saying. I just – this was helpful,
but does not address the risks. I mean, we are taking a stand. We haven’t
tended to take a – Now, maybe find on recommendation that the reason we didn’t
do something was that it didn’t –

MR. BLAIR: Harry, you kind of caught us, because in this part of the
letter, we really didn’t want to – we wanted to try to avoid conclusions. We
wanted to indicate, up front, in the letter who we heard from and give a
general background and then save our observations and recommendations for the –
you know, the last part of the letter, and we violated that a little bit here,
and let me tell you why and then you can decide whether or not it should stay
or it should go or we should do something different. Okay?

The reason that that statement is in there about the NIST/OMB piece in the
appendix, which we really thought was important and needed to be in the
appendix, because it was the only assessment of what the risk of the
information is, and, I’ll tell you, a number of us who looked at that looked at
it two to three times before we realized that, oh, this is for the open
internet, not for the environment we are generally looking at, and so we
thought we would point that out, so that people didn’t go looking at that chart
and misinterpret it. So maybe we overreacted to the misinterpretation, but that
was why that phrase was in there.

MS. FRIEDMAN: I think this guidance to several potential audiences of this
letter, having this there, instead of moving some of this to the back is – this
is probably a good place to have it. We considered this, thinking about some of
the potential audiences for this letter and how if we put – moved this to the
recommendations or something in the back that it might get missed somehow.

MS. GREENBERG: Well, couldn’t you keep up front that you had heard from
them and – and just factual, and, then – I don’t know. I haven’t seen the
observations yet or recommendations, but it may fit in somewhere there that –
you know – that the NIST and the OMB standards – or actually to just say that
their standards were for the open internet, and then you could say somewhere in
the observations that a lot of the security issues raised by the NIST or the
OMB standards have been mitigated by the e-prescribing networks, rather than
putting it up here, but there is a problem with putting it up here, I think.

DR. COHN: Well, you know, I actually – I guess I disagree. I mean, we can
see how it plays out as we go forward, because this is not making a
recommendation. This is just an observation that we investigated this, thought
initially it was going to be helpful and then decided it really didn’t because
it didn’t deal with this dimension. We aren’t giving a positive recommendation
about what it is we are doing, and I don’t know that we’d want in our
recommendation to say we investigated the OMB information, didn’t find it to be
helpful, because – That just doesn’t seem like an appropriate recommendation.

MS. GREENBERG: Well, I am not suggesting it as a recommendation. I was
thinking maybe it belonged in the observations, but I guess –

DR. COHN: Well –

MS. GREENBERG: – the reason we got into this was because you are commenting
on some testimony and then not commenting on other testimony, and I suppose you
could just say, if it did, in that second paragraph that you heard from these
other – heard testimony that also informed our recommendations or something. If
they did.

MR. BLAIR: Could I react? Because I think Harry’s reaction and Marjorie’s
reaction, I think, have convinced me that we should be consistent, and that
appears as if it is a value judgment. It looks to people as if it is a
recommendation, and maybe we don’t need it in the body here. Maybe as long as
that is clear when we introduce it in the appendix, maybe that would be
adequate. No? No? No? You feel differently, Simon?

DR. COHN: Well, I guess I am going to have to stop for just a second,
because I am not even sure what we are arguing about right at this moment. So –
the blue words.

I think something like that needs to be there. So we can decide in further
versions. This is the first read-through. We can decide if somewhere after we
get into our recommendations whether it might fit better somewhere else, but I
don’t think – I think we agree with the gist of what that is saying, and if we
can decide there’s a better place, that’s fine, but I can’t think of any other
– If we don’t say something like that, then I’m left with, well, why are we
even – why do we even have the section?

Harry.

MR. REYNOLDS: Not on your – but in the same section, but not on your
comment.

DR. COHN: Okay. Please.

MR. REYNOLDS: I’m fine with what you said.

I actually found the OMB and the NIST helpful – very helpful – and will be
helpful as we try to deal with the difference between true PKI and true what we
want to come up with, and so that level of standards that they had where, if we
recommended two-level authentication and – thing else, I think that was really
the basis for it. So I would – if it is staying, I would like to strengthen it
that although it didn’t pertain, it sure did give us a structure.

MR. BLAIR: Uh, that’s a good way to –

MR. REYNOLDS: It sure gave us a structure to think.

MS. FRIEDMAN: A useful framework for analysis.

MR. REYNOLDS: That’s perfect. Okay. Because I don’t want to shortchange
that either. I think that whole discussion really put us in context as to how
if we couldn’t absolutely do PKI directly, we could recommend something else
that would still get the same job done. So I’m good.

DR. COHN: Yes. Harry, thank you. Jeff’s nodding his head. That’s a good
sign.

Other comments?

MS. AMATAYAKUL: Okay. The last section in this background is industry
experience.

Testifiers that currently use PKI are experimenting with it in environments
that are relatively limited in scope and, in general, use only certain aspects
of PKI. These testifiers describe considerable overhead in their implementation
of PKI and the lack of PKI product interoperability.

DR. COHN: Could I ask a question here? Is it testifiers that currently use
PKI in healthcare? Is that what we are talking about? Okay. Good. Because I – I
actually think, in some areas, it is actually pretty far advanced.

MS. AMATAYAKUL: The next section – next paragraph is, Testimony from the
e-prescribing networks, software developers and prescription transaction
standards developers express concerns that the use of PKI would impair the
ability of the e-prescribing networks to reformat or update the version of the
prescription, if necessary, before it is sent to the dispenser, create severe
performance problems, due to the overhead of managing PKI across disparate
entities, impose significant additional costs in an industry which is
struggling to establish an adequate business case for e-prescribing, and delay
the adoption of the use of e-prescribing as a result of the cost and burden to
install and maintain a system.

MS. FRIEDMAN: Do we want to put in there something about the complexity of
the process? I’m not sure if that second bullet is –

MR. REYNOLDS: Yes, I agree.

MS. FRIEDMAN: – the place to do it, but say, create severe performance
problems due to the complexity and overhead?

DR. COHN: Now, let me ask another question here – and I am trying to
remember all the various testimony, but I am trying to remember that – I don’t
think any one – another piece that I heard from the industry was is that even
despite all the technology and all of that, people – no one thought it was
necessary and that they had no evidence of any –

MS. FRIEDMAN: System failures?

DR. COHN: – either system failures or abuse going on.

MS. FRIEDMAN: This is – what problem are we trying to solve? That today’s
environment has levels of security –

DR. COHN: Yes.

MS. FRIEDMAN: – and authentication already built in and millions of
prescriptions are flowing through the pipes without problems.

DR. COHN: Yes –

MS. FRIEDMAN: This is – actually PKI is not ready for prime time, but it is
actually overkill.

MS. AMATAYAKUL: Something to the effect of PKI would not add any greater
level of security than is called for at this time or something to that effect,
is that what you are sort of saying, Simon?

MR. BLAIR: I’m not sure –

MR. REYNOLDS: I couldn’t say that.

MR. BLAIR: – that that –

MR. REYNOLDS: I couldn’t say that.

MR. BLAIR: I’m not sure that that was clear. I think that there were folks
that were – I think the items that we have mentioned there were the reason why
PKI would not be appropriate to be imposed now – in the business case, the
performance, the transactions – but in terms of whether or not it would provide
better authentication and security, I don’t think we heard people saying that
it wouldn’t do that.

MS. FRIEDMAN: I don’t think that was the point. I thought the point was
that things – that while, yes, it provides greater security and authentication
and all that good stuff, that things are working very well now, that the –

MR. BLAIR: Yes, but – maybe – my thought is that what are we measuring that
– I don’t know.

MS. FRIEDMAN: We did not hear anything about the sky is falling, in terms
of network failures and bad things happening out there.

DR. COHN: Yes. Ross, do you have help here?

MR. MARTIN: Ross Martin with Pfizer.

I think a lot of what people were saying was about the incremental benefit
does not justify the additional costs and burden that is imposed by PKI. It is
not that it doesn’t authenticate more robustly than other solutions. It is just
very expensive and very costly to – it is very intensive compared with what is
being done now, and is there a burning need for more? Are we doing okay with
what we have? I think that’s the – and what you said before about it being –
eventually, sure, PKI will probably be done for everything, or something like
it, but do we want to start that way and risk reducing it to option entirely?

MR. BLAIR: Would it be possible if this could be something that we make a
note and come back to whether we really want to make a statement like that here
in this background section, after we have gone through the rest and the
observations and the recommendations? Because I think we touch on these as we
get down further.

MS. AMATAYAKUL: I have suggested one way of not making it a separate
bullet, but saying, where we have imposed significant additional costs, we
could add, With insufficient incremental value in an industry which is
struggling to establish an adequate business case for e-prescribing?

DR. COHN: Yes. You know, I guess I think it is a separate issue, and the
only reason I think about that one is is that I think – you know, the only
thing you are describing up here is a cost argument, and I think that there is
actually a sort of a quality piece here that where there’s a question about
insufficient incremental value, knowing that, quite frankly, that that – the
value of equation is not just the business case for those trying to establish
e-prescribing. Obviously, it effects the prescriber, the dispenser, et cetera,
et cetera, and they are, obviously, only pieces of the whole process.

MS. FRIEDMAN: Everybody in the chain.

DR. COHN: Yes.

MR. REYNOLDS: Didn’t we really hear that in this multi-step process that is
called e-prescribing most of the people that felt that PKI was the answer found
it very helpful in a point-to-point situation, not in a multi-step process that
this is – that is what I thought I heard.

MS. FRIEDMAN: There weren’t – It’s still in its infancy, actually.

MR. REYNOLDS: Yes, but, I mean, if you believe in PKI, once it leaves me,
somebody else opens it, I know they did it, and we are talking from the
prescriber to the dispenser, and if you’ve got five or six hand-offs, PKI, as
it is explained in the normal world, blows up. It’s great point to point, but
it doesn’t necessarily work through all those steps without a whole lot of
overhead and a whole lot of burden.

DR. COHN: Lynne, are you going to help us?

MS. GILBERTSON: Lynne Gilbertson with NCPDP.

Your comment, Simon, triggered something.

There were comments about the current environment, the actual prescribing
event. There were not cases of fraud and abuse happening, and so it gets back
to – it doesn’t have to do with cost. It doesn’t have to do with system
technology. It has to do with the fact that are people trying to skirt the
system or do something illegal?

There was also the concepts about no matter how tight you made the
electronic and the computerized sections, the fact that the human equation was
still involved was where the fraud and abuse was going to happen, a doctor or a
pharmacist who is trying to do something illegally, and then there were
comments also made about how there are statutes and regulations and laws to
cover if the human element tries to do something illegal. So I wonder if that
needs to all be included as part of where you were headed, and those are just
some of the things that kicked off in my mind when you said that.

MR. BLAIR: I could feel comfortable with adding something the way you said
it is there’s no evidence, to date, using the e-prescribing – well, there’s no
evidence today that there is fraud or abuse or that current methods are
inadequate. There’s no evidence at this point. If that statement is made, then
–

MR. BROOK: Yes, Richard Brook again.

I testified several weeks ago that we had done over 19 million transactions
at this time, and we have never had a situation where somebody has broken into
the network.

And to what Lynne said, we had a situation what took place at a physician’s
office where somebody illegally entered prescriptions into the system, but
going through our point-to-points, our encrypted lines, the VPNs, whatever the
methodology is, with direct connects to some of the pharmacies, we have never
run into a situation where that line has been broken where somebody was able to
get in there and do something illegal.

MR. BLAIR: Would that make – you know, would that cover, Maria and Simon,
the thought you were trying to add?

DR. COHN: I think that this begins to help a lot. It just adds another
dimension to the argument, which, I think, is, in some ways, as compelling as
the cost and hassle factor.

MS. FRIEDMAN: And maybe not for the letter, but just, again, to note that
PKI does not solve what Lynne very nicely called the human equation, what I
call stupid human tricks – (laughter) – where people take their key to the wall
and stuff like that. You know, we keep doing the same things all the time.

MS. AMATAYAKUL: I need to work on this, because the beginning of this is
PKI would – So I have to –

DR. COHN: Yes, well, maybe it’s not a bullet. Maybe it’s an additional
sentence.

MS. AMATAYAKUL: Yes.

DR. COHN: Okay. And I guess we’ll have to sleep on that, but if there’s – I
mean, I was trying to think if there’s anything more that we –

MR. REYNOLDS: We go back to the first –

DR. COHN: First bullet?

MR. REYNOLDS: Yes.

DR. COHN: Sure.

MR. REYNOLDS: No, second one. Creates severe performance problems due to
the complexity and overhead of managing across disparate entities. Can it be
done over these disparate entities for e-prescribing?

DR. COHN: Well, sure, it could be, as long as you didn’t touch anything.

MS. AMATAYAKUL: What’s the –

MR. REYNOLDS: Yes, PKI, really – I sent it as a doctor. Somebody passes it
along, and when the pharmacy opens it, then it is still my signature, and each
of them – well, never mind. I won’t belabor it, but I –

DR. COHN: Yes.

MR. REYNOLDS: I still struggle that that can even work and I worry about
that –

DR. COHN: You know, there is one other thing that – Harry talking sort of
led me to think about this one. You know, I think the other point we are trying
to make on all of this is – has to do with sort of time, I mean, some of the
stuff today versus some point in the future where things are more robust and
all of this, and I think we need to somehow – I don’t know that our position in
the PKI would never be applicable in healthcare or never applicable to
e-prescribing.

MR. BLAIR: Matter of fact, let her go on a little bit more, because she is
going to get to some sentences –

DR. COHN: Oh –

MR. BLAIR: – you know, today.

DR. COHN: Sure, please go ahead.

MR. BLAIR: Yes –

MS. FORQUET: Lori Forquet with ASTM.

Just want to make sure you are tying those bullet points, and when you
separate that last one from one of the bullet points it is bound to that this
is the testimony heard from those three parties, and please avoid statements
within there, because I don’t think there was testimony stating it won’t work
with so many pass-offs. There was concern that it would be complicated with the
pass-offs.

DR. COHN: Thank you.

MS. AMATAYAKUL: The last part of this background is just the following
recommendations: Focus on the development and adoption of industry guidelines
to ensure that health information is protected over the e-prescribing networks
and evaluation of the feasibility for adopting stronger authentication and
non-repudiation methods.

MR. BLAIR: There is a sentence there that you missed.

MS. AMATAYAKUL: No.

MR. BLAIR: No? Hum.

DR. COHN: Yes.

Now, are people generally comfortable? I mean, we obviously have a couple
more versions of the introductory material prior to recommendations.

Now, I know how strange it sounds, but I am actually going to suggest that
maybe we not get into the recommendations or begin to brainstorm on them today,
and that we, instead, start working on those tomorrow. Is that – I mean, I know
it’s unheard of that the chair lets us out 14 minutes early. So is everyone
okay with that?

What we will do is to – and everybody is like – doesn’t know what to say.

SPEAKER: We’re shocked.

DR. COHN: We’re shocked. Exactly.

I just want to remind everyone that we will reconvene at 8:30 tomorrow
morning.

I want to thank Maria, Jeff, Margret and others for their help this
afternoon, and the meeting will be adjourned until tomorrow morning.

(Whereupon, the meeting adjourned at 4:47 p.m., to reconvene the following
morning).