[This Transcript is Unedited]



Meeting of Subcommittee on Standards and Security

August 18, 2004

Hubert H. Humphrey Building
Room 705A
200 Independence Avenue, S.W.
Washington, D.C. 20201

Proceedings By:
CASET Associates, Ltd.
10201 Lee Highway, Suite 160
Fairfax, Virginia 22030
(703) 352-0091


Welcome and Introductions

Subcommittee Discussion

P R O C E E D I N G S  [8:57 a.m.]

Agenda Item:  Welcome and Introductions

DR. COHN:  I want to apologize to everyone.  We are running a little late, in the process of printing out some documents to use as references as we start the work of the day.

Good morning.  I want to call this meeting to order.  This is the second day of three days of meetings on the Subcommittee on Standards and Security of the National Committee on Vital and Health Statistics.  The committee is the main public advisory body to the U.S. Department of Health and Human Services on national health information policy.

I am Simon Cohn, Chairman of the subcommittee and National Director for Health Information Policy for Kaiser Permanente.  I want to welcome fellow subcommittee members for the second day, HHS staff and others here in person.  I also want to welcome those listening in on the Internet.  As always, I want to remind everyone to speak clearly and into the microphone, so we can hear you in this room and so those on the Internet also can hear us.

Today and tomorrow, the subcommittee will deliberate in open session, discussing observations and draft recommendations we may wish to communicate to the full committee at their September meeting.  These are intended, assuming we reach consensus agreement on these observations and recommendations, to be a key part of our first set of recommendations, in fact, the recommendations to the NCVHS and for a set of recommendations on e-prescribing.

I think as we have commented numerous times, this is the first set of recommendations, and we intend to have additional hearings this fall, winter, and we will talk about those dates in just a minute, coming forward with a second set of recommendations for March for the full NCVHS committee to consider at that meeting.

As always, I want to thank Jeff Blair, our Vice Chair, for his leadership in moving this agenda forward.  I especially want to thank him for version ten of the work plan.  Hopefully we will be able to stop in the double digits on the work plan at this point, but it is hard to know.  I also want to thank Maria Friedman for her help in terms of putting all of this together.  I want to thank Margaret A., and I will stop trying to pronounce your last name, for being our consultant on the project, and for your hard work in terms of helping us put all of this together.

I want to emphasize that this is an open session.  For each of the major topic areas, we will have subcommittee discussion.  I want to keep things balanced, but we do want to provide an opportunity for those here, especially the experts who are joining us today, to make comments as is appropriate.  We would ask any comments to be short and to the point, because we do have a fair amount to cover today.  As the Chair, I will exert some prerogative as we try to manage that, to make sure that we do move through the agenda, and I would ask for everybody’s forbearance as we do this process.

We will also hold open sessions, both at the end of the morning and at the end of the afternoon for more general comments  that those in attendance may wish to make.

Finally, for those on the Internet, we of course welcome e-mail and letter comments on issues coming before the committee.

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 issues coming before us today, would you please so publicly indicate in your introduction.  Then we will handle some housekeeping issues, and then by that time we will move into discussions on observation leading to recommended actions.


MR. BLAIR:  Thank you.  Jeff Blair, Medical Records Institute, Vice Chair of the Subcommittee on Standards and Security.  The organizations that I am a member of are ASTM, AMIA, HL-7 and HIMS.  The only thing that might be a potential conflict of interest might be the discussion of ASTM CCR within our topic area here.

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

DR. HUFF:  Stan Huff, member of the committee and subcommittee.  I am with Intermountain Health Care and the University of Utah in Salt Lake City, and I have a potential conflict of interest with HL-7.  So I can be a resource for questions of fact, but would need to recuse myself from votes that would adopt HL-7, et cetera.

DR. FITZMAURICE:  Michael Fitzmaurice, Agency for Health Care Research and Quality.  I am liaison to the full committee and staff to the Subcommittee on Standards and Security.

PARTICIPANT:  — federal EPA, staff to the subcommittee.

MS. BURKE-BEBEE:  Susie Bebee, ASPE, staff to the subcommittee.

MS. AMATAYAKUL:  Margaret Amatayakul, independent consultant assisting the committee.

MS. AULD:  Vivian Auld, National Library of Medicine, staff to the subcommittee.

DR. GREENBERG:  Marjorie Greenberg, National Center for Health Statistics, CDC, and Executive Secretary to the committee.

MS. TRUDEL:  Karen Trudel, Center for Medicaid and Medicare Services, staff to the subcommittee.

MR. REYNOLDS:  Harry Reynolds, Blue Cross Blue Shield of North Carolina, member of the committee.

DR. WARREN:  Judy Warren, University of Kansas School of Nursing, member of the subcommittee.  If we have any votes about HL-7, I’ll need to recuse myself, as i am co-chair of the Patient Care Technical Committee.

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

(The remainder of the introductions were performed off mike.)

DR. COHN:  I want to thank you all for joining us.  I do obviously need to disclose, in case CPT comes up, that I am a member of the CPT editorial panel.  I think that is a long way from e-prescribing, but you never know, so I will publicly disclose that, and will recuse myself from any discussion or votes related to that.

Before we move into e-prescribing, I think we are still waiting for the document to print out.  I do want to announce some hearing dates and meeting dates for the subcommittee.  As many of you are aware, in the midst of everything yesterday, we were also polling subcommittee members and staff for their availability in 2005.  We will be sending out e-mails about this also.  But I thought I would at least announce it and let you get it on your calendars.

What we have established is a January set of hearings, which will be on the 19th and 20th of January.  I did observe that all of you polled for early February dates.  Even though I am hoping that we are not going to need to have hearings on those dates, I am for the moment going to ask you to hold those dates, given that depending on how many issues the subcommittee decides it needs to deal with relating to e-prescribing for the March second set of recommendations, we may need that time.   So what I am going to ask you to do is to hold February 1 and 2 as additional dates, just in case.  The hope is that we will be able to release those dates in the near future, but at least to allow some flexibility and to avoid a number of three-day hearings, which might be otherwise the result.

We have also established dates in April, for April 6 and 7.  We attempted to poll for May and early June dates, and at this point we are unsuccessful.  I will be establishing new dates in polling the subcommittee for May to early June dates, but we have established some July dates, which will be July 26 and 27.

So anyway, those are the dates.  The intent is to move back to a two-day format if at all possible.  I see some of you by your facial expression are being hopeful on this one.

Now, I think I will turn it over to Jeff for some introductory comments.  I do want to note that there is a request for some — let me see if I can get that sentence out; it is still early in the morning for me.  HL-7 has requested that they be able to clarify some of their testimony from yesterday, so given that we are going to be talking about message formats as the first order of business, I am going to suggest that after the setting of the day that we allow them to make whatever clarifications they need to.  They will be proceed on by a discussion of message format standards.

So Jeff, would you like to start off the morning?

Agenda Item: Subcommittee Discussion

MR. BLAIR:  You bet you.  Good morning, everyone.  For those of you who have been with us steadily, I commend you for your loyalty, fortitude and steadfastness.

I think that many of you have observed, or many of you who may be new before the subcommittee have realized, that there is a lot more to e-prescribing than might have been apparent when we first took a look at this activity.  We have had to look at a number of dates.  We have had to add additional testifiers.  We have had to examine additional issues.  As Simon pointed out, there are only so many things that we will come to consensus on by September 1 and 2, and some of those things we already know are going to be deferred for either hearings or testimonies in the fall, October, December, January, hopefully not February, but maybe.

The only thing I would add at this point is to try to help you put today’s event into context.  We have gone through hearings in March, May and July, where we heard folks identify what standards they are using, what standards are available, what gaps there are, what limitations there are.  After our July hearings, we then tried to share that with the developers of the message format standards, and the identifiers and the terminologies, to ask them to what extent can they address the limitations in gaps.  We had wonderful responses from everyone.  I think the cooperation that we have received from the industry and from Health and Human Services and everybody involved has been fantastic.  By August 13, we received written responses which Margaret Amatayakul has compiled over the weekend.

Then in addition, we focused on some of our primary issues yesterday in terms of how those limitations and gaps could be addressed, and whether they can be addressed in the near term or long term, near term being defined whether those could be addressed in time to be a part of the demonstration projects that would begin January 1, 2006.

As far as today is concerned, Margaret worked last night to try to see if she could organize some of the findings and observations, so that we could try to see if we could expedite and organize the way we deal with our observations and recommendations today and tomorrow morning, if it goes to tomorrow morning.

As Simon indicated, we are going to try to stay focused on these things, but there is time scheduled at different times for open mike, and our Chair, Simon, has been very, very good at discerning when additional comments may be necessary, although from a time standpoint, we will have to manage that.

So those are my only comments.  Simon, let me turn it back to you.

DR. COHN:  As I said, I think what our intent was — we know there are three main issues:  message format standards, terminologies and identifiers.  I will defer to Margaret whether we handle second the identifiers or the terminologies, but I do think that we need to start out with the message format standards, just because that feels like the most concrete of things, but that seems like a very good place to start.

MR. BLAIR:  We might mention what we are doing with important issues and the history.

DR. COHN:  Jeff is reminding me that we will also be handling recommendations and observations and recommendations relating to important issues and other policy considerations probably following all of that.  So we have no lack of issues to go through.

Now, given that we are talking about message format standards, I have had a request for — Scott Robertson, you needed to clarify some of your testimony, or wanted to clarify?  If you want to come to a microphone and introduce yourself.

MR. ROBERTSON:  Scott Robertson, representing HL-7.  Again, I should state this again, for full disclosure, I am with Kaiser Permanente, but I am speaking on behalf of HL-7.

It was brought to my attention yesterday afternoon outside of the meeting room that there may have been some confusion between my testimony and responses and questions in terms of, are we recommending a standard be chosen or two standards be chosen.  My written testimony has, NCPDP and HL-7 should be accepted and acknowledged as standards.

In response to questions, what I recall saying, which was not written down, was that yes, the selection of a standard in a domain is something that we can support.  The problem with that, and where the ambiguity comes in, is that the concept of a domain is not clearly defined.

In reference to that, I contacted the HL-7 chair, Mark Schafferman, to discuss what HL-7 could say to try to clarify this situation, what we could suggest to try to clarify this situation.  I have spoken with my NCPDP colleagues to make sure that we can both support this position.

What we are basically coming down to in HL-7 is recommending that the committee state that the standard for e-prescribing communications, new prescriptions, refill requests, orders, all of those communications between the prescriber and the pharmacy, be stated as, they should be conducted in NCPDP Script, with specific exceptions.

MR. BLAIR:  Can I help with one piece?


MR. BLAIR:  When you say with the pharmacy, are you talking about retail and commercial pharmacies now, or institutional pharmacies?  That is a very important clarification.

MR. ROBERTSON:  That is part of my clarification.  That is the big point, because that is where the big gray area is.  We think we have come up with what I would call language, but I am not proposing to give you something like that  I am giving you the concept at this point.

If you want me to give it as language, I can do that, but the concept is, exceptions for HL-7 are allowed, but not required, but they are allowed in existing implementations that already have these e-prescribing functions working with HL-7 messages, basically a grandfather clause, let people continue to do what they are doing.

The exception that Jeff was specifically asking for is, in new installations — and I just want to get the language that I had so I can say it reasonably — new implementations in which the prescribers and pharmacies are closely associated in a health care organization.  This could include where the prescriber and pharmacy are in the same physical location, or where they are in different locations, but within one health care organization.

The test for health care organization is that there is a business — the business is recognized and has an organizational structure that at some point in that business structure, the two players in this case, the prescriber and the pharmacy, can be considered to be working for the same organization.

Now, there are strange things that come up in that.  I am hoping that that will help clarify what is an integrated health care program, for want of a better term.  There are people like the VA, that have pharmacy as part of their operation.  There are a number of others, Cleveland Clinic Foundation.  I could run off a couple more.  I’m trying to avoid going into specifics in terms of organizations, but the idea being that there is some recognized level where the two actors are considered part of the same organization.

I also wanted to include an exception.  There may be cases where more than likely, the prescriber can talk HL-7, because they have some software that can speak HL-7, and the pharmacy, as most pharmacies, speaks NCPDB Script.  I don’t know if it needs to be specifically called out, endorsing the fact that bridging technology, the mapping that we are doing, is a relevant way of addressing that situation if the prescriber does not have the capability, resources or control over their software, that they shouldn’t be penalized for the fact that they aren’t able to conform, that there is technology that can allow them to adequately communicate with the pharmacy.

One last point, which I concede might be a very contentious point, would be the idea of allowing an exception for cases in which — where new novel installations are being put in place that are going beyond the information capabilities that might be available in Script.

I can’t really give you an example of this.  This is really an idea that came from Mark Schafferman.  We don’t want to do anything right now that is going to box us all in.  We don’t see the selection of NCPDP Script as a limiting factor, as the primary communications protocol.  But on the other hand, we don’t want to come up to a situation where we have said, we are going to do X, and there is no way to adequately evolve beyond that.  Whether that needs to be explicitly stated in the recommendations, whether that needs to be specifically stated in the regulations, or if it can just be handled as an exceptional case pilot project, is really up to the committee.  I’m just bringing the point forward.

I really appreciate the opportunity to provide what I hope is just a clarification.  I am not trying to change the direction that I think the committee is going in.

Thank you very much.  Any questions?

DR. COHN:  Michael.

DR. FITZMAURICE:  A clarification.  You are asking for an exception that is a grandfather clause.  So are you saying that as of a date certain, probably to be determined by the Secretary or something, HL-7 would recommend that beyond that date, all communications between the prescriber and the pharmacy for purposes of e-prescribing be done in Script?

MR. ROBERTSON:  With the exception of these defined, or however we defined health care organizations that hold pharmacies and prescribers together, or if the term intra-institutional works.  I’m not really concerned too much about it, just the idea that there are these entities where they have a great deal of communication already set up or supported, and the addition of e-prescribing is a minor addition, but if you have to go to a different protocol, that might require a lot more resources.  Not an infinitely lot more resources, just more resources.

DR. FITZMAURICE:  Is there a time certain when you anticipate a single standard of communicating between a prescriber and the pharmacy?

MR. ROBERTSON:  I would really hope — sometimes I think too far out in the future, but there will be a time when it is not a question of NCPDP or Script or whatever.  There will be a time when we have effectively evolved into one standard.  It may not be HL-7, it may not be NCPDP.  It may be one or the other.  It may be something that looks a whole lot like both of them squished together.

Yes, eventually I think there can be one standard without any exceptional cases.

DR. FITZMAURICE:  Thank you.

DR. COHN:  Harry.

MR. REYNOLDS:  Scott, when you talk about this exception, it is only in communication within that entity.

MR. ROBERTSON:  Within one institution.

MR. REYNOLDS:  Let’s take an example.  A person gets discharged from a hospital, goes to the hospital-based pharmacy, which is actually an outpatient prescription, as they are leaving, in many cases.  That communication could be from the hospital to that pharmacy, using HL-7 in your example, but when that pharmacy communicates to the PBMs, it is under whatever standard is set up outside of that.  So once they branch outside of the institution, it follows this.  It doesn’t continue to be a different kind of transaction.

MR. ROBERTSON:  Yes, that is correct.  I’ll take that one step further.  If for some reason the prescriber, probably at the request of the patient, were to be sending that prescription not to that hospital’s outpatient pharmacy, but to a more recognized retail pharmacy down the street, the expectation is, that communication would be an NCPDP.

MR. REYNOLDS:  Thank you.

DR. COHN:  Michael.

DR. FITZMAURICE:  In your recommendation for HL-7, do you recommend that the data elements be identical between the two standards, and you are talking just about the message format?

MR. ROBERTSON:  Within the context of e-prescribing, the data requirements can be supported by either standard.  That is what the mapping effort is looking at, making sure that we know that ORC field four is NCPDP segment something-something.  That is the formal mapping that will be endorsed that both the SDOs is working on.

If there is additional information that can be put into NCPDP Script or can be put in an HL-7 message, that would be useful to somebody that can understand it.  But I can’t expect that the retail pharmacy is going to be able to deal with something that is not able to be carried in the message.  So it would be limited to that subset of the data elements that are relevant to the communication.

DR. FITZMAURICE:  I think that is a yes.

MR. ROBERTSON:  Sorry, yes, it is a yes.

DR. COHN:  Susie.

MS. BURKE-BEBEE:  Just to clarify that, are you saying that as a result of the mapping, there may be a subset of data that would be standard?  I’m sorry?

DR. GREENBERG:  Couldn’t hear you.

MS. BURKE-BEBEE:  I’m sorry.  As a result of the mapping, are you saying that there would be a subset that would become standard of the two joining together?

MR. ROBERTSON:  I really shouldn’t present myself as an expert on NCPDP Script.  I’m limiting myself just to Script at the moment.  But I believe we will be able to map everything that Script has defined data elements for into the HL-7 standard.  It would be incumbent on anybody communicating with a pharmacy to  support those data elements.  If it is in Script, it is pretty straightforward; it is a standard.  If it is in HL-7, the only thing that that HL-7 — the required communication would be based upon that minimal mapping.

DR. STEINDEL:  Scott, I just wish to point out that this mapping goes a little bit beyond data elements, that there is a requirement here to make sure the two standards stay in synch in terms of the code sets that they actually use and the versions of the code sets as well.

MR. ROBERTSON:  Yes, that is true, and that is part of our effort.  I have taken a simplistic way of describing the mapping at the moment.  Thank you for pointing that out.

DR. COHN:  Michael.

DR. FITZMAURICE:  I think I heard in your description of changes in the testimony, you would also not like to see a prohibition with experimenting against new functionality where it might be beneficial to both sides.  You didn’t say it in those words, but I —

MR. ROBERTSON:  I like your words a little better than mine.  Yes, that was my intent.  I like the way it sounded, the way you said it.

DR. FITZMAURICE:  Shucks, ‘twernt nothing.

DR. COHN:  May I have a comment, and Jeff also?  As I am listening to you about the grandfather piece, which I am having some concern with, but it appears to me that you are bringing up the issue that we don’t want to negatively affect or disadvantage early adopters.  As we move forward with this, whatever happens or whatever CMS decides, there needs to be some attention paid to supporting and enabling any transitions that might occur, or they may at least have to look at the business cases.

I am listening to this, and I’m just not sure that the answer is necessarily grandfathering in.  But we can all talk about that.

MR. ROBERTSON:  For the most part, I would think that the early adopters that have chosen HL-7 are probably going to also be in the health care organization intra-institutional exception.

DR. COHN:  Okay, so that may not be an issue.

MR. ROBERTSON:  However, that may not be entirely true. There could be some VHR vendor that incorporated an e-prescribing functionality that decided to do it in HL-7.  I wouldn’t want to exclude them unnecessarily.

DR. COHN:  Thank you for that clarification.  Steve.

DR. STEINDEL:  Simon, based on the testimony that we heard, I’m not sure if there is any significance beyond maybe single digits of messages going to retail pharmacies not associated with any type of institution in HL-7 format.  We are aware of a tremendous amount of prescription information going to pharmacies.  Like, virtually all of VA’s prescription information is going to HL-7.  I would imagine Kaiser’s might be as well.  But those are all internal.  I think they come under what you have described.  But I’m not sure any that we are aware of that is going from an institution or a prescriber to a retail pharmacy.  All the retail pharmacy chains, et cetera, in NCPDP that we have heard from have never indicated that HL-7 is being used at that level.  Are you aware of any?

MR. ROBERTSON:  Actually, that business case came forward as one of the drivers for the mapping project that we have entered into now.  The NCPDP and HL-7 have talked about doing a mapping between standards for some time, but along with MMA and the committee’s inquiries, Cleveland Clinic has a specific business case — Cleveland Clinic Foundation has a specific business case to communicate to local retail pharmacies that are not otherwise directly affiliated.

DR. STEINDEL:  Yes, but I think that business case comes on discharge prescriptions, which we are aware of is a situation that we do have to handle.

MR. ROBERTSON:  Yes.  I am not aware of an individual practice physician communicating to Rite-Aid pharmacy.  I am not aware of those situations in particular.

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

MR. BLAIR:  This may be an oversimplification, and if it is, then correct me.  It seems as if the complexity in this area comes when we look at it from the prescriber standpoint, as opposed to from the dispenser.  It just seems to me that if we look at a retail or commercial pharmacy, it appears to me that everybody is accepting the fact that any prescription to a retail or commercial  pharmacy should be sent in NCPDP Script.  But what you are asking for is that if the pharmacy is in an institutional setting or part of some type of an enterprise where HL-7 is being used, you are asking that we do recognize the fact that HL-7 messages are being used within those institutional settings to an institutional pharmacy.  Is that too simple, or does that hit the nail on the head?

MR. ROBERTSON:  I think it is very close.  I still have a bit of an issue, or I think there could be an issue on the definition of institutional.  But your perspective that you have used to state this is acceptable to me.

MR. BLAIR:  You said that it was too simple.  So could you point out where it might be too simple?

DR. STEINDEL:  I think it comes to about 90, 95 percent of what we are looking at, but there might be a little bit of area that I can see that it might not fit.  I’m not sure where yet.

MR. ROBERTSON:  In Jeff’s version, you mean?

DR. STEINDEL:  Yes, Jeff’s version.

MR. ROBERTSON:  That is my —

DR. STEINDEL:  One area that I am aware of a little bit is with regard to DoD, who works with retail pharmacies, and also works internally.  Internally, their ordering is handled by whatever messaging system their computers are currently using,  which is probably HL-7.  I’m just not totally certain.

But I do know that when we were discussing drug terminologies with respect to CHI and drug messaging, DoD pointed out that they are the only federal organization right now that does use NCPDP Script in a limited fashion when they go to retail pharmacies.

MR. ROBERTSON:  So that still is consistent then.

DR. STEINDEL:  That still is consistent with what he is saying.  So I think it might fit in all cases, but —

DR. COHN:  And I think we have Lynn Gilbertson here to hopefully provide some additional clarification.

MS. GILBERTSON:  Lynn Gilbertson from NCPDP.  One of the things that I testified for is, we had quite a bit of participation in a paper that I supplied to the committee, that long term care wanted Script to be enhanced to meet their business needs.  So it is hard to say what an institution is, but I wanted to make sure we didn’t forget LTCs.

DR. COHN:  Harry.

MR. REYNOLDS:  One more clarification, Scott.  Scott, one more clarification, see if you agree.  About that institution, you mentioned an institution and communicating back and forth.  But that institution will also have to have the capability of sending NCPDP script to an outside pharmacy, because under the Medicare law, that patient can have the right to go to any pharmacy that is involved.  Do you agree with that?


MR. REYNOLDS:  So in other words, we are not just saying that the institution only communicates in HL-7.  It communicates with itself in HL-7 if the patient has decided to go to that pharmacy.  And if the patient hasn’t, then they will communicate with NCPDP Script, which is what the outside world is agreeing to.


DR. COHN:  I want to thank you for your clarification.  We should also thank Mark Schafferman for — it sounds like much of this is his clarification.

MR. ROBERTSON:  Much of this is his clarification.  I wanted to make sure that I was really speaking for HL-7 and not for my own little med info world of things.

DR. COHN:  We should thank Mark for his assistance in terms of helping with the clarification, and thank you for being with us.

MR. ROBERTSON:  Thank you for this time.

DR. COHN:  Margaret, I think we are at a point where — do you want to help lead us through some of the things that you have been capturing as we move along, and we can discuss them?  I think it is discussing point by point.  Is this the process that you and Jeff both agreed to?  Then the question is, what else are we missing, et cetera.

MS. AMATAYAKUL:  What I tried to do was capture as succinctly as I could the key observation relative to messages, terminology, identifiers and important related issues, and recorded them here on the left side, and then attempted to draft some suggestions for actions that you might want to recommend.

What I have done is raise the size of the font so you can see this here in comparison to what you have got on paper.  I also inserted rows for myself, so that as we look at a particular point here, I can be taking some notes below that.

DR. COHN:  Would you do us a favor and not move things up and down a lot?  I can’t see any reason why you have to go up and down on this one.


MR. BLAIR:  It doesn’t bother me, Simon.

DR. COHN:  I know, Jeff.  That is why I was not waiting for you to make the observation.

MS. AMATAYAKUL:  I will go as slowly up and down as I can.

Simon, I don’t know if it would be worth glancing through the document in its entirety first or taking point by point, simply because you might say something is missing, and it is actually in another point later on, to kind of scan through first.

DR. COHN:  Sure.  Why don’t you talk about the observation at that level, and then we can go back and go one by one.

MS. AMATAYAKUL:  Okay.  So the first observation related to the fact that NCPDP Script for new prescription changes, renewals, cancel and fill status are existing standards used to varying extents today.  So that is pretty widely adopted.  I just added the information that Scott gave us about the HL-7 and the business case, so I have inserted that following number one there, so I’ve got 1A and 1B here.

I have also identified as an observation that the eligibility, inquiry and response via the ASC, X-12 and 270-271 version 40-10 capability exists, but only in about 24 percent of provider organizations overall, and 45 percent are payor systems, which I got from the HINS Phoenix survey that was just published a week ago.

There are some gaps in 270-271 that Lisa provided us with.  There are some work-arounds.  Versions 40-50 and 50-10 are available to address some of these industry needs.

Formulary and benefit coverage information from payors, PBMs to prescribers is currently performed with proprietary messages, and Rx Hub is submitting its transfer protocol to NCPDP to become an ANSE accredited standard.

NCPDP Script has inquiry and response components for each function.  So for special situations in which an acknowledgement of receipt is required, NCPDP has a request for a return receipt for use by  prescribers.

NCPDP has initiated work on its directory that would identify prescribers, nursing facilities and pharmacies that are able to accept electronic prescribing transactions.  ASC X-12 also has the 274 provider roster, and Rx Hub is submitting its NCPDP base provider list to NCPDP.  I would want to make sure that those three are somewhat equivalent, and that is what they are working in, because I wasn’t completely sure about the Rx Hub.

Requests for prior authorization from dispenser to prescriber exists in NCPDP, although it is not widely used.  This is a request for the prescriber to get a prior authorization.  It is not a request for prior authorization.

DR. COHN:  I’m sorry, say that again?

MS. AMATAYAKUL:  The request for prior authorization from the dispenser to the prescriber exists in NCPDP.  In other words, the dispenser is saying, hey, prescriber, we need a prior auth.  This is not a request for a prior auth from the prescriber to the payor PBM.

However, ASC X-12N 278 does provide for prior authorization in general, but very limited support for prior authorization of drugs.

DR. COHN:  I thought I was understanding, now I am confused.  So basically you are referring to NCPDP as the dispenser to PBM prior authorization message?


DR. COHN:  You weren’t?  Okay, I guess I am confused.  I very much get that the 278 is a vehicle under consideration from the prescriber to a payor.  What are you saying beyond that?

MS. AMATAYAKUL:  Sometimes the dispenser will identify that this particular drug that they have a script for is in need for a prior authorization.  There is a standard in Script that allows the dispenser to send a message to the prescriber to say, hey, prescriber, we need a prior auth.  But I wanted to clarify that that is not a request for a prior auth to the payor.

DR. COHN:  Okay.

MR. BLAIR:  Could I elaborate a little bit?  Maybe I could clarify.

DR. COHN:  Yes.  Jeff is exposing my ignorance.

MR. BLAIR:  No, I’m sorry, I don’t mean to do that, no, believe me.  Of all people to expose ignorance, I’m not the one.

What this is saying is that within NCPDP Script, they do provide a message from the pharmacy.  She used the word dispenser; maybe that was the hangup there.  From the retail pharmacy back to the prescriber saying, by the way, I can’t fill this until you request a prior authorization.  So that is the message in NCPDP Script that goes from the pharmacy to the prescriber.

Then if the prescriber is to request a prior auth, then we have got to turn to X-12 to communication the prior auth between the prescriber and the payor.

DR. COHN:  I guess maybe the reason I was confused was because if this was important, I didn’t see it in any way listed in the observations.  So I was trying to figure out if this was an important clarification, why it had to be as an individual discussion item.  It didn’t appear to be an observation.  I wasn’t sure whether there was a recommendation that came out of this.  That was really more my confusion about why it was being brought up.

MR. BLAIR:  What she is taking us through right now are the observation.  She is not hitting the recommendations yet.

DR. COHN:  Okay, so maybe I’m confused.  What Jeff just said, is that somewhere in an observation here?

MS. AMATAYAKUL:  Right here, request for prior auth from dispenser to prescriber.

DR. COHN:  Okay, I’m sorry.

MS. AMATAYAKUL:  The exchange in medication history may occur at multiple points, among prescribers, between patients and prescribers, between payors and prescribers, between prescribers and dispensers.  Medication history from payors to prescribers is currently performed in multiple proprietary formats.  Rx Hub has a protocol based on NCPDP and HL-7, and is submitting to NCPDP for balloting.  Medication history among prescribers may be performed using HL-7.  NASTM CCR is a new standard that supports the content for exchange between prescribers, such as an electronic health record and with patients such as in a personal health record.

Then the last in the message category is, medical history from prescriber to dispenser may be a desirable transaction, but will require an electronic health record or other clinical computing capability that is not commonly existing in the industry today.  Early efforts are being made by HL-7, ASTM, CCR and others to promote this functionality.

So I hope I have captured all the messages flying around.

DR. COHN:  Shall we just ask to start off, are there other obvious messages or pieces that we are missing?

MR. BLAIR:  You probably hit this, but maybe I was drifting off or not focused.  Tell me about the messages again that communicate formularies, and who they go to and how they are being done now.  I know that you hit those, but would you hit those again for me?

MS. AMATAYAKUL:  I have formulary and benefit coverage information from payors to prescribers currently performed with proprietary messages, and Rx Hub is submitting its formulary and benefit information and file transfer protocol to NCPDP.

Now, the one thing I did not include here was where the prescriber was viewing a formulary that was preloaded into their point of care device from a drug knowledge base, because it appeared from the content of the summary that that was primarily a download and then a screen lookup.  So I didn’t see it as a message.

MR. BLAIR:  So what have we got there, five or six major groupings of observations from which —


MR. BLAIR:  Eight, okay, and those are all of the e-prescribing messages.

DR. COHN:  Harry has a comment.

MR. REYNOLDS:  So right now, we are just making sure we have the right list, not evaluating what it says, is that correct?

DR. COHN:  Exactly.  We are not wordsmithing, but the first question is — hopefully we won’t wordsmith much today, though we will see — but the bit issue is, are we missing some  major observation, some major piece here.

MR. BLAIR:  One of the things that I like about what you have done is, this tracks very directly to the MMA requirements.  I think that really helps us a lot.  Then you consolidated them.  So basically you have given observations on the status of the ability of the industry to meet the MMA requirements with messages.

DR. COHN:  This will not be the last opportunity we have to add additional observations, but we want to make sure there are not glowing holes as we move forward.

Should we start going from number one, and start going through in slightly greater detail then?

MS. AMATAYAKUL:  For the NCPDP Script new prescription changes, renewal, cancel and fill status, there are existing standards today.  I have listed these out separately as 11213, because in both the testimony yesterday as well as the responses to the worksheet, NCPDP did some have suggestions that were slightly different with respect to each of the functions.

So I have 1.1, adopt NCPDP Script renewal function for refills as the initial standard, as an easy transition, developing educational material that demonstrates an ROI, like an early win.

1.2 is to adopt NCPDP Script, new prescription function, determining the impact of Medicare as payor from pilot.

1.3, adopt NCPDP Script’s cancel function, developing educational material that demonstrates benefits.

1.4, adopt NCPDP Script’s change function, developing educational material on benefits.

1.5, support development of further guidance by NCPDP to adopt Script’s fill status function for pilot testing and use case development.

1.6, test the capability for inclusion of decision rationale for drug choice when off-formulary is selected, that exists today in NCPDP Script.  Then I put for longer term as opposed to near term.

Then I added some stuff about HL-7, but maybe we could digest this first before we go to the HL-7.

MR. BLAIR:  You wound up saying, develop education.  On each of these recommended actions, could you clarify — maybe you already have — are these all actions to be taken by HHS?

MS. AMATAYAKUL:  Jeff, that is a good question.  Last night I started writing CMS should do, CMS should do, and then I thought, is it CMS or HHS?  Or is it ever going to be anybody other than CMS or HHS?  I didn’t really know, so I didn’t put.

MR. BLAIR:  That is okay, because it tends to go to the Secretary.

DR. COHN:  It goes to the Secretary, anyway.

MR. BLAIR:  Right.  The next thing is, in terms of the educational, could you elaborate just a little bit more what you are thinking of and what needs to be done with that?


MR. BLAIR:  Who needs to do it?

MS. AMATAYAKUL:  And maybe Lynn Gilbertson can embellish, too.  But it sounded to me like what she was saying was that there are some of these functions, for example, the cancel function and the change function and the fill status notification function, are not widely used today.  Even though they exist and they are usable, they aren’t being used.

So a value of pilot testing them would be to develop educational material to get a return on investment that could help the industry understand why these are important to adopt, anticipating some of the problems we had with the HIPAA transactions, perhaps.

I understood, but perhaps we should get clarification, that NCPDP would be willing to assist in developing the educational material, and maybe structuring some of the ROI from the pilot.  I also have later on in important related issues the need to insure that the pilot has some metrics in it.

MR. BLAIR:  I’m still drilling down a little bit more here.  Are you implying that the role of HHS in this would be to fund or support the educational materials?

MS. AMATAYAKUL:  I think that is an option, a part of the pilots.

MR. BLAIR:  Maybe I am getting too close to the wordsmithing, and we don’t have to go into the —

DR. COHN:  No, the question is, do we want to discuss that.  A number of other people had questions.  I don’t know whether we want to bring that up as an issue.  Why don’t we talk about this particular issue and see what we want to say there?  The question is whether we want to recommend that CMS fund or otherwise.  Harry, and then Steve.

DR. STEINDEL:  I just have a general point.  I think educational material and education with regard to this whole process is a general statement.  We probably should just defer it to the end, because this is going to apply to everything.

MR. BLAIR:  Maybe yes, maybe no.  If it turns out that it is everything, then yes, but I think that it may turn out that there are different actions needed for different things.  But if that doesn’t turn out to be true, then I would support what you say.

DR. COHN:  Steve, I think what you are saying is that we need to have education — that the same statement applies wherever we need to have education?

DR. STEINDEL:  Yes.  I think it is really obvious that we need to have some education if we are going to put this program into effect.  We have learned that from HIPAA and we have developed educational material for that.  I think it is a general course of action that should just occur in a basic paragraph someplace.

DR. COHN:  It sounds like there are a couple of people who have comments on this.  Harry has a comment on it also.

MR. REYNOLDS:  The key thing on each of these recommendations is, we did in the original letter we wrote to the Secretary talk about the demos, and we talked about that they would prove the ROI.  So rather than just developing the educational material that demonstrates the ROI, the pilot should validate the ROI, and then we create educational materials being able to explain that to everyone else.

But we have heard a lot of testimony that there is ROI, but we haven’t seen a pilot or anything else prove that yet.  We agreed in our letter to the Secretary that we would pilot these and come back with the results of those pilots.

MS. AMATAYAKUL:  This particular observation here is in the important related issues.  There are lessons learned from HIPAA regarding both the need for standards and the flexibility to respond to industry needs and technology changes.  Pilot testing should insure not only that new standards can be implemented, but provide ROI information, education and implementation guidance and other elements that contribute to success on widespread adoption, with some recommended action, potentially doing things like incorporate the ability to respond to industry needs for revisions and e-prescribing standards that are regulated, identify specific goals, objectives and metrics to construct and evaluate pilots.

MR. REYNOLDS:  If that is the catchall for pilots, then why do we mention pilots in 1.2 and the other things?  In other words, my point is, if we are going to make a general observation that this is going to be piloted and here is what is going to happen and so on, then singling out places that will be pilot and others that pilot is not mentioned seems —

DR. COHN:  We need to have other people comment on that.  Part of the question is, does everything need a pilot?  I think we observed yesterday that the refill renewal function —

MR. REYNOLDS:  If you take an example, the renewal function to me needs to be a part of the pilot.  We are not piloting it to know whether it is accepted or not.  There are two differences.  Things are included in a pilot to make sure — what you would be piloting here, that there is an ROI, not piloting that we accept it as a standard, whereas some of these others, where there may be data elements missing, we are piloting to understand what may or may not be missing or how it does or doesn’t work.

That is where I am coming from.  That is why this whole thing of ROI and everything comes into play for me.

DR. COHN:  Sure.  I forget whether Marie or Karen is next.

MS. FRIEDMAN:  I think Karen is next.

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

DR. HUFF:  It is a little broader, so you can rule it out of order.  To me, a simpler way to do this, which goes along with what Steve and some others have said, I see a lot of correlation with the observation number four.  You could accomplish pretty much the same thing by saying, look, we recommend adoption  of all of the Script transactions.  They are broken out here, and then the acknowledgements are mentioned down in four.  You can just say, look, we want to do all of those transactions and then just in that same category, after you say, we want to adopt Script, say, and we want to produce educational materials for Script, end of story.

MS. AMATAYAKUL:  The only one area that might be a little bit different is that the new prescription function — they wanted to investigate the impact of Medicare, and then of course later on under terminology we have the Rx Norm.  The fill status seemed to need a little extra.

DR. HUFF:  I guess I would amend what I said to say, and then if there are particular things that we want to test in the demonstrations, we could list those specific things that we were trying to test in the demonstration.  So you would end up with three things.

So you would say, this is what we are adopting.  We want educational materials, and this is what we hope we could test in a demonstration.

DR. COHN:  It sounds like Lynn has a comment that may help, and then we will go back to Marie and Karen.

MS. GILBERTSON:  Just a quick one on 1.2.  Yes, we did have testimony about Medicare involved in part of the pilot because of the payor, but they are not impacted in the new prescription function.  That would be down below in the eligibilities and the formulary and things like that.  A new prescription going from a doctor to a pharmacy does not have a payor in there somewhere.  So we need to move him down to where the payor is affected.

DR. COHN:  Thank you.  So we get rid of that particular little spot there.  Karen, and then Maria.

MS. TRUDEL:  I just want to reinforce the importance of delineating things that do not and do need to be included in the pilot.  The law does require that anything that is piloted be done in 2006, there be a report to Congress in 2007, the regulation in 2008.  So anything that is going to be pilot tested will not be in widespread use under Part D of Medicare until 2008, at the earliest.

That said, MMA also gives the Secretary the ability to adopt standards that don’t need to be pilot tested, standards that are already in widespread use that you can skip that part of the process.  That makes the time line much shorter.  Then we are talking about 2006, if we can get the regulation process working.

When the Secretary announced at the health IT summit last month that he really wanted to advance this process and do things faster in terms of e-prescribing, the feedback that he got was absolutely wonderful.  Representative Johnson thought it was a good idea, Senator Frist.  There was a lot of very positive feedback there.

So I just want to make sure that the committee is aware that that is kind of an important threshold decision.

The other question that I had — and I think I understand where Lynn is coming from, saying that Medicare is unknown territory in e-prescribing, you don’t know what is going to happen, and we may put in requirements that would cause things that are already working well to not work anymore.

I would suggest that as the subcommittee thinks about what it wants to say in terms of pilot-no pilot, that we be thinking about, does the standard work today, in widespread use, and if there is a concern about Medicare possibly putting in some requirement that could cause things to go haywire, perhaps that could be a caveat to a recommendation, instead of making it an assumption that somehow, we are going to throw a monkey wrench into the works and nothing will work anymore, even though it has been working perfectly well.

So I would just make that comment.

DR. COHN:  Thank you for reminding us that nothing is simple.

MS. TRUDEL:  Sorry.

DR. COHN:  No, I brought this up earlier, but thank you for redoubling it from yesterday.  Marie, do you have a comment?

MS. FRIEDMAN:  Yes, I have a couple.  I would like to underscore what Karen and others have said.  I think the way that this is structured captures what was said.

I think it misses apples and oranges, for purposes of our recommendations to the Secretary.  MMA requires us to recommend standards.  While information on educational materials is important and I believe in it with all my heart, I don’t think it is a standard.  So I would defer some of those other observations to another section, and think about them conceptually a different way.

A second thing is to underscore what Karen just said about trying to figure out what is in widespread use today that could be good to go without pilots, because we have heard that.  The way this is structured, to suggest that everything needs to be piloted will actually slow things down considerably, and might throw the monkey wrench into the way business is conducted now.  There is stuff out there that is in widespread use that is good to go, and I think we need to make those distinctions.

DR. COHN:  Steve.

DR. STEINDEL:  I have a question for Karen.  It relates to the statement that you made about standards in widespread use under the requirements of MMA.

Now, there are standards in widespread use that are for ordering a prescription from a pharmacy, but these standards do not support a lot of the other requirements of MMA.  So for instance, if we suggest that we can use NCPDP whatever, Script, new prescription, et cetera, it is currently in use, and we have heard that it is in current widespread use, but we also have heard that items like the SIG, which is important, is free text and has no standardization.  We have heard that the medicines are ordered in free text and have no standardization.  We have heard that there is no decision support involved with it, so DUR cannot be done using the present form of Script.

Now, Script itself meets the needs for sending a prescription from a prescriber to a dispenser.  But it doesn’t support the other requirements of MMA at this time.  What does that mean with respect to, it is in widespread use, it supports MMA, and does or does not need to be piloted?

MS. TRUDEL:  I think that is one of the things that we need to work through.  If there is for instance a use to which the NCPDP Script is now being put, people have come up with work-arounds that make it work.  Is that something that we can say, it is okay to go?

I think I am looking at Lynn there.  There may be things within a prescription like the SIG segment that are not standardized to the extent that we might ultimately want, but does that mean that the transaction is not workable for the business purpose?

DR. COHN:  Jeff, did you have a comment on this specifically?   Then Lynn, maybe you can respond also.

MR. BLAIR:  I think that we could accommodate what Karen and Maria are saying without compromising the point that Steve was making.

That is, I think that I am getting back to the way Margaret has characterized these things and grouped these things.  If we just go directly to the standards and clump them and wind up saying this is good to go or whatever, we are going to lose how these relate to the requirements.

So I think we could do both.  I think we could wind up with this structure, saying here is the requirements, here are our observations about them.  Given these observations, there are NCPDP Script standards that are in use today, but they don’t meet a number of these requirements, here are the things that need to be done so that they can be included in the pilot test, so we can at least get these standards closer, if not completely meeting the MMA requirements, closer to them.

So to me, it is a matter of the integrity of the process to be able to say and be accountable for, to what degree do our recommendations meet the MMA requirements.  So I don’t want to lose that, but I think that within this framework, you can still wind up saying that something is in use today, and you could also be explicit, to wind up saying exactly which messages — because NCPDP Script has a bunch of messages — which ones of those need to be in the pilot test.

DR. COHN:  I know Karen wants to comment.  I guess I would agree, and I think Karen is also observing that none of these are ready to go, or there is nothing explicit here about that.

MS. TRUDEL:  I thought it would be helpful to read the exact language of the statute on this subject.  It is the exception to the pilot project.  It says, pilot testing standard is not required where there already is adequate industry experience with such standards as determined by the Secretary after consultation with affected standards setting organizations and industry users.

DR. COHN:  I know Lynn wants to make a comment.  I would also reflect that as I said, I have likened some of this stuff to a car before.  The question gets to be, we have got a car, we realize MMA is requiring new shock absorbers and better lights and a better rearview mirror.  All those things probably need to be tested and piloted.  But that doesn’t mean the car can’t get you to work the next day, and all these things be improvements as we go along.

The question of course we need to ask ourselves is, is this really a car that can move, can it get us to work, which places can it get us to, and which places can it not.

Now, Lynn was going to make a comment.


DR. COHN:  Karen did that, okay.  Steve and Stan.

DR. STEINDEL:  I have a question regarding this.  If we say that these standards are in widespread use, industry acceptance, it meets the requirements that are spelled out in MMA, and then we  put another paragraph and use Simon’s analogy, we would like to have power windows and air conditioning in there as well to meet the other requirements of MMA, that may not be piloted, because that is up to the discretion of the Secretary at this point.  If we say it doesn’t meet the industry acceptance, that is forcing into a pilot.   So I think we are caught in a conundrum there.

DR. COHN:  Karen wants to respond.

DR. STEINDEL:  I am hoping she will.

DR. COHN:  Then we have Stan, Maria, and we will just go down the line here.

MS. TRUDEL:  First of all, I didn’t say it was going to be easy.  But second of all, to keep going with the car analogy, I think there are some thresholds that the subcommittee is going to have to wrestle with.  Using your car analogy, if the car does not have power windows or a CD player, you may not want to just never turn the key in the ignition until you have it perfect.  On the other hand, if the car doesn’t have any brakes. that could be a threshold issue.

DR. STEINDEL:  Exactly, and patient safety is more an analogy to does it have the brakes yet.

DR. COHN:  Well, it depends on what we mean by that.  I don’t think that anyone thinks that you have to have a structured SIG as an equivalent to brakes, for example.

DR. STEINDEL:  Yes, I guess the analogies are sort of losing it.

DR. COHN:  No, actually I don’t think they are.  I think the question gets to be, at what level.  There are things that are definite improvements that help encourage and support X, and do we not allow anything to happen before that.  That is what Karen is bringing up.

Stan, you are next, then Maria, then we will go all the way down the line.

DR. HUFF:  I just go along with something Harry said.  I think we want to make the distinction between something that would be tested in a demonstration and something that is just participating.

For instance, I think we are in a situation where we would say we want to adopt Script, it is in widespread use, and so that is the feeling of the committee and the evidence that we have heard.  Script would be used in any demonstration because it is an integral part of what has to be there to make any of the other parts meaningful in the test.

Another nuance to that then is saying that using Script in the demo, where we now are incorporating a structured SIG, or where we are conjecturing that Rx Norm would be the code in that prescription, maybe we should construe that as a test of Rx Norm, not as a test of NCPDP Script, and that would alleviate some of the difficulty of how we state it.

DR. COHN:  Stan, thank you.  We have Maria and then Jeff also.

MS. FRIEDMAN:  I just won’t want to let the perfect drive out the good here.  I really endorse what Stan is saying.  I think there are levels of functionality that are required by the MMA that are gaps in the functionalities that we have heard through testimony.  It ain’t perfect, but it is working out there today.

I think the trick of this is going to be, to make that cut, how we structure the recommendation to say, let’s go with Script, for example, because it is an integral part of what is being done today.  It is not a test of Script per se.  We may want to combine it with these tests of functionalities, but again, let’s not let the perfect drive out the good.

DR. COHN:  Then we have Marjorie and then Kathy and then Jeff.

DR. GREENBERG:  I’m next?

DR. COHN:  And then Michael.

DR. GREENBERG:  Am I next?

DR. COHN:  Yes, you’re next.

DR. GREENBERG:  This may not be a necessary comment, but I feel compelled to say it.  Regarding how you want to address the need for educational materials, I certainly do think that any standard that isn’t supported by adequate guidelines is not a full standard or instructions.  What you may be suggesting regarding educational materials may go well beyond the instructions and guidelines that currently exist.

I am loath to in any way disagree with my colleague Maria, for whom I have great esteem, but I think the subcommittee has been holding hearings now for several months, has gathered a huge amount of information, and I think whatever you can say explicitly or recommend regarding educational materials and further guidelines, et cetera, is really critical.

I think we have learned through the experience of HIPAA standards that just having the standards out there is not necessarily enough.  So where you want to put it in your recommendations, I don’t know, but explicit and detailed stuff about educational materials I think is very valuable.

DR. COHN:  And I think we need to have Margaret hold that.  I want to make sure that every conversation right now is on the issue we were just talking about.

DR. GREENBERG:  I was responding to something that was said.

DR. COHN:  That’s fine.

MS. AMATAYAKUL:  May I respond to that real quick?

DR. COHN:  No, because I would like you to hold that.  I think we are right now talking about, is everything a pilot, what can we go forward with and say is a real standard.  I do think we need to talk about education.  That is one of Margaret’s areas up there.  I don’t think we have to resolve that right now, but I do think we need to resolve the issue about how we are describing this, and are we agreeing with Stan or are we saying something else here.

DR. GREENBERG:  I guess in a way it is related to that, because — I may have misinterpreted what you said, but the possibility of being able to roll something out without a pilot might be enhanced by better educational materials.

DR. COHN:  Marie, do you feel you need to respond to this?

MS. FRIEDMAN:  Yes.  What I said was that I heartily endorse the educational outreach materials.  I spent most of my life doing that.  I just didn’t think mixing it in with the standards per se was — it just muddied the issue.

I think there should be very strong recommendations about outreach and education.  I would make that a separate section and pull that out of the recommendation on the standards per se.

DR. COHN:  Kathy, you’re next.  You have been waiting very patiently.

MS. HOLLINGER:  I just wanted to say that even old standards, if you have to integrate new elements or if you are looking at trying to work towards integration of many different pieces are really new standards.

So even though they are tried and tested and true under current conditions, we may be trying to integrate several different pieces, some of them new, some of them old, in order to make this work and work well.  I have seen a lot of pieces that are old and some that are new, and some of the terminologies from NDFRT and Rx Norm are new, some of them have not been put to use for this purpose yet, or some of the old standards for the scripts or some of the messaging standards are older and well established.

So I would just look at it from that perspective.  If I go back to the car analogy, what we have been talking about so far from where I sit is the parts of the car that haven’t been completely assembled yet.  Upon assembly, they do need to be tested and tried.  So they do take on a context once they are assembled from when they are separate.

DR. COHN:  Jeff, I think you are next.

MR. BLAIR:  In trying to express my thoughts, sometimes I try to wind up saying, what is underneath this, what is the real issue, what is the real concern.

I don’t think there is anybody that — I think there is a lot of convergence now as to what the standards need to be and what they are.  So to me, that is kind of a given.  That is going to be in the recommendation letter:  Here are the standards that are available, here are the ones that are in use.

My concern is that if we just have a sentence or paragraph that just simply says that without the caveats, I am concerned with the actions that will be taken if we do that out of context.

So my thought was that we do this in a context which is focused on saying, okay, these standards are in use, but right together in the same area, not separated in another part of the recommendation letter, but they fell short of the MMA requirements in this area, this area, this area and this area, and here are the actions that we need, so that in the pilot test we can go ahead and make as much progress as possible.

What I have just said is where I think the value of NCVHS lies.  I think this is the heart of the matter.  We add virtually no value if we just look at what is out there today and say, here is what is out there today, it works.  Well, it does work.  It works well.  A lot of good work has been done, and we don’t want to stop it.  We want to encourage that.  But the federal government doesn’t need  to spend money on a demonstration project, because the private sector is going forward on its own and making good progress, and for us to wind up getting caught up in that, I think we have lost where our added value really is.

That is my thought.

DR. COHN:  I think Harry is next, and then Lynn, and then Steve.

MR. REYNOLDS:  Michael has his hand up.

DR. COHN:  I’m sorry, Michael.  Harry and then Michael.

MR. REYNOLDS:  You asked the question, does somebody agree with Steve or Karen.  I think it is too early to agree with anyone yet.  I play off of Kathy’s comment a little bit.

As you look through these, we are talking about pieces of the car.  NCPDP Script might be the engine.  SIG is one, medication history is one.  So I am anxious to see as we go through these and decide what we decide, on which of these we agree.

I’ll just use an example.  If you take the formulary and if we come out of this discussion today that we are recommending that the Rx Hub proprietary format go through the NCPDP process and become an ANSE standard, then the Secretary or anyone else would have the opportunity in this car we are going to drive out of the showroom  to put that in the car, right now, and go ahead and drive the car.  But we are not necessarily saying it is done until it becomes an ANSE standard.

That is why I can’t vote on either side yet, because once we pick the pieces and we decide what car we are willing to drive out of the showroom, using your example — and right now we have gone to one, and then we are jumping to no pilots or pilots.  I think we have got a list here that is a very short list.  After all this testimony, this is a pretty short list to get us to where we want to go.

So I agree with both of them, but can’t vote either way at the moment.

DR. COHN:  I think we are going to need to talk about them individually.  Michael has a comment.  Hopefully it is on this point.

DR. FITZMAURICE:  In looking through what the law requires, the Secretary on September 2, 2005 would recognize the existing standards and send a signal to the industry, this is how I see things pulling together.  But as Jeff says, it doesn’t meet all the requirements of the MMA.  The Secretary can be pilot testing not only the standards that are developed, but can evaluated the other standards.  That is, not to pilot test them, but they would be included as part of the evaluation.  It is hard to do an ROI if you only have the new standards and not the whole gamut of standards.

Then the Secretary doesn’t have to promulgate, i.e., go through the regulatory process and formally adopt them, until  2008.  In the meantime, what NCVHS has done is given recommendations based upon industry testimony and user testimony, here is how we see it putting together, these standards are ready to be driven out of the showroom.  They don’t fulfill all the requirements.  We say to the Secretary, here is what has to be done to make them fulfill the requirements.  Then somewhere in MMA, I hope there is money to help support the financing of meeting the requirements of MMA.

So all along, the NCVHS can be guiding this, but also it serves as a signal that the industry can begin adopting and putting into place some of these things with some assurances, though not 100 percent, that the world isn’t going to change drastically from  what they told the NCVHS they need and what we see as the view.

So I see that we can proceed, the industry can proceed, and the Secretary is going to have a large job to do.  But we can make the recommendations without being held back, and industry can adopt them as they wish.

DR. COHN:  Karen.

MS. TRUDEL:  I believe we were getting hung up on the fact that by chance, the recommendations have been drafted in a way that presupposes whether a particular standard is ready or not.  I would suggest that we step back from that, and rather than saying this one is ready to go, this one should be piloted, this one needs educational material, let’s just talk about the standards.  is there something that does this?  Does it do it well or does it have gaps?  Let’s go all the way through that, then come back and see what we have, which is basically what Harry is saying, and come up with a second round of, it appears that if you put these particular standards together, you could get e-prescribing off the ground sooner than 2008.

DR. COHN:  Lynn has a comment.

MS. GILBERTSON:  I think I am agreeing with how Jeff stated it, but when I was reading the spreadsheet, I was thinking there were more columns out to the right that would say in use, needs to be piloted, needs educational study or something like that, so that we would know what the marching orders were.

The recommended action says adopt, but I don’t necessarily know that that is the action item that would be taking place.  It really is, during the pilot period, provide studies of usage.  Another one might be, during the pilot prove the concept of this new standard that is out there, things like that, where it is really action.

Above this, you might say, adopt Script, adopt HL-7, adopt X-12, yadda yadda.  But these are actually what business functions are you expecting to perform and how do you expect us to either operate in 2006 or be ready for 2008 or whatever those different columns are of action.

So it was just a thought.

DR. COHN:  I think Steve, and Kelly, you had a comment, too.  Then I would like to try to bring things — because I think we are beginning to talk about the same thing, except for of course Steve.

DR. STEINDEL:  Except for of course Steve.  I think that is the observation I am going to make.  Karen brought us to the law and is focusing on the section of the law about putting pilots into place.  But there is also another section of the law that spells out very specifically what the program requirements are for MMA.

What we are recommending is something that has to fit both the program requirement section and the standards section, before we can talk about the pilot section.

I’m not sure we have a choice on what options can be put in the car.  Congress has told us what they want the car to look like.  I like what Mike is saying with a little bit of modification, that we have a set of standards that can be used for this aspect of what the program requirements are.  We need to add to the standards like Kathy pointed out, which creates new standards that need to be tested, these aspects, so that it meets the other requirements of MMA.

Now, the Secretary can decide how to phase this in.  If he wants to phase this in as a certain set that is meeting some of the requirements but not all of it, and Congress is willing to accept that, that is fine.  But maybe we should word the letter somewhat like Mike was talking about.

DR. COHN:  I think maybe we are all coming together.  I was harkening back to what Stan was talking about, about testing added functionality.  So I’m not sure that what you are saying is so different than what Mike said or what Stan said earlier.

DR. STEINDEL:  I don’t think it is so different.  It is just a matter of wording differences.  The implication is that the Secretary can waive standards if they meet the requirements and are in widespread use, but I don’t know if we have found any standards that meet all of the program requirements of MMA and are in widespread use.  We have found standards that meet some of the requirements.

DR. COHN:  That’s fine.  I think the Secretary has some option about — there is nothing in the law, and I have read it enough that I have almost memorized it — that says that the Secretary has to have a comprehensive, complete set of standards proved before 2006, or can’t prove any of them.

DR. STEINDEL:  I think that is the way we should word the letter.

DR. COHN:  Yes.  Kelly, you were probably going to talk about this.

MS. CRONIN:  That was my exact point.  I think the intent was trying to regulate — and Karen can probably speak to this in more detail, but to regulate two times here.  In the near term, in the next several months, we are interested in working out a regulation where the scope would be standards that are mature enough, and perhaps don’t meet all the requirements, but they do address some needs, and they are certainly in widespread use.  So they don’t meet the test of still needing industry testing, which is the qualification for the pilot.

But I also agree with what Mike was saying.  If there is a pilot with a fairly rigorous evaluation, that you would do that with the whole program in mind and all the functionalities in mind, and perhaps focus the evaluation component on the standards that are lesser developed or new, or perhaps not in widespread use.

But I think the intent is for this recommendation letter to clearly spell out what should be or what you believe to be the scope of this initial NPRM, and perhaps even qualify things, like the need to do an ANSE accredited standard for formulary and benefit representation.

DR. COHN:  You haven’t even been here for the entire testimonies.  That is very good.

DR. FITZMAURICE:  Can the Secretary just recognize that these standards exist, and they look to be pretty good without going through the regulatory process?  Put out a press release on September 1, so you avoid the regulation?

MS. CRONIN:  No, I think there is a clear strategy, and Karen — again, this is really her responsibility.  But it has also been in the Secretary’s office, the clear intention to regulate more aggressively, with the idea that this will help drive adoption, particularly because the benefit is going to be implemented in ’06.

So we want to be ready to go in ’06.  We expect the prescription drug plan sponsors who will be delivering the benefit by ’06 to be offering these programs, because there will be a legal requirement to comply with these standards.

DR. COHN:  I want to move us back around a little bit, because I think the good news is that Margaret has now shown us some additional tabs over here on the side, that she renamed as we were talking.  But they are probably not bad.  Since we are now talking about observations, recommended actions, I think formerly she had something called near term and long term, and now they are suddenly ready, which I think is probably what you meant by near term, as well as things that might need to be proved in pilot.

So I think we are beginning to — I do agree with Harry and Kathy that we need to reflect as we go through all of this.  It is something that as we get done, going back and looking at it again, going back to my analogy of a car, are these pieces of a car, is this a whole car that just needs a rearview mirror and better lights and better shock absorbers, along Stan’s idea of additional functionality.  There really is a piece here that is, is there enough there that is worthy of saying that is there, or are we talking about a good axle or whatever.

MR. BLAIR:  Can I make a suggestion that I think will get us around this?

DR. COHN:  Yes, please.

MR. BLAIR:  I think that if we wind up using the word standard as ready, that there is going to be a lot of dissension on that.  However, if instead we wind up indicating what standards are in use today, then I think you get ready agreement, and you have many more standards that are identified on that, and you are not making a value judgment as to whether or not they meet MA requirements.  You are simply identifying, these are the ones  that are in use today.  I think we can come to consensus quickly, and then move on to the other pieces.

DR. COHN:  Karen, before we do that, can you re-read that sentence or two from the reg?  See whether that differentiation helps us or not.

MS. TRUDEL:  Pilot testing of standards is not required where there already is adequate industry experience with such standards, as determined by the Secretary after consultation with affected standards setting organizations and industry users.

DR. COHN:  So Jeff, given that, I think your comments are probably useful.  Margaret, shall we start going back through this again?  I’m sorry, Steve, I was too quick.

DR. STEINDEL:  I would like to ask a clarification question of Kelly.  When you were making your statement earlier, you used the word benefits.  You used it several times.  Were you talking about benefits with respect to Part D in total and not just the area that is involved with e-prescribing?

MS. CRONIN:  Yes, in general terms I was talking just about the delivery of this new outpatient drug benefit, or Part D.

DR. STEINDEL:  Because there is a little subtle difference there.  There are standards required under the e-prescribing portion that we have already identified may not achieve the benefits that the law wants.

MS. CRONIN:  Right.  It is a step in the right direction.  I think the intent really is to create the requirement for these PDP sponsors to offer in a program.  There may be varying degrees of functionality in doing that, but they will have that obligation.

DR. STEINDEL:  I think we are in synch.

DR. COHN:  I have been reminded that it is time for our morning break, which I think is probably a good idea at this point.  I think we have had a relatively full conversation about what it is we are supposed to be doing here.  Jeff, I think your idea of, as we go through these, making sure they are the right things, talk about whether things are in use, and whether things are not in use or need to be piloted — I think you have prove there, pilot, whatever.  I think it may be a useful thing as we go through, along with making sure that we have the right recommendations.

So when we get back from about a ten-minute break, we can launch into that process.  Thank you.

(Brief recess.)

DR. COHN:  I just want to review with everyone where we are, and then we will start getting into the work.  I think we have had a broad discussion about what it is we are supposed to do, and what types of recommendations we should be making.

I think at this point, we have been talking about the observations.  I think we have also agreed about the idea that we have recommended actions, but now we are also recognizing that there is a piece that is called, a standard is widely used, and this might be something that we might consider to recommend to the Secretary as something that in and of itself doesn’t need to be piloted in 2006, but that could actually be adopted beforehand.

We also put up a category that says — and this was an odd one, but we at least try it on for size, which is that it has come to our attention that additional functionalities in some of these standards, where the function actually works, but g it ma be that the business case is lacking, so therefore it is not widely used.  So we just have to be aware that this may be a separate set of recommendations, saying these additional functions, once again based on what we come up with when we say, these are functions that appear to be usable, but there has not been a business case established, and that the Secretary’s discretion might be appropriate for this or for a piloting.  Then finally, things that clearly need to be piloted.

Margaret has also stuck something up that says beyond, and I will leave that to your imagination, what beyond means.  But I think you all get the general gist there.

So is that okay with everyone?  Jeff.

MR. BLAIR:  I think we could leave it.  I just don’t want to restrict what could fall under the column of widely used, because people wind up saying it is not widely, it is used.   I think that my thinking is that the point of what is there is that it is in use today by those folks that are using it.  Maybe it is five percent or ten percent or 15 percent of the industry.

At some point, I think we have to be careful so we don’t have people argue about the word widely.  But I think we are in agreement of what we want.  Simon, were you able to —

DR. COHN:  Yes, I was just going to try to find the legislation and once again read it for the 150th time.  Can you just read that particular section, or do you want me to take it from you and —

MR. BLAIR:  Because some people could stand there and say nothing is widely used yet, but we still want to have things in there that are in use.

DR. COHN:  Let me just read the exception here again, just to refresh everybody’s memory, which Kelly has already read to us two times.  It says, pilot testing of standards is not required under clause one, where there is already adequate industry experience with such standards, as determined by the Secretary after consultation with affected standards setting organizations and industry users.

MR. BLAIR:  Is that adequate usage or —

DR. COHN:  Yes, adequate industry experience.

MR. BLAIR:  There you go.

DR. COHN:  So we now have adequate industry experience up there.  That is obviously a judgment for the committee to make.

MR. BLAIR:  Good.

DR. COHN:  Michael.

DR. FITZMAURICE:  There may be an issue that we would want to discuss.  That is, we have seen with HIPAA that once you put a standard in a regulation, boy, it is hard to change it.  Yet, I see these standards as changing to adopt to the functionality that will be required along the way.

So I would say that we would find these standards are in industry use, they are widely accepted, and the committee might recommend that the Secretary recognize them, announce them, and maybe discuss whether we want the Secretary to adopt them as firmly as HIPAA standards.  Even though these standards might not be needed to be pilot tested, it might be wise to suggest to the Secretary that they be incorporated in the pilots so that the pilot testing of the new standards or the developed standards be done in a framework where everything is working together.

DR. COHN:  Michael, I’m not sure if I consider this to be an associated issue, but I don’t think it is an issue that we need to deal with right this second.  What I am hearing is that it is a statement of wanting industry to have some flexibility moving forward.  I am also hearing maybe some underlying frustration with the speed and flexibility of the HIPAA process.  But I don’t think that impacts what we are doing right now.

DR. FITZMAURICE:  I think you are right, Simon.

DR. COHN:  Let’s make sure we don’t forget it, though.

DR. FITZMAURICE:  It is not the frustration, it is wanting to anticipate a smoother process.

DR. COHN:  So when we talk about the related issues, let’s make sure not to forget that one.  Stan.

DR. HUFF:  Now that we have this new pattern, I am  trying to make sure I understand how this translates into what we put in the letter, which is what I assume we are working toward.

DR. COHN:  Sure.

DR. HUFF:  Is this a proposal essentially that we would put in the letter, that quote-unquote this is a standard that is in use?  That would be our recommendation?

DR. COHN:  That might be.  I would say we need to go standard by standard and reflect on whether there is adequate industry experience.  We could frame it as, there is adequate industry experience with it, we recommend that you go forward with it without piloting.  Then there will be places where we will say — this is a hypothetical; if we were to decide that, we might say X, Y and Z, there is new functionality that is required that needs to be piloted.  That may be what we would say, or if we saw that some of it was inadequate industry experience, other functions work but are not in industry use, we might say, this one there is adequate industry experience with, these pieces appear to be very functional modules, but for a variety of reasons including ROI or otherwise, industry has not been using them a lot.

That might be what we would say.  Does that help?  That is what I am thinking might come up in the letter.  Do you have an issue?  Is there some big hole here?

DR. HUFF:  I guess if that first column means that we are going to recommend it for adoption without needing to go through a trial, why don’t we just name that column that, rather than naming it in use?

DR. COHN:  I think Jeff had proposed that we do it this way, because he thought it would avoid the argument of —

DR. HUFF:  My point is that if we mark it this way and what we really mean by it is that that is going to translate into doesn’t need pilot when we put it in the recommendation, then we might as well talk about whether we want this one piloted or not at the time.

DR. COHN:  Jeff?

MR. BLAIR:  That is a good point.  I think that a few other folks have pointed out the fact that if we recognize that a standard is in use today, and it is a basic foundation, and then we are also indicating where there are things that need to be improved that should be in the pilot, we don’t want those two to be mutually exclusive, because the pilot probably also needs to include things that are in use today, along with whatever improvements can be made by January 1.

So is that what you are getting at?  We don’t want to exclude them.  They are not mutually exclusive.

DR. COHN:  I think I will speak for Stan.  I thought his framework earlier for discussing these things as things that are additional functionality in many cases of these standards that need to be piloted — isn’t that how you described this, Stan?

DR. HUFF:  Yes.  Another way you can characterize this is to say, there is no question about this standard being adopted.  From our perspective, there is no question.

Just take NCPDP Script.  I think at least my view would be, there is no question that it is going to be adopted.  Looking at a lot of the other issues, could there be modifications, could there be new versions.  Yes, that needs to happen.  Couldn’t we say basically, these are the things that we want to adopt, and these are the things that need to be tested as part of the pilot.

In other words, going back to what we said before, that means that NCPDP Script will participate in the pilot, but it is not being tested in the pilot.  What is being tested is maybe the efficacy or how that works combined with Rx Norm or how it works combined with the new structured SIG.  Those things are being tested in the pilot, but this isn’t.

So I guess what I am saying is a methodology where we would say, NCPDP Script is adopted, and we could put other caveats about recommendations about things that we want enhanced or whatever, that then we put in and say, in the pilot we are going to test whether Rx Norm works well with NCPDP Script, but Script is adopted

MR. BLAIR:  I feel very uncomfortable with that approach. The reason that I do is, I feel like it tends to trivialize the value of what we are doing.

Several months ago, we pretty much knew what the foundation — I am putting the word foundation in front of the standards, the standards that can provide the foundation, so we knew what they are.  If that is where our emphasis is, here is the standards that are being adopted, and then you wind up later on saying, oh, by the way, here are the bells and whistles, then you have also trivialized where I think the value is.

The NCVHS has an opportunity to take those basic foundation standards, which we could identify, we could say, these are in use and we recommend them as foundation standards, or basic standards to build on.  In some cases, they have to be ANSE accredited, in some cases there may be some other pieces harmonized or whatever, or with some caveats.

So I don’t have a problem with identifying what is in use today and saying that they are foundational.  I just have a different emphasis, because I feel like our added value is to accelerate the industry to do as much as possible in this next year, so that those projects are moving the ball forward and not just being a test of —

DR. FITZMAURICE:  How is that different than what I said?

DR. COHN:  Yes, Jeff, I don’t think there is any difference here.  I think it is basically maybe a slight wordsmithing on emphasis.  What I think we are all hearing is that, assuming we all agree, there are foundation pieces, and then there are pieces that need testing.  I think we can wordsmith how we frame that.  It has to do with where in the letter all of these things appear.

Obviously if anything that needs to be tested is on page 25 in eight-point type, that is one statement.  If it is connected with the main recommendations, I think that is another way to present this in a slightly better fashion.

I would also remind everyone that even though — Jeff, I know you have been the leader on this activity.  I don’t think many of us felt that we knew a couple of months ago exactly what all the answers were.  Indeed, I discovered some of the answers yesterday, at least from my view.  So I think we should not trivialize the issues related to identifying what are a reasonable set of foundation standards to advise the Secretary on, going forward.

MR. BLAIR:  I accept that.

DR. COHN:  Ross, did you want to make a brief comment?

MR. MARTIN:  Yes, just a brief comment.  One of the impacts that you can have by stating specifically foundation standards in particular areas is, as we do have gaps, and there are questions about where we might do the further work, you want to be able to give signals to the market that say, we know that NCPDP Script will be one of the standards that we will adopt, and we know that there are areas where it doesn’t meet all the needs of MMA, but we expect that a Script standard in this category, if it is formulary management or presentation, that kind of thing.  That way we know which standards organizations we should be introducing that stuff.  If Rx Norm is going to be submitting its stuff, we have to know which place to do it in.

There are some choices to make here.  It is not — I don’t think it trivializes your work at all. In fact, I think it helps spur development faster, because we know exactly what we need to do in which categories, in which standards organizations going forward to fill those gaps.  We have a clear time line.  We have a clear mandate to do that.

I think just saying we recognize that there are standards out in use doesn’t weaken that.  I think it will slow development, especially for those who aren’t going to wait for final standards to go ahead and develop to what is there today.  That is just an opinion.

DR. COHN:  Thank you.  I think we have talked enough about this.  Everybody is beginning to say the same thing, which is probably good.  Shall we start actually doing the work?  I know, no fun.

Margaret, would you like to begin to lead us through this, and NCPDP Script?

MS. AMATAYAKUL:  Can I just review these categories, adoptive, adequate?  Adopt is adequate industry experience, adopt because there is adequate industry experience.  Function words, minimally used, include in pilot, test in pilot and beyond.

DR. COHN:  I’m not sure we had decided what we were going to do with that middle category about being tested in pilot, but we can argue about that as we go along.

MS. AMATAYAKUL:  1.1, NCPDP Script renewal function for refills.

DR. COHN:  I’ll look around.  Are people comfortable with that being a ready to go piece?  You don’t think it is a ready to go piece?

MR. BLAIR:  When we say that it is adequate, to just say that anything that is in use today is going to be adopted, when I think of the messages for example for the formularies, I think that what we want to say is that it is in use today, and that we strongly encourage that to be ANSE accredited before we would recommend adoption.

DR. COHN:  Jeff, we are only talking about Script now.  We can change the categorize as we move forward.  So let’s talk about Script.  Let’s try to do this with a known ANSE standard that we have been dealing with, and see where we get to.  Kathy.

MS. HOLLINGER:  A quick comment. The fact that it has been widely used, does that mean that it is adequate for the use?  I guess I would like to see a list of what is the target that we are aiming at, what are the MMA requirements for the functionalities.

I wrote a quick list here.  I think maybe it addresses certain aspects, but not all of them.  But there are some areas that it may need to be developed further in, in order to achieve the target which MMA says.  This gets at Karen’s comment earlier.

DR. COHN:  Karen.

MS. TRUDEL:  I think it is misleading to try to apply those MMA requirements to any given proposed standard or code set or messaging, because the complete list of functionality can’t be accomplished unless you put them all together.  I think we realize that we can’t put them all together yet and meet all of it, but that is not a reason for not trying to go forward.

I think where it might be helpful to utilize Kathy’s list is in looking at some of these things, to say we realize that in saying that this is a standard that the industry has experience with, we are hearing that it doesn’t do this as well as it could.

DR. COHN:  So it sounds like we should move forward.

MS. AMATAYAKUL:  NCPDP Script, new prescription function.  It is adoptive, adequate industry experience, but there is added functionality that could be included, and that part of it should be tested and piloted.  Such as, —

DR. COHN:  Such as.

MS. AMATAYAKUL:  And then we got into Rx Norm and that, which we have got as a separate recommendation.

DR. COHN:  So this is structured SIGs.


DR. COHN:  So that is a TBD.  So the base standard itself is, this function is functional, but we would need augmentation.  You are going to see that in a lot of these things, so I don’t know that that is going to be — maybe put that in parentheses.  I would remove that as a testing pilot.

MS. AMATAYAKUL:  Say that again?

DR. COHN:  I would in the recommendations put what you just said there in parentheses, and I would remove test in pilot, since you are going to handle that in a separate section.  At least, I presume that is going to be there.  Anything that is going to be additional functionality is going to be needed to be tested and piloted, by definition.

MS. AMATAYAKUL:  Can I ask just a format question?  So you would like to see the new prescription function, because it is adequately used in the industry, as an adopt.  Then what I have in parentheses, the added functionality be tested, move that away from that and put in a separate recommendation, not associated at all with this here.

DR. COHN:  We can decide how it plays.  I think in parentheses is fine right there, but to do what you are describing, there is going to be a lot of redundancy, because you are going to use Rx Norm and test in a number of different places.  So you have made the note, we see the note.

MS. AMATAYAKUL:  I guess what I am trying to ask is — because it seems to me that this is the point of contention, that there are some functions that are ready to go today, but then there are also functions that are ready to go, but we would still like to see some additional functionality.  That additional functionality needs to be tested and piloted.

So there may be some functions that are ready to go and don’t have any additional functionality we have to worry about, such as the renewal and refill.  That is where I was trying to make the distinction, so that if somebody saw this list and didn’t read further on down, they would recognize that there is still added functionality.

DR. COHN:  Is that a wordsmithing issue that we need to deal with in the document?  Michael and Steve.

DR. FITZMAURICE:  I think I go along with Margaret on the spreadsheet that is showing the functionality that we are talking about that is needed.  I think on the recommendations that will eventually come out, we probably won’t mention the functionality in the specific recommendation, except further down we will mention, here are some additional things that we think need to be added.

But for this purpose, it means that we won’t forget why we said this.

DR. COHN:  That’s fine.

DR. FITZMAURICE:  And that’s all.

DR. COHN:  Steve, comment?  Do you agree with Mike?

DR. STEINDEL:  I think I agree with Mike.  If you just want to put the mark there for memory holding for the purpose of this discussion, I’m fine.  But my specific statement is, I believe we are totally correct in saying NCPDP Script new prescription function, we have adequate industry experience.

Now, the parts of it that we don’t have adequate industry experience are still going to fit into the message function.  The carrier is going to be the same.  So when we discuss the parts that need to be added, and we say it needs to be tested in pilot, that is where we are emphasizing the other part.

DR. COHN:  Harry.

MR. REYNOLDS:  I personally have a problem with putting things in here that change — in other words, in my opinion, we are accepting the NCPDP Script new prescription function.  Later on in the document, we talk about key pieces of this thing that are not structured.  But the Secretary could decide to implement current NCPDP Script using the current structure whether the SIG changes or not.

So I think we are cluttering it up.  We are putting a lot of addenda and caveats in it.  We have got a list.  We have got nine or ten things listed here that we are going to decide on independently, after we decide on them independently as to what we are going to do.  Whether we come back and change the way we word this or anything else, I think we are listing a lot of additional things I think don’t necessarily add to the final decision.  We are accepting NCPDP Script new prescription.  If they want to go with that, he can go with that.

DR. COHN:  Judy and then Karen.

DR. WARREN:  What I see as this spreadsheet is, this is a tool for us to organize our recommendations for the letter.  Therefore, for me, with all the dialogue that we do and back and forth, I really need a placeholder on 1.2 about SIGs.  I don’t expect that to go anywhere else in there.  To me, that is just a little memory twig, it is not anything else.

So I would like to keep it there, so that I don’t forget that at some point, we need to address Rx Norm in a sense.  That will move somewhere else, but let’s keep this as a tool instead of the letter.  It will make it easier for us.

DR. COHN:  Karen.

MS. TRUDEL:  I was going to echo that same thing.  I think the structure of the document itself is becoming a distraction.  The discussion is going on, but we can’t seem to capture it because we are so distracted by the document.  I’m not exactly sure what to do about that, but maybe we don’t want to work with this document right now.  Maybe we want to have a discussion and take with notes, and then put the information back into the document when we are finished.

DR. COHN:  Actually, I think this is a pretty good frame, myself.  I don’t know how others feel about this.

DR. WARREN:  I guess for me, if I think about this as a document, I start getting real confused.  When I think of this as a tool for us to begin to think about a document later on, it is very constructive.  So I need the memory.

So let’s just talk about this as a tool that focuses the discussion, and not worry about the letter that we have to write later on.

MR. REYNOLDS:  And Simon, if there are people that need the memory jog in the tool, it is not worth discussing.

DR. COHN:  So it is in parens.  So are we fine with that, leaving it in parens the way it is?


DR. COHN:  Okay.  1.3.

MS. AMATAYAKUL:  1.3, adopt NCPDP Script cancel function.  This function is not in widespread use right now, but it is ready to go as a function.  The function works.  Change function, same.

DR. GREENBERG:  Can you read that second comment?

MS. AMATAYAKUL:  The function works, minimally used, include in pilot.

DR. GREENBERG:  You do want to include it in the pilot?

DR. COHN:  I don’t know whether we do or not.

DR. GREENBERG:  So how does it differ from test in pilot?

DR. COHN:  I thought the include in pilot was unnecessary.  I thought we were just describing this as something mature, and we could figure out whether it is a pilot issue or not.

DR. FITZMAURICE:  We can cover that later.

DR. COHN:  Steve.

DR. STEINDEL:  The NCPDP experts would have to correct me if this is wrong, but I believe the change function would probably have to reflect the changes that we would want in the new prescription as well.

MS. GILBERTSON:  If you determined those functions were necessary for the pilot.  But as someone stated earlier, if we never get to a point of standardized/codified/whatever to a SIG, it will continue just fine as is in the text.

So if it is important enough to be added in, yes, then change will follow right along with it.

DR. STEINDEL:  Yes, this is for our senile memory purposes.

DR. COHN:  So does this capture it appropriately?


DR. COHN:  Lynn, did you have a comment that you wanted to make about any of these statuses, or are you all discussing about whether or not some of these are actually in greater use?

MS. GILBERTSON:  I’m just asking them to double check.

DR. COHN:  That’s great, thank you, that is a useful piece.  1.5.

MS. AMATAYAKUL:  Support development of further guidance by NCPDP to adopt Scripts.  Fills status function for pilot testing and use case development.  Test and pilot.

DR. COHN:  Steve.

DR. STEINDEL:  Simon, I believe some of the comments that concern the fill status extended beyond NCPDP.  There was a misuse that might concern privacy, so it might actually be out of the domain of NCPDP.  We need to support the development of further guidance of the fill status, but I don’t know if we should just rely on NCPDP to do that.

DR. COHN:  Maria?

MS. FRIEDMAN:  It seems to me that the way that that is worded isn’t consistent with the rest.  This is the first time that we talked about fill status, so we don’t know whether it is in widespread use or whatever.  So just to make it consistent with the others.

MR. WHITTEMORE:  I don’t think it is accurate to state that it is in widespread use.

MS. FRIEDMAN:  That is my point.  The way that is worded isn’t reflective of the way we have structured the other pieces.  It is hard for me to tell from reading this whether it is not in widespread use or not, that’s all

MS. AMATAYAKUL:  It is not in widespread use.


DR. FITZMAURICE:  Not used, but ready?

MS. AMATAYAKUL:  So the action is to support development of further guidance?

DR. FITZMAURICE:  What does that mean?  How about we just say, it is not used but ready, based upon the testimony.

DR. COHN:  Lynn.

MS. GILBERTSON:  One of the things that we did testify is pretty much the language Margaret used as I review it.  What we are going to do is beef up the imp guide so that it has more information on the use of these transactions, how you would use it, why you would use it, but we needed a little bit more interaction like with the vendor software, when you kick off a fill status, for example.  You don’t kick off a fill status notification when the medication is filled, because it may never have been picked up.  So there are different actions at the pharmacy that would kick off this transaction better, any other guidance we can offer.

So that is one thing the NCPDP took back as an action item, that we would add more beef to what is out there.

DR. COHN:  Judy, you have a comment?

DR. WARREN:  I think I understand what Maria is getting at.  Support development of further guidance really means that NCPDP needs to beef up your implementation guidelines, which is part of that standard.  So I don’t think we need to say you have got to beef up your standard.  I think all we do is just start there with, adopt the Script’s fill status, because you are going to be evolving that standard over time and changing your implementation guidelines.  That is all part and parcel.

So let’s not confuse it with the others.  It is out there, somebody is using it.  It may not be in wide use, but at least it is there and we can make recommendations that it be included in the pilots.  Is that what you are saying?

MS. FRIEDMAN:  It is like 1.4 above.  What Judy just said is exactly like 1.4 above.  It is there, it can be used, but you have the memory jog.

DR. COHN:  Harry.

MR. REYNOLDS:  We have had two or three comments in the last couple of hearings about privacy related to that.  If we are developing a process that shows the doctor the medication history, this doesn’t matter.  If I am bringing down to the doctor when they are doing the prescription what medications a person is on, that medication is not going to be out there at the PBM if it wasn’t filled.

Trying to argue over privacy, we have a totally different function that is handing that doctor every prescription that person has had filled anyhow, so this is a little different.  I’m not sure that is an issue.

DR. HUFF:  I think I would agree with you, but I think I would have some different assumptions.  My assumption is that the medication history always originates from the PBM, because the PBM may not know the whole history.  As you pointed out, there are over the counter drugs and other things that could have been given by oral history to the provider that contribute to the medication history.

So I think it is a good issue, but it is an issue that needs to be explored in the future.  I wouldn’t preclude using this transaction based on just one use case.

MR. REYNOLDS:  No, I’m not precluding that.  I am trying to take the privacy issue off the table.  If we are going to — during the filling process, if we are going to hand the medications from the PBM, nobody has asked the patient or anybody else about that.  So I’m taking privacy off, I’m not taking this off as much.

We keep bringing up privacy, like maybe we shouldn’t do this because of privacy.  We have eliminated privacy when you download somebody’s medication history, because that doctor is going to see what they got from four or five other doctors.  So I just think we are confusing this with privacy.  We need to take privacy out of play.  It keeps coming up, and I think it is not an issue.  We have set up a process that basically takes privacy away.

MR. BLAIR:  I’m concerned. I tend to not worry on a personal basis about all of those different medications that I might be taking being aggregated by a PBM and sent to my pharmacist.  I personally don’t worry about that.  But I think that there are members of privacy advocacy groups, and I am a little concerned that this is in public session here, that would take this and make this an issue.

I don’t know how to deal with it, but I don’t think that we could say that it is not an issue.

DR. COHN:  I think Margaret has something here that she wants to show us, please.

MS. AMATAYAKUL:  In the important related issues, we do have one area there on paper privacy and security protections applied to e-prescribing.

So just as a for example in this case that you are discussing, if the patient wanted to restrict the flow of medication history from a payor, they could request that of a payor under the HIPAA privacy rule.

DR. COHN:  I think you probably have that reflected here as whatever needs to be discussed or talked about, what you are describing as an important related action.  There are probably other things that we are not aware of.  I find it a little odd — and you have to remember, I am a physician, so perhaps I come at it with a physician view, but I am struggling to think about the privacy ramifications of the physician-patient relationship and the level of concern there is that a patient would not want to disclose to a physician medications that had been prescribed that he may or may not be on.  Then again, I am a physician.  Steve.

DR. STEINDEL:  Simon, my comment on this with respect to privacy was not that we should remove this, or I was making any comment about the validity of anything.  It was just a comment  that this should be solely in the domain of NCPDP, the discussion of the fill status notification, that there are issues that are associated with the fill status notification that was raised in testimony here that may be outside of the domain of NCPDP.

PARTICIPANT:  Simon, I don’t disagree with that.

DR. COHN:  So how would we change this then?

DR. STEINDEL:  This is like Judy pointed out.  This is our tool, and what we should do is add the parenthetical remark, so that we know when we are constructing a letter whether or not to add that sentence.

DR. COHN:  You are absolutely right.  So are we comfortable with what we have here?  Okay, let’s move on.

MS. AMATAYAKUL:  This one, test the capability of inclusion of decision rationale for drug choice when formularies are selected that exist today in NCPDP Script.  This may be more appropriately put down into the areas where we want to describe what we want to test.  But I didn’t want to lose it, so I will put a parentheses around it so that we can move it.

DR. STEINDEL:  Is that really a message?

MS. AMATAYAKUL:  No, it is really a piece of a message.

DR. COHN:  Judy.

DR. WARREN:  But if you are going to include a decision rationale, then there has to be a spot in the message, isn’t that what we are talking about here?

DR. STEINDEL:  I think it exists.

DR. COHN:  Lynn, it looks like you are thinking about that.

MS. GILBERTSON:  The DUE fields already exist, so it is maybe a testing functionality that exists.

DR. WARREN:  But wasn’t this in response to the e-mail that we got from Montana, that they are now doing — the boards of pharmacy are looking at having rationale included for some of the drugs?  Wasn’t this part of that?

MS. AMATAYAKUL:  There was some testimony that suggested that this inclusion would help eliminate or reduce phone calls.  I think the National Association of Chain Drug Stores had also commented that it would be helpful.

DR. COHN:  I think this actually was not only formulary, it also was when there was allergy identified.  Wasn’t that another reason that would indicate that the physician knew the allergy, but decided to override the decision support added?  That was my memory.

MR. MARTIN:  That’s right.  As soon as you start doing a DUR process within the office, and you are expecting it to reduce phone communications or any kind of communications between pharmacy and the office, you have to be able to let the doctor say, asked and answered in the message.  That way they won’t get a callback from the pharmacy, because they have already, as you said, looked at the drug-drug interaction problem, whatever those might be.

DR. WARREN:  And all those exist right now.  Do we have it in the right place as part of the standard?

MS. GILBERTSON:  It is a part of all the different transactions.  Any time a drug is transported in the transaction, whether you are a new prescription or a refill or whatever, the DUE information exists.  So it is not like it is a separate transaction, it is not a separate function.  It is all part of the information shared between the prescriber and the dispenser.

DR. WARREN:  So this is what goes in a free text field.

MS. GILBERTSON:  No, these are actual fields for level of effort, professional service code, result of effort, those kind of fields.

DR. STEINDEL:  I think in response to Judy’s comment, probably with respect to the letter when we construct it, it needs to go in a separate section.  It is not part of the message, it is part of the content.

DR. COHN:  Yes, it is more of a use case.  Isn’t that what we are talking about?  The standard is there, it has the stuff, it is just not used this way.  So this is really a use case investigation to see if this would be helpful at all, I guess.

MS. GILBERTSON:  I think it would be helpful to know whether or not it has ever been evaluated rigorously.

MS. CRONIN:  So does it really serve to inform drug utilization, or does it minimize phone calls.

DR. STEINDEL:  I don’t think so.

MS. GILBERTSON:  You are saying it exists, so —

DR. STEINDEL:  No, what I am saying is, the place in the standard exists.  We have the boxcar to put this information in.

MS. CRONIN:  This is not something that has already been put into use.

DR. STEINDEL:  I don’t think it has ever been really tested.

MS. GILBERTSON:  As of right now, probably the answer is no.  There were implementations in the late ’90s, early 2000’s, that used this pretty widely, but they are no longer in business.

MS. CRONIN:  In those implementations, did they ever look at this specific function?

MS. GILBERTSON:  I’m sorry?

MS. CRONIN:  You are saying that it was implemented in the late ’90s, but there was nothing ever reported out of that, in terms of what —

MS. GILBERTSON:  Not that I am really aware of.  I can’t point you to a paper or anything like that, no.

MS. CRONIN:  Do you know anything,  Ken?

MR. WHITTEMORE:  I was very involved with the NCPDP work group that was handling DUR back during the ’90s.  I am not aware of any complete use of those fields as they were originally created.  I think this is a worthy area for a pilot.

DR. COHN:  So I think we have it reflected there as you have described.  Is there any other NCPDP pieces here that we need to be looking at, so that we consider that we have done  NCPDP, and we can reflect on this?

MS. AMATAYAKUL:  Yes.  Let’s look at the formulary and benefit coverage information from payors to prescribers, because Rx Hub is going to go to NCPDP.

DR. COHN:  Two is the X-12 transaction?

MS. AMATAYAKUL:  Two is X-12.

DR. COHN:  So go through this one again.

MS. AMATAYAKUL:  Do you want me to read the observation over?

DR. COHN:  Yes, please.

MS. AMATAYAKUL:  NCPDP standard has inquiry and response — sorry, wrong spot.  Formulary and benefit coverage information from payors, PBMs, tool prescribers currently performed with proprietary messages, Rx Hub is submitting its formulary and benefit information, file transfer protocol to NCPDP to become an ANSE accreted standard.

At the same time, Medicare is a new payor in the retail pharmacy arena.  Based on experience with HIPAA transactions, it is possible the Medicare contractors may be required to have new or different requirements from those in existing standards.

The recommended action is, support the facilitation of NCPDP promulgating a formulary and benefit information file transfer standard, using input from Rx Hub.  Test and pilot, especially with respect to Medicare as payor.  Just ignore these X’s that I have put on here, because now that we have changed the headings, they don’t apply anymore, so we will have to add the X’s in later.

DR. COHN:  So is this a test and pilot?

MS. AMATAYAKUL:  Support the facilitation of getting the standard and test.

DR. COHN:  So it will actually go to test and pilot, in other words.


DR. COHN:  Comments about this one?  This one is going to require a significant amount of wordsmithing.  Harry, you were going to comment?

MR. REYNOLDS:  I won’t wordsmith it now, but I thought we had talked about, that NCPDP would work to adopt the Rx, different than with input from Rx, though.  That is different.

MS. GILBERTSON:  Not to confuse the issue, but this is the task force that is going to put together a different entity.  Rx Hub is one of them, Medi Media is another, people who are in the business of sending formulary and other information around.

So there is one issue if you want to accept for the pilot, for example, the Rx Hub in use or the Medi Media in use, whatever.  The other is the bringing together of those two plus any other industry organizations involved in this who want to build a consensus based standard.  It may not look very different when we are done.  I just don’t know until the different groups get together and start sharing their thoughts.  I didn’t answer your question.

MR. REYNOLDS:  No.  I thought yesterday when we left, the discussion was that Rx Hub’s would be the basis.  We would recommend that they come to NCPDP and get that made an ANSE standard.

The discussion was that Rx Hub was going to come and work with NCPDP and go through the process to develop a standard.

MS. GILBERTSON:  That was discussed, yes.

DR. STEINDEL:  That is different.

MR. REYNOLDS:  That is not what that is saying.  That is saying with input from Rx Hub, which is an equivalent input.  We are not using it as a base.  Yesterday I thought we came out of here that that would be the basis of the discussion, not the basis of the answer.  That is all I’m saying.

I just want to make sure we understand — because what this says is that it is just going in as a free for all into the standards process, which will take dramatically longer than if you start from a basis and make sure you get everybody’s input.  That is all I’m trying to understand.  This is Karen’s earlier point; what are we doing.

MR. MARTIN:  Rx Hub will certainly submit a DURF ultimately through this work group process, and their will be what you consider the straw man or straw person or whatever, straw thing, that we will look at to see if that is — it could be accepted unmodified, it may need to be modified to address every player’s need, but maybe somewhat different from Rx Hub’s perceived need for the industry, or it could be accepted carte blanche.

But that is what happens in the standards process.  We are not going to sit around the room and say, let’s make something up.  There is going to be something to start with which is an Rx Hub model, but it may be modified somewhat.

MS. AMATAYAKUL:  I’ll read this version.  Support the facilitation of NCPDP promulgating a formulary and benefit information file transfer standard using Rx Hub protocols as a basis to achieve industry consensus.

MR. REYNOLDS:  That is what we agreed to yesterday.

MS. AMATAYAKUL:  Test and pilot.

MR. REYNOLDS:  And I’m fine with that.

DR. COHN:  Terry, do you agree?

MS. BYRNE:  I do agree with that.  My concern on this one is, we do have a standard that is widely in use today, 30 participants that we testified yesterday are already using this.  Not that I am saying that is the standard that will get approved at some point, but I want it to still be clear from your tool here that the standard that we are using is widely in use.

Will that become the exact standard in two years?  Probably not.  It will probably be modified.  The question is, what are we going to use in the demonstration project, and to understand that, we do have really good experience with this transaction so far.

DR. COHN:  Judy.

DR. WARREN:  I think some of the concern we have is, Rx Hub is not a standard.  It is a proprietary solution to a problem that is in widespread use.  Somewhere we need to reflect that.

There are many strategies for playing in the standards development game, and you will figure that out as you go along.  But I do think that one of the strengths that Rx Hub has provided is the ability to develop the solution, to deploy it, and have a lot of success with it.  You have proven that the approach works, which is a real leg up in developing standards.

I think what we are struggling with is, our direction is to recommend standards, not to recommend solutions.  So we really need to support this taking of some of the Rx Hub’s solutions and putting them through an ANSE process so that we can get there and make that recommendation.

So does that help?

DR. COHN:  Let’s see what Kelly has to say.  You are talking about an industry solution as opposed to an ANSE accredited standard.

DR. WARREN:  I think we need to be careful of that.  One of the things that I try to be very strict with is use of terminology.  We start talking about informatic standards, most of us agree that the ANSE definition is the definition, and we need to stay straight with that, and not confuse it with other things.

Rx Hub can be considered a de facto standard like Microsoft is.  It doesn’t mean that Microsoft is an ANSE standard.  It just means they’ve got a really good market share.  I think that is what you are taking a look at.  It also doesn’t mean that the approaches don’t work.  They obviously do, or you wouldn’t have a good market share.

DR. COHN:  I think Kelly has a comment, and then I also have a comment.

MS. CRONIN:  I think that is an important observation, but I think we also need to be mindful of what standards or solutions that hope to be developed into ANSE accredited standards will truly enable some of the more important functionality.

I would just ask that the committee carefully consider something around this particular area for the NPRM that is going to be introduced this fall, primarily because we would be much better off if there is some clear direction on how to present formulary and benefit information in early ’06.  Perhaps there could be some consideration given to regulating only if something is through the ANSE accreditation process and obviously has had input from all stakeholders.

DR. WARREN:  So does this statement capture what you just said?

MS. CRONIN:  Yes, I think so.  It is just that this is more — orients us what to do.  I think the pilot — there is still some room to consider what could be done in the NRPM this fall, so endorsing what has been tested and what is being used today, but what will come out of that ANSE accreditation process.

DR. COHN:  I guess I am confused now a little bit.  You are trying to think of some way to accelerate this not to a pilot, but to an actual, this is in industry use, and with adequate industry experience.  I am hearing you —

MS. CRONIN:  I think that is one way to interpret it, but if the committee doe snot have consensus on that, then perhaps there could be more careful consideration to accepting a standard that would go through the accreditation process, but not accept it or endorse it as inclusion of the initial NRPM as it is right now.

DR. COHN:  That is where we would have to look at the subcommittee and see what their sense is.  My sense of this one, and it is an important wordsmithing set of activities, but I consider this to be a very important functionality.  I actually want to thank Rx Hub for moving this from their work into a standards development process.

Now, we need as a committee — and we need to probably advise the Secretary on ways to at least from my view accelerate the successful work of NCPDP to move this into a standard in a way that allow for appropriate standards body input and process, but also is an expeditious accelerated — whatever term we want to use for motion forward on all of this stuff.

I don’t know the exact wording, but I do know that the work of the Rx Hub to put it into NCPDP should not be considered to be a step backwards.  But on the other hand, I am having a hard time saying — not knowing what the outcome of all of this is, it is hard to approve it as a standard that has adequate history experience.  I have to ask the other subcommittee members.

This is also, what can HHS do, working with the standards development community to really accelerate this thing in a way that preserves the process.  I have to reflect on things like — I know I mentioned this to Lynn yesterday; I don’t think she liked the concept a whole lot, but one has to think about things like the HL-7 functional standards effort, where HL-7 worked very closely with the standards development organization, provided a lot of support.  It wasn’t like turning on a light bulb, but certainly was able to preserve the process as well as to accelerate the outcome.  Then it becomes the sort of things that I think we need to think about.

Stan, this may be only my view.

DR. HUFF:  No, I agree.  I would just add that it is probably not anything to do with the suitability of this for use.  But it has to do with the process going forward, because without this being part of an open consensus standards process, you have no guarantee that it couldn’t change next month without adequate input from the industry.

I think that is what you are looking for, is that protection that comes from the open consensus process, where everybody has an opportunity to vote.  So it is not deemed as unsuitable for use; it is that we really want that additional protection of process that would say it doesn’t change arbitrarily in the future, based on a minority opinion of a small group of people.

DR. COHN:  Yes, but the balance of that needs to be that this has to be an expeditious process.

DR. HUFF:  Right.

DR. COHN:  I don’t know the wording.  I have in mind the concept, and I think this gets to be action items.

MR. REYNOLDS:  What I am trying to do in looking at this, keep in mind what Karen had to say this morning, and the position the Secretary might or might not be in as far as moving forward.

That statement right there says that there needs to be a standard.  That is a fact.  But that statement also allows the Secretary if he chose to, to do a pilot using that, what is already out there, to prove the overall functionality of what is going on.

DR. COHN:  But what Kelly was bringing up was not the issue of a pilot.  She was bringing this up as an issue of something that has adequate industry experience and wouldn’t even require a pilot.

MR. REYNOLDS:  You could go there with that, too.  I’m not going there, but I’m saying you could possibly do it.

MR. BLAIR:  I’m just going to speak for myself on this.  I happen to feel as if what Rx Hub has done is really excellent.  It is something that works, it is pragmatic, it is out there.  It is an excellent solution, and it has support from at least a certain potion — but I don’t know the portion of the industry that has not had the opportunity to participate.  Have all of the e-prescribing vendors been active in their concerns and requirements in this?  Have the EHR vendors been active in their requirements on this?  That is not clear to me.

So my thought is that our responsibility in terms of recommending standards puts a very heavy weight on the fact that they are consensus based standards, that they have gone through that open process that are consensus based.  Otherwise, I don’t consider them to be standards, and certainly not nationally recommended.

That doesn’t mean that everything that goes into e-prescribing has to be a national standard.  The message solutions that Rx Hub has developed, they work.  They work whether or not they are designated as standard.  But my thinking is that it is our responsibility to promote openness and a consensus process and a maintenance process and continuity, and therefore, if we were to recommend it as a standard, I would want to see it ANSE accredited, and I would want everyone to support that transition of that process to make that as fast as possible, so that it could be fully recognized as a standard, because I think that has added value to the industry as a whole.

DR. COHN:  Other comments?  I see everybody nodding their heads.  Is there something we are missing here?  To me, this is like a big — you used the word acceleration on 3.1 as opposed to whatever you might have had there before.  So you have given us a bigger word.

I think there may be in parentheses a whole bunch of action items that maybe you have neglected that have to do with strong recommendations to HHS to work with NCPDP to facilitate that acceleration, both in personnel, time, money, whatever it may take.      I think we would also want to say that this is an area — I may be speaking for myself on this one — where if indeed the is starts moving forward, we think that this is important enough where we would want to have reports to the NCVHS on a regular basis to track the progress of it.

I am speaking alone on that one.  I am seeing heads nodding.  I am seeing a majority of the subcommittee’s heads nodding, to make sure this doesn’t get derailed.

MS. CRONIN:  Just one point I think might be useful, too, beyond recommending a more aggressive or accelerated schedule to go through that whole process, then what would be the recommended action coming out of that process to include in a demonstration or in a pilot, or would at that point it be ready for consideration for regulation.

DR. COHN:  Boy, isn’t that z question.  Steve and then Ross.

DR. STEINDEL:  If I can just comment to Kelly, I think the answer to that is a definite yes.  It depends on what comes out of the standard.  If it is vastly modified from what went in, probably we would want to test it before it went into regulation.  But if it was essentially unchanged, we may feel comfortable with saying you can go ahead and use it, because it has been in widespread use.  So I think that depends on what goes through the sausage mill of the consensus process.

DR. COHN:  Ross, do you have a comment?

MR. MARTIN:  I am thinking that there is a very big difference between the committee recommending and recognizing proprietary solutions that could be used in a pilot to inform the standards making process, than recommending a proprietary solution  as a standard that would then be mandated.

That is what I am hearing a little bit.  Maybe, Jeff, I misinterpreted your comments a little earlier, but that anybody would ever be required under a mandated regulation to adopt a process that has not gone through the standards process would be of great concern.

I can imagine you mandating a function and saying your tool, because there is no standard in this place, and let’s just use the formulary as an example, you have to be able to show formulary.  We recognize that the standards process has not gone far enough to lock that into place yet.  So for now, a function is required, but we are not going to tell you exactly how to do it.  Then assuming we get that standard locked in through NCPDP, and saying that NCPDP should be the ones to do the formulary standard is I think very appropriate.  Again, it drives the market to go in that direction, put all their eggs in that basket.

Did I misunderstand what the comments going back and forth were?

DR. COHN:  I think we have all been saying that it is not a standard until it is approved.  What we would really like to do is to get — and I think Steve actually said this very well, that depending on what happens with the NCPDP process, if it is minimal changes, the industry recommends it as something that has adequate industry use, it could be mandated by the Secretary.  If it has major changes, anybody would want to pilot it.

As I say all this stuff, yes, clearly if you wanted to pilot something, you would want to pilot what was going to be that new standard.

MR. MARTIN:  I just want to be very clear that I understand that, would this body ever recommend to HHS a non-ANSE accredited standard that would then be mandated?  You would only say that —

DR. COHN:  I’m not sure I even want to answer that, because I never say never.

MR. MARTIN:  Let me just tell you what the repercussions could be.  If you ultimately recommend —

DR. COHN:  I think we know what the repercussions could be.

DR. STEINDEL:  And I think, Ross, that is a very hard question to ask of this body, because this body makes recommendations to the Secretary in lots of areas, some of which your statement might be very appropriate, and other areas it might not be.

DR. COHN:  Marjorie.

DR. GREENBERG:  I just wanted to point out that you were correct obviously not to answer that, but that wasn’t what I wanted to point out.

DR. COHN:  Ross, I apologize for stopping you.

DR. GREENBERG:  It depends on what the application is, but none of the vocabulary or terminology standards are ANSE accredited, that have been recommended by the committee, because it is a different process.  Certain processes have been required or recommended, but none of them as terminologies or classifications as ANSE accredited.

But in the case of messaging standards, I think all of the ones that have been mandated are.

DR. COHN:  As I say, I think certainly the solution here from my view is the federal government working closely with NCPDP to figure out ways to accelerate the process while maintaining the integrity of the process.  Sorry, Lynn.

MS. GILBERTSON:  I don’t have any problem with that.

DR. COHN:  Good.  Steve.

DR. STEINDEL:  This is actually a question concerning the word accelerate.  Having been one of the people that was intensely involved in acceleration of the HL-7 EHR draft standards for trial use, there is one area that we can only accelerate slightly, and that is the part of the ANSE process, the voting and what is required by ANSE in this time frame and that time frame, et cetera.

If you have certain processes within NCPDP to meet standards, like it can only be brought up within so many days of a meeting or something like that, there is nothing we can do to accelerate that process.  We just have to work within that.

Now, what we can accelerate, and what HHS can help facilitate acceleration, is what the did with the DHR standard, is what went on between those dates.

DR. GREENBERG:  Right, exactly.

DR. COHN:  And obviously, we don’t need to micro manage how HHS might help accelerate all of this.

Now, I think what we have heard is a couple of things here.  I think we have at least in our mind that this is an important recommendation, and is beyond just adopt or whatever.  I think we also are probably saying that —

MS. AMATAYAKUL:  Shall I read it, Simon?

DR. COHN:  There is a piece here that is missing, I think.  Why don’t you read it, and I’ll mention the piece that I think is missing.

MS. AMATAYAKUL:  Support acceleration of NCPDP promulgating a formulary and benefit information in file transfer as an ANSE standard, using Rx Hub protocol as the basis to achieve industry consensus.  HHS should work with NCPDP to facilitate providing personnel, time and financial resources with regular reporting to NCVHS in order to prepare further recommendations relative to adoption.  Consider additions that may be required by Medicare as payor.

DR. COHN:  I think the other piece that I am going to suggest is that NCVHS will provide updates to you about whether the results of this work are appropriate for regulation or piloting.

MR. BLAIR:  Say that again?

DR. COHN:  NCVHS will provide advice, updates, whatever we want to describe it, to the Secretary — I think this is what Kelly was asking for, could this outcome be something that didn’t need to be piloted, but could go to regulation.  I think depending on what we are seeing, we could advise the Secretary about what the appropriateness of this functionality and the standards as it comes out should be in terms of what the Secretary should do.  I think we have a right to either pilot or, depending on what the industry says and all of this stuff, it could go into regulation, depending on the timeliness and all of that.

I am just thinking about adding that as another piece here.  Jeff, are you comfortable with that?

MR. BLAIR:  For now. Let’s go ahead and leave it.

DR. COHN:  Argue about it later.

MR. BLAIR:  Well, yes.  I guess I feel like there is tremendous consensus here.  Is that redundantly repetitious?  There is consensus.  I think everyone including Rx Hub seems to be feeling like it is the right thing to do for it to become ANSE accredited.  I think the issue that Kelly is expressing is that she is a little concerned that the process doesn’t move fast enough, will it be ready for January 1.  I think that many of us might share that concern, too, but I think what our focus should be is trying to make that successful, to do whatever we can, to accelerate that process so it will be ready for January 1, because that is the right thing to do now.

DR. COHN:  And Kelly is asking for January 1 for what?

DR. GREENBERG:  Of what year?

DR. COHN:  For piloting or —

MR. BLAIR:  2006 to participate in the pilot.

DR. COHN:  I think what Kelly is asking is, depending on what comes out, whether it actually needs piloting, or can it be mandated by the Secretary, is what she is asking.

MR. BLAIR:  The only thing that I feel is that if it is not a standard, if it is not ANSE accredited, in my mind it is not a standard.

DR. COHN:  Jeff, I think that is what we said.

MR. BLAIR:  Yes, but the Secretary is perfectly free to go forward and we don’t have to have everything as standard.  There are portions of the e-prescribing solutions that don’t necessarily have to be standards.

MS. CRONIN:  There is another effort going on in the private sector with certification that could also be addressing some of these issues.  But my only thinking behind this really is, if we miss our opportunity to make reference to this in a meaningful way, and the NPRM coming out in the next several months,  then it may be several years down the road before this ends up becoming something that would go into a formal government requirement.

That period of time may allow for other competing standards to come up.  Maybe that is a good thing, maybe that is a bad thing, but I think there is a very important need to get this right and to have standardized representation of very important information to inform decision making, and make sure that beneficiaries get the most cost effective alternative that minimizes  their out of pocket costs.  If we can get that right in ’06, that would be the ideal.

So that is the goal we are shooting for.  We want to do that in a responsible way so that we are not hurting anyone in the process, but at the same time, we don’t want to drag it out into what could be a much more lengthy process that could take several years.

MR. BLAIR:  Correct. I think I share all of your thoughts here.  The only thing that I worry about is, if you are winding up holding out the possibility that, well, don’t worry about making the ANSE accreditation process by the deadline, because we are going to wind up mandating it, anyway, then that takes away the incentive for everyone to work hard to try to achieve that.  So that is the only concern that I would have.

DR. COHN:  It is time for lunch.  Are we okay to stop for the moment?  I had actually anticipated that Margaret was going to handle four and not three here, but we have just taken a big gulp here, because this is one of the major issues.

We will reflect on that.  I think we have come up with what I think is a consensus of the subcommittee in terms of the advice, direction, guidance and next steps, but it is something I am sure we will further discuss.  Hopefully when we get back after lunch, we can move on to other parts of the Script standard, so we can complete the Script standard discussion, and then reflect on what we want to do with that, and then move on to the other parts of the message formatting standards.

We will take an hour’s break.  We will get back together at 1:20.

[The meeting recessed for lunch at 12:21 p.m., to reconvene at 1:20 p.m.]

A F T E R N O O N   S E S S I O N   [1:35 p.m.]

DR. COHN:  I think we have got some good work done this morning.  I think at this point we are going to continue on NCPDP Script, and there was another general NCPDP recommendation that we will take a look at, both observations and recommendation.  Then we will take a second when e are done with that area to go back and look and make sure that we are all comfortable with what it is we want to do about this, or what general statements we want to make.  Then we can begin to look at some of the other standards.


MS. AMATAYAKUL:  Just to review the observation, NCPDP Script standard has inquiry and response components for each function.  For special situations in which an acknowledgement of receipt is required, NCPDP Script has a request for return receipt for use by prescribers.

The proposed recommended action might be something like, adopt NCPDP Script request for return receipt function.  That function works and is in minimal use, I believe.

DR. COHN:  Harry.

MR. REYNOLDS:  I have a question of understanding.  Inquiry and response.  Who is inquiring and what is the response?

MS. AMATAYAKUL:  For each of the ones that we had looked at earlier, for example, the renewal, there would be a refill request and response.  So there would always be some interaction on both sides, the prescriber and the dispenser.  Whoever initiates it, there will be a corresponding response back.

DR. COHN:  So this is return receipt?

MS. AMATAYAKUL:  Yes.  This particular one is a return receipt.  There were some people earlier on who said we need a return receipt for every single one.

DR. COHN:  Why don’t we call this return receipt?

MR. REYNOLDS:  Again, thinking back to the standards that exist, when you start talking about inquiry and response, you could slide into 276s.  But I think we need to make sure we don’t leave it out there so that — inquiry is a key word in the standards world.

MS. GILBERTSON:  That is actually the official name of the transaction, is a verify. But I would really recommend that we just strike this row.  This is a transaction that is there, similar to a password change or other housekeeping transactions that exist if the trading partners need to use them.

But as I explained yesterday, if the transport vehicle might be the USPS, if you want to have an extra service of verification, you can choose to use that transaction in the standard.  But it provides no extra functionality, other than a return receipt might.  So I am not really not sure if it needs to be in here.

DR. COHN:  Are you saying that this may be a little picayune?

MR. REYNOLDS:  Yes, I think this is a level of detail that we don’t want to get at in a letter to the Secretary.

DR. COHN:  Others in agreement with striking this as a — it is not at the level where we need to be recommending action?  Jeff, are you okay?

MR. BLAIR:  Fine.

DR. COHN:  Item five.

MS. AMATAYAKUL:  Five.  NCPDP has initiated work on a directory that would identify prescribers, nursing facilities and pharmacies that are able to accept electronic prescribing transactions.  ASC, X-12N, 274 provider roster is an existing standard that may be useful in supporting this transaction.

Rx Hub is submitting its own NCPDP based provider list to NCPDP.  So the possible action might be to support adoption of a directory that would identify prescribers, nursing facilities and pharmacies that are able to accept electronic prescribing transactions.

DR. COHN:  Maria.

MS. FRIEDMAN:  I am missing a few pieces here.  To me, the issue on this is, do we have a prescriber I.D.

MS. AMATAYAKUL:  Identifiers are later in the list.

MS. FRIEDMAN:  Okay.  I guess I am just out of phase.  This is out of phase for me.

DR. COHN:  It is not out of phase, but i am seeing this, and I was pointing out what NCPDP had recommended.  I think you have identified appropriately that — I don’t know about initiated work.  What I saw here was, NCPDP will facilitate bringing the industry together to modify the NCPDP Script standard, exploit the communication of prescribers, nursing facilities and pharmacies that are available electronically.  The process is estimated to take one year.  So it sounds like this is another one of those Rx Hub standards that is being —

MS. GILBERTSON:  If I am not mistaken, Art testified yesterday, but mentioned that another company, ProxyMed, had actually done the work, and Proxymed will be bringing it forward.

DR. COHN:  Right.  There is also this basic research utilization, the 274 provider transaction.  I am wondering what we are doing.  Is there a belief that first of all that provider directories are important?  I think there is a view that that would be a useful functionality if people learn how to send them.

DR. HUFF:  I was thinking that this one was one that we could recommend in March, rather than trying to recommend it right now.

MR. BLAIR:  I’m a little lost.  What is the message?

DR. COHN:  It is provider directory.

DR. HUFF:  It is actually transactions that load — it is a format for giving somebody a directory of pharmacies and prescribers and nursing facilities.

DR. COHN:  It is fine to defer this, but what would be do between now and March about this?

DR. HUFF:  One of the questions is, how badly is this one needed.  It isn’t apparent to me.  This normally — I would guess what is happening today is, if they don’t use the standards, somebody just says, here is the file, and they figure out how to load it.  This is an order of magnitude more simple sort of activity usually than prescription transactions or the other things that we are talking about.

DR. COHN:  So basically what you would like to hear between now and March a little more from NCPDP about it, also maybe hearing from X-12 —

DR. HUFF:  About how close what they have got would be to meeting this need, that sort of thing.

DR. COHN:  Other feelings?  Harry.

MR. REYNOLDS:  I would agree that there is already an existing standard out there.  It would be good for us to hear why we wouldn’t use that existing standard with the 274 before we would recommend adding something else out there.

DR. HUFF:  I’d be welcome to somebody correcting me about whether this is a hard interface, and if this presents a hardship if we don’t take action until March, then I would be amenable —

DR. COHN:  Well, our only action appears to be recommending that it be tested and piloted at this point.  So I’m not sure if that is not the strongest action that one can take.  I see Terry coming up.

MS. BYRNE:  I’m not sure that it is not as important as say formularies.  However, the only hardship is that for vendors that are doing electronic prescribing today that have to work with multiple entities like Rx Hub and Sure Scripts and ProxyMed, they will have to develop to multiple standards, because we are all using some version of this, but none of them are the same.  They are close.  So that is the only hardship, that everybody who is working with multiple networks test and develop multiple standards.

DR. COHN:  Michael.

DR. FITZMAURICE:  Terry, who should pull people together to do this?  Who should pull people together to say let’s have a — and we agree on a single format.  Is it NCPDP?  Is it somebody else?

MS. BYRNE:  I think it is very appropriate for NCPDP to do it, because NCPDP is pretty much the entity that works with prescribers and pharmacies, and that is who is providing this information.

MS. GILBERTSON:  And I believe the only reason we brought up the 274 is because we know it is there.  We don’t really know anything about it.  So we were just trying to let the committee know that there was something out there.

MS. BYRNE:  And the 274 is an X-12 transaction, which is not technically used by pharmacies.  Also in talking with Lisa Miller yesterday, I don’t believe — and also in doing some research around X-12, we can’t find anybody who is using this transaction, not to say that there isn’t somebody.  It was probably designed more for providers, physicians from a health plan, somebody who is participating in a health plan versus somebody who is participating in trading information for electronic prescribing.

DR. COHN:  Marie.

MS. FRIEDMAN:  There is no standard, so this is in some ways another related issue.  There is no standard.  This is nice to have, but not critical.

DR. COHN:  This is a modification or an improvement to the NCPDP Script standard, to enable this.  So far it sounds to me like it isn’t Rx Hub, but it is a ProxyMed being brought forward as a possible standard, or at least an additional functionality for NCPDP Script.

MS. GILBERTSON:  Right.  As Terry mentioned yesterday or during testimony, NCPDP had worked on what we called the provider broadcast a few years ago.  ProxyMed was one of the entities leading the charge, and lack of resources fell by the wayside.  It wasn’t as high a business as the other transactions, so it went dormant.  But it will now wake up.

DR. COHN:  I guess what we have on the floor is either trying to do something with this or deferring this particular issue to further advice in March.  Preference of the subcommittee?  Mike.

DR. FITZMAURICE:  I guess the committee could recommend that this is a gap, and a gap with a potential solution.  The hope is that the industry will band together and come up with a standards solution for it.  That would be useful for the committee to do.

DR. COHN:  So that begins to look like the recommendation that is being brought forward there.

MR. REYNOLDS:  I still struggle. There is a standard out there, whether or not it fits this particular situation is the question.  If it does, the process is dramatically different, starting all over with that bottom sentence.  In other words, you have already got a standard that has passed.  It has already been through everything it needs to go through.  If we make a clear directional change to not use it, there will be people who will want to know why.

So I don’t think you can quietly walk way from a standard that appears to be in that same realm, killing it and going to something new.  I don’t think we can accidently walk away from it.

DR. COHN:  Then we should hear from X-12 and hear about the 274.

MR. REYNOLDS:  I think most of the people in the room will look at the piece by the time we get to see it.

DR. COHN:  So Margaret, would you make a note for us then that it isn’t just March 2004, but it is parentheses, need to hear from X-12 regarding 274 prior to making a recommendation?  I think we are hearing from NCPDP, but we can also hear from ProxyMed certainly.

DR. HUFF:  Just technical on the wording.  It is really not a directory that we are standardizing.  We are standardizing how to import or transport a directory.

DR. COHN:  Is this a directory format?  Or are we even doing that?

DR. HUFF:  What’s that?

DR. COHN:  A directory format?

DR. HUFF:  Yes, or directory loading format or something like that.

MR. BLAIR:  But is it the format or is it the message?  I don’t know.

DR. HUFF:  I’m using format as a surrogate for message.

DR. COHN:  Maybe between now and March we can figure this out.

DR. WARREN:  What I am unclear about, is NCPDP already working on developing a standard for this?

MS. GILBERTSON:  We had.  The standard was pretty much packaged and it was shelved, so that particular information will be brought back forward for November.

DR. WARREN:  So maybe what we need to hear, as long as we are looking at X-12 and ProxyMed is also see what NCPDP is doing in this.

DR. COHN:  Exactly.  So those are our next steps.  We will just remember to do them between now and March so that we can deal with this recommendation.

How many other — I’m trying to see whether we should stop now and look at our script or whether we should take a look at all the other NCPDP pieces.

MS. AMATAYAKUL:  There are two items for X-12.

DR. COHN:  I don’t want to do X-12 for the moment yet.  We talk about X-12 and NCPDP.  We are making the process up as we go.  Should we just continue going forward and then we will begin to look back on this at some point?

DR. HUFF:  If I understand right, we are going to lose the First Data Bank representative at 3 o’clock.  I would like to get to some of our coded terminology recommendations before he leaves, in case he has important input, if we can do that.  I don’t know if that works.

DR. COHN:  I think if we move forward, we should be able to do that.  We will just make sure that as soon as we are done with this area, we move into terminology.  I’m sitting here looking at the clock, and realizing that we only have an hour.

MS. AMATAYAKUL:  Let’s go to terminology.

DR. COHN:  Were you hoping that we would handle this Rx Norm to NDC as it says as a beginning piece?

DR. HUFF:  I think this is the cart before the horse, because we need to say something about whether we want Rx Norm first, or this mapping isn’t needed at all.  I would say the first question is, do we think Rx Norm is a good terminology for drug ordering.

DR. COHN:  That probably is a good first question, isn’t it?

MR. BLAIR:  We received a lot of testimony, people saying that would be a good solution.

DR. COHN:  I think we heard a lot of testimony saying it might be a solution.  I don’t think anybody has tested it yet.  I think what we are seeing here is that probably a recommended action here is to test Rx Norm in a message to see about its usefulness.  I’m sort of making it up here, and it may be somewhere else here.

MS. TRUDEL:  I just wanted to raise an issue that was brought up yesterday, where someone suggested that if we are going to make a recommendation for Rx Norm, we talk about which terms or levels of Rx Norm, rather than — it describes a lot of different concepts apparently, and it would be helpful to put some parameters around what we are recommending.

DR. HUFF:  Yes, so keying off of both of those things, I think what I am saying is that we would suggest that we use Rx Norm specifically for the code for the ordered medication in the Scripts transactions.

A related thing is that we would suggest using NDC codes for the filled drug code, to represent the drug once the prescription was filled.  Then the corollary to that is, or the follow-on, is that if people are amenable to that suggestion, then what particular parts of Rx Norm.  The part that I am most familiar with is the clinical drug part.  I probably need some further education about — the other function that you want is, you want the ability to order a proprietary thing and say fill as ordered, so that you can order that exact brand.  But I’m not sure what the main parts of Rx Norm are that would allow that to happen, or if they are appropriate or exist or whatever.  Actually it would have been nice to have Stuart here.

MR. BLAIR:  In addition to the requirements that you just articulated, Stan, don’t we also need to make sure that we have the portions of Rx Norm that will facilitate mapping among the drug knowledge bases, and maybe the person from First Data Bank, since he is still here, could help us.

MR. ROBINSON:  The first question I would have for the committee in looking at the act is, what business applications are we trying to support within the Script standard with inclusion of interoperability identifier.

So I would agree with what Stan was saying.  What you are trying to represent is the concept of the idea of the drug in the prescriber’s mind and we are replacing with a code what normally would be on a piece of paper.  So that really is the semantic. Within Rx Norm speak. there are two different concepts that fulfill that need.  One is the semantic clinical drug, which is the representation of the ingredients.  The salt is dropped when it is irrelevant from an order entry perspective, more from a hospital environment, but it also includes the strength of the formulation and the dosage form.

So there is a generic representation of that.  You may see for example Fluoroxitene 20 milligram, oral capsules, and there is also a brand representation of that which would be for example Prozac 20 milligram oral capsules.  If you are trying to represent what physicians typically write today, you need both, because sometimes it is written as a brand with dispense as written, sometimes it is written as the generic.

The distinction I was making earlier when we were talking to a couple of people on the committee was that level of abstraction is good for order entry.  It is codifying the concept that the prescriber intends, but at which point does the pharmacist really need to eyeball the code and the text and for order fulfillment find the inventory that would fulfill that order.

So I think — and this is the point I was making yesterday — right now what we get from you is the relationship of the generic to the NDC, but there is not a relationship of the brand to the NDC.  In talking to Randy Levin, we believe we may not get that type of a link until the structured product label is published.

So what that means for the drug knowledge base vendors is that we will need to take that brand representation and do some additional linking on our own.  I’m not sure if that helped.

I guess what I am recommending, if the order is being written in generic terms, it is the generic version of that code, the semantic clinical drug.  If it is being written in brand terms, then it is the semantic brand drug.  My associate, Karen Eckert, may have some comments on that, too.

DR. COHN:  Lynn has that process laid through here.  Comment, please.

MS. GILBERTSON:  One of the things that I could think of as the recommendation at least between now and X date is what we discussed, the NLM project that we talked about yesterday, of taking 100 different examples or whatever of what a prescriber would prescribe, in all different kinds of environments, so that you get the odd as well as the common, and mapping it all the way through the e-prescribing system, and being able to report back to the committee on what we found, because that will show us where there are gaps.

It may show us that we have got a total win and the picture is clear, but we would be able to at least report back with specifics rather than thoughts that there might be some conversions that have to take place and gaps that might occur.

DR. COHN:  I am trying to figure out to what level of granularity and specificity we want to get here.  I think the truth is that testing the availability or the ability of Rx Norm to work in NCPDP Script for prescription might — it might need to happen, but whether or not we say them in a recommendation —

DR. HUFF:  That might be a sufficient statement for the letter.

DR. COHN:  So let’s keep that in mind, as opposed to solving the entire problem.

MS. ECKERT:  Karen Eckert from MediSpan.  I have two slides.  I don’t know if they would be helpful or not.  One is a diagram that is from the National Library of Medicine and their information regarding Rx Norm.  It shows all the different term types of Rx Norm, to know what is out there.

Then I have the same sort of information in words.  Would that be helpful for me to share with you?

DR. COHN:  You have the same information in what?

MS. ECKERT:  The same information but in words instead of in a diagram.

DR. COHN:  Sure.

DR. HUFF:  I think that is entirely appropriate if we want to pursue this and say which parts of Rx Norm, or if we just want to make a recommendation that we go ahead with the experiment, than that detail would happen in the experiment rather than happening here, which is probably the better approach, actually.

DR. COHN:  Yes, that was my — given that we have heard extensively about all the various — it is not quite terminology overload, but there is that opportunity there.

MS. GILBERTSON:  And there is the other side of the equation.  If you go back up to the messages and say new prescriptions, et cetera, are working today, they are working with text, and that is what a prescription does, pull down or scribble or whatever.

So depending on how far along the low-hanging fruit you wanted to go, you have the functionality today.  Is it perfect, is it codified, is it whatever?  No, but at what level along your time line do you want that, now versus later.

DR. COHN:  I think what we were talking about was that this was a pilot for 2006.  I don’t think anybody had been suggesting that this is recommended for the Secretary to be identifying as a standard prior to piloting.

DR. HUFF:  That is true.  The point again is that it is not arbitrary where we cut off.  To get the benefit that is asked for in the legislation, this needs to be a coded thing so that you can do automated drug utilization review.

So that is not a question.  It needs to be coded.  It is not sufficient to have it stay forever as a textual item, I think.  I am asserting my own opinion there.

DR. COHN:  Can you give us the quick one on this one?

MS. ECKERT:  George can help me on this, too.  We both use this diagram.  This was where my question was yesterday, or my comment.  The first level on the top is the ingredient name, either the citrozene or the citrozene dihydrochloride.  They also have the brand name version, with the text type of brand name, which is Zertec.

If you come down to the third box on the left, that is the semantic clinical drug that we have talked about quite a bit, and was the original impetus of Rx Norm when it was providing the mapping between proprietary concepts by the different drug compendia.  That is showing you the citrozene, the strength, and the product route included with this dosage form, showing this in oral tablet of five milligrams.

They have also, on the third column on the right-hand side, associated that with the brand name.  So it is showing the Zertec name.  That is the type of semantic brand drug.  Then they have a different form, where it is just the drug with its strength and measure, and another place where it is the drug with its dosage form, without regards to the strength.

But these are all different term types available within Rx Norm.

PARTICIPANT:  So what does that mean as term type?  Which of these blue boxes has a unique identifier?

MS. ECKERT:  Currently the semantic clinical drug is the one that is the interoperability, that links MetaSpan concepts to First Data Bank.

DR. GREENBERG:  Nobody can hear you if you’re not talking into a mike.

MS. ECKERT:  Semantic clinical drug, the SED, that is what links MetaSpan to FTB to Moltem to Micrometics.

DR. COHN:  Having said that, is there a tremendous value to informing the Secretary of that?

DR. HUFF:  I don’t think so.  I think it helps with our agreement and understanding of what is going on, but I think that is a detail that we work through in the experiment that has been proposed, or the test that has been proposed.

DR. COHN:  So let’s go back to — so basically, we have talked about number one and its use within NCPDP Script.  What do we need to say — and I will apologize, since Margaret is getting herself back connected again — about its use with drug knowledge bases and mapping to them?  Is there a specific recommendation?  I guess that is the current recommendation we had up there.  Let’s just take a look at that for a moment.

MS. AMATAYAKUL:  So shall I read this?

DR. COHN:  Sure.

MS. AMATAYAKUL:  Test Rx Norm using adequate number of samples for the ordered medication, that is, semantic clinical drug and semantic branded drug, strength and dosage form, the NCPDP Script transactions, and map throughout the system to use of NDC to represent the full drug.

MR. BLAIR:  What was the last words?

MS. AMATAYAKUL:  And map throughout the system to the use of the NDC to represent the filled drug.

MR. BLAIR:  Maybe that is more than one activity.  There is a mapping to NDC, that is one piece, and then the other piece is mapping of the semantic clinical drug among the drug knowledge base vendors.  That is a different mapping.  So maybe this is two different recommendations?

DR. HUFF:  I don’t even think we want to — in this particular case, I think I would make this more abstract and say we want to do an experiment to see whether Rx Norm works throughout the whole process.  That is really what they are saying.  If you order with Rx Norm and you pass that prescription to a pharmacy, can they fill the pharmacy, make the map to the NDC code, can they send back a refill and renewal?

There is a whole bunch of detail in that experiment.  I would just say what we want to do is do that experiment and not break this down into each individual step that would be part of that experiment.

MR. SCOTT:  The outcome you are looking for is the code to give you a link to an NDC that is consistent with the concept that the prescriber had.  That is the measure of success at the end of the day.

DR. COHN:  Randy.

MR. LEVIN:  To do the experiment and to see if that works, there is one thing.  To put this into a function, the map to the NDC is a difficult step.  Right now, there is not — with our current processes of creating NDCs and keeping track of NDCs, we won’t be able to keep Rx Norm up to date.

DR. COHN:  So Randy, help me with this one.  Are you suggesting that we not go forward with a recommendation about Rx Norm or anything else in this area, because it is not going to be perfect yet?

MR. LEVIN:  If you want to keep Rx Norm up to date, you are going to need the process of generating national drug codes.

DR. COHN:  I don’t know that that is true.  I understood from the NLM — and maybe Vivian will jump in here — I understand they have contracts between NLM and drug knowledge base vendors to update the NDCs, since we know the FDA is not able to supply them in a contemporaneous fashion.  Am I overstating that?

MS. AULD:  The NDC codes that NLM has are the ones that come from FDA.  We have the others through the drug knowledge bases, but our source for NDC codes is the FDA.  So I think we should go ahead with the test just to make sure that Rx Norm as it currently exists will work with NCPDP Script in the future, but it has to be with the knowledge that there is going to be gaps that of those NDC codes that are on the fringes.

MR. SCOTT:  Just to be clear, the link to NDC is actually going to occur as I see it through the proprietary name based identifier that is organic to the pharmacy management system, not from what is published through UMLS.

So what is going to happen is, the Prozac 20 milligram oral capsule is going to be translated — let’s imagine for example that it is a physician management system that uses MediSpan.  That concept will be translated to Prozac 20 milligram oral capsules in Rx Norm terms.  That then needs to be translated to — let’s say the pharmacy system uses First Data Bank.  That is translated to an equivalent concept, and then as the relationship between First Data Bank’s name based identifier in our NDC file that is going to present candidates for order fulfillment.

Our reliance at this point in time on the NDC file to an Rx Norm concept would be there to help us manage our links.

DR. HUFF:  So just to make sure I understood that, what I was envisioning is that this could happen in two phases.  The second phase might not ever happen.  In fact, the first phase would be one where we are ordering Rx Norm codes, but the real mapping to NDC codes is happening through the proprietary knowledge bases as described.

You can imagine at some future date that people would have developed equivalent knowledge bases directly against Rx Norm or the public known identifier, whatever that is.  At that point, then it would be essential to have the mapping from that code directly to the NDC codes.

I think the reality today is that both prescribing systems as well as the pharmacy systems are assuming that they have access to one or more of the proprietary knowledge bases, and that is going to be the reality for several years to come.  So I think it gives us time to worry about making the NDC code mapping process real time in that time period.

DR. COHN:  Jeff had a comment, Harry had a comment, Lynn’s hand is up.  Let’s see if we can come to some agreement on what we are going to do.  Jeff.

MR. BLAIR:  You skipped one.

DR. COHN:  Harry.

MR. REYNOLDS:  I have a specific question and then a concept issue.  Does NCPDP Script, new prescription, have the capability of handling the NDC number and/or an Rx Norm number?

MS. GILBERTSON:  It currently handles a drug qualifier, which could be an NDC, a UPC, an HRI, a GPI, whatever you wanted, and the actual I.D. itself.  It does not at this point handle a code value for Rx Norm, because we aren’t sure yet until we get through with the exercise exactly what the standard needs to support.  If it is as simple as a code, we will add it, but I want to make sure we know whether we need to add two or three fields, or whether we need to add some instructions that say, if you are using text or a code value, use these fields, but if you are using Rx Norm, this stuff is all implied, so you don’t have to submit these fields, that kind of cleanup stuff.

MR. REYNOLDS:  Where I am going with the question is, if you think again back in the doctor’s office, they are going to be — whether it is a laptop or it is a PDA, they are going to be seeing the name of the drug.  They are going to pick the name of the drug.  They are not going to pick a code, right?

DR. HUFF:  Right.

MR. REYNOLDS:  So you have got the name of the drug, and once that is selected, that is going to get sent to the pharmacy in some way.

The way to look at it is, if both of those codes are captured at that time, then we have got a record that if Rx Norm is not ready, then it is in NDC.  But the point is, the doctor is picking a name, and somebody else is putting the codes in, not the doctor.

DR. HUFF:  In all of the systems the way they work now, that name is already attached to a code, so when they pick the name, they are picking the code.  It is invisible.

MR. REYNOLDS:  But that is my point.  So we have picked the standard transaction.  If that standard transaction allows for both, or can allow for both, then you could implement it, if you are only putting in NDC now, and as Rx Norm grows, and the mapping grows, both of them could be in there.  So that throughout the process, as you think of the whole process end to end, some people would want the Rx Norm and some people would want the NDC.  If we are trying to set standards, if the standard allows both of them to be in there, you end up picking the name of the drug, then —

DR. HUFF:  But there is some disconnect.  Most ordering either does or you would like to have happen generically rather than brand oriented.  NDCs by definition only handle brand ordering.  So what is really happening today is that people are choosing not an NDC code, but they are choosing one of the proprietary codes.  What is happening is that it is linked to a GCN sequence code or the equivalent code in the proprietary terminology.  That is what everybody is doing today, I think.

MR. REYNOLDS:  But again, they are picking the name of the drug.

DR. HUFF:  They are picking the name of the drug, and the code is not an NDC code, though.

MR. REYNOLDS:  I agree.

DR. HUFF:  It is a proprietary code that represents essentially the equivalent of a clinical drug, a generic clinical drug name.  You all correct me if I start lying, Karen.

MR. SCOTT:  My suggestion was, this may not be a battle you have to fight right now.  I think my recommendation with Script would be that they be allowed to pull in as a qualifier reference to either administrative code or standard terminology as deemed by HHS, and leave some latitude there.

So if you have got a representative NDC, which some systems are doing, that is what they want to send, as systems evolve they may want to send an Rx Norm concept.  We can’t anticipate all situations.  There may be just simply text that is sent for compound prescriptions and things like that.

So I think leaving some wiggle room for the standards to evolve over time, and just loosely make a recommendation that it be consistent with that might be sufficient.

DR. COHN:  I think we may be talking about two slightly different things.  I think on the one hand, we have Script functioning as it is now, and I think what you are describing now is the reality of Script in current day.

What we are talking about right here is a very specific pilot test that relates to the use of a structured, codified medication that could be selected by a physician and prescribed.  So that is a little different than what you are describing right now.

MR. SCOTT:  The pilot test though should probably be able to handle all three situations where there is a textual text based prescription, where a representative NDC is used, or perhaps a quarnine, and alternatively an Rx Norm.  Maybe each of those scenarios is something that needs to be supported and tested.

DR. COHN:  Lynn, you had your hand up.  I’ll let you chime in at this point.

MS. GILBERTSON:  Just to comment, because I don’t have the solution at this point, and I don’t like to make comments like that.  Not all things that are prescribed will translate eventually into an Rx Norm code and an NDC.  Recognizing that in a pharmacy system, UPCs and HRIs are used, so there is going to be a gap, because not everything can be handled through Rx Norm.

I don’t have a solution at this point, other than the text that is there and the proprietary codes that are used right now.  But it is something to at least recognize.  If you made a comment, you would probably get comments back from the public, saying that is not all they write for.

MR. REYNOLDS:  Where I was going was, if you have the three different ways of dealing with it, and I am talking as an implementer now, if in some cases they are going to order strictly on free text, as somebody just said, and that is going to exist forever, the foreseeable future that we are dealing with.

The second is whether or not we would want an Rx Norm, and the third is whether or not there is an NDC.  Philosophically, if the standard allows all three of those capabilities, when it goes to the pharmacy, the pharmacy uses NDC and it is not there, somebody was going to use an Rx Norm and it is not there, then they take the test.  I’m not sure that you have it handled.

If the standard can handle it, then obviously what comes in as the information is what everybody is going to process with.  But the standard is fine.  The standard would support all three.  The text is mandatory.  We use that as the least common denominator.  We always use the least common denominator.  So the text is mandatory.  You assume an NDC would be there, and then you build the Rx Norm or whatever else we decide as it grows.  It allows you to go.  That is what is happening right now in the paper scripts.

DR. COHN:  We know that NCPDP Script handles free text, in terms of prescription and all of that.  I think you said it handles — when somebody knows the code and is having a hard time, but that is a vanishingly rare phenomenon, I would imagine, that you would have the same NCPDP code —

MS. GILBERTSON:  Some of the testifiers, from the prescriber point of view they used a subset of the full NDC to give you a — is it the drug class?  Is that the right terminology?

DR. COHN:  Yes.

MS. GILBERTSON:  To at least relay the information that way.

MR. SCOTT:  They are just taking a representative one.


DR. COHN:  And then they are doing that, okay.  Versus Rx Norm, which is the same sort of thing, which is what we are talking about now.

Let me just move up a step and talk about what we are trying to do, and then let’s talk more about the recommendations.  We are so far down into the weeds right now that I can see dirt.  I’m not even sure I can see weeds anymore.

The point of this recommendation is that we felt it was important to start seeing structure and encoding for the medication being prescribed.  I think we have all observed, even there are work-arounds for NDC, which is effectively not using NDC, but making believe you are using NDC.  But basically, there are work-arounds for getting that coding and specificity that we have all been seeking.  We think that Rx Norm may be a better solution, but the bottom line is that I think we are trying to move away from free text as the 99.9 percent solution on how you send the name of a medication.  That is what we are talking about here.


DR. COHN:  So what it sounds to me like is that we need a pilot that evaluates probably the use of Rx Norm and other approaches to getting that structure.  We need to make sure that the NCPDP Script can incorporate these for the test.  I think that is what we need to do, and then we need to evaluate what happens.      Am I missing something here?

MR. BLAIR:  Just to clarify, when you said a pilot, are you talking about the demonstration project in January 1, 2006?

DR. COHN:  The test and pilot, yes.  That is what I am referring to.

DR. HUFF:  The only thing I would add to what you said, Simon, is that I think, at least my own personal motivation is that — the first motivation is what you said, that we want to move from having a textual item to a coded item.  The second thing that I think is an important principle is, we would like that code to be a non-proprietary open standard, rather than being a proprietary code.  That is the reason for proposing Rx Norm as opposed to choosing between MetaSpan and First Data Bank or one of the other knowledge base vendors.

MS. GILBERTSON:  I think when you say the pilot, there is work that has to be done beforehand to show these examples, show these walk-throughs, just on white boards and paper, so that we know what we are actually going to build for the demonstration pilots, right?

DR. COHN:  Right, but that is all the pilots.

MS. GILBERTSON:  Right, but we don’t want to start the exercise of trying to figure out how this might work January 2006.  It needs to be a clear goal while marching in the same step for the pilot to prove whether it is going to work or not.

DR. COHN:  Yes, but I think that is going to be true of any pilot that we do.  But I was observing that I think there are two pieces.  One is a piece on the part of NCPDP to enable the pilot, and then the other piece is the testing of these various ways of codifying diagnoses.

You are sitting back there very patiently.  Did you have a comment?  Maybe you are even going to say it better than I did.

MS. HELM:  Jill Helm from All Scripts.  I believe that we were the ones who testified back in May regarding the need for an additional identifier beyond the NDC.  So just in terms of what our hopes would be in the end goal, it is to provide physicians greater freedom to electronically communicate their prescription intentions beyond what they have today.

We happen to be one of those organizations that uses NDC numbers today, but there are both some challenges on our end as well as on the physician side, in that they don’t have the freedom to prescribe necessarily the medications that they would like to prescribe, because they don’t tie to NDC numbers.  Yes, they can send them as text, but then the challenge is, they don’t participate in DUR, drug allergies and all of those other goals that we have as well.  But to have more options available and to be able to translate between those different proprietary identifiers.

It is important not only for the initial prescription generation, but as we think about refill renewals and change requests, there has to be a way to translate information from the pharmacy that it meaningful to the pharmacy into an information set that is meaningful for the physician.  So that was our goal.

DR. COHN:  I appreciate that.  We have been talking about the medication history, which is an issue.  If that were ever to come down exactly with — NDC code is simply not very usable to try to understand things.

So anyway, Margaret, are you confused?  Randy?

MR. LEVIN:  Just a question.  You were talking about that physicians don’t have the codes for prescribing all products.  What products are using the NDC?  What products are you referring to?

MS. HELM:  Sure.  Generics are a great example.  A physician may prescribe amoxycillin 250 milligram capsules for a patient.  Today without sending free text, our only option is to translate that to a particular NDC number that is associated with a manufacturer of a product that is called amoxycillin 250 milligram capsules.

Now, the pharmacy when they receive that is unclear whether the physician really wanted that particular generic manufacturer’s amoxycillin, or whether any amoxycillin will suffice.

MR. LEVIN:  You aren’t referring to things other than drugs for prescribing, like supplies or non-drugs?

MS. HELM:  That is a challenge, but in my mind I don’t necessarily look to Rx Norm to solve those challenges.  What I am hopeful of in the pilot is that those are some of the issues that we will find that will be identified there as well.

MR. LEVIN:  Would it be something that would be useful to have something similar to an NDC or Rx Norm, if it covers those other products?

MS. HELM:  Absolutely.  To broaden the scope away from prescription drugs to orderable items, which is a concept that we talk about internally within my company, would be a very laudable goal.

MR. LEVIN:  That is something that you would want as well, to get away from the text, for these things that are not just drugs, that are orderable items.

DR. COHN:  I think the National Library of Medicine mentioned that in their testimony and in their written materials, about the limitations.  So we certainly agree with that.

MS. AULD:  Stuart listed those as limitations, and was saying that we would be amenable to working out a way to incorporate those into Rx Norm if that would be useful, if there isn’t a more appropriate place.

MR. LEVIN:  Would that be something that would also be more appropriate for something similar to NDCs, to incorporate those things?

DR. COHN:  Let’s look back and see what our recommended actions are here.  Let’s try to get back to these things, since some of that is discussed in here.  Can you help us?

MS. AMATAYAKUL:  Yes.  This is a long sentence, but I hope I have gotten the gist, and we can break it down.  Pilot test Rx Norm, and you said, and other approaches to achieve structure.  Is that really what you wanted?

DR. COHN:  Other approaches?

MS. AMATAYAKUL:  That is what you said, and I wrote it down.  Is that feasible,or is it just Rx Norm?

DR. COHN:  We can either entitle this as pilot testing Rx Norm or pilot test approaches to structuring drugs.  Stan, do you have a comment on that?  What do you think is the scope on this?

DR. HUFF:  I would be happy just with Rx Norm.  If we go with structured, then I think you have to include Rx Norm as one of the things that would be tested.  I’m not sure, as people speculate about what would be used instead of Rx Norm, —

MR. LEVIN:  I think the key is open source, for example, if an open source is something from the FDA, such as the product I.D. which takes the packaging strip off the NDC.  From the brand perspective, that would be a possibility.

DR. HUFF:  At a minimum, it would be a combination of things, right?  It would be a combination of different names that you went through yesterday.

MR. LEVIN:  Like for ingredients, compound.

DR. HUFF:  You would need ingredients, you would need — you don’t have the clinical drugs, right?

MR. LEVIN:  Right.

DR. HUFF:  But you do have the branded.

MR. LEVIN:  The core nine, which is at the drug product level.  There is a code at the drug product level and the package product level.  So there would be ingredient, the Rx Norm, the clinical drug, and then the drug product and packaged product.  It would be four different types.

DR. COHN:  Marjorie, did you have a comment?

DR. GREENBERG:  I am assuming, but correct me if I’m wrong, that the goal is to be consistent with previous recommendations on core terminologies for drugs?

MR. LEVIN:  That is correct.

DR. GREENBERG:  So I don’t know that we want to open this up to quote-unquote other structures.  I realize that some of those recommendations may not be immediately implementable in these e-prescribing standards, but I would think we would want to stay consistent.

DR. HUFF:  That was the reason for my comment that the administrative code set could be acceptable as well, which incudes the NDC.

MR. LEVIN:  Yes, the NDC and the ingredient were part of the previous —

DR. COHN:  I am just questioning in my own mind whether somebody is renaming NDCs to make them applicable as a coded structure for the earlier medications, whether or not that gets us to where we want to be.  In my own view, that doesn’t seem like a very reasonable solution unless the FDA is going to want to support having multiple different meanings for each of its NDC codes.

MR. LEVIN:  If someone wants to order a specific drug product, you use a portion of the NDC code to do that.

DR. HUFF:  But be careful.  Don’t call that portion of that thing the NDC code.  Its right name is something else.  It is your branded drug, right?

MR. LEVIN:  It is the label code and the product code of the NDC.

DR. HUFF:  So give it a different name.  Don’t call that the NDC code.

DR. GREENBERG:  You had a different term for it yesterday.

DR. COHN:  Stan, I need your help to move through this one.

DR. HUFF:  I like the recommendation the way it was written.  The only distinction is, coming back to what Lynn said, there are probably two things here that — we would like to encourage the planning for this to be done now, and then this to be tested in the pilot.  But maybe we don’t need to separate it out along  the lines of what you were saying.  But I like the recommendation the way it reads now.

DR. COHN:  The only thing I would remove would be, it says using an adequate number of samples, which I find to be — I don’t think we really need that in a recommendation, since it is pretty obvious, if we are piloting.

So basically, the pilot for Rx Norm, for the ordered medication in NCPDP Script, and determine that it can flow through to an NDC that represents the prescriber’s intent at the dispenser.  I think we have got the idea there.

MS. AMATAYAKUL:  I have to clean it up.

DR. COHN:  I know, we are not wordsmithing at this moment.  Is this what we are talking about?  So that is one.

Then the other piece had to do with the mapping?

MS. AMATAYAKUL:  I don’t know how this is different, test the ability of the available Rx Norm codes to map to NDC.

DR. HUFF:  Again, the difference is that that top one assumes that you have Rx Norm, and that is all you have.  The second one is assuming — at least my assumption is, the experiment that we are doing is not one that assumes that the drug knowledge base vendors are not in the application.

The mapping in the second one is happening because there is a mapping from Rx Norm to the proprietary terminology, and all of the knowledge base to the mapping in the NDC codes happens within their knowledge base.  That is why you don’t need the mapping specifically from Rx Norm directly to the NDC codes in the 9A experiment.

MS. AMATAYAKUL:  So test the availability of available Rx Norm codes to directly map to NDC.

DR. HUFF:  Yes.  I don’t think 9B is needed right now.

DR. COHN:  I was going to agree with that, about whether that is really something that needs to be piloted or tested.  Should we remove that?  Harry.

MR. BLAIR:  I don’t know what 9B is.

DR. COHN:  Oh, sorry.

MS. AMATAYAKUL:  Test the ability of the available Rx Norm codes to directly map to NDC.

MR. BLAIR:  Why don’t we want that to happen?

DR. HUFF:  Because it wouldn’t be used right now.  I don’t think it would be used at least for three years.  Until you have knowledge bases that are developed against Rx Norm directly, it is not useful.

MR. BLAIR:  So it is premature?

DR. HUFF:  I think it is premature.  I think at some point that will be a good experiment to do, but I think it is premature right now.

MR. BLAIR:  We also had a category of longer term recommendations.  Instead of deleting this, wouldn’t this go into a longer term requirement?

DR. HUFF:  Yes, we could use that beyond column.

DR. COHN:  I know Harry has a question.  I have a question about whether or not — I am wondering if this is a pilot test or is this an expectation at some point going forward that this will happen.  This to me isn’t a pilot test.  This would be an expectation, once FDA does its thing and NLM does its thing, this would — am I off on that?

DR. HUFF:  No, I think you are exactly right.

DR. COHN:  We are changing it from a pilot to an anticipation of, we have the expectation that the government will do this.   Harry, did you have a comment?

MR. REYNOLDS:  Yes.  I was in the weeds for a purpose.  I want to ask my question one more time.  I’m not a terminology guy, but I am an implementer.  We have a question up here.  If we say that we want all drugs to map to Rx Norm, if you don’t, do you get a halfway solution or two-thirds solution or something?  What happens if you don’t?

We have heard that you will have some text.  I was not in the earlier — I understand the process, I used to write pharmacy systems, but my point is, if you are always going to have text, if you are always going to have ones that Rx Norm doesn’t match, and you are always going to have some other identifier, recommending that everything go to Rx Norm, I haven’t heard a soul in here say you can get there.

MR. BLAIR:  I don’t think we said everything. I think we want to get as close as we can.  My understanding of the reason for that is that down the road, it facilitates a higher level of DUR comparisons and interoperability.  So I think it is a matter of a goal rather than a journey, to move as much as we can towards that, with the understanding that we will never get 100 percent.

DR. COHN:  I see Vivian maybe moving forward with some wordsmithing here.

MS. AULD:  The long term goal is when the FDA is able to provide us with the information for SPLs and NDCs on a regular, essentially immediate basis, then you will not — we will be updating Rx Norm like every day or something to that effect.  So you won’t need, except on rare exceptions, to use the text field, because it would all be in there as a pulldown.

MR. REYNOLDS:  I understand, but between now and when we get to this magic place we are going to, are we saying that there is no standard prior to that?  Or are we saying you can use any of the three?  That is all I was getting to.  If the Secretary wants to start pilots, if you start a pilot right now, you are not going to get much mapping.  Let’s say it only does 20 percent of the drugs or 30 percent of the drugs.

DR. HUFF:  No, no, no.  There is some misunderstanding here.

MR. REYNOLDS:  That is what I am trying to get to.

DR. HUFF:  Go back.  Stuart’s testimony, and I think it is accurate, he took a thousand of the most common ordered things in the VA, and there were 26 of them that were missed.  So we are down to 99.9 percent of the things that are in Rx Norm.

MR. REYNOLDS:  I withdraw my comment.  Some of the discussion and testimony backs him up.

DR. HUFF:  That overestimates it in a sense, but you are in the 99 percent-plus category of things that are already in Rx Norm.  There will be things missing, but there are things missing in what I am using today.  There are things missing in First Data Bank that I want to order, too, and you fall back to the text.  But we are talking the one percent category, not 80 percent category.

MR. REYNOLDS:  Now that that is on the table, that is a viable position.  I am just making sure I understand what we are voting for.

DR. COHN:  Is there something about this?  I’d like us to get back to this.

MS. AMATAYAKUL:  I don’t know whether you want to add this, but just on this discussion, would it be appropriate to add, support movement to adoption of an open source structure to coding clinical drugs to enable automated DUR, by supporting FDA, NLM, something, which I haven’t finished yet.

MR. BLAIR:  Open source I think means many different things to many different people.  Is there something we could — what do you mean by that?

MS. AMATAYAKUL:  Non-proprietary structured coding.

DR. COHN:  I am confused.  I don’t even know what you are talking about.

MS. AMATAYAKUL:  I’m just trying to capture the issue that we don’t have.

DR. HUFF:  Maybe this is what you are trying to get at.  What is clear is, besides doing the experiment, as a taxpayer what I would really like is the work that is being done at the FDA to be exactly correlated with Rx Norm, so that we know who we are getting those codes from.  If we end up — if the orderables are a combination of those two things, then it is very clear what the defined relationship is between what the FDA is doing and what the National Library is doing with Rx Norm.  That could include the relationship where they are passing daily NDC code mappings to the National Library of Medicine or doing other things.

So that is the issue that I saw, but whether we want to put it here or not is another question entirely.  Absolutely we want the NLM and the FDA to have the resources we are asking them to do.  That is without question.

MR. BLAIR:  Can I get some —

DR. COHN:  Is it on this one, Jeff?

MR. BLAIR:  Exactly.  Exactly on the phrase that Stan just entered.  I’m not exactly sure how to craft this, and we may not even try to craft it now.  We may just put a placeholder in here, but if it is possible to accelerate progress towards the availability of all of Rx Norm and the FDA progress with the structured product labeling, all of these pieces?  These things are needed.

They seem to be on a longer time frame than other activities.  Maybe if there are resources or funding or something that could help the FDA and the NLM both to move faster and to coordinate, maybe that should be something that is explored by HHS.

DR. COHN:  Thoughts, comments?  Maria.

MS. FRIEDMAN:  I am totally confused.  Are we voting for Rx Norm with the condition that it is used in the pilot?  I’m confused what we are doing.

MR. BLAIR:  We’re past that.

MS. FRIEDMAN:  I may have missed it.

MR. BLAIR:  I thought we were going to other —

MS. FRIEDMAN:  I just want to make sure I know where we are here.

DR. COHN:  What Jeff was suggesting was a subsidiary recommendation calling for an acceleration of the progress towards structured product label, everything about that.  Vivian, do you have a comment?

MS. AULD:  It is my understanding that the NLM side of this is in place at this point.  What is needed is on the FDA side, the drug listing rule to be put in place, and that if possible should be accelerated.  Once that is in place, then we will get the SPL, and everything should get what Jeff is talking about.

MR. BLAIR:  It would be constructed to encourage that acceleration.

DR. FITZMAURICE:  Are we talking about trying to accelerate somebody implementing a regulation that already exists?

MR. BLAIR:  No, for FDA to promulgate its structured product labeling.

DR. FITZMAURICE:  Maybe we should ask Randy what the current status is of the regulation that would enable this to happen.

MR. LEVIN:  I can’t comment on that.  But as we talked about before, there are problems with the generation of the NDC that need to be corrected.  That would mean, to do that would have to be a change within our current rules that regulate drug listing.  So I think that is what Vivian was talking about, that those rules would have to be changed or updated or whatever.

MS. AULD:  To expedite the fixing of those rules.

DR. COHN:  I am reminded, didn’t we write a letter on that last fall, that we could pull out a paragraph or so about that, so we again call for — wasn’t this the issue that the full committee brought forward?  So we hopefully have some language, Jeff, to help us with that, especially given that we don’t know the current status.  But I think our understanding of the current status is that there is no rule, there is no notice of proposed rulemaking for changes, and there certainly is no rule out there.

MR. LEVIN:  That is correct.

DR. COHN:  Thank you.  That is what we needed to know.  Thank you for the question, Mike.

Now, let’s go back and ask — because I think Maria was confused about what we were recommending around Rx Norm fundamentally.

MR. BLAIR:  We already recommended Rx Norm back in November.  Do we recommend it again?

MS. FRIEDMAN:  That was my question.

DR. HUFF:  Yes, we recommend it for this specific use, and we are recommending that we do this specific pilot.

MS. FRIEDMAN:  That is the clarification that I was looking for.

DR. COHN:  Let’s ask another question again.

DR. WARREN:  Simon?

DR. COHN:  Yes, I’m sorry, Judy.

DR. WARREN:  Stan’s response confused me.  I thought that previously, when we talked about — I don’t even know now what we previously talked about.  The comment was made that we had already approved it, therefore we didn’t need to do it again.  So does that now hold true for this, since Rx Norm was approved last fall?

MR. BLAIR:  It was recommended.  HHS has not taken —

DR. WARREN:  But we still need to be specific and say use Rx Norm and e-prescribing, even though previously it had been recommended for use as a drug terminology?

DR. COHN:  Yes.  Remember, looking at our framework as well as the MMA regulations, are we suggesting this be adopted with adequate industry experience?  Or are we saying, we had previously recommended that Rx Norm do this.  We need to have a pilot to assure its adequacy and readiness to fulfill that role?  Is that what we are describing?

DR. HUFF:  Yes.  The point is, we approved Rx Norm for use in the PMRI, full stop.  We approved a bunch of other terminologies.  So based on what we have already approved, we just assumed that NDC codes could be used here, or something else.

We are saying, no, this is now a more specific recommendation than we made before.  We now are saying, Rx Norm should be used for the orderable drug in the Script standard, which we just talked about.  So it is consistent with but more specific than our previous recommendation.  I think that is necessary, because we approved other terminologies that at least conceivably could be used in its place.  So I think that is why it is an essential further specification of what we want to have happen.

MS. FRIEDMAN:  We need to capture what you just said.

MR. BLAIR:  Yes.  We just need to break each of these actions into discrete recommendations, so that they are not all lumped into one where it gets lost.

DR. COHN:  So does this answer your question, Maria?

MS. FRIEDMAN:  Yes, it does.

DR. COHN:  It answers your question, Judy.  Terry, did you have a question you needed to answer?


DR. COHN:  Now, what else do we need to say about Rx Norm?  Or are we finished with Rx Norm?

MS. AMATAYAKUL:  Do we want something about accelerating the promulgation of FDA’s drug listing rule?

DR. COHN:  Yes.  The comment I would make is, I think we have some wording.  We had previously gone on record.  We again call on the Secretary to accelerate, da da da da da, because it is an essential aspect to enable Rx Norm to be fully usable, or other nice words before that.

MS. FRIEDMAN:  I’d just like to capture what Stan said before about, we are endorsing the use of Rx Norm for a specific purpose.

MR. BLAIR:  Recommending?

MS. FRIEDMAN:  Recommending, because we have recommended it before in a broader context, and for this specific purpose we are recommending it.  From that flows all these other sub-pieces.  Having said that, we want to do a pilot, and the acceleration of some of the administrative processes in the Department.

DR. HUFF:  So the question is, do you want to take the time to break that out as individual line items on this worksheet?  Or leave that as work for —

DR. COHN:  For Margaret?

MS. AMATAYAKUL:  I have got three sentences, and I think I know I can break it up into four.

DR. COHN:  Okay, great.  We talked about importance of structure, adopting Rx Norm for that structured coding, longer term needs related to the viability and usability of Rx Norm, and then on FDA action, accelerating FDA action.  I think that is what we need to say at this point about all of this.

Now, Stan, I am being sensitive to your comment that one of our experts is leaving in 11 minutes.  Is there anything else here that we should go to directly now that we need any input on?  Is the structured SIG an important piece that we talk about now?

DR. HUFF:  It is an important piece, and I guess I can’t predict whether George would have things to contribute.  He probably would have things to contribute, but I think it is less controversial and less complicated than what we have been talking about with Rx Norm.

DR. COHN:  Since it is the next one, should we do that one?

MR. SCOTT:  The only comment I have is, the assumption that we will get to medication history, that an Rx Norm concept may be used to express medication history in lieu of an NDC as well.  I don’t know if that is all part of the Script standard.  It is meshing that with the standards themselves.

MS. GILBERTSON:  Well, if the medication history is coming from the PBM, it is going to be NDC, UPC, HRI based.

DR. HUFF:  My suggestion was going to be that we take more testimony and study medication lists more before we make a recommendation, and do that recommendation in March.

MR. BLAIR:  Could I echo that?  The medication history can be provided by PBMs.  However, medication history, if there is an electronic health record system, could be coming from the prescriber.

Then in addition, a third source is I think the pharmacies wind up collecting medication history as well.  So I concur with Stan that we have got to figure out what we are going to do with this.

DR. COHN:  I guess that is a combination on medication history.  But we certainly have heard your comment.  Karen.

MS. TRUDEL:  Just one last question regarding Rx Norm, about the drug portion.  Are we recommending that that be shown in a pilot, or is it being assumed it is going to be in peoples’ systems starting as soon as we put these recommendations out for e-prescribing?  Is it acknowledged that Rx Norm is not the first phase of low-hanging fruit that we can get on e-prescribing?

DR. COHN:  I think our view was that there is not adequate industry experience at this point with Rx Norm.  Therefore, our hope would be that it would be very clearly the candidate, and come out in 2009.  Hence, the pilots.  Adopting it without adequate industry experience would not be —

MS. ECKERT:  It is not just industry experience, even if the Rx Norm is already at the NLM and the drug listing information is available, and the drug compendia have mapped to it, and provided that in our database?  It is still a matter of all the system vendors programming and storing that information also, and incorporating it into their systems.  That is a big step that I just want people to realize is not cheap and is going to take some time.

DR. COHN:  So are you stating that there actually is adequate industry experience, but there are these other issues?  I didn’t think that was what you meant.

MS. ECKERT:  I’m just saying, once we do do the pilot test and it has proved the concept and it works, there is still a large phase to get it adopted.

DR. COHN:  Sure, I agree.  But I think we would agree that there is not adequate industry experience, so we probably don’t have to go to all the other places you were just going.  Mike.

DR. FITZMAURICE:  I think you probably took care of it.  I thought Karen was asking, are you giving us direction that it is okay for us to go ahead and do this, if it works.  That is probably not what you were asking.  You were asking, what does NCVHS think it is going to do with its recommendations.

MS. ECKERT:  Once there is proof of concept and we agree through the pilot test that it works, I just want you to realize that the rollout is time consuming and expensive, and take that into consideration.

DR. COHN:  Yes, we are very understanding.  Having been through HIPAA, we are very understanding of some of these things.  That is a good point.

My God, it is 2:54.  Do we want to take a ten-minute break?  Ten minutes, and we’ll get back.

(Brief recess.)

DR. COHN:  Let’s talk about expectations for the remainder of the afternoon.  We are in the midst of the terminology section, so I am suggesting that we tough it and go through it.

It is 3:15.  My hope is that we can make it through the — actually, this is a wild hope.  My hope is that we can make it through all of the identifiers, message format standards and terminologies by the end of the day, with the idea that we will have a chance to review it again tomorrow, and then also talk about related issues.  We may not make that, but I would like everyone to think that is what we are trying to drive to.  So if there are thins that we are talking about, or things you want to say that have nothing to do with anything here, think about that before you make a comment, just to keep us on track.

Now, certainly there is no statement that this is everything that needs to be said or even right here, but let’s at least try to make it through this, with the idea that we need to be putting placeholders in for other issues that may not be being addressed.  Everybody game so far?

Margaret, I think our next piece is six.

MS. AMATAYAKUL:  Yes, sir.  I didn’t take time to make an observation, but I think we know there is no structured SIG at the moment.  So the recommended action that I have crafted so far is, support NCPDP, HL-7 and others in addressing the SIG components, preserving the ability to incorporate free text when necessary.

Maybe we should say, addressing the SIG components in a standard, to create an ANSE standard.

DR. COHN:  I thought what we were talking about is a structured SIG.

PARTICIPANT:  Product label?

DR. COHN:  No, I’m talking about structured SIG, is what I think we are talking about, a structured and codified SIG.

MR. BLAIR:  Yes, even better.

DR. COHN:  That is what we are trying to get here.  I don’t know that it is itself a standard.  I hoped that it might be parts of standards.  This isn’t a separate standard.

DR. HUFF:  No, it would be part of one or both of those.

DR. COHN:  I think we all agree, I would posit that none of us think there is anything that is going to be mandated that has had adequate industry experience yet.  So I don’t think that is on the table, but I do think we are talking about something to test in the pilot.

My question for everyone is, looking at the recommended action, — and I am trying to think of the work that is going to go on to get a structured and codified SIG.  I was just looking at your recommendations, and admittedly there are a lot of them.  I’m trying to remember whether there were any recommended next steps on the SIG in the testimony you gave yesterday.

MS. GILBERTSON:  That NCPDP, HL-7 and other interested industry organizations, players, — in fact, the e-mails went out yesterday to get people together to start talking about it.  We have got a little bit of a model to start with, with the work that HL-7 has done, so we are not starting from ground zero.  So it is that activity.

Now, the standard can support whatever we end up deciding.  There is still the factor of what the doctor enters and standardization of that to translate into what can be codified and standardized to the SIG.  We will try to get industry participation in that, but I don’t have enough experience with that to know how far it will or won’t go.

DR. COHN:  So you sent out a note and are going to put together a meeting that involves all the various players and all of that on this.  Yes, Margaret.  I see somebody else coming up to clarify also.

MS. AMATAYAKUL:  Would we want a separate recommendation to support industry participation?  And is Lynn referring to more the prescriber participation, as opposed to the SGOs, the NCPDP, HL-7, et cetera?

DR. HUFF:  I would have thought we didn’t need a separate recommendation for that.  I thought that was explanation about the process that we were going to use to make sure all that worked.

DR. COHN:  Scott, would you like to introduce yourself?

MR. ROBERTSON:  Scott Robertson, Kaiser Permanente, representing HL-7.  I also want to remind the committee that there is the issue that we the SCOs will be working together and other industry groups will be working together to attempt to come up with a structured format for the SIG.  But there may come a point where when we start talking about the terminology, it may go outside of our realm of messaging standards and need to be picked up by some other organization that can maintain the terminology.  We really are not in the business of developing large scale terminologies and maintaining them, like NLM would be.

DR. HUFF:  These terminologies are in the scope for instance of HL-7’s vocabulary TC, speaking factually.  None of the things needed here are greater than like 100 elements.

MR. ROBERTSON:  There has been some discussion about exactly what the entire range of structured SIGs actually means.  The codified forms that we have been talking about so far should be relatively small, put together in such ways that you can come up with a very, very large set of coordinated strings.

DR. HUFF:  You are right, if you get to where you are trying to incorporate into the SIG the reason for the PRN.  So if you are trying to include for itch, for rash, for eye irritation, then you are into large scale terminology stuff.  But to represent PO, sub-Q, QID, that stuff, is well within the scope of what HL-7 can maintain.

MR. ROBERTSON:  I want to apologize.  I just want to bring this up to make sure it was noted.  Stan and I can go back and forth on these kind of discussions for a long time.  I am already cutting myself short.

DR. COHN:  Stan, we appreciate your factual comment.  Judy.

DR. WARREN:  I just had one question, and if I am out of line, let me know.  One of the concerns that I have on e-prescribing is when we write prescriptions from the hospital during discharge that go out to the pharmacy.  We now have JCHO coming out and making a regulation about what goes in a lot of the abbreviations that most of us have grown up with, that now we are no longer able to use for the hospitals to maintain accreditation.  So are they going to be involved?

DR. HUFF:  That is a known input to this process.

DR. WARREN:  So JCHO is going to be part of this process?

DR. HUFF:  We don’t really need them.  They have made the pronouncement.  We can read it off the web.  And basically there is the agreement that we are going to adhere to what they have said.

DR. COHN:  Any other comments about this one?  This is not a statement of how we are going to do it, but I think it is an important area.  We think it is important for patient safety and other uses.  I think it is something that needs to be piloted in 2006.

MR. BLAIR:  You just said what I was going to ask.  Does the recommendation — I’m sorry.

DR. COHN:  Yes, the recommendation does box the test and pilot in 2006.

MR. BLAIR:  For some SIG, okay.

MR. REYNOLDS:  Simon, we don’t have the doctors listed anywhere up there.  We talk about setting standards without the medical societies.  Does somebody need to be there, or are we just recommending that they do it, and that is fine?

DR. HUFF:  They are one of the classes of participants in HL-7.  Whether they will come or not — they typically don’t come, but they are invited.

MR. REYNOLDS:  I’m just wondering whether or not it would be helpful to note that up there, because their testimony that we heard was, they didn’t know how to get involved in HIPAA.

So I totally agree with you, but we heard clear testimony they weren’t involved, and this would be a call to action.

DR. HUFF:  I am agreeing with you, but I am wondering what you are recommending we do.

DR. COHN:  Let’s see what Margaret just stuck up there.

MS. AMATAYAKUL:  Support NCPDP, HL-7 and others, especially including prescriber community, in addressing —

MR. REYNOLDS:  That’s fine.  I think it just makes a point of it.

DR. COHN:  And the right point.  Next?

MS. AMATAYAKUL:  Stan and I huddled, and we added allergies, indications and units of measure to the terminologies.

DR. HUFF:  I don’t know whether we want to attack these or not.  Actually, instead of saying just allergies, say allergens, to make it clear that we are not talking about the manifestations of the allergy, but the things that you can be allergic to.

DR. COHN:  Scott, and then I have a question here also.

MR. ROBERTSON:  I’m wondering if allergens is anticipated to include other prior adverse reactions, not specifically allergic reactions.

DR. HUFF:  Yes, would be my answer.  That is probably a better — you could expand it to say allergens and other adverse reactions or something.  The English is still not quite right.

DR. COHN:  I think that is going to take a little bit of work in terms of figuring out what it means.  Now it sounds like it means surgical misadventures, too.  But I presume, Stan, you can help with that one.

Jeff, it looks like you had a —

MR. BLAIR:  Help me a little bit.  Where is this information coming from and where is it going to?

DR. HUFF:  This one, it is not clear to me, either.  There are transactions in HL-7 that cover this, but we haven’t talked about them, I don’t think, in testimony, have we?

MR. ROBERTSON:  No, not in testimony, but in the context of a new order coming from a prescriber to a pharmacy or a fill status event or a change request or various other things coming from a pharmacy back to the prescriber, there may be pertinent information relevant to allergies, other prior adverse reactions that would need to be sent in a codified form, which is what I suspect this terminology was intended for.

DR. COHN:  Stan, is this part of the medication history or medical history piece?

MR. ROBERTSON:  It could be part of the medical history.

DR. HUFF:  I think we should include this discussion.  But one question, hopefully quick, how is this handled?  Is it handled in Script?  Can you send allergies within the set of transactions that are in the Script standard?

MS. GILBERTSON:  I don’t believe so.  There is an observation segment that is placeholding this kind of information, but I don’t believe there is.  I can’t think of a code set that goes back to what the different allergies are, although there is the DUR information.  But if you wanted to say, I am allergic to red dye number three or something like that, there isn’t a specific field I can think of.

DR. HUFF:  So I think this would be a good one to group in with medication history.

DR. COHN:  So we will defer it to March then?

DR. HUFF:  Yes.

DR. COHN:  Comfortable?

DR. WARREN:  When we talk about grouping with medication history, are we talking about HL-7 messages and things like that?

DR. COHN:  I don’t think we have decided.  That is why we are holding the conversation until — we are not making recommendations on this until March.  I think this is an area that we will need to investigate further to come up with some recommendations for the Secretary.

Indications.  Field for diagnoses exist in NCPDP.

DR. HUFF:  In some parts of the testimony, people talked about filling in this field.  We talked about using one of the ICD-9 or ICD-10 or SNOMED or something else to codify the diagnosis that was associated with this prescription.

Again, this seems like something we should address, because if you don’t standardize this part, then a lot of what you are trying to do in terms of drug utilization use and knowing whether this drug is appropriate for the indication, if you don’t standardize this, then you can’t do that part.  But whether we want to do it today — I’m pretty strongly biased toward SNOMED.

What is not known is whether that is more granular than it needs to be.  I think it would meet the use case, but it might be that there are questions about —

DR. COHN:  This is like picking little pieces out of standards and throwing terminologies in.  I feel like we are sprinkling dust.  It feels a little funny, the way we are approaching this.  Yet, I don’t know whether to talk about it now — some of these are low-hanging fruit; we know what we want to recommend, but the question is, is throwing out these terms at all helpful.

I guess I heard NCPDP handles a variety of different terminologies, or does it even allow you to put diagnoses in when you are sending out — what sort of terminologies does it allow for this?

MS. GILBERTSON:  I don’t remember all the code values, but I know ICD-9 is in there, ICD-10 is in there.  CTT might be in there.  I can’t think what other values are in there.  It allows for the flexibility, not knowing which made the most sense to use.

MR. BLAIR:  Maria and Karen, are you both here?  Is there any possibility that there could be a little bit more definition — it doesn’t have to be in the next few weeks, but maybe in October or December or something, from HHS, and maybe you have to go back to Congress with the conference report.  When they said medical history and when they say medication history, maybe they don’t have it better defined, but one of the things that rolls around in my mind is, if medical history for example is only the diagnosis or is only the diagnosis and the allergens, or it is narrowly defined, there is a whole different set of issues that we would be dealing with.than if it is broader.

So is there some way for us to get some guidance on that, or is that up to us?

DR. COHN:  Maybe I’ll start answering while Karen is looking through all this stuff.  Having recently reviewed the conference report, I can tell you that there is nothing in the conference report —

MR. BLAIR:  Yes, I did that, too.

DR. COHN:  — that relates to this.  One would also observe that in the legislation, this is not necessarily on the same time frame.  So whatever we recommend here for medical history, also needs to include a recommendation on the time frame for either piloting or implementation.

So having said all of that, I would defer to Karen if she has some greater wisdom on this.  My own view is that the work here is not just identifying a set of standards, but it is figuring out what is low-hanging fruit and what is important, and the time frame for either pilot testing or straight implementation.

MR. BLAIR:  Then given your answer, Simon, do you think that we should consider the diagnosis coding as part of medical history, or is this a separate issue that we should go forward with and try to make a recommendation?

DR. COHN:  I think my own preference, at least for the moment, is that we try to defer until March, and we can circle back around to see if — in some ways, both of these issues, allergens and diagnoses, I don’t think are a big deal.  It just seems odd taken out of context to throw them in.

But as I say that, we are talking about allergens.  SNOMED to me is not quite an answer, knowing that SNOMED has hundreds of thousands of terms.  We would want to have a SNOMED micro glossary identifying what would be allowable values or something like that.  That would be the recommended action there.  I think we need to think it ourselves about if there is anything else that is a little more specific for these other things, or whether this is just listening out diagnoses.

DR. HUFF:  We actually have the issue of whether that could be an identifiable subset of the UMLS meta thesaurus, too, which might be my first — so I would put UMLS meta thesaurus there along with SNOMED, just as a memory jogger.

DR. COHN:  I think we have to look at what we previously recommended in this area to make sure it is consistent.

Karen, did you have a language that you wanted to share with us?

MS. TRUDEL:  I have some language here.  First of all, the statutes makes a distinction between medical history and everything else, and does it at separate timetables for those standards.  It doesn’t say what the timetables are.  It just says, effective on and after such date as the Secretary specifies, and after the establishment of appropriate standards to carry out the sub-paragraph.

So it very clearly separates the medical history notion from everything else that the PDP is going to be providing.  Then it goes on to say in terms of medical history information, the information that relates to medical history concerning the individual and related to a covered Part D drug being prescribed or dispensed upon request of the professional or pharmacist involved.

So there does appear to be an expectation that the flow of information is from the PDP to the pharmacy and the prescriber.  A lot of those other connections there, while they are certainly relevant, don’t appear to have been considered under Part D.

The other thing that it does talk about, and it makes a distinction between medical history and these other things, is information on the drug being prescribed or dispensed and other drugs listed on the medication history, including information on drug-drug interactions, warnings or cautions and when indicated dosage adjustments.

So that is essentially considered by the drafters as being different from medical history.  Those requirements are embedded in the other functionality, like eligibility and benefits, including formulary prior authorization requirements, et cetera.

DR. COHN:  Thank you.  Marie, did you have a question or comment?

MS. FRIEDMAN:  Just a real quick question.  In my mind, I guess I view medication history as a subset of medical history, in a way.

DR. COHN:  I actually think it is something else, but I think we are talking about holding both of them until March.  But the medication history is clearly something that is a little lower hanging than medical history.  But we can see where we are on that.

Stan, I just want to move us through, since we have added these additional ones.  Is there a need for — allergen we are talking holding until March, though we can reconsider tomorrow morning.  Diagnoses we are talking about potentially the same thing.  Is units of measures important as a solitary piece here, or should we just keep that as a notation?

DR. HUFF:  I don’t think we want to act on it today. I think it could show up as a part of the structured SIG, so it gets tangled with that.  So that is another reason not to mess with it right now.

DR. COHN:  Okay, good.

DR. HUFF:  The other thing is, it is pretty clear that the things you have listed there, the ANSE guides, are not sufficient for what we are doing.  The FDA has sets of those, HL-7 has sets of those.  So again, I think we defer that until March.

DR. COHN:  Margaret, do you have a comment before we move on to 11?

MS. AMATAYAKUL:  I just wanted to ask, X-12 also reported it had — no, NCPDP noted that it used X-12 units of measure. Is that different than —

DR. COHN:  I don’t think we need to have that discussion right now, because I think we are deferring all of these, and we will deal with them at some point later, if at all.

Item 11 is NCPDP Script and DUE.  Margaret, do you want to help us with this one?

MS. AMATAYAKUL:  NCPDP Script reported that they had fields for drug use evaluation, including drug coverage status, reason for service code, professional code, result of service code, co-agent I.D. and qualifier, which are all industry supported fields and values, and wanted to recommend incorporation of these code sets in the required transaction.

DR. COHN:  I have a really odd question here.  Aren’t they already part of the transactions?

MS. AMATAYAKUL:  Yes, but they were added as additional items, and I didn’t want to lose them on the spreadsheet.

DR. COHN:  So the recommendation here really is not a — this recommendation is subsumed under our message standard, or recommendations for going forward with NCPDP Script, isn’t that correct?

MS. GILBERTSON:  They are part of Script.  They are also part of telecomm.  They are DUR/DUE fields that we wanted to make you aware of.

DR. COHN:  I was getting to whatever it is we recommend, it would seem to me that if we were recommending NCPDP Script in the same way as internal code sets on HL-7 or X-12, it would just come along with that.  I think that these are the same things, unless  I am mistaken.

Other comments on this one?  What I am suggesting is that we can make a note that this is covered, but we don’t need to make a separate recommendation around this.  Am I losing people, or is there general agreement with what I am saying?

Should we move on to identifiers?  Then we can move back to message format standards with the remaining issues there.

MS. AMATAYAKUL:  Which did you want to do?

DR. COHN:  Let’s go into identifiers, because it is the next one.  Then we will loop back and finish up the message format standards.  I am feeling lucky today.

MS. AMATAYAKUL:  The first one is, we did not go through this entire list earlier.  Should we just go through and check that we have got all the identifiers and then discuss them one by one?

DR. COHN:  Sounds good.

MS. AMATAYAKUL:  We have the NCPDP provider identifier number.  In other words, we need to identify the dispenser.  13 is, we need to identify the prescriber, there is no single identifier.  14 is the authorized agent such as physician assistant and nurse and supervisor, which is co-signer.  NCPDP noted that that there is balloting with respect to supervisor, but I wasn’t sure what that field was actually called, and that there was not a business need yet for the authorized agent identifier.

Then in 15, I referenced patients from the perspective of both Medicare and the data elements set, and then the health plan I.D., then the formulary identifier.

DR. COHN:  So are we okay with that world, knowing that we may go through and knock off some of these?  Let’s talk about the provider.

MS. AMATAYAKUL:  You have the dispenser or the prescriber.

DR. COHN:  I would ask the pleasure of the subcommittee.  We can talk about both of them together or whether we should separate them out.  Let’s talk about the provider identifier.

MS. AMATAYAKUL:  Provider first?

DR. COHN:  Is that what you have first?

MS. AMATAYAKUL:  No, I have the dispenser first.

MR. BLAIR:  It might be easier if we hit the dispenser first, because I think that is the least controversial.

DR. COHN:  That is the one we are talking about.

MS. AMATAYAKUL:  Unfortunately, the term NCPDP provider identifier number is used to describe the dispenser rather than the prescriber.

DR. COHN:  So we are talking about the dispenser.  Let me start out by asking a question for Karen, just to clarify what I was hearing yesterday, since she was a testifier for this.

My understanding is that all pharmacies will have NPIs and that they will be required to put those on any sort of a telecomm or other transaction.  Did I hear correctly yesterday?

MS. TRUDEL:  Yes, we can enumerate pharmacies and sub-parts of chains.

DR. COHN:  And not only can you, but you will.


DR. COHN:  Thank you.  What do people think about this one?  I can make my comment, but maybe others have thoughts about how they want to frame this.

MR. BLAIR:  Karen, just so I understand, the NPI will identify a pharmacy organization.  When you say sub-parts of chains, does that mean physical locations of pharmacies?  Then the third thing is, does it also provide the identifier for the pharmacist as well?  Does it give all three?

MS. TRUDEL:  Yes, yes and yes.  So for instance, it could enumerate Walgreen’s as a chain, Walgreen’s at a particular address, and the pharmacist at that Walgreen’s.

DR. HUFF:  With three separate numbers?


DR. COHN:  But also it doesn’t necessarily put them together.


DR. COHN:  Correct, which is one of the value-addeds that NCPDP does with their various databases.  Do people have a proposal here?

MR. BLAIR:  Can I go back and ask Lynn on this before I make a proposal?

DR. COHN:  Sure.

MR. BLAIR:  Lynn, where are you?  Talk to me, Lynn.

MS. GILBERTSON:  I’m right here.

MR. BLAIR:  Lynn, what Karen has told us to me was very compelling from the standpoint that the NPI — national provider identifier — when it is available will be effectively law.  Every provider will need to use it.  Is there any reason why, after the NPI is available for dispensers, pharmacy chains, pharmacy locations and pharmacists, is there any reason why there would still be a need for the NCPDP provider identifier after that?

DR. COHN:  And let me just further question, as opposed to the databases, which of course will be tremendously valuable going forward.

MR. BLAIR:  And if there is, could it just be linked or what?

DR. HUFF:  If you would accept a friendly amendment, isn’t the real question whether there is a need to send a different identifier than the NIP in a Script transaction?

DR. COHN:  That sounds like a friendly amendment.

MS. GILBERTSON:  In the billing of claims and the conversations for electronic prescribing, we would need to identify the pharmacy.  I have not thought of a situation where you would need to name the organization, the chain or the buying group or some higher level.  It is not a requirement to know the pharmacist in either of those situations.

As to the longevity, with the NPI, I’m not  close to exactly what is going on for data element to data element mapping yet, but there would be a need, I could guesstimate, that if the NPI did not contain the cross references to all the different fields, the state licensures, things like that, that would be used at some point historically, right now currently, I would think we would have to maintain the NCPDP provider I.D. for that purpose.

There has also been, if I understand correctly, discussions of whether NCPDP could help facilitate the enrollment of pharmacies with CMS, that they could come to NCPDP and we would take care of getting them enrolled in the NPS system as a way of facilitating that for the industry.  So obviously, we would keep a copy back and forth.

Past that, I guess I would have to plead I’m out of my league.  I just don’t know.  I’m not that close to it.

DR. HUFF:  Just to be clear, what I am conjecturing in my mind is that there would be a need for that HC Idea database and for probably other kinds of —

MR. BLAIR:  But we are not talking about HC Idea now.

DR. COHN:  But it is the same thing

DR. HUFF:  Because part of what the database provides is, they know what pharmacies and what pharmacists too, I would presume, and some other things like that.

So the assumption basically is that the pharmacist would have access to that database through their application.  Presumably the way this would work is that you would just put the NPI ID in the transaction and then those other things that were needed to be known, like the DEA number and — you’re right, I skipped to a different subject, but anything else I needed to know about the pharmacist I would know from the database, which would mean that it wasn’t necessary to send a different ID in there, because basically I could get to that information by a cross map from the NPI to that database, to an individual in that database, either a person or a pharmacy or a prescriber.

MS. GILBERTSON:  Yes, you could get all that different cross mapping.  One of the things, just to be clear, we do not enumerate pharmacists.  There is much too much moving around to be able to tie them to a given location, things like that.  It is at the pharmacy level and above that.

MR. BLAIR:  Would it be okay, Simon, if we give Lynn until next week to really answer the question?  We kind of put her on the spot here.  Maybe next week she could indicate whether there is some special attribute of the NCPDP provider ID, which is for dispensers, which would require its continued use or mapping or something else like that.  We are still going to be deliberating this stuff next week, anyway, and this way we won’t make a mistake by just assuming that the NPI for pharmacies and pharmacists will be an ultimate replacement for it.

DR. COHN:  I think Harry had a question and then Karen.

MR. REYNOLDS:  I’ll let Karen.  Mine is probably related to that.

MS. TRUDEL:  Again, the same friendly amendment again.  I think the question is not, is it necessary to keep the identifier internally for mapping purposes, but is it necessary to actually show that identifier on the transactions.  As somebody who has designed databases, if I were running for instance the provider ID database, I might continue to assign numbers to people internally, but they would never know what they were.

I think the question is, what number needs to be on the script?  Is there something about the NCPDP provider ID that that identifier itself needs to be on the script, or can it be handled by cross mapping in the background.  So I think that is the question.

MS. GILBERTSON:  Also, we have to keep in mind that there are a whole mess of other transactions that are done in the industry that it has no regulatory authority over.  Those entities may continue to use the NCPDP provider ID to identify the pharmacy until the cows come home.

DR. COHN:  Sure, for those other transactions.


DR. COHN:  Harry, and then I have a comment.

MR. REYNOLDS:  I feel anyplace we can take the opportunity to take a standard that obviously identifies anybody in the health community, which NPI does, and use it consistently, is a good thing going forward.  I would struggle to leave even the dispenser out of that, because at some point you are going to have to come back and sweep them all up again.  That is how we get partial things done in the health industry, and then we have to go back and pick it up.

So if it is there, it is real, if it has got some limitations, it would be better to look at changing that at some point than it would be excluding things from it, and then everybody doing what they are doing.

DR. COHN:  I’m not sure that the work that NCPDP has done on provider identifier, whether the issues are any different.  What I was hearing yesterday is that NCPDP has done wonderful work in terms of developing and maintaining terrifically useful databases that connect things and allow pharmacies and others to function in a complex health care environment.

I don’t think any of us think that the NPI or the NPS solves all those problems, and I don’t think the solution is to augment any of that stuff.  I think the solution is to have those things continue providing the vital role that they are providing, which is this cross mapping issue and locating and all of this, with the number that is sent in commerce being a single number like NPI.

So I guess as I look at this one, it seems there is a purpose in all of this, and all of them fit together, and I don’t see any of it going away, except that what we are talking about is what number is going to be on an NCPDP transaction under HIPAA or under e-prescribing.

But if there is something we are missing here, we would appreciate some clarification.

MS. GILBERTSON:  Part of the intelligence in the NCPDP provider ID file is the relationships to other things, —

DR. COHN:  Exactly.

MS. GILBERTSON:  — the buying groups, the affiliation codes, things like that, that the NPS is not going to cover.

DR. COHN:  We wouldn’t even want them to do that.

MS. GILBERTSON:  Right, probably not, not in the national scope.  Those are the kind of things that would be added in there.

So the question you want to have answered by next week, what are the values to keeping the NCPDP provider ID database in place after the NPI or the mapping from the NIP to the NCPDP provider ID in the background.  Did I get that close?

DR. HUFF:  No, I think we clearly see the value of the database.  The only question is whether there is some value in sending your provider ID in the message, or if all needs are met by sending the NPI, assuming that you have a cross map from that to your database.  That is the real question, I think.  Actually, it is not a question just for you; I think anybody in the room or anybody listening on the Internet, if they have a use case, we would like to know about it.

MS. GILBERTSON:  From that way of spinning the question, I think that is the value.  Everyone has recognized that the NPI is where we are going to.  As long as you can cross map to the existing other identifiers that give you the functionality that you need that is not being met in the NPS, that is the whole value to the database environment, versus just one database to solve all our problems.

DR. COHN:  I think that is what —

MS. GILBERTSON:  So did I answer the question?

DR. COHN:  It sounds like you answered the question without even —

MR. BLAIR:  Are you saying then that the NCPDP provider ID which identifies the dispensers will still need to be in the NCPDP Script message from prescribers to dispensers?

MS. GILBERTSON:  I don’t believe so.  I believe the beauty is in the mapping.

MR. BLAIR:  Fine.

MS. GILBERTSON:  It says I can get to other things if I need them, but most of the function of the NCPDP provider ID is for the billing environments, not in the prescribing environments.

MR. BLAIR:  That’s great.  Then the other piece is the other side of the equation, which is with respect to HC Idea for prescriber IDs and the NPI coming out.  There may be a different answer there.  It sounds to me, if I understood things correctly from yesterday, the NPI is definitely going to be used to identify prescribers, and it had more capabilities than we thought, including some degree of identification of a proxy or location if not directly a location.  Will we as we set our recommendations need to consider HC Idea needs to be included in an NCPDP Script message, along with NPI?

MS. GILBERTSON:  I feel like I have the weight of the world on my shoulders.

MR. BLAIR:  If you want some more time, come back to us on this.

MS. GILBERTSON:  As from the testimony yesterday, it is recognized that the NPI is the ID going forward.  The beauty of the HC Idea is that it is going to cross reference all these fields again, and recognizing that we have a lot of historical data that is going to have old IDs on and stuff, that has got to be maintained, that going to a new database is not going to solve.  It will also have the multiple locations and things like that.

When you were talking about the location of the medical record, I thought that was an interesting discussion, to use the phone number, because there is always the catch that where the locatio of the medical record is has nothing to do with the physician’s physical or billing address or mailing address or whatever.

MR. BLAIR:  Although for this purpose as a proxy, it might serve reasonably well.  Anyway, I just didn’t want us to come up with an NCVHS recommendation on the prescriber side that — one part of the recommendation is going to be that we recommend the NPI.  Do we also need to recognize whether HC Idea needs to continue to be included, or is that something where we don’t have to make that statement?

MS. GILBERTSON:  I’m not sure.  I know there was some testimony that said HC Idea should be used as the ID for prescribers.  I don’t think it was turned around to say, if we use the NPI for prescribers and you could cross reference back to NC ID and all the other fields, would that be an appropriate solution.

So I am not quite sure.  I don’t know how quite to answer that.  My gut reaction is, yes, one ID would be nice, but there is concern that if the granularity of the NPI is not at the same level as what we need for e-prescribing, it is going to be the wrong solution to the problem.

DR. COHN:  Shall we ask Karen then?  Can you tell us a little more about the granularity in the NPI?  My understanding is that it would enumerate any prescriber.  Is that correct?

MS. TRUDEL:  As far as I can imagine, we could certainly enumerate any prescriber.  It permits us to enumerate physicians, MDs, optometrists, nurse practitioners, nurse anesthetists, registered nurses, whatever.  So in terms of the prescriber, the person, that we certainly cover.

Some of the discussion yesterday had to do with locations of for instance a clinic or a physician practice with multiple locations.  As I indicated, the NPI does not have — we started out with a location code that would be appended to the NPI.  We backed away from that as a result of the public comment.

I have to say, I am a little bit confused too about HC Idea and what its capabilities are, because I originally thought that it assigned codes to specific practice locations.  What I thought I heard Phil and Leanne say yesterday was that it collected all the addresses that it didn’t define in the code.  So I am a little confused about that.

DR. COHN:  I think that that is what we heard also, and I think that is why we were all asking questions.  I think unless we despite five or six questions mis-heard that, I think the value of having all the practice locations is terrific.  It isn’t the same as knowing which location the information necessarily came from.  Begin to give more thought of a way that you might be able to track down somebody or identify where it came from and where it went back.  Maybe I misunderstood.

MS. GILBERTSON:  I don’t know that level of detail as well.  From the folks who are implementing it, if there is anybody in the audience who wants to make a comment.

One of the things that we do have to be aware of is, you can enumerate the prescriber, but you need to know where to deliver the transaction to.  So there has to be some kind of cross reference to say NPI 123, where do I deliver it.  Is it office number one, office number two, office number three.  So I think that is missing in just using the NPI as a routing mechanism.

DR. HUFF:  Again, that is not what we are speculating we are going to do.  The assumption is that I can do routing based on mapping the NPI to information in the database that does the routing.  As long as I know who the individual is in the transaction, however that was known before — if I understood correctly yesterday, the testimony is that in the HC Idea, it doesn’t include the location.  You have in your database knowledge of the association of this person to locations.  That remains in the database, that is my assumption, and to get to that database, I go from my NPI to the HC Idea.  Then I have that information in the database, and I do what I always did.

DR. COHN:  I realize that this is not your part of the responsibility.  That is why we had Leanne and others come and testify to us yesterday.  I was going to move us into action items around all of this.

MS. BYRNE:  I just wanted to add one thing.  That is, I don’t believe that the HC Idea would be used for routing, either.  It is intended to identify who the prescriber is.  In most transactions that I am aware of, the routing is not done via that; it is done via the information that is in the envelope section of the transaction, which is quite different.

MS. GILBERTSON:  I just wanted to make sure we weren’t making that leap that said, I could figure out what location you’re at just by looking at your NPI.  That can’t be done.

DR. COHN:  Terry, did you want to testify or make a comment?

MS. BYRNE:  I feel like a geek, because I am looking at the HC Idea specs.  It does have a location code on it.  I thought it did.  That’s why I went and looked.

DR. COHN:  We all thought we asked that four or five times yesterday, and you were certainly here.

MS. BYRNE:  I heard that it did have a location code, but if it is not part of the key, I don’t know if that matters.  What matters from a developer’s perspective is, can you specify a location?  If a physician has multiple locations, and you know which code, it specifies primary, secondary, third office — because we haven’t implemented it yet, I don’t know how you would use it.  But my understanding of the file is that you could use it to specify different locations.  But I don’t know.  Because we haven’t implemented it, I can’t answer that.

DR. COHN:  And you do that now by — how do you specify locations now?

MS. BYRNE:  We actually have it in our provider directory.  We can specify multiple locations for a physician.  When pharmacy software wants to send a refill request to a specific location, they can see those different locations from their viewing the data.  That is part of that proprietary provider directory that we have and SureScripts has, and ProxyMed and others have.

DR. COHN:  So obviously there are other ways you have of using the database.

MR. BLAIR:  Maybe it boils down to, we can go forward to recommend the NPI for the dispenser ID and for the prescriber ID, but we include language in our recommendation that we recognize the need to protect the ability of NCP or the networks to do the mapping to other industry-accepted identifiers.  Would that work?

DR. HUFF:  I think it is true.  Do we need to say it?

MR. BLAIR:  That is basically my question.  If we don’t say it, will something happen that may impair that mapping at the right time, at the right point?  So that is what I was groping for.

DR. COHN:  I guess I maybe think about this in a slightly different way.  I think Margaret is looking for her notes or whatever so we can get back to the action items.  Did you find something that you want to comment on?

MS. AMATAYAKUL:  In response to Stan’s question about what does this HC Idea look like physically, the response was, there is one number that is given to an individual and dropped down in database that gives you multiple locations.  But then the secondary field for identifiers is NPI if required.  That is identifier, not location.

MR. REYNOLDS:  If you think about it, you are really talking about a base number with a suffix.  That is what that dropdown box is.  A dropdown box is just a suffix to the original number.  It says location one, location two, location three, so it is another field.

MR. BLAIR:  I have a feeling that they do need it.  They do need it for routing, because without that, they don’t know what location it is going to be sent to.  So just having the NPI for the prescriber area is not going to be sufficient to be able to direct that message to the prescriber, where that prescriber happens to be acting at that time.

That is the thing that I suspect.  That is why I keep asking this question.

DR. COHN:  And unfortunately, Lynn doesn’t know the answer.

MR. BLAIR:  Right.

DR. COHN:  The other piece is that this is something where one has to recognize that people are doing this without HC Idea currently.  So there must be some other ways to do this.  Michael.

DR. FITZMAURICE:  I would prefer to call it another field rather than a suffix or a prefix to an ID.  You need to identify the physician, then you need to identify the locations of service, or locations where that physician writes prescriptions, or prescriber writes prescriptions.  It is a second variable, just like home base address or billing address is a third variable.  There is no reason it needs to be embedded in the ID.

DR. COHN:  Maybe Lynn can answer an NCPDP Script question.  Is there a location field within NCPDP Script, or is it just the address and phone number?

MS. GILBERTSON:  The ID, address and phone number.

DR. COHN:  Is there?


DR. COHN:  So that is the level of specification already in there.


DR. COHN:  So nothing coded on that one, but all of that is there.

MS. GILBERTSON:  You are talking about describing the prescriber or the pharmacy?

DR. COHN:  Yes.

MS. GILBERTSON:  Because there are IDs for the pharmacy or for the prescriber, and then name, address, city, state, phone numbers.

MR. BLAIR:  I don’t have the same problem with the pharmacies, because the NPI seems to be able to describe it in even greater granularity than the NCPDP provider ID.      I think our concern is, do you need now or in the future something like HC Idea to get it to the right prescriber.

MS. GILBERTSON:  I can’t answer that, but it will definitely be one of the questions on my list.

DR. COHN:  Okay.  Let’s talk about some recommended actions here, beyond Lynn going back and finding out from NCPDP whether we understood correctly or not, or whether there was something that we were missing here.

I am hearing that regardless, we are recommending that CMS with all due haste enumerate prescribers to enable e-prescribing as well as dispensers.  I think that is an obvious piece here, and that this be a high priority for them in an initial enumeration.

I think I am also hearing that we are recommending that CMS work closely with NCPDP, both to — because it sounds like NCPDP can help them with the enumeration, but also so that we can make sure that there is no problems with the linkages that may occur as those numbers are enumerated.  Am I missing some of the things that need to be done here?  Okay.

Beyond that, I’m not sure I have any other recommendations.  It seems like a quasi-definite conclusion that the dispenser ID should be the NPI, and unless there is some additional information coming forward, that the prescriber ID ought to be the NPI.  But there may be some additional information that we haven’t received yet.

I think the other piece we would say here is, until such time as the NPI has been enumerated for dispensers and prescribers, there have to be — one either has to use the NCPDP provider ID or a locally defined number probably for the provider, for the physician, prescribing person.

MS. AMATAYAKUL:  So in the interim, the existing NCPDP number would be used.

DR. COHN:  For the provider ID.

MS. AMATAYAKUL:  For the prescriber?

DR. COHN:  No.

MS. AMATAYAKUL:  The dispenser.

DR. COHN:  As far as I understand the basic idea that is being developed and tested right now, if I were a provider I wouldn’t have a clue of even how to get one of these numbers.  But I’m sure somebody would likely have one on me at this point.

As I said, there may be some additional information that may cause us to revisit this one.

MR. REYNOLDS:  I want to put a period after — under NPI as a standard for ERX, period.  Why would we want to describe that they should bulk load and cross reference?  Where it says recommended action, why don’t we just say NPI as a standard for ERX and get rid of the rest of it?  Why do we want to explain bulk loads and cross reference?

DR. COHN:  I think we more fully define that later in the other pieces there.  Should we move on to authorized agent and keep going here?

DR. HUFF:  Yes.

MS. AMATAYAKUL:  Are you still feeling lucky?

DR. COHN:  I’m not feeling lucky, but it is like the doll that keeps coming up after you keep punching it.

DR. HUFF:  My thought on this one is that whoever this person is, they would have an NPI.  This is really a message issue.  In other words, you need a slot in the message to say who the cosigner is, but if such a slot exists, then it is the NPI that goes in there, right?

DR. COHN:  Lynn, what is the status on this one?  Does that exist, or is this being balloted right now?  The issue of the authorized agent and the supervisor.

MS. GILBERTSON:  The supervisor is either balloted to just completed.  I’m not sure which one that one is.  So that is basically done.

The authorized agent, we have positions for the name.  It is just the name field, as I recall.  There isn’t one for the ID because we didn’t have anything to put in there and a qualifier to go with it.  So we would need to add —

DR. COHN:  And you need a qualifier because of what?

MS. GILBERTSON:  We need an NPI if that is what is going to be used.

DR. COHN:  Thank you.

MS. GILBERTSON:  So this level is going to be enumerated on the NPI as well, Karen?  You can get one if you want one?

MS. TRUDEL:  If we are talking about a health care provider, yes.

MS. AMATAYAKUL:  So we would require both authorized agents and supervisors to obtain NPI for use in the e-prescribing.

DR. COHN:  Yes, though I have got to say, I can’t imagine somebody who was either supervising or authorizing a prescriber who isn’t themselves a prescriber.  There may be some occasion where that is not the case, but I can’t imagine that circumstance.  People are nodding their heads on that one.

MS. GILBERTSON:  These are two different entities, though, recognize.

DR. COHN:  Oh, yes, a different person.  But I’m saying, that person should have an NPI, is what I was thinking.  So are we okay with that?

MS. AMATAYAKUL:  This is what I have.  I just want to double check.  We are going to require authorized agents and supervisors to obtain NPI for use in the Rx.  They probably will.  NCPDP Script, add authorized agent and supervisor qualifiers to the NCPDP Script.

MS. GILBERTSON:  It is only the authorized agent.  The rest of it is already there.  I just don’t have an ID and a qualifier for an agent because I don’t have anything to enumerate them with.  The supervisor is already taken care of.

DR. COHN:  Great.

DR. HUFF:  I would change it to say we are not requiring that.  We are suggesting that we use the NPI as the identifier for the authorized agent.  It didn’t change the work, but it sounds better.

DR. COHN:  You’ve got to wordsmith sometimes.  So are we okay on 15?  We just did 14, so on to 15.  Margaret, help me with this one.

MS. AMATAYAKUL:  This is related to patient data.

DR. HUFF:  So we already have an action item to talk about patient identification, right?

DR. COHN:  Yes.  Let’s look at the observation and make sure we have the right recommended items.  We are saying that — what is it?  Tell me what the observation is here.

MS. AMATAYAKUL:  Insuring right patient is important, patient safety and efficiency measure.  Today, most patients are identified through matching effect of variable non-standard data elements, some including SSN, which the industry is moving away from.

DR. COHN:  So what we are saying is — the assumption would be, for the purposes of the MMA, the HIC number should be the number.  But as I am looking at people, the one question I would have is, knowing that the federal government — and this is a question for Karen — the federal government will be contracting with various health plans, PDPs and Medicare Advantage plans to administer this.  Normally, what those plans do is, they enumerate numbers for members.  That is typically how everyone in the world —

MS. TRUDEL:  Under Medicare, we will be contracting wit these organizations.  One of the criteria that we will provide is how you enumerate the beneficiaries.

DR. COHN:  You will do that?


DR. COHN:  As opposed to what you do now.

MS. TRUDEL:  No, that is what we do now.  All Medicare contractors use the HIC number.

DR. COHN:  If you have a contract with Medicare, you can identify the member and the member’s HIC number, but there is no requirement that I am aware of that anyone has to use the Medicare number internally.

I will look to one of my other colleagues in the business here, Harry, to see whether or not in his Medicare line of business, that everything is done with HIC numbers.  I am just speaking for my own organization, which is a large Medicare contractor.  We are aware of the HIC numbers, we track them and we follow them, but we don’t have the members use their own HIC numbers; we have them use their number, which is the number that has been given them by their health plan, which is associated with their medical record.

Harry, is this how it works in your world, too?

MR. REYNOLDS:  Yes.  When we used to process Medicare, yes.

DR. COHN:  Back in the old days when you processed Medicare.

MR. REYNOLDS:  Some plans do.  We don’t.

DR. COHN:  I am just speaking of industry practice now.  If you have a plan that is giving other insurance as a PDP provider, you are having to have two numbers?

MS. TRUDEL:  If there is an additional requirement there, let us know what it is.  If there is a need for all PDPs to identify patients the same way in these transactions, not necessarily in their own internal systems, but in these transactions, then we require that the Medicare ID be used for Medicare claims, without exception, by Medicare contractors.  We require that reporting from Medicare contractors to us in terms of claims data done using the Medicare ID.

So if there are certain specific things that you feel we need to make sure that we standardize, make a recommendation.

DR. COHN:  We unfortunately have not done — this is an area where we need to get some testimony on, to understand what the industry practices are.  I am speaking of my understanding of how my organization works, Harry is speaking of how his organization works.

I am just aware that HIC works for certain instances, but it didn’t occur to me that this was necessarily the solution for identifying numbers in an e-prescribing transaction, especially when that transaction is not going directly to CMS for initial adjudication, which this isn’t.  This would be reported to CMS.

So the question is, do we make a recommendation here, or do we need to evaluate the whole thing and come back with some recommendations on how this should be done, either though NPI or through other methodologies.

DR. FITZMAURICE:  Solely for making e-prescribing work, wouldn’t industry like to have an organization that has a unique number for each of its beneficiaries, one that could be used through all claims and all e-prescribing?  What better number would there be for making e-prescribing work?

DR. COHN:  I’m just speaking of my own experience.

DR. FITZMAURICE:  I understand.

DR. COHN:  But let me give you a slightly different view of the world.  A person is aged 50 or age 40 or age 30, affiliates with an insurance company or a health plan, is at that point given a number, and at that point doesn’t have a HIC.  In the ideal circumstances, and we know that is unfortunately too rare these days, but they age through their plan and then at a certain point they become eligible for Medicare.  They may be working with that same insurance company.  They may have a different plan at that point, but considering that there is an ongoing relationship with that insurance company, they very likely will stay with the same number at that point, as opposed to being given a whole different number at turning 65.

DR. FITZMAURICE:  That plan might want to keep their own number, but the patient will have a new number to the plan, and that is their HIC number.  If they want to send a claim to the Medicare program, they will use the HIC number.  So they will have another column on their database of numbers that is the HIC number.

DR. HUFF:  Yes, but if we come back to e-prescribing thing, I don’t think we want to put a burden on the prescribing system to know my HIC number in order to write a prescription for me.  I need to know that by the time I send a claim.  Why would I need to know that or care about that at the time of prescribing?

DR. FITZMAURICE:  I suppose if you wanted to link a patient history from that point on with the Medicare program for quality purposes — well, it is even more important to link them to all other previous medical history.

DR. HUFF:  I agree.

DR. FITZMAURICE:  That is the reason for the continuity of the number.

DR. COHN:  Karen.

MS. TRUDEL:  First of all, the prescriber has the health insurance claim number, because they wouldn’t be able to submit a claim to  Medicare for their service without it.

Second of all, if they don’t have the health insurance claim number, they can’t do an eligibility query, because the internal — whatever Kaiser Permanente number they may have isn’t going to get them necessarily what they need to know,  because now the person is a Medicare beneficiary.

DR. COHN:  Once again referring to the world as I know it, eligibility claims, at least in many cases, don’t go directly to CMS.  They actually go to insurers that are somewhere in the middle.  I would have to check with the industry to see if that in a standard fashion is done with HIC numbers.

MS. TRUDEL:  That would be interesting to know.  I would say to you that the only number that the physician knows today is the Medicare claim number, because that is the number on the Medicare card that the patient brings them.

DR. COHN:  As I said, I can only observe that this doesn’t — as you know, I don’t spend a lot of my time in fee for service Medicare, but I am aware that the Part D plan is not a fee for service Medicare plan.  It is being run by prescription drug plans and Medicare Advantage organizations.  So I think it may make it look a little different than traditional Medicare fee for service.

Having said that, it sounds to me like we need to — I am becoming convinced as we are talking here that we need to do some investigation on this.  What I was going to try to do initially was say, for Medicare the HIC number, but that isn’t quite right, and then we would do further investigation for all other uses.  But what I am being persuaded on, and what I am proposing, is that for uniquely identifying an individual, being it Medicare or others, we clearly need to do some hearings to hear both about the practices related to Medicare members, as well as the practices related to under 65s, which we were going to do, anyway.  You remember, one of our responsibilities out of the — you were all at the strategy summit on health information technology; we were asked along with privacy to look at the issue of uniquely identifying patients to support health care and all of this stuff, recognizing that it may have many different numbers, along the lines of what Rx Hub and others have done in terms of using NPIs and all of this.

So it sounds like we will need to reflect on that.  If we get some administrative information before, maybe it will allow us to give some recommendations before.

I apologize, I have been monopolizing this one, but this is one I thought I knew the answer on, and I am convincing myself that I don’t.  Do others have comments?  Harry, am I off on this one?

MR. REYNOLDS:  Again, Medicare has many different arrangements with people.  It becomes the health plan’s number. That is what I think you are relating to, versus a direct Medicare program managed solely by Medicare.  Then you’ve got the Medicare crossover.  So I think it is confusing enough, the fact of how people deal with Medicare and how Medicare has supplemental and Medicare Plus Choice and some other things, where there is little other risk.  It would be good to completely clarify it.

DR. COHN:  To understand it, yes.

MR. REYNOLDS:  But on this, since this will be a Medicare  program, and they do use the HIC number, it is hard not to agree with that.  But without further testimony, you can’t check it completely and go where everything is okay.

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

MS. FRIEDMAN:  I just have a wording change for 15.  The HIC number is built on the social security number plus this suffix.  So I would just stop it at, use Medicare HIC number, because the rest of that stuff is not quite right, the way you have it described.

DR. COHN:  We haven’t gotten very far through that one, have we?

MR. REYNOLDS:  The other thing that is different, Simon, as we go forward with this, is that right now, there is some eligibility done, but in this case you are bringing that person’s — you are not bringing the contract information back, which you do now.  You have somebody that holds the ID card, so you are bringing down the benefit for them.  If there are three of them, it is one set of benefits.

In this case, you are bringing down the medication history and you are bringing down other things for that particular person.  That is a difference over a lot of what we are doing now.  If my wife went in and they presented my Blue Cross card, her eligibility is the contract that we all have as a family.  Her medication history has got nothing to do with me, my daughter or anybody else.  It is only her.  That is where this next level of specification comes in, because now you are actually delivering her medication history, not my contract information because I am the card holder.  The eligibility is card holder level, it is not individual person.

So that is why this is really getting down to the bottom line of, I am talking about that person and exactly about that person, and no other person other than that person.  It is a little different than we have had to do in the past.  Edibility was usually a contract level.  This is drug history and other things  at that person, nobody else.  It doesn’t matter that I am the card holder.

DR. FITZMAURICE:  But Harry, when I first got insurance with Blue Cross in 1972, I wound up having nine children.  They didn’t know who my children were.  We just sent in bills, and they took the child’s name and the date of birth, and they just kept recording it.  Every so often they would say, are you sure all these kids are yours?  I’d say, yep.

MR. REYNOLDS:  But I think the difference is, you are dealing on a real time basis on a lot of this, and that person actually walks in.  So your child matters when you walk in on e-prescribing.  If I am going to deliver the medication history for your child, it doesn’t matter how many you’ve got, it better be the one, and it better be coming straight back.

DR. FITZMAURICE:  I fully agree with you, but I think it got to the individual level long before now, but it wasn’t as important to the health of the person.

DR. COHN:  Well, absent additional information or clarification, I can imagine that Karen and I will have further conversations about this one.  If we can somehow get it clarified between now and next week, we will come forward with some language, because I think this is primarily understanding how the world works and what sort of wording needs to be around Medicare, knowing that by definition we will need to do further investigation for the rest of the industry.

Number 16.

MS. AMATAYAKUL:  Sorry, I’m just making a note.

DR. COHN:  Sure.

MS. AMATAYAKUL:  Sixteen.  The HIPAA health plan ID is still being defined.  NCPDP telecomm N script includes unique identifiers to identify benefit programs.  So the recommended action might be something like, HIPAA health plan ID should be constructed to accommodate use and claims as well as eligibility and benefits and inquiries incorporating plan, pharmacy and subgroups.

DR. COHN:  Actually, I think I have a different recommendation here.  I think the recommendation is that CMS should move forthwith to enumerate all health plans under Part D benefits until such time as a HIPAA plan ID is established.

MS. GILBERTSON:  You are using existing PBMs to do business today as part of the PDPs in a lot of cases, right, Karen?

MS. TRUDEL:  I’m not sure that the PBMs would be the PDPs, or whether it would be business associates.

DR. COHN:  Consider that a health plan contracts with a PBM.

MS. GILBERTSON:  Oh, okay.  Sorry, semantics.  Where I am headed is to identify where to go get eligibility from.  It is the same entity that a pharmacy would submit a claim to.  We have identifiers, so we wouldn’t need to enumerate, we would need them to support the enumeration structures that are out there.


DR. COHN:  Please, go ahead.

MR. REYNOLDS:  What I would have added to that is the fact that we don’t have a plan ID, we need to adopt — I think that is what Lynn is saying — we need to adopt the current NCPDP standard, until such time as the plan ID is available, and it should do what is said under the recommendation.

So you have got a standard, you can use a standard.  When the plan ID comes out, it needs to make sure that it incorporates that standard, at least what is out there.  Because none of us know if that plan ID really, really is going to happen, and exactly when.  There is a standard out there that would allow everything to take off and get going right now.

DR. COHN:  It looks like we have somebody to clarify that for us.  I just want to take this opportunity to once again, as usual, thank our experts who are behind and around us here, for helping clarify as we move forward.  Terry.

MS. BYRNE:  As far as the processor ID for processing a claim, are you speaking of the BIN?  That is assigned to a PBM, not a health plan.  So you wouldn’t be able to identify the specific PDP within that.  The BIN is just the entity that processes the claim, so Express Scripts, mental health, et cetera.

MS. GILBERTSON:  Right, but they use the processor control number and the group numbers when appropriate to subdivide their benefit programs so you can get down to them.  If somebody is doing a 270-271 directly to a PBM, they need to be able to get to plan XYZ under Express Scripts in a similar fashion to how they would get to — a pharmacy would get to a claim at that same place.

MS. BYRNE:  But I don’t think it is the health plan.  I think the PBM identifies who the patient is with by the patient information that is on the claim.  Then they understand what health plan to process that under.  I don’t think that is identified on the claim itself.

MS. GILBERTSON:  Within the processor control number, the group, tell you what benefit package with this patient, whether it is step one or step two or step three.  I’m just trying to get to the point where we don’t have a whole new set of identifiers for the same benefit programs as where pharmacies are going to submit the claims, because then we will have a mess, because we will never be able to call X X again.

I’m not sure I have all the answers, but just as a consideration.  I don’t know how the PBMs that are involved in the PDP business are going to be doing all this.  But let’s not make new IDs if we don’t have to.

DR. COHN:  So I think we are saying two things then.  CMS should leverage currently existing BIN control numbers, PBM identifiers?  Is that what we are describing?

MS. GILBERTSON:  If you use the term BIN processor control number and group number, entities in the pharmacy environment know exactly what you are referring to.  Trying to munge those into a better semantic term always gets into trouble, whether it is a health plan, whether it is a benefit program, whether it is a processor, whether it is a payor.  It can be any of the above, depending on who you are talking to.

I don’t know how else to state it, other than those identifiers that are on the prescription drug card that identify where the pharmacy is supposed to direct the claim, I would think is the same place where the prescriber needs to direct a real time eligibility, unless they are getting a feed or something like that, which makes sense as well.

DR. COHN:  I wonder if we even need to go here, but go ahead, please.

MS. TRUDEL:  I don’t view that as a plan identifier, though.  What we are trying to identify in terms of MMA at least is the Medicare Advantage plan that has a drug program, the PDP or the fallback plan.  Any of those three could deal with any number of business associates, who also happen to be PBMs.  But the PBMs are not the people who are providing the benefit and absorbing the risk, necessarily.  It is going to depend on how the contract works.

MS. GILBERTSON:  I am just down to the level of, you need to submit a transaction, where is that transaction going to, and the identifiers that the pharmacy uses to bill a claim and the identifiers that a prescriber would use to check eligibility both go to the same place, to check eligibility or to bill a claim, and they should use the same identifiers so that X is X is X.

MS. TRUDEL:  But again, isn’t that in the envelope?

MS. GILBERTSON:  But when would you use anything but?  If it is not in the transaction level to route the claim to where I want it to go, what other purpose would I have to know the health plan, or whatever level?  I’m just trying to imagine why a pharmacy would care past where do I send this claim to be paid, or a prescriber, where do I ask for the eligibility information.  Past that, I don’t know that I care any further.

DR. COHN:  Scott.

MR. ROBERTSON:  Scott Robertson from Kaiser Permanente, not representing HL-7 at the moment.  Speaking from prior experience, I used to be one of the assistant administrators for the Southern California pharmacy application.  For a long, long time, we dealt with Kaiser, and we didn’t have to deal with anybody else, in what was a very simple time.  Then we had to start billing MediCal, which was California Medicaid, and its various variants.  Then there were a few other third parties that came into play.

I had to learn about BINs and processor control numbers and group IDs.  But from the prescription perspective, when that script came into me, all I needed to know was, is this a Kaiser prescription?  Is this supposed to be billed to Medicare or Cal Optima or one of these various things?  Knowing what plan it was, knowing that it is a particular type of Medicaid just led me to the point where I had the information about where I send the adjudication.

So I wouldn’t really expect the prescriber to give me the detailed information about how to submit the claim.  I just need to know who is the ultimate authority.  I have means then to communicate the particular plan identifier to the appropriate — actually, the plan identifier might be the group ID in the script message, but the BIN processor control number isn’t something that I necessarily expect the prescriber to spike to the pharmacy.

Just a perspective.

MS. GILBERTSON:  As we go down the road, I can see where there is probably some information how Rx Hub helps glean that for their vendors, that would be of help to the pharmacy, to know.  If you have gotten — heaven help us, but if you have gotten a prior authorization, it could be of help for the pharmacy to know who you got it from, and at the level of identifiers would be nice, rather than just a generic, here is the bunch.

As somebody who has gone through prescription programs that may have one name and the pharmacist has to figure out which one of the seven programs that mine have busted up, if we can provide those standard identifiers when it is analysis, I think it would be helpful.  Is there immediate need?  No, but I still have to do the eligibility and the claim transaction needs to use the same identifiers.

MR. ROBERTSON:  In brief response, I’ll keep it very brief, I wouldn’t necessarily throw away the information if the prescriber sends it to me, but as long as I know the ultimate payor, that should point me in the right direction on how to send the claim.  If they give it to me, it saves me a little time.

MS. BYRNE:  It is difficult because we have a different model than a direct model, but I would be careful not to specify that you have to direct the eligibility to a specific payor, because part of — a patient may be covered under multiple payors, and especially in the case of Medicare, you would want to know if that patient had another payor.

In nowhere in our model does the health plan have to be specified.  We find the patient and we find the coverage.  We would find out if they had Medicare under a PDP.  So I think we need to be careful about that, because eligibility to a PBM, if you know to direct it to a PBM, you don’t have to direct it to a specific health plan.

MR. BLAIR:  Are you saying we do not need to recommend a standard for health plans?

MS. BYRNE:  I don’t believe you do.  I don’t understand the value.

DR. COHN:  Harry, maybe you can help us.

MR. REYNOLDS:  I think we do need to recommend a standard for health plans, but the BIN processor control number and group ID are the routing message.  It is like a clearinghouse.  Right now, if somebody is sending something to WebMD, that is not a health plan.  A health plan is whoever owns the coverage.  That is the routing mechanism.

So if you look — I got my card out; it goes through paid and it goes through BNC Drug.  You couldn’t tell that is Blue Cross from North Carolina.  The card tells you that, but the transaction is routed through paid, and then ends up paying those with the stuff here, my health plan number.  Once Blue Cross of North Carolina comes up with a health plan number, there is going to be a single health plan.

I think to me, we are almost mixing two things here.  The health plan is in our case Blue Cross of North Carolina.  It is not this thing.  This is a routing code as to how to get that information back and forth between the people that are doing business.

Karen wants to know, if there is a report, if we were actually running Medicare D, she would want us to be able to send her the health plan ID, which is us versus other ones that will be all over the country, and she would know how much we paid for a Medicare beneficiary based on what we did and didn’t do.  That is a whole different business transaction than this is.

MS. BYRNE:  This is part of paid claims, right?


MS. BYRNE:  Then you would need to know that.

MR. REYNOLDS:  Okay, but that is what the health plan ID is supposed to be.  This other is the routing.  Basically, a PBM is a clearinghouse.  It does a little more than a clearinghouse; it actually processes it.

DR. COHN:  Comment, and then let’s try to figure out what it is we are exactly doing here.

PARTICIPANT:  I am from Health Solutions.  I just want to clarify that in the submission of a claim for processing by PBM, the health identifier is not required.  There are patient identifiers that get provided.  So I was concerned that that might have been misrepresented.  It sounds like we have got that clarity, that PBMs do not require health plan identification for the processing of claims on the submission.

MS. GILBERTSON:  I didn’t mean to assume that.  They need to get to who is going to pay the benefits.  So in your terminology, Harry, that is the routing information.  I am just not sure what is the business case for a pharmacy to submit a health plan ID, what are they getting out of it, or what does the prescriber get.

MR. REYNOLDS:  That is what I am saying.

MS. GILBERTSON:  So we are on the same page.

MR. REYNOLDS:  A health plan ID is derived from the routing information.  Then the patient information, that is attached.  It has got nothing to do with the health plan.

MS. GILBERTSON:  In the e-prescribing world, what do we need a health plan ID for?

DR. COHN:  That is where I was going.  Is this something that we should remove from this requirement?

MR. BLAIR:  It sounds like we are trying to fix something that is not broken.

MR. REYNOLDS:  When there is a report coming from Medicare as part of e-prescribing, in the end they are going to have to pay somebody.  They are going to want that health plan ID, that goes back and forth between the health plan that you contracted with and you to be the health plan ID, not some other magical thing that is made up.

MS. BYRNE:  But that is based on a paid claim.  It is not based on a prescription, right?

MR. REYNOLDS:  I know it isn’t, but it is the whole program.  Medicare D is a whole program, it is not just this piece here.

MS. BYRNE:  Right, but we haven’t addressed the telecommunications claim or the mechanism by which that information is going to be fed back to Medicare yet.  That has not been addressed at all, and that is where you need the information.

DR. COHN:  I am going to try to put this together a little bit.  Karen, I need your help on this one.  I am getting increasingly convinced that I don’t think that we need to have an action around all of this, or if so, I don’t know other than having you enumerate, which is not a standards action, but is an action that you guys need to do, anyway.

But I don’t think we need to talk about the BIN processor control number or group ID.  Thank you for writing that down, because I could never have said that otherwise.  I don’t know in the context of e-prescribing what the health plan ID is going to do for us.

If you can help me with that one?  From your view, is this an important issue that we need to be discussing in recommendations on e-prescribing?  You can think about it if you want to.  Do you want to think about it?

MS. TRUDEL:  Yes, I’ll think about it.

DR. COHN:  Okay.  Why don’t we — at this point, unless we come up with something over the next day, I think we need to defer that to March and maybe after, on what exactly is this going to do for us, having this, knowing that you need something, but whether or not it is a standard as opposed to just a number is the question here.  Is that okay with everyone?

The next piece, formulary identifier.  This had been suggested by one of our testifiers previously.  I’ll just review this quickly.  Formulary identifier being placed on a standards benefit card was suggested as a practical way to link a patient to a specific formulary list.  The question is, is this part of e-prescribing or not, would be helpful.  The recommended action, support development of a formulary identifier standard, which could be anywhere.

Questions, comments?

MR. BLAIR:  It is not clear to me.  Somebody might explain this to me, but I don’t understand why we need in our e-prescribing standards deliberations to standardize a formulary identifier, is there a need for us to do that.  I don’t understand.

MS. BYRNE:  I think it would be extremely difficult to standardize identifiers for formularies across multiple payors, A.  It is almost impossible.  I don’t know who would do that.

But also, that information, if you print it on a card, it changes.  So the benefit can change.  You still have the same group number, but the formulary that they are using is different.  So I don’t see this as a practical thing to do.

DR. COHN:  I want to say, just out of my own experience, I noticed awhile ago on my own health card that they removed benefit information, because it kept changing so quickly that they couldn’t keep track.  They got tired of sending me three or four cards a year.

MS. BYRNE:  It is expensive.

DR. COHN:  It gets expensive.  So this is my story.  Brian, please.

MR. BAMBERGER:  Brian Bamberger, MediMedia.  It was pointed out yesterday by someone from Express Scripts, who said that there are two groups of plans right now, one group that is submitting information through Rx Hub, where there is a need and a requirement for high granularity in defining the benefit, right down to the individual patient, the individual patient’s formulary.

There is another group that has not yet gone that far, and that have less granularity in their benefits.  They have maybe a single formulary, variable copays, and they haven’t made the commitment to broadcast eligibility information, maybe because they don’t have the systems or they don’t see the need yet, or the ROI isn’t there yet to go down to that granular level.

So in today’s world, where we are operating with our physicians that are using the data that we provide, the difficulty is in trying to figure out how do I connect — how does the patient connect themselves or the physician connect that patient to their individual formulary benefit.  I know the plan name, BCN, but I don’t know if it is the preferred or the preferred gold or what other benefit may be offered by that plan.  Today it is in text.  Or I have to loop back through an eligibility in a complex system; how do I get that information out today.

So that was the need that I expressed in several rounds of written and oral testimony.

DR. COHN:  And see that we don’t forget.  Very good point.  Comments, questions, suggestions on this one?

DR. HUFF:  I can just ask stupid questions.  I think I understand the problem in the linkage, but I don’t see how this solves the problem, because it is changing regularly.  So I understand the need, and I think we would like to address the need, but it seems like it would need to be addressed on the ability to ask some database for this patient, what is our formulary, and not print it on a card.

DR. COHN:  A la the NCPDP attempt to develop a formulary standard.

DR. HUFF:  Right.

DR. COHN:  Brian, comment?

MR. BAMBERGER:  I think in today’s world, part of the problem is, there is a lot of different nomenclature out there for what the formulary is   Internally in a plan, they may be using a code for it.

The marketing people, when they print the brochure, they use a snappy name for it, and the patient may understand it merely by a paper list that they may have gotten once, and not really realize what that name is, and because there is no common nomenclature around that, it really presents difficulty when a patient selects a plan and a physician tries to figure out what benefit or what drug list they should be using for that patient, or anyone else trying to figure out what is going on in that relationship.

The lists themselves, the drug lists, the formulary as a drug list, does not — they may change periodically, either added to or subtracted to; people are not necessarily jumping from one drug list to another mid-year or between years or whatever.  I think changes in health care coverage are more likely than formulary list changes.  Again, not going down to whether a specific therapeutic class like oral contraceptives is covered or not covered in any group within a formulary list.

DR. COHN:  I have a thought, and then Jeff may also chime in here.  I am very sympathetic.  Not everybody is doing e-prescribing tomorrow in a very sophisticated way.

However, I am wondering if rather than us calling this out as a separate item, I am wondering if this is something that needs to get resolved in the actual standard pieces that we have all been working on about this issue of the formulary going down to the provider.  Maybe there needs to be some transparency between that identification of that formulary and the health plans may at their discretion choose to put something on the card to represent that in the same way, so that it would be useful if a health plan decided to do it.

As I said, you heard my story, which is that every six months they do something to my plan.  So I am probably not a good example of that.  But I also agree that thee are other environments, people who don’t live in California where things are a lot more stable.  So I think in other environments this may work a lot better.

Jeff, do you have a comment?

MR. BLAIR:  Yes.  Brian, I think that you have identified a real concern.  But as of this stage, not only is there not a standard that we could turn to for the format of a formulary or standardized terminology for a formulary or a nomenclature for a formulary, to my knowledge the work hasn’t even begun yet.

My thought, and let’s see if the rest of the subcommittee would feel comfortable with this, I think that this is a long term issue that shouldn’t be lost.  We shouldn’t ignore it because it is not something that is going to be able to be ready for a demonstration project in 2006, but somehow, I think that we ought to indicate that HHS should encourage the development of — I’m not sure that is the right way to put it, but somehow we need to encourage the development of standardized formats for the formularies and nomenclatures for the formularies, not necessarily the content, not necessarily the rules.  I don’t want to mess with what payors do on their formularies, but standardized formats and nomenclatures, and that that be a long term goal, and maybe we can flesh it out more by our March recommendations.

DR. COHN:  Comments?  So I think what I am hearing is deferral to March and see if there is something that we can turn this into, and also get some advice.

I don’t think that we can leverage the work that is going to be done, which we all are going to be working on around this issue of getting the formularies electronically and bringing them down to the prescriber at the point of care.  It seems that this ought to be somehow a piece of that.  I’m not going to sa a big piece, but it feels like we are all talking about the same thing, even though some of this will be more manual applications of this.  But it needs to be consistent.

MS. BYRNE:  If you don’t mind, I would like to speak to that, because I do have some experience in that.  This is Terry again.

We originally developed our formularies with the intent of being able to get the information on eligibility.  But we also have implemented the CAQH model, which is more like the model that MediMedia has, where they distribute formularies at a plan level.  So we do feel your pain as well, we understand that.

But I do believe that through our work groups and through the implementation with CAQH, we have been able to help solve that problem by helping to name the formularies, analysis, et cetera.  So I do believe that that should be a part of the standard format task group discussion, and it is something that we should address.  We fully understand that not everybody is going to do an eligibility transaction before they have access to the formulary.

DR. COHN:  I heard last time that Brian was going to be joining this group.  This is the NCPDP group?

MS. BYRNE:  Yes, the task group.  I believe Brian has somebody who is going to participate on the group.

DR. COHN:  Is this a reasonable resolution to this one, in terms of recognizing at least that it has a home?  Comments?  I think I am hearing — it is something we may want to follow along, because we are going to be very interested in the formulary in terms of the standards work, so we will just add this to the list.

Brian, thank you.  Terry, thank you.  Now everybody take a deep breath.  We are on to number two, right?


DR. COHN:  Okay, Margaret, 1B.

MS. AMATAYAKUL:  This has to do with HL-7 harmonization.  This is the observation.  A business case may exist where intra organizational communications occur in HL-7, and may extend to retail pharmacies in some cases, especially for prescriptions written on discharge, or maybe example.

Scott suggested that there be exceptions to the adoption of the NCPDP Script standard in 1A, that HL-7 prescription message should be permanent in existing implementations between prescribers and dispensers so as not to negatively impact early adopters, new installations in which prescriber and dispenser are closely associated organizationally, in the case of a novel installation that extends functionality beyond NCPDP Script, and if the prescriber application uses HL-7, bridging technology should address the ability to communicate to dispenser of individual’s choice having NCPDP.  That last is a little confusing.

DR. COHN:  Comments?  Jeff and then Stan.

DR. HUFF:  I would  be more comfortable just saying that our members are the same enterprise instead of closely associated organization.  I think the intended meaning was the same, but what we are really saying is that these things are part of the same business entity.  We are not saying they are two businesses that have a close relationship, right?

MR. BLAIR:  Do you have more?

DR. HUFF:  No, that’s all.  The other part is wordy, but I think it says what we intended.

MR. BLAIR:  The only thing that I worry about in that — 95 percent of it I feel fine with — the only piece that I worry about is that I don’t think that that statement should be so broad that it would put us in a position where a retail commercial pharmacy needed to accept not just NCPDP Script messages from a prescriber, but would need to also accept HL-7 messages from a prescriber as well.

I don’t think that it wants to go that far, but it has got so many caveats that I am just afraid that it could be interpreted that way.  Am I wrong?

DR. HUFF:  I don’t think so.  It is wordy, but basically it says if it is already implemented, it can continue.  For new ones, the only new ones that would allow HL-7 are things within the same enterprise.  Outside the enterprise then it is NCPDP Script.

MR. BLAIR:  So you are saying, if there happens to be a retail pharmacy that is already accepting HL-7 —

DR. HUFF:  We are not going to blacklist them.  They can continue —

MR. BLAIR:  We are not going to ask them to rip it out.

DR. HUFF:  Right.

MR. BLAIR:  But anything that is new from now on, the message that the pharmacy receives would have to be received in NCPDP Script format.

DR. HUFF:  Right.

MR. ROBERTSON:  There is one little part to my recommendation that is not really reflected in here.  This referred to the e-prescribing messages, new order, refilled, those specific actions, where NCPDP Script is well established in the retail market.  But when you start talking about other messages, and the example that has been around off mike is medication history, I am not in a position where I want to exclude the possibility that HL-7 might be the preferred method of doing that.

In fact, if you are talking about CCR as another message  that a pharmacy eventually might want to receive, I’m not quite sure how CCR can be supported within NCPDP Script or otherwise, whereas CCR is being designed with HL-7 CDA.

So my focus was on those messages that NCPDP Script has a well established — or even if it is not widely used it is part of a recognized set of interactions that are supported.

MR. BLAIR:  What are the words in there that cover what you just said?

MR. ROBERTSON:  The only problem is that when Margaret took my words out, she took the part about the HL-7 exclusions.  But the sentence before said that NCPDP Script should be accepted as the standard for e-prescribing interactions.  I’m not quite sure how I said it.  I could pull up my computer again.  But it was the lead-in sentence that limited the scope.

She very accurately pulled out, including most of my wordiness which was part of the problem, pulled out the section on HL-7, but there was a small piece at the top that limited the scope.  I can go to get my computer.

MS. AMATAYAKUL:  Give me the rest of it and I’ll add it in.

DR. COHN:  Harry, did you have a comment?

MR. REYNOLDS:  Yes.  Can you go back to the observations?  I don’t like the wording much.  It may extend the retail pharmacies in some cases — I don’t think we agreed.  Any communication to a retail pharmacy from that institution should be in NCPDP, not HL-7.

Now, we have modified it on the right-hand side to say — if they are already in production now, that’s fine.  But I really believe we agreed this morning that whatever the institution is, if they are sending the script to something outside their institution, it has to go in NCPDP, not HL-7.

DR. COHN:  Terry.

MS. BYRNE:  I didn’t speak up this morning because I wasn’t exactly clear about the actual discussion.  I did talk to Scott and Lynn at lunch, and I have a couple of concerns.

First of all, we work with entities who have clearly said to us — we work with entities to speak NCPDP and entities who speak HL-7 both.  My concern is that if you have a hospital who is speaking HL-7 within its own hospital, and then they want to at discharge send a message to a retail or a mail service pharmacy, or even a specialty service pharmacy, you are going to force them to have to speak NCPDP.

I don’t understand why, if we have messages and we are recommending two different standards, why we have to recommend or force specific entities to support one standard. Why can’t they support the standard of their choice or multiple standards if they choose?

DR. HUFF:  I guess I would have said that what we are trying not to do is obligate retail pharmacies to support both, not to require them to support both.

MS. BYRNE:  But in doing that, you are obligating the sender to support both.

DR. HUFF:  Yes.  That is like the train business.  If the train tracks were different sizes, you are obligating some train companies to change the size of the wheels to fit on the standard uniform track.

The other issue that has been mentioned is that if you support two standards, then you have the potential for them to want to get in in advance of the other.  So now there are capabilities that I can do in one standard that I can’t do in the other, and you don’t have interoperability because of that.

So yes, I think it is actually very reasonable to say to hospitals, when you talk to a retail pharmacy, the requirement is on — again, with the caveats that are already here, if I am talking to myself, if I am talking to a pharmacy that is wholly owned by my same enterprise, I can speak to them any way that I want to, HL-7 or otherwise.  If I am speaking outside, I think it is unreasonable to try and ask every pharmacy be required to accept HL-7 or NCPDP Script.  I just think that is a bad design.

MS. BYRNE:  Right, but in giving them the option, you also have — and this happens in claims processing.  You have switches that can translate between versions, can translate between — and I thought that one of the reasons that we were doing the mapping exercise was to make sure that we were keeping these in synch.

DR. HUFF:  Exactly.  All we are saying is that in this case, the question that is unanswered in there is who is responsible to actually execute it.  Is it the sending party or the receiving party?  To me, it makes a lot of sense to put it on the sending side and say, you are responsible for doing the translation.  It shouldn’t be a lot of work, because the mapping is already explained.

In fact, you are not asking each hospital to do this, you are asking the vendor of the hospital system to do it.  So what you are asking is, Cerner and Sunquest and McKesson and those guys to do it once and propagate that translator to all the customers, so it doesn’t seem like an unreasonable burden.

MS. BYRNE:  Would you also say that it would be acceptable then for a network to do the same thing?

DR. HUFF:  Sure.

MS. BYRNE:  I think that is the piece that is missing then.  My biggest concern is with the medication history then, because that is exactly what we are doing for our partners.

DR. HUFF:  Yes, another way of saying that.  I think the goal is that we are not requiring retail pharmacies to implement both HL-7 and NCPDP Script.  So the hospital can do it, or if they want to contract with an intermediary to do it, as long as the message that gets to the pharmacy is NCPDP Script in the usual case, I think that is what we are trying to achieve.

MS. BYRNE:  Then is the plan as well to name the other types of pharmacies that are out there and specify which standard they have to use as well, such as specialty pharmacies or mail service pharmacies?

DR. HUFF:  I may be using the wrong term.  Basically any pharmacy that I send outside of my enterprise to, it would be NCPDP Script, any pharmacy.  So if I am using retail in too restricted a way, any prescription that went outside of my enterprise I would be responsible for translating that to NCPDP Script.

DR. COHN:  Yes, Ross.

MR. MARTIN:  One suggested way to approach this is to say that NCPDP Script is the default standard way of doing prescription transactions, and acknowledging that there are systems that will have both — in the case of a hospital system or a health care system, an integrated delivery network that will have a HL-7 based hospital system, prescribing system, and outpatient pharmacy that also uses HL-7.  We are not obligating them in that circumstance to use Script if both of them are happy using HL-7.  But they are obligated to either send out a Script message to play in the retail game and receive a Script message if they want to.  If they are a pharmacy that uses HL-7, we can’t obligate the physician to have to convert to HL-7, or the prescribing vendor to convert to HL-7.

So either they would do it themselves, or they would use an intermediary, like SureScripts or Rx Hub, to translate that for them to play in that retail space, if you will.  But you can’t obligate the Script user to have to speak HL-7, but you shouldn’t force the HL-7 user to have to change out if they are both using HL-7 and in their business agreements, they are okay with that.

DR. COHN:  I think up until the very last thing you said, I think we are in fundamental agreement, but I’m not sure about the very last comment you made.  Harry.

MR. REYNOLDS:  I am in total confusion over the way this is being reacted to.  We had a group of people sit up there saying they were all going to get together, go do the NCPDP standard.  We talked about issues with unanimity.  Now it is coming apart.  We are talking about standard SIGs, we are talking about other things, which one could get ahead of the other.  We are talking about Medicare beneficiaries, where you want them to get the same service no matter where they go or what the have, the same information coming back, same things dealt with, the same way you want to deal with all the people that you contract with.

I hear a lot of beeping on the trucks, people backing up.  I just want to make sure — I am seeing some heads say no, but you haven’t convinced me yet by what I hear.

DR. COHN:  I think Jeff wants to make a comment.

MR. BLAIR:  I’d like to chime in, and my backup lights are off, not backing up.

I feel very, very comfortable with the way Stan and Scott articulated where the boundaries should be.  I think that I could very concisely say that our recommendation would be that any prescriber message from wherever it is, if it is going to go to a retail pharmacy, a commercial pharmacy, any pharmacy outside of that institution, that they have to be able to receive that information in NCPDP Script format.  The exceptions were the ones that Scott noted, which was, if there is an existing retail or commercial pharmacy or outside pharmacy that is already receiving in it HL-7, fine, they don’t have to rip it out.  If it is within an institution, the same institution, they continue with HL-7, that is fine.

I think I have made that concise, and I think that is perfectly consistent with what Stan and Scott have said.

MR. ROBERTSON:  Speaking to Harry’s point, while it may sound like we are backing up, it is actually that we are trying to move forward in resolving one small part of what I think Lynn and I just understood, but it became an issue.  We were saying there are the two standards.  They have their domains that they work in.      The problem is, there is this overlapping gray area, and this is an attempt to define what you do in that gray area.  The attempt is to try to define retail pharmacy very broadly, but to allow those institutions that to some extent play the retail game, but they don’t have to re-engineer for internal messaging.

DR. HUFF:  We agree.  That is what we said.


DR. HUFF:  That is what Jeff just said.


DR. COHN:  Does everybody agree with —

MR. MARTIN:  The proviso is that you wouldn’t want to force the retail player, meaning a prescriber who wants to communicate with a retail pharmacy that uses HL-7, they would not be the ones that are obligated to do the translation into HL-7.

MR. ROBERTSON:  That was never in conjecture.

DR. COHN:  I think we had all said that anybody who accepts HL-7 would have to also be able to accept NCPDP Script, is what I thought I was hearing.

MR. MARTIN:  If they want to receive retail or if they want to send out.

DR. COHN:  I think that is what Jeff said.  I’m getting confused; everybody is saying the same thing there.

MR. MARTIN:  It is a default.

MR. BLAIR:  The only thing we are imposing is that a retail commercial pharmacy, any external pharmacy, they have to use NCPDP Script.  If they are already using HL-7, we are not going to ask them to rip it out, but anything that is new, it would be in NCPDP Script.

MS. AMATAYAKUL:  Marjorie, go ahead.

DR. GREENBERG:  I’m a bit with Harry.  I have heard two things which seem to be kind of contradictory.  One is, people are talking about internally, so like within a hospital or something.  I don’t think any of us have a problem with that.  But I have also heard that if a retail pharmacy is already accepting HL-7, and that retail pharmacy is external to the prescribing or provider organization, it can also continue to accept HL-7.  So that is not just within an organization that uses HL-7, but we are talking about some retail pharmacy out there that for some reason uses HL-7.  I don’t know if it also uses NCPDP Script or only uses HL-7.  Are there some that only use HL-7, retail pharmacies?  You are saying if you use a really broad definition of retail pharmacy, there might be somebody out there that uses HL-7?  Then we are back to two standards for external transactions.

MR. BLAIR:  I think the key phrase was that any new messages sent to a retail commercial pharmacy needs to be in NCPDP Script.

DR. GREENBERG:  Even if that retail pharmacy for some unknown reason just accepts HL-7?

DR. HUFF:  That’s right.

MR. BLAIR:  Even if they are already accepting HL-7.  Anything new is NCPDP Script.

DR. GREENBERG:  Anything new.  But we are saying, if there is an existing situation.

MR. ROBERTSON:  I do not know the existence or the extent of any existence of HL-7 messaging between non-enterprise systems, so a hospital here or a small chain pharmacy down the street using HL-7 messaging.  It is possible.  I just don’t know, I can’t give you an example of that.

But if they actually have that in place and are functioning today, my suggestion was, don’t force them to change.  If that retail pharmacy is working with other people and starts up new things, they are expected to — based upon these recommendations and trying to get the whole industry to go one direction, they would be expected to be speaking NCPDP to any new installations that they put in.

It is unlikely that that HL-7 messaging would be broad based, because usually HL-7 messages are between specific trading partners, specific communicating partners.  So you wouldn’t have this pharmacy doing it with a broad range of hospitals, a small number, maybe.

DR. GREENBERG:  Do we have an interoperability problem, though?  If there are external exchanges here using HL-7, whereas the vast majority are using NCPDP?

DR. HUFF:  We may be making this overly complicated for essentially one or two situations.  We are not talking about tomorrow, anyway.  We are talking by 2009 or something.

DR. GREENBERG:  So couldn’t they migrate to NCPDP by then?

DR. COHN:  Are we?

DR. HUFF:  By the time it is mandated for —

DR. COHN:  Let me just remind you that standards that are adopted as adequate industry experience don’t need to go through the same piloting and don’t need to wait for —

DR. HUFF:  So what are we talking about for those?  2007?  That is still okay.

DR. COHN:  That’s fine.

DR. HUFF:  Just based on the amount of conversation we have had about it, I think it is very clean if you say, if you are talking to yourself, you can talk however you want.  As soon as you talk outside your enterprise, you need to talk Script.  That seems enforceable, and it seems like we are overly complicating this to save one or two institutions that might be talking externally using HL-7 today.

DR. COHN:  I think I agree with you.  I’m going to let Terry comment, and then let’s see if there are some ways to clean this up.

MS. BYRNE:  I would ask for clarification on this, because after this conversation I am not as concerned about prescriptions as I am about some of the other script functionalities.  So are we talking about just new prescriptions here, and if so, could we clarify that?  That it is just new prescriptions?

We saw NCPDP Script; that encompasses several types of messages, and in future will encompass a couple more.

MR. BLAIR:  When we said new, I think we were saying any new installations of a prescriber or dispenser.

DR. HUFF:  The text actually says that.  But I think the intent is that these are only the prescription messages.  This would not apply to medical history or medication history, which are not resolved yet.

This was at least my understanding of this.  We were talking about prescriptions which would include new prescriptions, prescription refills, fill status, all of those things, but all of the things that are directly related to prescriptions, not to any other transactions like medication lists or medication history or medical history, et cetera, which are yet to be resolved.

MS. BYRNE:  Can we also assume it really only applies to the retail pharmacy side?  So we are just saying that retail pharmacy can not only speak in NCPDP Script, but if there is an EMR that talks HL-7, they convert it back to HL-7 for them, or convert it out of HL-7?

I am most concerned — just based on my experience, I am most concerned about the vendors.  They are the ones that end up having to do a lot of work, and they don’t have the money typically to develop to multiple standards.  We have been asked to help translate — they say we don’t talk Script, in the case of medication history, we talk HL-7.  Can you translate for us back to HL-7, and we map the data for them, so they don’t have to learn how to speak Script, because it is completely different than HL-7.

DR. HUFF:  Again, the goal is that — I’ll use retail pharmacy, recognizing that I may be using the wrong term.  Every external pharmacy, the goal is to have them only support NCPDP Script.  Whether the enterprise changes it or whether they contract with a network to do the exchanges, fine.  What we are really saying is that we want NCPDP Script be the only thing that is required for a pharmacy to receive e-prescribing information, prescription information.

DR. COHN:  Scott.

MR. ROBERTSON:  Which brings us back a little bit to the final portion of the recommendation, where it starts talking about bridging technology.  The point that I was trying to include in there was, with the statement that everything should be NCPDP whenever possible, that it would be allowable for that translation to occur by a third party, so that in the situations that we are talking about right now, if it is just not economically feasible, there is at least a path to conform to the intent of, NCPDP Script is the standard, even if you have to do a work-around.

DR. COHN:  So you are describing basically what Stan stated, which is that if you are sending, you have the responsibility to make sure it is in NCPDP Script, no matter what your system is.


DR. COHN:  You can bridge it, you can contract, you can do whatever.

MR. ROBERTSON:  It is your responsibility to make sure that it is in the language of the retail pharmacy.

DR. COHN:  Sounds like HIPAA to me.

DR. HUFF:  I’d like to make a comment.  There is a significant hidden cost in HIPAA, which is the magical world of translation.  When it is not economically feasible for somebody to use NCPDP, I would challenge equally that it is not economically feasible to pay a transaction charge that is not currently being paid every time you want to translate.

So I want to make sure that is in any discussion that is taken forward, because there are transaction charges that are hidden right now, that are adding costs to those small providers and others.  So I want to make sure — those translations aren’t free.

MR. ROBERTSON:  Part of economic feasibility is, which is cheaper, especially over time.  Yes, it might be a lot of money up front potentially to do programming revisions, but is that going to be offset by six months of translation charges?

DR. HUFF:  I don’t disagree.  You made a very clear statement that it was not economically feasible.  I am saying that each business case decides if it is or isn’t economically feasible.  You made a hard line statement that it would not be economically feasible.

MR. ROBERTSON:  I said if it was not economically feasible.

DR. COHN:  I think we are all beginning to say the same thing, I think.  I do think that there is going to be some wordsmithing to all of this.  I think most of us are concerned to make sure that this is clear enough and tight enough that people don’t drive Mack trucks through it.  I am concerned about this functionality beyond NCPDP Script.

What I would remind everyone, and I think we will be talking about it tomorrow, is the issues around mapping.  If indeed this goes forward, I think it speaks of the importance of good ongoing mapping capability.  I would put on the table, and we may or may not have a chance to talk about it tomorrow, but one of the questions gets to be, if indeed some of this stuff goes forward, does there need to be a much stronger merging and — I won’t say harmonization, I’m not quite sure what the right word is, but are we looking at interoperability here, and is that the requirement that we are beginning to set out for ourselves.

So I don’t have the answer for that one, but I would have us think about that one as maybe yet another part of the answer.  Ross, did you want to make a final comment?

MR. MARTIN:  One comment about new implementations.  I can still foresee a hospital system that has outpatient pharmacy services still implementing a pure HL-7 system, and they would still want to intercommunicate with HL-7 alone.

I think maybe the wording could be something like, as long as they fulfill the obligations of MMA, they fulfill the requirements, you are not concerned with how they talk amongst themselves and immediate business partners.  But the rest of the world they have to be able to talk to with NCPDP Script.

DR. HUFF:  Again, I think the requirement is that it has to be their own enterprise.  We don’t want them making business to business bilateral agreements.  You get into just like you do with HIPAA, where you have got some large health care provider, and this pharmacy says, I’ve got to do HL-7 because I want to keep their business, and now they are forced to do both standards.  We don’t want to do that, I don’t think.

MR. MARTIN:  And I agree.

DR. COHN:  I think we are all in agreement.  Should we move on to the next item then?

MS. AMATAYAKUL:  I am uncomfortable with what I highlighted in blue.  In the case of a novel installation that extends functionality beyond NCPDP Script —

DR. COHN:  I actually agree with you.

DR. HUFF:  I think we unanimously agree with you.  Take it out.  It’s not novel.  I think the issue that we may want to come back to is one that everybody — what would be ideal, for instance, is that some next version of Script is actually harmonized, and there is one version of it that is consistent, that was jointly done by HL-7 and NCPDP, if I were speculating in an ideal universe or something.

But regardless of that, I think what we do have to address somewhere is to say, we have got a set of functionality now, and we are adopting this standard.  At some point we know that we are going to want to adopt some next version of a standard, either this one or a different one.  We are going to want to be able to prototype new versions of something, and have the prototyping of that not be illegal or immoral or something.

DR. COHN:  I think that was a member of the HIPAA reg, which was that you could apply to do new things.

DR. GREENBERG:  Demonstration projects.

DR. COHN:  Demonstration projects, exactly.

DR. HUFF:  That applied to HIPAA, and we can apply the same thing here, but i wasn’t sure that the HIPAA statement covered what we are doing here.

DR. COHN:  I don’t think it does, but I think we should make a statement here.  I think you are absolutely right.  When I looked at that, I said, let’s make a general statement that there needs to be a way to do demonstrations.

MR. ROBERTSON:  And that was my intent.  As long as there is some way of trying new things without getting in trouble, that is all I’m trying to do.

DR. COHN:  Okay, sounds good.

MR. ROBERTSON:  Continued evolution.

DR. COHN:  Are we covered on this one?  I am still trying to chug ahead.  We will stop at six.  We’ve got 20 minutes.  We have a couple more items here.  Item two is the issue of eligibility, inquiry and response.

MS. GILBERTSON:  Is it really worth trying to chug through all of this in 20 minutes?

DR. COHN:  Let’s try to go through another one or two.

MS. AMATAYAKUL:  Eligibility?

DR. COHN:  Please.

MS. AMATAYAKUL:  The observation is, eligibility, inquiry and response by ASC, X-12 and 270-271, version 4010, capability exists in 25 percent of providers overall and 47 percent of payor systems.  There are some gaps in 270-271 with respect to pharmacy eligibility and with respect to benefits.  Work-arounds for some of the gaps have been developed.  Versions 40, 50 and 5010 are also available — and I should say, available or in development — to address industry needs as identified during HIPAA implementations, and in the initial phases of understanding e-prescribing.

At the same time, Medicare is a new payor in the retail pharmacy arena.  Based on experience with HIPAA transactions, it is possible that Medicare contractors may be required to adopt new or different requirements from those in existing standards.

So the suggestions for recommended actions might be, support NCPDP’s effort to create a guidance document to make the pharmacy ID card information to the appropriate fields on the ASC, X-12N, 270-271, and for X-12N to enhance 270-271 for use in e-prescribing.  Also, test ASC, X-12N, 270-271 eligibility version 4050 or 5010 in pilot, and incorporate the TA1 997 and 999 transaction components and requirements for adoption of the ASC, X-12N, 270-271 to eliminate work-arounds for receipt acknowledgement.

DR. COHN:  Harry.

MR. REYNOLDS:  I have a comment under observation.  I think the percentages are very misleading.  A 100 percent of payors are required to be ready.  These surveys, most people aren’t spending any time answering them.  It is required for all payors.  So I think the 47 percent is dramatically misleading, because at some point they are going to be out of compliance if they can’t do it.

I think if we just made a positive statement that this transaction is a required transaction if eligibility is going to go on in the health care community, which Medicare is requiring and maybe they are doing.  I’m not sure putting percentages out there does anything.  We are just stating a fact.  It is a standard transaction, it has been adopted, and here we go, would be my recommendation.  I think anything other than that is —

DR. HUFF:  Simon?

DR. COHN:  Yes, Stan.

DR. HUFF:  Could somebody educate me about the pharmacy ID card?  That is a thing that I am not familiar with.  What is that?

DR. COHN:  Do we want to remove that?

MS. GILBERTSON:  That was one of the recommendations.  There are some models for how to facilitate getting eligibility probably that are much more efficient, but we were thinking in terms of 270-271 being named, and that there is an implementation guide of which some fact sheets were shared with you in probably the March testimony, of what a standard pharmacy ID card is.

That basically once again is to provide basic information for a pharmacy to know how to get a claim to the right payor.  So it has got the BIN, the processor control number, the group number and some information like that in it.

Even though we don’t expect that patients will walk into their docs with their pharmacy ID card very often, we thought it might be helpful to say, if they do present this card to you and you are going to do an online eligibility check, here is a guidance for the vendors of where you put the input data from the card into the X-12 standard.

So it was just to float a thought through, that maybe we could close a little gap there before it started.

DR. HUFF:  So maybe we could just solve that by breaking it out as two recommendations, one that says we want to use 270-271 and secondly, that that could be facilitated by mapping the pharmacy ID card information into the right slots for that.

MR. REYNOLDS:  Could we rephrase that to say the pharmacy information on the ID card, some people bring in a health card that has pharmacy information.

Is D going to have a pharmacy ID card?  It will be a separate card.  So I guess I would stand corrected for Part D, but I think in general throughout the industry, people just don’t have them.

DR. COHN:  And obviously, that first sentence is the important piece.  This is 2.1 and 2.2 is basically what I think we are describing here.

I think what we did here was, people in industry using 270-271.  Certainly it needs some work-arounds, which are there, but it is certainly there and ready to go.  What we are seeing is that future versions need to be piloted; they probably need to be piloted, anyway.

Do we need to do the 4050 or the 5010 as part of e-prescribing, or is that just a general statement that they need to be piloted?

MS. AMATAYAKUL:  It seemed that the 5010 was providing tighter e-prescribing functionality, that the 270-271 4010 still had work-arounds in it.

MS. BYRNES:  Can I comment on that?

DR. COHN:  Sue.

MS. BYRNES:  I think we are the only ones that testified that we were doing this for e-prescribing.  So I think I can say that we don’t have any work-arounds in place, and it is sufficient.  There are no gaps in that transaction that I am aware of.

Secondly, I’m not sure that we need to call out the TA-1, the 997, the 999.  We have implemented those, but they are no different than using error and status for script.  So it is basically part of the 270-271.  So just like Aaron Sasky part of script, I don’t know that it is necessary to call out using those transactions.  If you are using them legally, you should be using the response transactions, the acknowledgement transactions, with them.

DR. COHN:  Margaret.

MS. AMATAYAKUL:  Yesterday Lisa was expressing concern that the TA-1 and the 997 exist today for all the other HIPAA transactions, but because it wasn’t mandated, it wasn’t being used.  Work-around were still being used.  That is why she thought an explicit recommendation would be helpful.

MS. GILBERTSON:  But what is happening in those is, those are one-way transactions that you never get a response to.  So you send a batch, 837, and you never hear, did you get it, are the syntax correct, all that, in real time.  You send a 270 request, you are expected to see that 270 response immediately, telling you something is wrong, or fix it.

So I think it was mixing the transactions.  Yes, all those are in place, but only because I sent it one way, but I never got any response back.  The eligibility doesn’t work that same way.

MS. BYRNE:  I think Lisa was also speaking mostly — I did get clarification from her at lunch or afterwards.  I thought she was talking about our implementation, because our implementation is the only one I know for e-prescribing.  She was talking about medical implementations.

So I don’t know if we want to mix that within this.  I think we can just clearly say for electronic e-prescribing you need to use the 270-271 or pharmacy benefit eligibility legally, or however you want to say that.

MR. REYNOLDS:  I think more and more, the 270-271 process is moving to real time.  If it is required by this group that you get a 270, you send them a TA-1 or 997, and then you send them the answer right behind it.  I’m not sure that is the electronic environment that we are all heading towards.

MS. BYRNE:  The TA-1 and the 997 are error messages.

MR. REYNOLDS:  They also are — 007 goes back acknowledging that you got it, even if there were not errors.

MS. BYRNE:  From a batch mode.

MS. GILBERTSON:  So if you sent a 270 and it had problems, can you get a 271 back that tells you there are errors?  Or do you have to get another transaction?

MS. BYRNE:  There are three different ways to send a real-time error on a 270.  One is a TA-1, and that means I couldn’t interpret your message.  The 997 says I could interpret it some, but there was some bad data in it.  You can send a 271 with errors, which specifies errors within the transaction.

So if you are using the transaction the way it is supposed to be implemented, you would be implementing these transactions.

DR. HUFF:  What if we didn’t make it a separate item, but up there where we said adopt 270-271, including proper use of TA blah, blah, blah?

MS. BYRNE:  I think that would be appropriate.  I was just calling up the fact that we didn’t do that for Script.  So if you are going to do that, you might want to do it for all, because Script also has error messages and status, et cetera.  That is kind of a lot of detail.

DR. HUFF:  I think we actually do do it.  At least right now, we have those four where it said.  We have item four, which we are moving towards at a rapid pace.  It says exactly that, that we should use those other acknowledgement parts of Script.

MS. GILBERTSON:  But we killed four.

DR. COHN:  I think we ar actually okay going forward.  Harry.

MR. REYNOLDS:  I also feel that — back where you mentioned the 4050 and 5010, whatever we put in there should be the next version.  That is going to be expected to be implemented by HIPAA in general.  Otherwise, you are going to have people putting in the 4050 for health and the 5010 for Script.  I think that would not be a service to the industry.

DR. COHN:  I think your wordsmithing is a very excellent division here.  So are we comfortable with what we have said here so far?  I think the simpler, the better on a lot of this stuff.  The Secretary, if he chooses to read it, —

DR. HUFF:  I agree with you, Terry, that that first statement is redundant.  But I don’t feel bad about being redundant when it is that short.

MS. BYRNE:  There is still the problem saying there are gaps in there.

DR. COHN:  Margaret, are you okay there?


DR. COHN:  Yes.


DR. COHN:  Good.  I think this will be the last item we discuss, since we are down to seven minutes to six.  What other items here do we have?  We talked about prior authorization for tomorrow, is that right?

MS. AMATAYAKUL:  Six, seven, eight.

DR. HUFF:  And then important related issues.


MR. BLAIR:  So we are going to start at six tomorrow?

DR. COHN:  I want to thank everyone.  I know this is not easy work.  The devil is in the detail.  On the other hand, we have got to keep moving it up from incredible detail, because that is not necessary.

Tomorrow morning we will start at 8:30 sharp.  We will be starting to work through the remainder of the message format standards, and moving up to other related items, and probably swing back to look back once again at what we are saying around the message format standards as a whole.

I want to thank everybody, and see you tomorrow morning early.

[Whereupon, the meeting was recessed at 5:56 p.m., to reconvene Thursday, August 19, 2004 at 8:30 a.m.]