[This Transcript is Unedited]
Silver Spring Hilton
8727 Colesville Road
Silver Spring, MD 20910
Agenda Item: Call to Order, Welcome - Dr. Cohn
DR. COHN: Good morning, I want to call this meeting to order, this is the first day of three days of hearings of the Subcommittee on Standards and Security of the National Committee on Vital and Health Statistics. The committee is the main public advisory committee to the U.S. Department of Health and Human Services on national health information policy. I am Simon Cohn, chairman of the subcommittee, and in my day job the national director of health information policy for Kaiser Permanente. I want to welcome fellow subcommittee members, HHS staff and we want to thank you for your help in terms of putting these three days together, and others here in person. I also want to welcome those who are listening in on the internet. I obviously want to remind everyone because we are being broadcast to speak clearly and into the microphone, and we'll probably remind people as we go on about this from time to time.
Obviously there's a lot to cover over the next three days. This morning we'll be hearing from the Combined Healthcare Information Initiative with the proposed final reports this morning on medications, immunizations, and a piece related to interventions and procedures. After lunch we move into preliminary reports on genes and proteins, text based reports including history and physical, billing and encounters, and other such issues as which CHI will bring forward to us for discussion. Now the purpose I think as you all know of this conversation is to really provide input into the CHI process before final reports are indeed final, and an early comment period on the preliminary reports, as well as to make sure that there was public comment on their deliberations, so we obviously want to thank you for coming and joining us.
Now following this by mid afternoon we'll be talking about the HIPAA regs, both an update of what's to come in terms of additional final rules and preliminary rules, and we'll be talking some about the actual implementation so far of the recently finalized and implemented administrative and financial transactions. Finally at the end of the afternoon we begin our conversations on ICD-10, this conversation will continue tomorrow morning and into the afternoon. Tomorrow morning we will have testimony related to ICD-10, both CM and PCS, and then in the afternoon we'll be having subcommittee discussions on next steps and potentially what our recommendations may be to the full committee.
On Thursday we will move into our PMRI letter and our recommendations on core terminologies and hopefully we'll be adjourning by early afternoon on Thursday. Obviously these agendas do depend on timeliness of discussion and either the items may move forward if we finish with an item early or they may be pushed back if there is need for additional decision making.
I obviously want to emphasize that this is an open session. Those in attendance are of course welcome to make brief remarks if what they want to say is germane to the topic being discussed. We will also try to make sure at the end of every session that we have time for open comment, so if those in the audience have time, have something they want to say we're more then happy to have them make open comment at the end of each morning and afternoon session. Finally for those on the internet we welcome email and letter comments on issues coming before the subcommittee.
Now with that let's have introductions around the table and then around the room. For those on the National Committee I would ask if there are any issues coming before us today for which you need to publicly recuse yourself to please so state in your introduction. Jeff?
MR. BLAIR: Jeff Blair, Medical Records Institute, vice chair of the subcommittee, I'm a member of AMIA, HL7, HIMSIS, and I'm not aware of anything that I need to recuse myself of.
DR. STEINDEL: Steve Steindel, Centers for Disease Control and Prevention, staff to the subcommittee and liaison to the full committee.
DR. HUFF: Stan Huff with Intermountain Health Care and University of Utah in Salt Lake City, a member of the committee. If we discuss topics that pertain to LOINC or HL7 or ICD-10-PCS then I need to recuse myself from those issues.
DR. COHN: Stan, just to clarify, is that in your formal appointment statement, those various recusals?
DR. HUFF: Yes.
DR. COHN: Okay.
DR. ZUBELDIA: Kepa Zubeldia with Claredi Corporation, member of the committee and subcommittee, and there is nothing that I think I need to recuse myself of.
MS. GREENBERG: Marjorie Greenberg, National Center for Health Statistics, CDC, and executive secretary to the committee.
DR. COHN: Marjorie welcome back.
MS. GREENBERG: Thank you.
MS. AULD: Vivian Auld, National Library of Medicine, staff to the subcommittee.
DR. SORACE: Jim Sorace, CMS.
MS. WARK: Cynthia Wark, Consolidated Health Informatics Initiative.
DR. BROWN: Steve Brown, the Department of Veterans Affairs.
MR. BURKE: John Burke, Office of the Secretary, staff to the subcommittee.
DR. MCDONALD: Clem McDonald, Indiana University, Regenstrief Institute, I'm members of many of the organizations relevant but the only one I'm an officer or direct control is LOINC.
MS. GRAHAM: Gail Graham, staff to the committee, Department of Veterans Affairs.
MS. FRIEDMAN: Maria Friedman, CMS, staff to the subcommittee.
MS. BRADFORD: Alicia Bradford, Consolidated Health Informatics Initiative.
MS. NOSTEIN(?): Karen Nostein with the CHI Initiative.
MS. LESH: Kathy Lesh, the Kevar Company.
MS. ECKERT: Karen Eckert from MediSpan.
MS. WILLIAMSON: Michelle Williamson, CDC, National Center for Health Statistics.
DR. BICKFORD: Carol Bickford, American Nurses Association.
MS. SMITH: Louise Smith, McKesson Corporation.
MS. PROPHET-BOWMAN: Sue Prophet-Bowman, American Health Information Management Association.
MR. KYLE: Frank Kyle, American Dental Association.
MS. BLOOMROSEN(?): Marilyn Bloomrosen, the Altirom(?) Institute.
MS. WALTWORTH: Shelly Waltworth, Unicorn Medical.
MR. LEVIN: Randy Levin, Food and Drug Administration.
MR. LINCOLN: Mike Lincoln, Department of Veterans Affairs.
MS. PICKETT: Donna Pickett, National Center for Health Statistics and staff to the subcommittee.
MS. SQUIRE: Marietta Squire, CDC, NCHS, and staff to the subcommittee.
MS. WILHIDE: Cheryl Wilhide, Magna Systems.
MS. HAWKING: Linda Hawking, Magna Systems.
MS. WHITE: Gracie White, NCHS.
DR. COHN: Okay, now before I hand it over to Cynthia Wark to give us an agenda review because there are some changes in terms of ordering of this morning I obviously first of all need to publicly announce and recuse myself in relationship to CPT, I think as you all know I'm a member of the editorial panel and if we will be talking about any CPT issues today I will be publicly recusing myself. Now having said that I do want to move to the items that we're going to be discussing and talk about a little bit about process before we move into ordering of the items. We are going to be discussing a number of domain issues related to CHI today and I apologize, it's Consolidated Health Informatics Initiative not combined. To make things flow smoothly, I think you all remember at our previous meetings we developed a template letter and so as part of our discussions is what we're going to try to do is at the end of each domain discussion we're going to try to capture our thoughts and recommendations about each of the domains as we go along before we move on to the next, so I want to make sure that the subcommittee understands that. Steve Steindel will be sort of the holder of the draft letter but as we listen to the issues and begin to discuss recommendations and thoughts on them just be aware that we'll be trying to capture them as recommendations back to CHI. Now Cynthia would you like to briefly discuss the ordering of today's discussion items and the changes?
MS. WARK: Yes, thank you Simon, and I'll start by giving an overview of the status of CHI and then we will move into the reports, we do have a total of eight reports to give today, three finals and five preliminaries. And I ask for your indulgence as we modify the order of the presentations to fit other scheduling needs of the presenters. The five preliminary reports that we will present today are genes and proteins, encounters, billing/financial, text based reports, and history and physical. The three final reports will be immunizations, laboratory interventions and procedures, and medications.
I'll go over our schedule and the progress that we're making to date and then we'll be move into the presentations. Here is the schedule that you've seen before and we are making every earnest effort we can to stay to the schedule and complete all of our work covering all of the 24 domains in the portfolio by the end of the calendar year, and as you can see here for October we are really come down the home stretch and everyone is working very hard to stay on schedule. The status of our recommendations are we have three domains with recommendations in clearance, and I won't spend a lot of time talking about those because the committee heard presentations on these in August and September and we have had subsequent presentations with the CHI Council and throughout the participating agencies. The recommendations for lab result contents, demographics and units are in clearance now. The domains with recommendations that are ready are immunizations, medications, and laboratory interventions and procedures.
The next round of recommendations that we expect to have ready for subsequent meetings are interventions and procedures non-laboratory, diagnosis and problem list, anatomy/physiology, nursing, medical devices and supplies. Domains that have teams deployed are encounters, disability, population health, text based reports, history and physical, multi-media, genes and proteins, billing/financial, and chemicals. Those teams that you, domains that you have not heard preliminary reports on previously or today we are trying very hard to have available for the next meeting.
This is a graphic of our process, as a reminder we have four main stages, the first being deploy the teams. We form subject matter expert teams, define the scope, and identify the candidate terminologies. The second phase is where the workgroup does the analysis and feedback, evaluating the terminologies and assessing deployment. It is at that point following that stage that we come with a preliminary report to the subcommittee. The third stage is Council consensus, this consists of a technical presentation to the CHI Council, department and agency review and feedback, and then the CHI Council establishes a consensus and at that point we come back with a final report. Assuming that we are all in sync on these recommendations we move on to standards adoption with government wide policy rollout and agency department specific implementation by an architecture.
Reminder again that we have a range of possible recommendations, it is not expected that we will come with solutions for all domains. We may come saying that there is a perfect terminology that needs only ever grading to maintain. We may come saying that there is an imperfect terminology with defined weaknesses and actions needed to fix those weaknesses. Or we may come saying that there is no solution in that domain at this time, we recommend development of a solution and make suggestions about to whom that activity would be directed.
Now I'm going to ask Dr. Jim Sorace to give two presentations, the first being a final report on laboratory interventions and procedures, and the second being a preliminary on genes and proteins.
Agenda Item: CHI Final Reports - Interventions and Procedures - Dr. Sorace
DR. SORACE: Nice to meet you everybody. I'm part of the CHI team and it's been a fascinating experience. So we had a very nice team, we had representation from the VA, Indian Health Service, DOD, and also strong representation from the NIH. And basically we took a look at a very simple view of the standard for laboratory test orders and that is how should a laboratory test name be represented in the computer. We took sort of a machine oriented view of this and then we worked backwards to the end user. And basically what we wanted to know was is there a good vocabulary for test names and names for test panels that does clinical and anatomical pathology, so we tried to cover the gauntlet of the laboratory. And we looked at LOINC, CPT, and SNOMED CT predominantly and basically we're recommending LOINC, we found that it's a very well-focused vocabulary for this area. It has a large degree of penetration already into the marketplace and into the government, for example, both IHS and the VA are using it for lab test result names.
We did however make an additional recommendation. Basically you have a very interesting interface problem here in that people don't sit down and order highly granular LOINC codes and nobody comes to you and says serum sodium in a hepronized tube versus serum sodium in gray top tube or a red top tube, and in order to address those needs they'll be some sort of, you need to have some sort of hierarchy or structure that coding naïve users can interface with and ask for more general test information as they try to drill down to the actual panel or test that the health care institution offers. It also felt, and I know LOINC is working on this, that the actual issue as to how you name test panels is a very important. People in hospital environments don't tend to order anatomized tests, they tend to look for SMA 20's or SMA 6's, which they have some anticipation from their training contains certain types of tests, and we've asked that they go back, they're currently working on this and I think they have over 50 panels currently named and standardized, but again, work with people in the laboratory community as to how panels should be designed and maybe add a little bit of flexibility in terms of panels that are sort of designed based on very epidemiological data that may be very rapidly updated. And also that we need to give people the ability to navigate on these issues and it becomes more of a user or an expert system interface that this vocabulary will have to support.
And I think that's it.
DR. COHN: Questions, comments? I think some people who are knowledgeable about this probably can't actually ask questions but, Kepa?
DR. ZUBELDIA: Have you looked at the mapping between LOINC and CPT for billing purposes?
DR. SORACE: We've made a recommendation and I believe that LOINC is looking into that as well in terms of trying to construct such a map so that you'd have a ready access to billing codes method.
DR. COHN: Is that a clarification? Sure, whichever of you would like to comment.
DR. HUFF: There have been a number of people who have been interested in the CPT/LOINC mapping and both commercial companies and people who are just trying to create systems. And a lot of the mapping is also known by LOINC because LOINC is loaded into the UMLS but Betsy was pursuing formal discussions with the AMA before we did sort of anything official in terms of publishing that or doing something with it. So it's sort of like a lot is known and probably more could be done but we were sort of waiting to make sure that the AMA felt okay about publishing that before we did any kind of official publishing and mapping.
DR. COHN: Clem, did you have a comment also?
DR. MCDONALD: Well, I just wanted to say I think currently there are, we hear of various mappings around but I think it'd only be legal if somebody already held a license to CPT. I've actually made a contact at AMA to try to discuss this kind of thing and one option might be they would distribute the mapping, or they could manage the licensing and we wouldn't have to worry about it. That's not what they said, that's just in my mind. But yes, I think something could be done and it should be done.
DR. COHN: Okay, I guess I'm wondering if there's, yeah, Steve.
DR. STEINDEL: Simon, this discussion is consistent with the recommendation number five, which just says recognizing there are copyright issues the availability of the map from LOINC to CPT codes would be helpful to produce administrative data claims from clinical applications, so I believe it's already covered in the recommendations.
DR. COHN: Jim can you help me a little bit on this issue of panels and all of this stuff and I guess I'm, I recognize that as a very difficult issue because of course everybody has their own panels, at least that's been sort of an experience that I've had historically that from facility to facility there seem to be slightly different panels or gradations and certainly different providers have their own preferences about what are absolutely essentially elements of the panel of your choice. Now I guess I was trying in my own mind figure out whether this is the space that LOINC ought to cover or whether it's something else, and if something else what is that and I'm just, referencing back to my understanding of what LOINC was, which was not the orderable terminology I didn't think, but more once you ordered something really the very great specifics of what you got back and maybe I misunderstood, I could be misstating the true intent of LOINC, do I see Stan nodding his head so I think I do have that at least historic memory correct, am I wrong?
DR. HUFF: No, I was agreeing that you were wrong.
DR. COHN: Okay, well Stan clarify this for me then before we even ask Jim.
DR. HUFF: No, LOINC intended from the start to be orderable codes and the strategy was that in order to know what to order you needed to understand the results because usually the definition of what's ordered depends on what it is you want to get back. And then the definition of panels was also intended from the start but in order to have great panel names or battery names correctly you need to know what, you need to name accurately what they contain before you can say what the panel is. And LOINC has been working for probably six or seven years on batteries and there are a lot of detailed issues about defining batteries that have to do with optionality, and especially when you get into calculating values, I mean you can almost sort of define required in different ways based on what you want to measure and which other ones you want to calculate, so when you get into hematology things, etc. Being as I'm recusing myself I guess I can't talk about the quality, but just to state that LOINC has worked on and is intended and in fact has done panels for a long time and that was always the intent and always within scope of the work.
DR. COHN: Okay, Steven and then I guess I have a follow-up question
DR. STEINDEL: I guess I can comment a little bit on it since I'm not really involved except as the CDC liaison to the LOINC. We at CDC see panels as very important from both points of view, the ability to group tests that are related together for reporting purposes and for ordering purposes, and we've always looked at this as a basic fundamental feature of LOINC from both sides. And from LOINCs presence in this area previously to about a year or so ago the way they were thinking about panels was from two approaches, one was a totally prescriptive approach where all the tests in the panel would be defined as a LOINC batter, which naturally everyone realizes is a very limited group of tests, and I don't even know if any had been defined that way, there may be a couple. The other has been a battery with a set of tests that are defined that should be part of that batter and an additional set of optional tests that could be used by the facility to flesh out the battery, and this is the common way a CBC is defined in LOINC, where a few tests are always part of the CBC like hemoglobin endomatacrit(?), and then there's a whole set of tests that may or may not appear depending on how a laboratory chooses to do CBC, and that's the more common form of battery. The surveillance community, the epidemiological community posed to LOINC a while, about a year ago, was we have run into the situation where we would like a panel that serves as just an open bucket that we can put test results into and while this sounds somewhat strange it's primarily because in the areas of microbiology, especially in emerging investigations of new outbreaks, the order that comes in is test poor and the actual tests that are being done may change very rapidly as we develop new testing conditions as we understand the organism better, and so creating a defined panel in that respect is very difficult, so we've asked LOINC to consider the idea of adding a third form of the panel, which is the open panel, and they are considering that and working on ways to add it.
DR. MCDONALD: This is a comment not specifically about LOINC but the panel world has gotten a little bit easier since CMS has forbidden certain kinds of chemistry panels, so there's really just eight, six I think, defined billable chemistry panels, there's still lots of panel problems, you're right, I mean because Dr. Jones' favorite panel, which never be a standardization thing, that would have to be redefined in terms of its elements I think for the transmission because there's too many different unique ones to just personal preferences.
DR. COHN: I guess I should ask is there anybody in the audience who desires to make a comment on this particular issue? Other comments from the subcommittee?
DR. STEINDEL: Simon, Jim pointed this out but I just would like to emphasize it. The other thing that the CHI recommendations is asking for from LOINC and they're already in the process of doing this is to create some more generic test codes so that might be designated and that some of these test codes be designated as the preferred ordering form. As the example Jim gave generally speaking when you order a laboratory test for sodium you don't define it to the level a LOINC code actually exists today and we would prefer a test that you can order a more generic sodium, and that LOINC designate that more generic sodium for ordering form. And then the lab would report in the more detailed form.
DR. ZUBELDIA: So Steve let me understand that, are you asking for a modification to the LOINC structure to have some indication that this is the ordering form as opposed to a reporting form?
DR. STEINDEL: No, actually it doesn't change the LOINC structure in that sense but it does change the LOINC structure in that they would add a field to LOINC database and indicate in it whether this was a preferred ordering form or not, and there's some discussion at LOINC as to whether it's better to indicate the preferring ordering form or indicate it in the reverse fashion within the database, but LOINC has already agreed to do that. Kepa, just to clear up, any LOINC code could still be orderable.
DR. COHN: I guess actually I have one, hopefully it's the last question, but I was just looking through the recommendations and conditions and I am sort of aware that there's I think been work to do at least interfacing between LOINC and SNOMED, and obviously SNOMED now has a contractual relationship with the federal government. Does that in any way need to be referenced here, is that a value added, any sort of a condition, completely out of scope for the discussion?
DR. SORACE: I think LOINC is working directly with UMLS to map LOINC codes within UMLS. There was an earlier version of LOINC, to the best of my knowledge, that was downloaded and mapped into SNOMED a while ago. LOINCs added a more, considerably more content since then. So we're simply trying to keep things driven down one standard pathway and avoid forking it.
DR. MCDONALD: It's true that we're dialogue, we have an active dialogue with SNOMED and there is an intention that there will be some kind of orders defined and some kind of connections defined, but the details of that are still under discussion and whether it's through squeeze principally or who does which I think it will, something will happen but I don't know that we can predict exactly what that will be now, today, it might prejudice it to specify what it might be.
DR. COHN: Kepa and then Steve.
DR. ZUBELDIA: I would like to make an editorial comment, even though Stanley may have to recuse himself from voting on this topic I would like to take advantage of his full expertise. Rather then recusing himself from the discussion I would like to benefit from the discussions and his knowledge if at all possible.
MR. BLAIR: Do we need, Marjorie are you able to clarify the latitude that we have because I've often had much the same feeling with respect to the information that would be available to us related to LOINC or HL7, is it the rules that they have to recuse themselves from comments or is it just the rules that they have to recuse themselves from voting?
MS. GREENBERG: It's discussion and voting but factual or informational questions certainly can be responded to.
MR. BLAIR: So in other words we can ask a question for clarification and they can reply.
MS. GREENBERG: Yeah, that's only reasonable, but if you get into a discussion of should we agree with this, should we recommend something else, etc., that they have to recuse themselves from that.
MR. BLAIR: So in other words an advocacy position is not appropriate.
DR. COHN: And I look to Marjorie to keep us clean on that stuff and certainly help us steer on the appropriate side of that particular rule. So Steve?
DR. STEINDEL: Simon I'm asking you from a keeping of the editorial comment, are you asking the subcommittee if they would like to add another item to the conditional list that the SNOMED/LOINC map be maintained and updated and worded just generically like that, or just leave it to the conditions that are? The actual recommendation is conditional based on the first two elements that they list and I would propose adding this as item number eight in the list if the subcommittee wishes to add it.
DR. COHN: I guess what I'm wondering about and I would ask the subcommittee for their views, it just seems that we in some ways want to emphasize the importance of, I'm not sure whether the right term is mapping or mapping it as appropriate integration, or whatever we want to cal lit to important administrative and clinical core terminologies, I mean I think we see it, you had mentioned it in relationship to number five, I'm bringing it up on this one, we'll probably want to reference it when we talk about our PMRI core terminologies and I'm just, I guess I'm sort of looking, I'm not sure what the right wording is on this one but I think we've all commented and indeed the PMRI core terminology letter comments about the importance of maintaining and having the availability of mappings to make everything work. Thoughts, comments, Clem?
DR. MCDONALD: Well, I'm just not sure what I can say, I think we can readily commit to mapping to UMLS because we already have agreements to do that in both sides and it's a three open standard. We are not ourselves can't commit to mapping to something that we don't have copyright control over, so we can't specify anything else to say we can do it because it's not in our control. In other words we can't say we can map to CPT, we can't say we can map to SNOMED, we can only say we can map, we actually are underway doing that.
DR. SORACE: We discussed this and I think that since UMLS I think is really the repository of ultimate mapping and if a domain vocabulary can collaborate directly with UMLS in its field they can extend it to the other vocabularies, I mean that's clearly what the UMLS model is, is to take domain vocabularies and try to map them into a broader medical lexicon. So from our point of view mapping into UMLS with one strong standard bearing organization really has advantages because there's one place to go to keep the standard, there's one path to get it mapped to any other vocabulary, including SNOMED, that may be in UMLS.
DR. COHN: Good point. Comments, questions? I mean maybe we should not pursue this particular bullet.
DR. STEINDEL: In the PMRI, the draft PMRI letter that we will --
DR. COHN: References all this.
DR. STEINDEL: Specifically states that that's one of the jobs of the National Library of Medicine in the future is to maintain these maps.
DR. COHN: So maybe since both these letters are going to come out the same day maybe saying it only once is probably appropriate. Okay, any further discussion of this particular item? Am I hearing any disagreement to acceptance of these recommendations, supporting these recommendations? Actually it's endorsement, concurrence with these recommendations.
MR. BLAIR: Do you need a motion of any kind or you want to just --
DR. SIMON: Why don't we at the end go through our concurrence and other issues when we just finalize whatever the total letter we're saying, I'm not seeing that we need a specific vote on this item right now.
Agenda Item: CHI Preliminary Reports - Genes and Proteins - Dr. Sorace
DR. SORACE: Well laboratory orders is relatively straightforward, and now we're going to step off into the preliminary report on genes and proteins, and basically I handed out, Cynthia handed out this word document and I'd basically like to just work from that. We basically divided this field, we took a top down approach, we did not at any point in this saying like building on genes how do we get to MRNAs and then how do we get to proteins and then how do we get to pathways and then how do we get to diseases. We took more of the view of a health care worker in the future has a clinical problem and they want to research what is known in terms of genes and proteins about this disease state. And so we took a deliberately top down approach because we wanted to make sure that ultimately we serve a health care information infrastructure.
And we actually broke it down into three different categories, you can see them beginning on the second and third pages, and they're inherited genetic variation. And this is basically applications of genomics to disease, so we have a patient and they have a certain set of genes in their genome and do they have a known genetic disease, do they have a predisposition to a disease, or do they have any sort of inherited phenotype, for example, they're a slow metabolizer of a drug and it may be part of their treatment plan. But in this group we're just saying that whatever it is it's inherited through the genome, any way but it's inherited.
The next thing we thought was that patients don't just get genes but they suffer genetic damage and then they get diseases, and malignancy is a classic example of this. So you have the entire issue of acquired genetic changes in patient's tissues, for example colon cancer, or skin cancer, and you're picking up genetic changes now in the patient's DNA at the tissue site as a passive inheritance and this is causing disease.
And then the third thing and in some ways I think the most interesting is you get into the entire issue of infectious diseases, what genomes are important for health care and it becomes very clear that it's not just the human one but you get into all the issues of medically important human pathogens. So we took this structure and we were working on it and we're trying to flesh each one of these knowledge domains out to see what standards might be there to help meet the needs of an end user in the health care setting, five, six, seven years from now.
And basically I listed a bunch of databases on the very front page where you can sort of map these into these three categories, and we became aware of the existence of several large genetic databases that focus on human disease and they include OMIM, GeneTest, and Huge Net, which is out of the CDC, and these are good examples of human genetic disease databases. OMIM, for example, is widely referenced and it's run out of Hopkins and basically they review the literature and they look at what allele's are associated with what inherited phenotype and they assign that an OMIM number, and that's actually the beginnings of a medical vocabulary in terms of genes and diseases. It's very rich in content and if you have a chance go to the OMIM URL and start looking at it, it's extremely well referenced. GeneTest collaborates with OMIM and they have a different site skew, they're looking at issues that include like where can you order a test for this disease as well as writing their own gene summaries.
Now within this domain of genetic diseases we're having a dialogue within our own group because these sources like OMIM and GeneTest are largely free text and there are many people who will point out that that is not really a well structured vocabulary for disease, and that other language sources like UMLS offer the actual vocabulary. On the other hand OMIM and GeneTest are creating this de facto standard and doing the enormous work of trying to assign phenotype to genotype, which is an absolutely crucial step if you're going to have a medical vocabulary move forward. And they actually I think find that it's difficult using some of the vocabularies to describe all the phenotypes that they see out there and they're trying, and so they tend to use free text. I think a lot of the work in this particular domain of the committee will revolve around how OMIM and GeneTest, which are both under NLM contract, what types of structures can be best given to their database perhaps that they can find tolerable to work with the 10,000 fire hoses of genetic information and genes and diseases and papers which they're trying to keep up with but that allow more structured retrieval and structuring of data in this field. And we're trying to follow up with the NLM and both with OMIM and GeneTest and Huge Net to get a better feel for how that might be done.
I'd like to switch quickly to acquired genomic mutations. Currently NCI has the Genome Cancer Micro Anatomy Project, they're trying to populate a lot of chromosomal alterations that are associated with malignancy, and they haven't yet started anything that actually looks at point mutations. Part of this is just that it's very translational research and any given tumor may have like 1,000 point mutations and we know to measure P-53, so it would be sort of a skewed database. We've discussed this and they have representation in our committee and they're thinking about trying to put more of this data in some of their data initiatives and see if they could have something ready in about six months to be really reviewed.
So where I'm going to go to the third branch of this thing, which is infectious disease, and let me just drop back and say that any human disease is a complex issue and you ultimately end up searching across all three of these branches. For example an HIV patient who has a lymphoma might well require that all three of these branches be searched in some manner. And having been a blood banker for a few years in my career infectious disease needs to be looked at really carefully and we are aware of, and I think what we're finding is that we need to be very certain that people are developing vocabularies that allow things like plasmids to be annotated and searched for pathogenicity as well as for antibotic resistance. There are I think some scattered, I shouldn't say scattered, there are groups at the CDC that are aware of this and I think it would be fair to start trying to develop standards within NIAID that has a pathogen anatomy branch, and we're trying to see and so far we haven't really seen it, if there's a clear plan for how to annotate genomes in terms of pathogenicity data, including things like resistance to disease or whether or not it makes it toxin, that's consistent in such a way that you can start building a real cohesive set of databases regarding infectious organisms. At the moment everything is highly scattered so we have like the HIV community doing a lot of work and you can go and put in like HIV sequences and get back drug resistance, but there's no real uniform format or querying format or vocabulary approach that we've been able to find across organisms. And we're still early in this venture so there may be something out there that we haven't found, so it's caveated, but we've asked people, we have representation from NCBI say on our committee and I think we start to talk more directly with NIAID and a few of the other investigators within the NIH to see where this actually stands.
So those are our thoughts at the moment, we've broken it down really more or less along a top down clinical approach rather then a bottom up basic science approach. I think we can see where within genetic variation with collaboration in terms of how to structure existing databases, you can make great strides and really move towards a standards. With acquired genetic changes, that's still an open field I think, there's some work going on at the NCI that may close that gap and it's, it would take about six months to see what the prototype might be. And then within genetic diseases broadly, I think that's actually a very difficult but important area, and we plan to move very aggressively and try to figure out if indeed there is some sort of way of structure this data so that you can have a genetic database of other organisms that allows people to find out pathogenicity and treatment data, and identification data and link it to public health and surveillance data.
So that's where we stand.
DR. COHN: I'm sorry, and when did you say you'd have this work done?
DR. SORACE: Another eight weeks.
DR. COHN: Clem.
DR. COHN: There's one gene related thing you didn't mention is HUGO(?), and I was curious, HUGO I think is the one that assigns the identifier that's been accepted for the gene, when it's identified.
DR. SORACE: Actually in terms of identifying the genes and naming the genes, that's sort of done and we decided to try to take more of a top down approach in terms of end users in the health care setting but you are correct, HUGO is sort of the keeper of the genomic nomenclature. It's much more difficult to find a keeper of the protein nomenclature, that's still --
DR. MCDONALD: And it gets worse to because axons and the control segments and all the rest, but I guess I just would suggest so that people wouldn't when they read this report, would appreciate all the research you did just would probably be reasonable to include it and with that caveat, that this is just, so people aren't saying hey --
DR. SORACE: I understand.
DR. COHN: Other questions from the subcommittee?
DR. STEINDEL: Jim there are two regulatory movements or actually one is not really a regulatory movement, it's an implementation of regulation that the group might want to consider if nothing else to say that they don't need to think about it. The first one is FDA is tightening down on the ASR reagents that are being used by most laboratories for present testing for genes, and you may want to look into their investigations in that area and what they're thinking. And the second is the proposed CLIO(?) regulations for genetics subtests as a special category of laboratory testing and the regulations that they're proposing in that area where they'll be defining what a genetic test is.
DR. COHN: Carol Bickford?
DR. BICKFORD: Carol Bickford, American Nurses Association. I have a question in relation to those of us concerned about occupational health and exposure to chemicals, toxins, environmental entities. Where do those fit in the structure in relation to the genetic changes, do they fit under the infectious disease component or acquired genetic disease or changes, whatever? And I'm asking because I got the sense pretty much of this was focused on infectious disease but there are occupational components. Where do those fit?
DR. SORACE: We've had discussions about environmental exposures and really they fit under all three. The first one is that if the individual has a susceptibility trait that would be in inherited genetic diseases, so to the extent that there may be a sub population that does really get the malignancy from this organic chemical. Beyond that you move into well just exposures to infectious agents, which are actually prevalent in the health care setting, so you're in infectious disease. And then after that you move on into acquired genetic mutations. If you're trying to say for example that this particular malignancy was associated with aflatoxin(?), there's a very specific mutation, I think it's P-53 that's known to cause padocellular carcinoma, so you'd have to genotype the tumor and say that this particular patient or health care worker had this particular mutation in their tumor so that would be strong epidemiological evidence of that.
Your question I think though does raise a really controversial, well, a real important one, in that, but probably one that's sort of outside our direct domain, and that is that everything involves environmental backgrounds and clearly the CDC and infectious disease wants to put in these databases lots of epidemiological data so that they can follow say antibody levels and antibody tighters(?) in populations. But you also run into the fact that basically for example the reason why liver cancer is so much more prevalent in certain areas of Asia is simply because you have a high rate of hepatitis infections that causes the liver cell to proliferate, and proliferating cells are vulnerable to mutagens and then you have a mutagen that causes this P-53 mutation and you get liver cancer. So it all comes together in a mechanism of disease but we more or less are just looking in those three areas to try to make sure that at least those three areas are there.
DR. COHN: Other questions? Kepa?
DR. ZUBELDIA: Going back to what Clem was saying in your scope it shows that the gene nomenclature is out of scope, are you going to look into recommending HUGO as the standard gene nomenclature or --
DR. SORACE: Well, I think what we may do is slightly different in terms, but I'll bring this back up at the committee because we have representation from NCVI and I can get access to NLM as well. I think basically it's going to be the group at the National Institutes of Health, predominantly NCBI, that's going to be keeping the genome itself, and that they'll be responsible for both nomenclature issues as well as maintaining the sequence of it. Because all of that gets to be necessary to go in and look at developing tests and mapping and the issues that you need to support the thing, and to duplicate that effort would be an incredibly enormous probable waste, so I think we may look more at how the genome itself gets maintained and put that in our report. And I don't know that we want to formally adopt that NCBI be used as the standard for the genome or really more or less recognize that it is indeed the de facto standard.
DR. COHN: Please introduce yourself.
MS. LESH: Kathy Lesh from the Kevar Company. I know you say that some things can fit into multiple categories but where do you see birth or a combinatorial type of changes fitting in, whether they'd be under the acquired because they're not really inherited --
DR. SORACE: Can you give me an example?
MS. LESH: Just some of the alterations of when the genes get together --
DR. SORACE: Polygenetic traits?
MS. LESH: Something like that or whether it be some of the enzyme disorders that may be genetically inclined but they're not inherited.
DR. SORACE: Well, I mean one of the, to answer your question we recognize that any human disease can easily spread to all three categories. I think that it's, I think much of what you're referring to is a genetic predisposition to an illness, which may not, two things, first off you may have polygenetic traits, and that is just more then one gene takes to cause them, and again that would be under inherited genetic diseases because you have to inherit them. The second issue I'm getting the feeling has to do more with penetrants or one where you can get the genes but you may not get the disease, that may require some environmental interaction or some other just probabilistic roll of the dice that unfortunately effects some individuals more then another. And that's really still an inherited genetic trait but it gets to what the phenotype is, and it actually gets to a fairly complicated underlying question with UMLS and vocabulary developers and that is how do you represent probabilistic phenotypes, so block one gene mutation doesn't necessarily mean you'll get breast or ovarian cancer, so those are the types of issues which we sort of want to discuss with NLM in the not too distant future and to see if we can get some feedback because those are the kind of issues that you hear from the people who maintain like OMIM and GeneTest is that we start aggregating all this literature but then we get to something that's more probabilistic or isn't well quantitated like how do you quantitate microsoftly(?) or some physical finding, and I think that's actually going to turn out to be one of the areas where the dialogue of this committee goes.
MS. LESH: So does that cover the genetic mutations?
DR. SORACE: Well, it covers how you establish a, it's a question as to how do you assign a phenotype to a set of genetic mutations. This is a preliminary view of the thing so, yes, Steve?
DR. STEINDEL: Jim, I'm hearing a lot that is delving into the research realm, that work still needs to be done in this area, and what I'd like to hear is how are you going to focus the final report that NCVHS is going to see as a CHI recommendation. It sounds like that you may not be able to actually recommend any type of particular terminology except maybe in the most minimal sense, and that your recommendation will probably be that more research and work be done in this area to develop consensus over the next ten years. Is that a fair summary?
DR. SORACE: I think it's a fair summary, I think if we do it right it can happen faster then ten years. So I would hope that whatever we do results in a beneficial change that sort of accelerates the formulation of standards. I think it's very clear that there's not going to be any one standard in any of these fields that's a slam dunk, but I think we can provide some thoughts as to what the final standards might be written to, I don't want to say specifications but maybe just a more clear idea as to how these standards could be focused. For example GeneTest is beginning to adopt some very primitive but useful I think approaches that they're trying to just XML tag their gene reports, and just developing sort of a set of documents that would allow you to mark it up and include UMLS vocabulary where it's appropriate within the tags and also as tag names, could be just extremely useful to accelerate this thing. And I think that we're sort of at the point where some kick starting would help a lot.
DR. ZUBELDIA: My question is exactly on that, on the XML tags, is that something that, how do you see that going? Is that something that should be going to a set of recommendations to HL7 to develop tags for genetic parsing or parsing of the genetic description of the text? Or is that something that should be independent from HL7 or another standard body creating that sort of parsing mechanism?
DR. SORACE: Well, I don't know that we're really, it's really appropriate for us to try to pick the body because I think what we're trying to do right now is just survey the field and determine what is there and what might be done, so I think it's still open to the scientific community. Clearly I think the NLM will have a major role in it in that many of these databases are under contract to them, and are funded in that manner.
MS. LESH: Kathy Lesh from the Kevar Company again. This looks just more disease oriented, how about using the genes and proteins in the treatment arena, gene transfer, which has brought, Steve's comment brought that to mind.
DR. SORACE: You mean gene therapy?
MS. LESH: It's called gene transfer, yeah.
DR. SORACE: Well, we're not, we're trying to I think look at the more broad issue of diagnosis and treatment of diseases, and currently gene therapy is not extensively used. Having said that it's an acquired genetic trait variation and you're acquiring it through gene transfer, so you'd have an inherited genetic variation, or an acquired genetic variation that's leading to your breast cancer say, and you're treating it through a treatment modality that's an acquired genetic variation so I mean I can map anything to this, I don't mean to be flippant but I'm just saying that that I think actually would be an acquired genetic variation that's being used to treat a disease.
MS. LESH: But in a lot of the gene transfer trials the gene itself isn't being integrated into the body, it's just being used as a vector in order to be able to get something into the body so it's not integrated into the human genome.
DR. SORACE: I'm sorry, you mean they have the vector that's used for treating the disease, that's used for inserting, and then what's the gene that you're putting in the vector that's supposed to have the therapeutic effect?
MS. LESH: Right, the gene is attached to --
DR. SORACE: A vector.
MS. LESH: Yes, whether it be plasmid, whether it be a virus or something to that effect, but then it's producing a protein in the body but the gene itself is never integrated into the body.
DR. SORACE: Well it's integrated into some tissue in the body and it's an acquired genetic change in that tissue.
DR. COHN: Well, thank God this is a preliminary discussion. Steve?
DR. STEINDEL: I had just a quick comment to Kepa concerning his comment on XML tags and I would just like him to make a note to hold that comment until we hear the preliminary report from CHI on text based reports because it may be a very appropriate comment with respect to that. And just a quick comment on what Kathy was just asking, I think from the way we've broken up the world of domains and everything I would look at gene therapy as a procedure and possibly the genes that are inserted as a drug, and what we would be using as the gene nomenclature to name the drug that we are using for insertion purposes.
DR. SORACE: It's a therapeutic modality, I mean there's no doubt about it.
DR. STEINDEL: And it just falls in with the other therapeutic modalities, it just happens to be a very unusual one.
DR. COHN: Other comments? Now be aware this is a preliminary report, it's really more of a brainstorming and hopefully providing CHI input as they go off to sort of further define and clarify the scope, so certainly there's no decision making at this point but I think we are being reminded how complex this area is. Carol Bickford?
DR. BICKFORD: Carol Bickford, American Nurses Association. When we take a look at the matrix at the top which identifies the scope for this domain's work it identifies that protein and gene nomenclature were not within the structure. Are there key groups working these pieces or are there, is there work that needs to be accomplished to help flesh out this environment that wasn't within the scope of this charge for this short term work activity?
DR. SORACE: Well there's YUGO and there are gene nomenclature organizations that are keeping track of the DNA level work in this field. And they've mostly grown up from the basic science environment, but they are the workgroup that is actually directly tasked with sequencing the human genome and trying to catalogue its variation. So I think it's very, so I think that basically that's reasonably well established. In terms of protein names it's unbelievably hard to figure that out and there's basically no one out there who I think can lay legitimate claim to being a protein nomenclature, so we haven't looked directly at the nomenclature issue for two reasons. The first one, in one setting there's a pretty reasonable answer and that's YUGO and their nomenclature for keeping the human genome, and it's going to be a de facto standard, I mean it's just going to happen. And the other reason is that we couldn't possibly figure it out either and it's going to be a really complicated issue I think trying to figure out what is it, we've spoken with the people say at Georgetown who are in the protein information resource and a few of the other protein people out there and it's an unbelievably complicated issue, like do you name every protein by its function, do you name every protein by its post translational modifications, there can be numerous for each protein, what do you do when you don't know any of that, how do you name all the splice variance, so we don't see anything within the timeframe of CHI now that's going to address that.
DR. COHN: Jim, I guess for my own, myself knowing that we will at some point have you come back and give us a final report, like Carol I think I see things obviously in and out of scope, but probably in this report some reference just as you just articulated about things that may be out of scope but which are clearly a standard versus what isn't since all of these are interrelated, not certainly the same level of what do you do now, but certainly some reference to that would be helpful to give people a complete picture of this area.
DR. SORACE: Well, know with an asterisk, and then explain.
DR. COHN: That would be fine. I wish you good luck on this one, I think this is, I thought medications were sort of complex but I think I need three dimensional modeling software in my computer to figure this one out.
DR. SORACE: I think there's really two major things I'm going to flesh out, I want to figure out what the plans are for annotating microbiological genomes and keeping them, and I'd also like to get a better idea as to what the relationships are going to be between OMIM and GeneTest and some of these people who are putting in substantial amounts of work, and how they might best mark it up and store it. Because there really are tradeoffs, I mean you want it well structured, and we have this debate in the committee, you need to link the phenotype or you don't really have a genetic language. On the other hand it's a mammoth job to keep up with the literature and they run into all these issues in terms of using structured vocabularies, so there has to be a bit of a dialogue on that.
DR. BICKFORD: Carol Bickford, what I'm hearing is we have some silos that are occurring, is there a cross pollination and sharing best practices, building on previous lessons learned, some group work that says we're going to have this set up as our standard between the three of us and make it move more effectively? Is there any of that work going on?
DR. SORACE: Well, we're making extensive contacts to try to address that specific issue and there's this sort of coherent plan in terms of how to bring these resources together. So that's what I'm saying, basically in the next few weeks we just want to focus especially on what I just outlined to try to determine if we can answer that question. There may be so I don't want to be judgmental, it's sometimes hard to find the right contacts, we're looking, so I think we will in a few more weeks, I have a very good idea on that.
DR. COHN: Well, Jim, I want to thank you for updating us on your plans, I think obviously this is the work of CHI is to sort of figure out if there is a game plan, what's really there, if there's something that clearly needs to be done, so I'll be looking forward to seeing what you report.
DR. SORACE: It's been very fun actually.
DR. COHN: Well, good. Do others have comments on this particular issue? As I said we don't, this is not where we're really making a recommendation, it's more that we're just trying to discuss it in public, see if there's other directions or other things that you ought to be doing so we aren't surprised by the final report. Other comments from the subcommittee? Okay, now I do know, I think at this point we're going to transition into medications, but I'm going to suggest, we're getting close to a break-time, I'm going to suggest we take like a ten minute break and then we'll start on the medication area if that's okay.
[Brief break.]
DR. COHN: Please be seated, we're going to get started with the second session. Okay, I think we move into our discussion on the medication final report at this point, Steve Brown you're leading.
Agenda Item: CHI Final Reports - Medications - Dr. Brown
DR. BROWN: Thanks for having me, I guess it's back since May when we gave our preliminary report and I've covered a lot of ground since then. And I guess everybody has a briefing packet, people have seen some of this. So as I mentioned back in May we presented a preliminary report after we'd done some initial analysis, so since the preliminary we've finished the analyses and written the reports and have gone through the regular CHI process, and have come to, I guess we had two presentations within CHI and I think came to consensus within the Council on these recommendations, so this is the slide that you saw earlier of just our regular CHI processes.
I had the honor of being the team lead, we had representation from CDC, DOD, the Food and Drug Administration, Indian Health Service, which I had to do as INHS because Power Point insisted on respelling IHS to HIS every time, and of course the National Library of Medicine.
Initially as I had previously described we cast a wide net and asked the group members what areas of medication related terminologies would be important to them and their agencies or administrations. We initially defined drugs from the FD&C Act, products intended for the diagnosis, cure, mitigation, treatment or prevention of disease, manifestations or symptoms of disease, or altered structure or function of the body, so basically sort of as broad as it gets.
As presented we initially had a number of areas that were considered to be in scope, including active ingredient, clinical drug, manufactured dosage forms, drug products, including a finished dosage form, packaging, labeling, section headers, populations, drug classifications. We initially also performed analysis for adverse events, those were gene administration, indications, contraindications, and aspects of pharmacokinetics and pharmcodynamics. So rather then trying to come to one overall answer to this complicated problem we decided to break it down into smaller problems and see if we could come to solutions via that approach.
What we ended up doing is breaking it down into those that we felt we could consider in scope and provide some reasonable analysis for, and those that were, that we considered to be out of scope for a number of reasons, largely though with the out of scope elements it was because these were much more complicated things then just medications, for instance an adverse event, whether it was a combination of a problem or some symptom, some drug, and some patient in some sort of situation, so it requires a more complex information model as well as a lot of terminologies, and those were the areas that we tried to shy away from as just being too hard to solve in this timeframe. So what I'll do is go through the areas that we ended up doing analyses on that ended up being in scope in these eight areas.
So active ingredients are a substance responsible for the effects of a medication. We considered a lot of options. The FDA's established name, USAN names, USP-NF names, CAS numbers, MolFile structures, IUPAC names, any of the other adopted, I'm sorry, approved names from international organizations. There's actually quite a broad range of these kinds of things. We selected the FDA's established name for active ingredient and UNII codes as soon as they were available, and the reason that we selected the FDA established name actually has some applicability to many of the other domains as well, sort of an underlying concern or respect for the FDA's regulatory authority in a lot of these areas. So to say that we would use names for active ingredients in an area it's obviously an area that FDA regulates, we just couldn't do, so as it turns out we pretty much had to go with what the FDA had to say or be in conflict with their regulatory authority. So we elected to go with the established names, which are I think for all intents and purposes identical with USAN names at this point in time, but there's a sort of a question of who the responsible body is and the UNII code issue, they are not presently broadly available but will be so at some point, I know that FDA is working hard on those. And we declined to have a coding scheme, to select a coding scheme until those became available.
DR. COHN: I thought maybe given the number of different recommendations and pieces, do we want to talk about each one, is this still related to this particular recommendation? Okay, then I'll hold my question until you're done with this particular --
DR. BROWN: Do you want to go through sub-domain by sub-domain, or how would you like to do it?
DR. MCDONALD: If I might make a suggestion, it appears, I would almost suggest we go through a lot of these at once, they're very related and you might just catch a breath, you might want to go all eight or nine of them.
DR. COHN: Okay, I will hold my question then until the end.
DR. MCDONALD: Unless others disagree.
DR. BROWN: In the area of ingredients we made some secondary recommendations in the odd event that there maybe not yet a FDA established name, sometimes these things happen in different sequences in different bodies, but for the most part the FDA established name should cover 99.99 percent of things and if some other agency gets to it first and we're lacking these would function as back-ups. Like I say it's because of FDA regulatory authority, widespread use, everybody uses these and they're all familiar to everyone, and UNII codes when available will be freely available through NLM and thought that was a nice thing.
I'm going to just sort of move through it because there's quite a bit.
MR. BLAIR: Steve? They've got a wonderful air conditioner in this room but it has this background noise so if you could scoot just a little closer to the microphone it would help me a lot.
MR. BROWN: Okay.
MR. BLAIR: That's great, thank you.
MR. BROWN: I want to move on then to clinical drug, and a clinical drug is what we called a pharmaceutical preparation with its components, defined as active ingredients and strengths, and a dose form as given to the patient but not necessarily as manufactured, so an example is amoxicillin 250 milligram tablet, and as I mentioned it includes the dose form as administered.
We considered a number of alternatives, the Semantic Clinical Drug section of RxNorm as distributed in the UMLS. There's certain core clinical drug portions of SNOMED CT, and a number of representations at the clinical drug level by commercial vendors such as First Databank, Multum, Micromedex, MediSpan, a number of others. We selected the Semantic Clinical Drug portion of RxNorm for clinical drugs. We did this because it's a public domain system developed with NLM in conjunction with FDA and VA, a number of other collaborators, and in consultation with HL7 it has mappings to the VA's NDF, to FDB, Micromedex, MediSpan and Multum and all these mappings are freely available.
At the time we wrote this RxNorm was still under development and we recommended on a provisional basis, but them saying it's no longer under development of production system.
DR. COHN: I'm sorry, what did you just say?
DR. BROWN: At the time we wrote this the recommendation was since RxNorm was still “under development” we recommended on a provisional basis for the NLM saying it's no longer under development but a production system.
DR. COHN: Okay, so you are now recommending this not on a conditional basis because it is a production system, or you are recommending it as a provisional basis?
DR. BROWN: It will be for the Library to determine when it's a production system, I believe that's, I wish Betsy were here to say that, I know she's, I think is working on getting some documentation together but that would be a question for, I don't know, Vivian do you want to field it?
MS. AULD: No. I'll find out.
DR. COHN: Vivian, could you get closer to the microphone?
MS. AULD: I'll find out.
DR. STEINDEL: Simon, the CHI recommendation as presented to the Council is that this condition on Rx-Norm is that it's provisional until it becomes production, and if it is in production we can just presume that the condition has been met and it's now recommended. And I think we'll find out something about that on Thursday.
DR. BROWN: Manufactured dose forms is the next area we considered and as distinct from clinical, from a dose form as administered, so that's really within the FDA's purview, so a dose form is a way of identifying a drug in its physical form based on physical appearance, prior to dispensing maybe something with how the product is administered, so for example, lyophilized powder for reconstitution is not anything you ever give to a patient but if you're the FDA and trying to keep track of what manufacturers are producing that matters a whole lot.
We looked at approved drug products from the Orange Book, the FDA/CDER Data Standards Manual and some of the dose forms within HL7. We selected the Data Standards Manual from FDA/CDER, it's available on their website for manufactured dosage forms. It's authoritative, it's appropriately granular, it has definitions, and everybody is using it, and it's the FDA's to regulate. They really do need to be careful about when something is described in the same way or different ways from other products and this is sort of what they do.
Drug products is the next area we considered, and a drug product is one or more finished dosage forms, each of which may contain one or more ingredient, so we're sort of moving in levels of abstraction. We looked at NDC's, FDB's offering, Drug Facts and Comparisons, Micromedex. We selected the FDA's National Drug Code for product names and codes. It's broadly used, used in transactions, it's used domestically and internationally, it's already a HIPAA approved code set. We are of course aware that NDC's have opportunities to improve and encouraged the FDA to improve them and further revise them, also in their co-generation processes to address some of the well known issues that I won't further belabor.
Medication packages is the next area that we looked at, that's any sort of container or wrapping in which the drug is enclosed for delivery or display. There was really only one option in this area that we were able to find and that's the FDA's Data Standards Manual for Packaging, so we selected it. Again this is an important regulatory issue for the Food and Drug Administration and the respected regulatory authority.
Label section headers was another area that we looked at, so when I've talked labels, the package inserts, those horrible four point font things that are hammered into tiny, ever tinier pieces of paper and stuffed in the package, and as we've talked about the FDA is moving towards electronic representation of labels and we wanted to make sure that there was out there the header sections that would be used for those labels, electronic labels I should say. Those are now in LOINC, that's where they are, I know there's been a lot of work by FDA and the LOINC committee, and some with HL7 honestly in this area, it's up for balloting, and so we put that forward for label section headers and didn't, wanting to sort of prevent any confusion as electronic labeling comes out. So it's on ballots because the HL7 membership I guess next month now.
Special populations is the next domain that we looked at and this is, we included this because in my special populations they're defined as those where there might be different labeling considerations for dosage or contraindications, warnings, related to specific drug products, so don't take this drug if you're pregnant kind of thing. There were any number of alternatives we could have gone through because a lot of it's based on demographics so there's 15 of the things we considered for any number of these, there's sort of a limitless number of ways to slice and dice these recommendations. What we were able to come to was the use of HL7 vocabulary tables for race, ethnicity and gender, and deferred on all the other areas because there's just too many of them to try to get through to cover all cases but thought this would be some progress at any rate. And this does align nicely with the other demographics choices that we've made in CHI.
Also there's been race and ethnicity as one of those discussions that you get into and it always, you can talk about it endlessly and not come to any conclusions but there's been a collaboration with CDC, Census and HL7, it fits in nicely with what OMB says, and that's why we went with it.
The final area we made recommendations, some recommendations on was the area of drug classifications. And I think one of the important things to recognize is that there's any number of clinically relevant methods for classifying medications as well as methods for classifying that maybe have other uses for administration and the like. And one of the sort of take away conclusions for drug classifications was that they need to be use case based and that it would probably never be possible to come up with any single classification that was useful in all instances. These are some examples of drug classifications, by no means exhaustive, but they're used for analysis tasks.
So what we did was we made a very limited recommendation with the hopes of advancing patient safety and that is the hierarchies from NDF-RT for mechanism of action and physiologic effect and deferred other areas. The reason we did that was in hopes of spurring some sharing for decision support rules in patient safety issues, so if you watch TV you've probably heard don't take drug X concurrently with an MAO inhibitor, and something like Rx-Norm may provide what drug X is in this statement, but the idea of having what an MAO inhibitor is or some sort of terminology for that we thought would be highly useful and that's why we recommended these particular classifications. They're available at no cost and they're non-proprietary and all that sort of stuff. We recognized though that these are very specific uses and there may be any other number of classifications that would meet other needs that we wouldn't be able to comment on and we certainly don't preclude the use of other drug classifications, just for trying to talk about these very particular areas that we thought that would be useful.
So that's the high speed through the eight recommendations that we made.
DR. COHN: Okay, given the number of sub-domains I think I'm going to suggest that maybe we talk through them sort of sub-domain recommendation by sub-domain recommendation, though I guess I would defer to everyone about how best to do that. But certainly I would start off with just the first sub-domain, which is active ingredients. You're obviously making a comment about any UNII codes, the conditional upon their availability in 2003 and obviously we're, last time I looked at the end of October 2003, at this point is that still a, is that going to be available in the next month, month and a half?
DR. BROWN: I think it's fair to say an initial publication of those is within scope, a comprehensive list of UNII codes is probably not within scope for 2003.
MS. WARK: Perhaps I could clarify that this is really a responsibility of FDA and the CHI recommendation is contingent upon FDA completing the work needed to be done to make the codes freely available to the public, so there are some steps to be taken and some work to be done and until that takes place we're not really in a position in CHI, it's FDA's responsibility to determine what needs to be done, what kind of resources they need and to complete that work. It's analogous to the distribution issue earlier that NLM has responsibility for, so I don't think it's fair for us to give a particular date by which FDA would have their work completed and certainly --
DR. COHN: Cynthia I do understand what you're saying, you have to be aware that obviously being on, even though I'm part of the advisory committee I tend to look at the government as being sort of responsible, sort of end to end processes here. Now I'm not in any way arguing with the comment, I'm actually wondering whether, the reason I'm probing here is I'm wondering whether part of our recommendation is yes we accept this but we also urge the FDA to speedily complete publication of the UNII and I couldn't tell given that we have a month and a half left in 2003 whether that was an issue or not, and that was what I was probing. And maybe Randy, since we have a FDA representative in the audience might want to tell us whether or not this would be done before we'll be balloting on this.
MS. WARK: And while Randy is coming to the microphone let me just say that yes we only have a little bit of time remaining in this calendar year. That doesn't mean that all of these recommendations would be through the entire approval process and ready for implementation, so there is certainly more work to be done beyond December of his year.
DR. LEVIN: At the FDA we have a system in place that we generate codes for our ingredients, we've been doing this for quite some time and now we want to make this publicly available, so what we have been doing is improving our processes so that we can have a comprehensive up to date system. We have the system in place, we generated actually 2,000 codes already, we're taking our legacy data and putting it into the system, and now we're going to work it into our processes so that when investigational drugs come to us that we'll assign them codes and that way we'll keep our system up to date and comprehensive. So we have, it's a matter of putting the processes in place, that will take us into next year to do that.
DR. BROWN: You know what I said was we expect an initial publication of those that are done but it's an ongoing work.
DR. LEVIN: We already have put over 2,000 in our system.
DR. STEINDEL: Randy can I summarize as being probably the only barrier to this is resources, the ability to get it done, that the procedures are all in place and there's no regulatory restriction to do this?
DR. LEVIN: There are no regulatory restrictions, it is a resource issue.
MR. BLAIR: Just a point of clarification, Randy. With the existing resources you have 2,000 codes in place, and I know that you'll never be totally complete but when would you consider that you would be effectively complete and up to date? Is it going to be three months, six months, nine months, a year?
DR. LEVIN: It's hard for me to say exactly but I would say we're aiming for within this next year that we would have, we'd have the processes in place where a new investigational drug would come in, it would be assigned a code, it would be put into the system.
MR. BLAIR: Thank you.
DR. LEVIN: Steve, I should just mention that there was something about resources, there is a, this also is tied to our process for listing products, not every product comes through as investigational, some of our products have already been out in the market and we capture that, those ingredients through our listing process, and to improve our listing process would require a regulatory change. But to be able to generate the ingredient codes for most of our ingredients we will not need a regulatory change, just one to clarify --
MR. BLAIR: Is this the kind of situation where it's meaningful for NCVHS to recommend additional resources, is this something where, can you effectively get somebody in place in the timeframe, or within a year, I don't know it's going to be even helpful, will it be helpful or do you think it won't?
DR. LEVIN: You mean for additional resources?
MR. BLAIR: Yeah, to move things forward.
DR. LEVIN: I don't know quite what I can say about that.
MR. BLAIR: I'm sorry, I must be asking the wrong question.
DR. COHN: I think you'll have to use your judgment on that. Stan?
DR. HUFF: Could you clarify what it is you need regulatory change to do?
DR. LEVIN: Well, this ties into some of the things that we'll see later on but --
DR. HUFF: If it's better we can just defer that.
DR. COHN: Okay, other comments about this first area? Steve?
DR. STEINDEL: Simon, just from a scribe point of view, I capture a sense that we are urging additional resource because if this is adopted as a PMRI standard on Thursday we will be making that statement then.
DR. COHN: Yes, I think that probably is sort of an appropriate thing within this document also.
DR. STEINDEL: And in this document also?
DR. COHN: I'd say so, I mean I'm looking, I think we're in agreement with this first piece but recognizing that the conditional aspects of this are all resource constrained we obviously want to ensure that there's appropriate resources to expedite this.
DR. ZUBELDIA: I've heard the comment from Jeff about, or it was maybe you Steve, about completion, and perhaps the recommendation should be that it's conditional upon being substantially complete, we know it'll be never absolutely complete but that it be recommended, because right now it's in the development and to adopt something that may not be substantially complete ever could be a problem.
DR. COHN: Other question or comment?
MS. ECKERT: Yes, Karen Eckert with MediSpan. My only question was we talked about initial publication of the codes. Is that going to be through the NLM or through the FDA's website?
DR. LEVIN: We could publish the codes either, both places.
DR. COHN: Okay, so the answer is yes.
MS. ECKERT: I just didn't know if there was additional resource issue going through the NLM --
DR. LEVIN: I can't speak for how NLM will publish the codes but I would imagine that they can, but for us we can easily publish the codes on our website.
DR. COHN: Further comments on this section before we move to the next sub-domain? Steve?
DR. STEINDEL: Simon, this is just a procedural question, since we're recommending that additional resources be provided, Cynthia, is this appropriate to put in a CHI recommendation or should the NCVHS comment in the cover letter that this is what it recommends?
MS. WARK: The issue has been raised within CHI and is a part of the provision that has been provided to HHS decision makers, so there is an awareness that more work needs to be done and resources need to be dedicated to complete the task.
DR. STEINDEL: So it would be appropriate to add that, the need for additional resource to meet this condition in the actual recommendation itself, as one of the conditions.
MS. WARK: Yes.
DR. STEINDEL: Okay, I'll work on some wording for that and we'll go over it.
DR. COHN: I guess I thought I understood what we were doing, now I am confused. So you're, rather then having this be in the cover letter, you're including this as a recommendation coming out of the government?
DR. STEINDEL: The way we have worded the cover letter is such that we have one of two ways we can approach it. In the cover letter the way we would word it is that the NCVHS concurs with the recommendation of the medication domain with the modifications as attached, and then we make the modification in the document that we attach.
DR. COHN: Well actually I guess I would question this one being hidden into the document on standards, I think that what we have just been talking about is a comment that yes we approve this but, then there's a sub-bullet, we urge funding and that's actually in the cover letter, I don't think it's an appropriate thing in the body. I mean CHI can decide whether or not they include requests for additional funding within the body of their recommendations but I think this needs to be in the cover letter.
Other comments before we move on? Kepa?
DR. ZUBELDIA: Is the recommendation that the adoption of the UNII be conditional to the codes being substantially complete by the end of 2004? Given additional resources and all that?
DR. MCDONALD: Can I make a comment? If I understand this right, and we probably should get some people who know it better, some of these chemicals we don't know exactly what they are and so there's a challenge with defining them with the specificity you can define crystallized chemicals, and I just don't know whether that can be met by 2004. So I'd just be worried about making such a severe, depending on what you mean by substantially complete.
DR. ZUBELDIA: It will never be absolutely complete --
DR. MCDONALD: I mean I support what you're saying is that we want to get some urgency on this, I just don't know how to say it given the complications.
DR. COHN: I'm not sure what to do with all of this, I think Kepa was intending to push along this recommendation but I think the concern you have Clem is that the wording he was suggesting actually creates potentially a barrier.
DR. MCDONALD: I mean why don't we say is we'd urge due haste, haste so we get the completion as fast as possible.
DR. COHN: And I think Kepa was making a point that there's never completion.
DR. ZUBELDIA: There's two aspects to it, one is that yes, it should be substantially complete as soon as possible, the other is if that completion is not feasible then the recommendation may be to look at something else, I don't know what else but --
DR. MCDONALD: Well I think it's really going to provide us an escape clause, I think it's going to boil down to for these categories where you can't chemically identify a single entity, they may have to just find some approximation and just make up a number, so I guess your goal is still good.
MR. BLAIR: Could we use the words accelerate the availability?
DR. COHN: Yeah, I think that that's probably a better concept since we get, we get into a problem with these substantially complete, what exactly that means, or completion period since these are dynamic code sets as we know. So Steve have you captured some of those thoughts?
DR. STEINDEL: Yes, I've added a comment to the letter and we'll discuss the sentence before or after we finish this whether it will be modified or unmodified, we further note the need for additional funding to be provided to the Food and Drug Administration, I have to reword this a little bit, to provide a publicly available version of their unique ingredient identifier codes before the end of 2004.
MR. BLAIR: See that's the piece that I really want to know clarification on, and maybe we could give them more latitude on that and also not pin it down because I don't know how fast the federal government can move, I mean even if they're expeditious and provide the resources, whether that is realistic or not and I think it may not, so my thought was just to indicate that what we want to do is accelerate the availability and let them move as quickly as possible without putting a deadline on it.
DR. COHN: Yeah, Jeff I think you had talked about expediting the publication of, was the term I think you had used which I thought actually sounded better then --
MR. BLAIR: Accelerate the availability --
DR. COHN: Well expedite --
MR. BLAIR: Expedite, that's fine, too.
DR. COHN: And I think we want to remove 2004 just as being, I mean in my view it should be 2003 but then again I'm an emergency physician. Okay, we can go over that, we'll see what other areas need expedited funding. Anything else before we move on?
MS. GRAHAM: Can we also add, so it doesn't look just like a short term project, and the continued resources to maintain this as a national standard?
MR. BLAIR: Yes, yes.
DR. COHN: Okay, thank you.
DR. HUFF: I think the only comment I would have trying to address Kepa's, I think the situation we're in is that they're, to really solve the problem there is no good alternative, I mean this needs to be a part of the FDA's process and it needs to be done this way and so I understand in one sense you don't want to say we're going to do this until you're sure you can do it. On the other hand it's the only way it can be done that will really be appropriate and correct and sustainable and so what we really want to do is just make sure they have the resources to do it and encourage them to do it as fast as they can.
DR. COHN: Steve, I'll let you wordsmith on that one, I think we've gotten a sense of where we need to be on that domain or sub-domain. Do we want to move on to the next domain? Okay, the next one is clinical drugs. Any comments about that?
DR. ZUBELDIA: I have a question on this, this sounds to me like it's a prescribable drug, right, and I see in the out of scope the dosage and administration, and there's a lot of overlap between the two and I'm not sure why one is out of scope and the other one is not.
DR. HUFF: I think I can address some of that. The clinical drugs are meant to be the names of generically orderable things, and the differences that a given pill has a particular dose and form, so you can say five milligrams and that's different then the prescription because you could say give two of these pills or give one of those pills and so then the amount that you give with the prescription is actually a different amount then may be contained in individual pills associated with that. And so the whole rest of the issue, so clinical drugs are just focused on specifically what is generically contained in that, in sort of that generic, that orderable thing and then the rest of those issues are contained in what would be data fields in electronic prescription. You're right, they are related, but they're different data fields.
DR. BROWN: There's also a considerable amount of terminology in the dosage and administration sections of a medication label that we did not want to address.
DR. HUFF: You need to get closer to the microphone.
DR. BROWN: There's a considerable amount of information and terminology that would be needed to cover the dosage and administration sections of a label that we didn't want to get into either.
DR. ZUBELDIA: But it's just as necessary as this and I'm having trouble understanding why it's out of scope.
DR. BROWN: We didn't think that we could answer those questions because it would require a more extensive information model and we would be modeling the whole world basically at that point so we decided to do something and try to be successful at it rather then model the world and not be successful at all. Not to say that some group can't do that.
DR. COHN: Other comments or questions on this one?
MR. BLAIR: Steve, do you feel like that's an area where you would suggest other word be done? I mean should that be part of the recommendation that additional resources be applied to extend that work beyond what you've been able to do?
DR. BROWN: Well, I think there's two levels here, one is that there's the CHI work looking at the area and then there's others, there's actual work that needs to be done doing all the modeling, so these are different animals. And absolutely the work needs to be done to create the information models and terminology models to structure the data in the labels. And that's a different activity and surely encouraging that work go on so that information is available directly from say the FDA in as structured a form as needed or usably possible would be great. But again there's the sort of the difference in the CHI thing, which is to survey what's out there, what could we possibly make recommendations on that are useful versus this sort of de novo modeling that needs to be done.
DR. COHN: Is there something here that we want to pursue or are we happy with the information?
DR. ZUBELDIA: On this sub-domain it's fine.
DR. COHN: You think the sub-domain is fine? Okay, and we'll reflect on if there's something more we want to say on these other issues at the end. Okay, anything else on this particular area? Okay, manufactured dosage form. Any comments? The next one is drug product, including finished drug form, actually finished dosage form, I'm sorry. Does anybody have any objection to that one?
DR. HUFF: I don't have any objection, if I understood the previous discussions though there's a difference between the finished dosage form and things that are named at the level of the NDC code, because the NDC code basically includes packaging, so you get a different NDC code for each size of package whereas the pills in all the packages could be the same. And so there's, I mean at least within Intermountain Health Care it would be nice to have the finished dose form because that's usually what you give to the patient, so in administration that's what you want, the NDC is actually not exactly the correct semantic fit for what you give to the patient. So I guess I'm happy with NDC being chosen for what it does but I don't think it covers the finished dosage form use case. It seems to me that should either be placed out of scope and said that's work that we're going to do another day, not, I think it's improper to say that adopting the NDC codes covers the finished dosage form use case.
DR. COHN: Since your argument is that describing this as including finished dosage form is an incorrect statement of what NDC is, which I would sort of agree, I mean Steve, comment? Or is there a different meaning of finished dosage form here?
DR. BROWN: Now the domain that we ended up using this for was drug product, which is one or more finished dosage forms, and probably, I don't know if Randy wants to comment on why we went in this particular direction versus finished, we had long discussions about finished dosage form versus drug product and all that kind of stuff.
DR. LEVIN: To address Stan's issue about the NDC and the drug product, we're looking at a code for the drug product in addition to the one for the product and its packaging, so that will include, we're proposing a code for the product. Now the product usually is confined to one finished dosage form. Sometimes there are products that have multiple finished dosage forms and there we're still focusing on the product, not necessarily each finished dosage form.
DR. COHN: Clem, is there a question?
DR. MCDONALD: Well, it's a question/comment. I think that I agree with everybody, but what I'm not sure if I'm hearing that we just want to change labels around, the way I think of it is you've got this inventoriable thing, which includes the package size, the brand and everything, that to me now is associated with the NDC code. And you have this other thing that's less, it's less granular then that, which is what you hand a patient and that's associated with something, and I'm only thinking that we're going to get labeling confusion as you'd be worried about if we aren't careful about which name we use for what and that's what you're worried about Stan I think, too.
DR. HUFF: It sounds like they're doing the right thing, I was just kind of confused by the wording of this, it sounded like the way this is worded that adopting the NDC code gives us the finished product codes, or the finished dosage form.
PARTICIPANT: It gives you product codes.
DR. LEVIN: We'll have product codes and a packaged product code, the latter one is the NDC, that could also be a product code as well.
DR. MCDONALD: Maybe, you could maybe clarify the words that says what everybody means.
DR. STEINDEL: The present recommendation, CHI recommendation where it says final recommendations, it says to improve and revise NDC codes and NDC code generation processes in an expeditious fashion to address well known issues. Would it pay for NCVHS to ask that those well known issues be enumerated? I mean this seems to be one well known issue.
DR. MCDONALD: I don't think this issue is that issue, if I can be so bold, there are a set of issues for the packaged product code, which is what the NDC, unless you're going to change, I guess I'm still confused by this.
DR. LEVIN: Currently the NDC has potentially imbedded in it a product code, the NDC is of three components, one component is the labeler, who labeled the actual product, combine the labeler code with the next series of digits is a code that could be unique to that product, it has nothing to do with the packaging. The last part of the NDC is the coding for the package itself.
DR. MCDONALD: But currently it doesn't and can't.
DR. LEVIN: Currently there are issues with that, the way that the NDC is assigned, some of those issues prevent it from being, those are the problems that people are discussing. In order to correct those problems that would require a regulatory change in the way we gather listing information.
DR. ZUBELDIA: Would you consider renaming this sub-domain, instead of the sub-domain name being drug product includes finished dosage form, to being labeled drug product or packaged drug product? And clarify that finished dosage form? Because that would make a clear distinction.
DR. COHN: Kepa, good point, I think that that's really what we've been sitting here talking about. Is that an editorial suggestion or an editorial change in all of this? Do we ask Randy for that question or what?
DR. STEINDEL: Let's change it to labeled drug product --
DR. COHN: Obviously include just drug product --
DR. HUFF: I think packaged is the more common usage.
DR. STEINDEL: Should we leave the phrase including finished drug form? Is there any comment from the office?
DR. MCDONALD: There are two separate code streams that people would like, the finished package, the packaged thing and this other thing. I think the world would like both. I'm just worried about the way it's stated now and the way you said it that they may get tangled if you're going to use one for both.
DR. HUFF: I think he's intending to make both of those codes and I think the only confusion was the way this was stated. So this recommendation as far as I understand it is really just talking about using the NDC code as the identifier for packaged drugs, and you're going to make in some subsequent, some other, I think you already have them maybe but at some point you're going to publish the things that are the codes for the finished dosage forms.
DR. LEVIN: For the product.
DR. HUFF: For the product and the finished dosage forms.
DR. LEVIN: The product usually consists of a single finished dosage form, sometimes a product may be multiple finished dosage forms, so it would be a code where we were looking at and as described here would be for the product, not necessarily each finished dosage form.
DR. HUFF: Are you going to make codes for the finished dosage forms as well?
DR. LEVIN: Right now there are codes for finished dosage form called, for like solid dosage forms, called imprint codes, and this is related to a series, it's the appearance of the product but we don't have a specific code, that wasn't in there but --
DR. COHN: I guess I'm going to suggest that we get rid of this finished drug form, period, and just call it a packaged drug product. I mean in some ways we're discussing HIPAA if you think about, since this is already part of the HIPAA regs. What?
DR. ZUBELDIA: Look at the next sub-domain, medication package.
DR. MCDONALD: That's the package without anything in it.
DR. COHN: So what do we want to describe this, is this a drug product?
DR. HUFF: Packaged drug I think is the --
DR. COHN: Packaged drug product?
MR. BLAIR: -- including finished drug --
DR. COHN: No, I think we're getting rid of including finished drug dosage form, that's what I think we're trying to get rid of.
DR. BROWN: On the slides, is that okay then, it's drug product, one or more finished dosage forms, each of which contains one or more ingredients, so call it drug product, without the finished dosage forms, I think that's kind of in the table because we initially had the finished dosage form in the early part of the analysis and ended up coming here as a way, so I was trying to sort of convey that it was rolled into a product but not a unique code for a finished dosage form and I guess that didn't come through clearly.
DR. COHN: So I think we're just talking about renaming the sub-domain, which I think is primarily editorial, and that's the second sentence that further describes it, which I don't think anybody has any objection with. Are we okay, Kepa, Clem?
DR. MCDONALD: I'm going to suggest as a separate subject when we finish this one that we also open whether we should explicitly identify some of the weaknesses in NDC to make it easier to make those changes. Just a follow-up on Steve's comment, not as part of this discussion.
DR. COHN: Well, are we okay with what we have here so far? I presume that this will actually be a just a strike-out in the recommendation, so I think we're okay with that. Now Steve had actually previously commented and I don't know whether it was to us or to Cynthia that this issue having to do with the well known issues related to NDC needed to be addressed. Now I don't think we have time to begin to elucidate the issues of NDC, or if we did I would probably want to have public comment on a long description of issues related to NDC. But the question is is do we want to, I mean Steve you had actually asked Cynthia was this a reference that they ought to in some way in this recommendation mention those or was that somehow responsibility that you felt was for the NCVHS?
DR. STEINDEL: Well, it's been accepted at CHI Council with the wording to address well known issues and I was raising the question of since the comment came up about packaged versus non-packaged drug forms and Randy mentioned that was one of the areas they were addressing in the NDC codes is if we wanted to add some other well known issues to this list or just leave it vague. Since CHI Council is happy with it being left vague at this point in time and as you know this is a very complex issue and is already being looked at by the regulatory agency, I would assume that they might come back to us if they require further guidance on the well known issues. But I'm perfectly happy leaving it the way CHI Council has it.
DR. ZUBELDIA: My comment is that I would not call them issues, seriously, I would call them features because what may be an issue for this application is of great benefit for inventory, so it's not necessarily an issue, just a characteristic of the NDC.
DR. COHN: Yeah, certainly I know there are many repackagers who consider that some of the flexibility inherent in NDC is really a feature. So we have to be careful about what we describe as an issue without asking other people for their thoughts. Stan?
DR. HUFF: I agree that's true of some things but I can't imagine any situation where NDC codes that have been reused with a different meaning later is a good thing. There's some things that I think truly are not just features, they're dangerous.
DR. COHN: Okay, well let me ask everyone, I mean do we want to get into this right now or do we want to hold this for another conversation? Steve?
DR. STEINDEL: I would prefer to leave it as issues at this point, I think one of the things that we're looking at that is not directly addressed here is the use of NDC codes for areas that the FDA does not regulate and consequently does not and probably will not generate NDC codes for, that manufacturers are prone to generate now on an ad hoc basis. To me that's an issue.
DR. HUFF: I guess one question I would ask Randy and others who are here is whether in fact at future some date we want to have hearings about, whether that would be useful input to the FDA if we had some hearings about NDC code features. I mean would that be useful to you Randy or is that more or less then what would be useful?
DR. LEVIN: We've had a lot of input about the NDC features, though the issues about, coming back to something we were talking about before about supporting the clinical drug code, supporting the product code and the NDC would require a change to our regulatory process and if the sooner that would happen the sooner that these things would be addressed, so you have to keep that in mind as well.
DR. COHN: Stan to short circuit this discussion I think that this is actually not something to ask Randy, it's really something we need to ask ourselves, and what I've done is based on the conversation, I keep an issues list, and it's something I think we'll need to talk about whether or not we want to look at in 2004 as sort of post HIPAA implementation, recognizing that this piece is already there, it's already being implemented so we don't need to CHI to sort of see where it all is, though some of the CHI recommendations handle other piece, so the question is what are the other pieces that need to be dealt with. Steve?
DR. STEINDEL: Simon, NDC is an approved HIPAA code set and if FDA was proposing modifications to a approved HIPAA code set it would come back to us in the normal chain of events to look at and comment.
DR. COHN: Okay, can we move on to the next area here? I think we're in medication packaging. Any issues, any comments about this one?
PARTICIPANT: No.
DR. COHN: Then I think we're moving to the next one which is labeling section headers. I guess I have a question about this one, I guess I maybe, I probably should have asked Stan, I guess he can factually comment. This recommendation confuses me a little bit only because the final recommendation is, it seems to be sort of a circular comment which it says LOINC clinical SPL section terminology upon incorporation with the terminology into LOINC, and so I thought that maybe that was a little circular. Is that already in clinical LOINC and it just isn't in regular LOINC?
DR. HUFF: Yeah, they were happening in parallel, they're actually there now, they're actually in the current version of LOINC.
DR. COHN: I see, so where did they come from before they were in LOINC?
DR. HUFF: They came from the FDA through HL7, through the structured document committee of HL7. So the FDA and other interested parties at HL7 said as part of proposing this standard for HL7 said these are the label sections that we want, we would like those label sections to be in LOINC so that came from basically the FDA and the people in the structured document committee of HL7 posed a set of codes to LOINC which were subsequently included in the LOINC terminology.
DR. COHN: Okay, thank you. Kepa, do you have a question?
DR. ZUBELDIA: So that brings my question that I noted here, it says that there's no other terminology for labeling, and I thought that the FDA already had some labeling requirements, very specific labeling requirements for manufacturers --
DR. BROWN: They're the ones behind all this anyhow, they've been working through HL7 and LOINC.
DR. ZUBELDIA: But I don't see any mention of the FDA in this, and I think it probably would benefit to know that it is in fact the FDA expressed as LOINC codes but it is the FDA requirements for labeling expressed as LOINC codes.
DR. BROWN: I've actually generated so many reports out of this, it's in at least some of the reports, that it's the FDA that's working with these other organizations.
DR. COHN: Randy, did you want to clarify?
DR. LEVIN: Just to reiterate what Steve just said, that the, we're working on the structured product label balloted in HL7 and to improve the way that we communicate the packaging, the information, drug information in the package inserts, but we're the ones who were proposing how to structure the label in the HL7 group. Does that answer your question?
DR. ZUBELDIA: It does, my comment is that there should be some recognition here that the FDA is together on this with HL7, that it's not just HL7 because I think people will know that there is FDA labeling requirements today, they're going to say what happened, they did not look at them or what happened to them.
DR. COHN: Other comments on this particular section? I think we're agreeing with the recommendations, it's just wordsmithing, or suggesting some wordsmithing around that.
DR. BROWN: Cutting and pasting of additional sections into those that have already been cutted and pasted. It's in there.
DR. COHN: Next one is special populations, and I actually have no, my understanding is is that the HL7 vocabulary tables on this are consistent with the OMB directives and I think the only piece of final recommendation, and once again this is probably a cutting and pasting again, might be just of a recommendation, in the recommendation that the vocabulary tables, which are consistent with the OMB directives, is the recommendation, just to keep things consistent, and not to allow anybody to in any way think that there's an inconsistency. Am I misunderstanding that one?
DR. STEINDEL: No, you're fine, actually we have accepted this recommendation as part of the demographics domain already and if we're going to do anything to modify we should just make a note that this has been previously accepted as part of the demographics domain recommendation, otherwise I would just leave it alone.
DR. ZUBELDIA: I thought that we also discussed that the HL7 tables should be migrated to other standards such as X-12.
DR. STEINDEL: That's in the demographics recommendation.
DR. ZUBELDIA: Should we echo that here, too?
MS. WARK: I think we can probably just refer to the demographics recommendations.
DR. STEINDEL: I would just probably add in the final recommendation section that this is the same recommendation that was made by the demographics domain, any differences should be reconciled.
DR. COHN: Okay, sounds good. Anything else on this one? Okay, final area is drug classifications, and specifically the mechanism of action and physiologic effects. Comments? We haven't worn everybody down have we? Everybody satisfied with this recommendation as it is? Just look at it carefully because we'll be seeing some of this probably coming back on Wednesday for our PMRI, Thursday, thank you, Thursday with our PMRI letter. Okay, any other comments about our medication recommendations?
DR. ZUBELDIA: I would like to go back to the dosage and administration not being in scope, and I would like to recommend that they consider putting that in scope.
DR. COHN: You mean in a future --
DR. ZUBELDIA: For future, that they look into putting that into scope in the future.
MS. WARK: The process that we're using for those areas that are out of scope but would benefit from additional work is to put together what we're calling a phase two work plan and it would benefit us for you to say that in your feedback to us so that we would include that.
DR. COHN: Okay, the holder of our pen I think has left the table, I'm hoping that he will add that to our recommendations about sort of next phase action. Are there other things related to these medication recommendations, either around the table or around the room? Any comments? Steve did you hear that last recommendation to be inserted into our letter?
DR. STEINDEL: No, I was looking for another recommendation to insert into our letter. What was this one?
DR. COHN: The recommendation from Kepa which I think we generally agreed with was that we recommended as a next phase activity that the dosage and administration sub-domains be investigated by CHI. Kepa, did I say that appropriately? Okay. Now I guess I should ask, should we review these, now Steve I guess I would ask your help on this one, are these recommendations that we should be reviewing at this point, or do you feel you need a little while to --
DR. STEINDEL: There was a, some of the recommendations regarding the NDC code changes that we talked about a few minutes ago will require regulatory change, the NDC codes are regulated.
DR. COHN: What recommendations? I don't think we referenced anything that required regulatory changes. I thought we talked about well known problems, which we I think left at that.
DR. STEINDEL: What we said was the CHI recommendations that we would also like the full Council to encourage the FDA to improve and revise NDC codes and NDC code generation process in an expeditious fashion to address well known issues. Well, that's a very bland statement but for them to improve and revise NDC codes is actually a regulatory change and we might want to make a note that the Secretary be expeditious in starting this regulatory change process.
DR. COHN: I guess maybe I'm confused because I thought NDC was a HIPAA code set and I think we're the ones who would probably need to start by holding hearings on it, am I wrong about that or is this something different?
DR. STEINDEL: We could start the process, being expeditious in the process by being proactive in holding hearings on the process but it can go one of two ways, the NCVHS can recognize this as an issue that's going to come up with respect to the HIPAA code process and be preemptive and start holding hearings to advise the Secretary as to what type of regulatory changes might be needed, or to wait until the FDA, until the Department published an NPRM with the changes in the NDC codes that the FDA would like to see and then react to it. We can do it one of two ways, it might be better for us to be a little bit proactive.
DR. ZUBELDIA: There's two different phases to this, NDC under HIPAA is adopted by the Secretary as issued by the FDA, so what Steve is talking about is that to recommend that the FDA change the mode by which it issues and produces the NDC's is different from the HIPAA recommendation whether to adopt or not adopt the NDC's.
DR. COHN: I guess it's true that there's some that's probably, what I think we're sort of hearing is that there's some that the FDA can do that really don't change the nature of the NDC and that there are others that really get into going to another version of NDC effectively, which probably are our concern. So Steve what are you thinking, I guess the question is whether or not, I mean one option is for us to make a note that says we will be following up on this in 2004. Randy do you have a comment that you want to make?
DR. LEVIN: Just to go back, the regulatory changes might not even change the characteristic of the code at all, but there's a lot of things about the process, how the code is obtained and how it's assigned and those are things that they might have to be changed in order to change, improve the way that the code is handled, without changing the code set in any way.
DR. COHN: Okay, agreed, so Steve you're suggesting we need to somehow reference this? After we said we weren't?
DR. STEINDEL: I think it might be worthwhile to reference in the letter a comment that if we're in agreement that some of these as discussed it earlier, well known issues of NDC codes be addressed in an expeditious fashion, it might be good to call it out in the letter.
DR. COHN: Comments? Jeff?
MR. BLAIR: The only thing that I'm thinking of is that even though our intent is to have this be expeditious, if we phase this in a manner which limits the ability of the FDA to move quickly on the things that it can do then we have been counter productive, so I just wanted to make sure that we weren't being so expansive that we start to move things into a regulatory process where it doesn't have to be.
DR. COHN: I guess I'm trying, I'm sitting here thinking about what the right answer is here and I've been trying very hard to not have us spend significant time on going back to all the various bullets we've heard over the last while that relate to issues related to the FDA process. Now if there's some way to reference at a high level that we think that there needs, the NDC process needs to be evaluated and may require some regulatory improvements, something like that, I don't think I have an issue with that but I'd be concerned if we got to a much greater level then that. Thoughts from the subcommittee? Randy do you have comments on this one? Steve do you have comments? It sounds to me like you want to work on this one and we'll talk about it after our final piece. Okay, what we'll be doing then is to reviewing the letter later on in the afternoon. Now I think we're about at, we're precisely at lunch time. We're going to break for an hour, we are a final report shy of where we thought we were going to be, which is immunizations. On the other hand we finished up one of the preliminary reports so I think we're running about on schedule. What we will be doing is reviewing the preliminary reports after lunch, we will be hopefully reviewing the comments that we want to make in the letter and either reaching agreement or making modifications to it, and that will be sort of the completion of this setting and discussions on CHI. So with that why don't we break for an hour and we'll see everybody back at 1:00. Thank you.
[Whereupon at 12:00 p.m. the meeting was recessed to reconvene at 1:00 p.m. the same afternoon, October 28, 2003.]
DR. COHN: Okay, I want to welcome Karen Trudel, you'll be talking to us about HIPAA later on today, but welcome, glad you're here. I think this afternoon we lead off with Cynthia Wark talking about the final report on immunizations, and then we'll from there move into a number of the other preliminary reports. Correct?
Agenda Item: CHI Final Reports - Immunizations - Ms. Wark
MS. WARK: Yes, thank you. And I will start off by presenting the final recommendations for the immunizations workgroup, and this workgroup was led by Jason Goldwater from CMS and he sends his regrets that he cannot be here at this time so I will go ahead and present the recommendations on behalf of the workgroup. As you see on the slide it was a very diverse workgroup with representatives from CDC, USAID, FDA, DOD, Indian Health Service, and CMS.
The work of immunizations was split into two, what I would call a part A and a part B, part A being a reassessment of the messaging standard that was previously adopted with the first round of recommendations in March, and the workgroup looked at the immunization guide for immunization data transactions that is available from CDC and made the recommendation that HL7 be the standard protocol for immunizations based on the National Immunization Program Guidelines, or guide I should say. They then went on to look at the vocabulary standards for immunizations and they defined the domain as the implementation of a data standard for immunizations would provide an organized and streamlined means of communicating between federal partners by offering a real time means of transferring information regarding immunization encounters, vaccine events, patient records, and other immunization related information.
The scope of immunizations was defined rather broadly and encompassed patient safety, it would include standardization to and from providers and users of vaccine information such as primary care physicians and schools, and would provide an up to date standardized method of communication to keep vaccination records current and complete. The alternatives considered were HL7 version 2.4 and SNOMED CT.
The recommendations are in addition to the HL7 version 2.4 and higher for the messaging standard to use the CVX, which is vaccines administered, and MVX, manufacturers of vaccines codes from the HL7 for the clinical vocabulary. The conditions and gaps were identified as adverse event reporting would be revisited within 12 to 18 months, the immunization workgroup was working in concert with the medication domain and felt that they could not adequately address all of these sub-domains being consistent with what the subcommittee heard this morning on the medications domain because a lot of the sub-domains for medications, things like dosage and administration, are the same for immunizations, therefore they were differing to a lot of the work that the medication workgroup did and in the case of adverse event reporting said that we should revisit that within 12 to 18 months.
So that is the recommendation being HL7 version 2.4 for the messaging, and the CVX and MVX codes from the HL7 clinical vocabulary. And I know that you have the final paper from the workgroup, which does elaborate more on the range of coverage of the CVX and MVX codes and the acquisition from HL7. And the condition section basically outlines the existence of the use of the standards, the ones that are being recommended by the National Immunization Program now, and that any future work done in the areas of adverse event reporting would be done in concert with the medications workgroup. So I will stop there.
DR. COHN: Comments, questions from the subcommittee? Let me just start with I think a question which, and obviously I'm just looking at the, I'm obviously still trying to read through the whole thing but as I read this what we're doing is talking about accepting or endorsing the HL7 version 2.4 message format standard and I guess now it's a question about the terminologies within all of that. My understanding is is that that standard references these externally maintained code sets, the CVS and the MVX --
MS. WARK: That's correct --
DR. COHN: -- which is maintained by CDC, and that those are really the key elements that we're really concerned about. Am I --
MS. WARK: That's correct.
DR. COHN: And so those codes are really codes from CDC as opposed to being from HL7.
MS. WARK: Yes.
DR. COHN: Comments, questions, Kepa?
DR. ZUBELDIA: I have a question, in the final recommendation it says for the messaging standard HL7 2.4 as defined by CDC and this Guide version 2.31.
MS. WARK: That is the version of the Guide as opposed to the version of the HL7.
DR. ZUBELDIA: Well, then it says of the HL7 standard protocol version 2.1, so I'm getting confused as to which version is this, 2.4 or 2.31.
MS. WARK: I cannot answer that, I believe it is HL7 2.4 and I thought that the version 2.3.1 referred to the version of the Immunization Guide, not the version of HL7, but is that, I think Steve's making a motion that that's correct.
DR. STEINDEL: The answer is yes, and I'll explain. The present version of the Guide is 2.3.1 as issued by CDC. The immunization messages were written for HL7 version 2.1, they have remained unchanged through version 2.4, so currently the immunization messages that you do see in version 2.4 are the ones that we refer to in version 2.1, but of course HL7 just is consistent with this version number so you can't say the 2.1 HL7 messages, so that's why it's worded that way. And CDC has made a commitment to revise the various immunization identification codes so CDC is comfortable with the way this is written and participated in its development.
MS. WARK: I would be happy to revise this so that it is clearer that that's the version of the Guide and that the standard is, to clarify the difference between the HL7 standard and the version of the Guide. Essentially the immunizations workgroup aside from these specific tables for the vaccines and the manufacturers deferred to the medications workgroup to see what kind of headway they could make and then agreed to revisit the adverse event reporting along with other efforts that would look at that in the future.
DR. COHN: I'm sorry, we're all sort of quickly trying to read, the conditions here are longer then the actual report and I was trying to see if they were really conditions or just a commentary, they seem to be primarily commentary.
MS. WARK: Primarily commentary. I think it's explaining why some of the sub-domains that one might consider natural that are related to immunizations weren't covered by these two tables.
DR. COHN: Other comments or questions? Kepa.
DR. ZUBELDIA: These conditions add to the confusion in the version numbers that I had earlier when it says this culminated in the publication of Implementation Guide version 2.3.1 in June of 1999, subsequently updated as version 2.1 in September of 2002, so I guess the version number went backwards in the subsequent update.
MS. WARK: We'll work on correct those, thank you for pointing that out.
DR. COHN: Any other comments by the subcommittee? Okay, am I to take this quiet as either the fact that everybody is still trying to read the conditional statements or that there's any argument? I think what we want to do is effectively endorse these recommendations, concur with these recommendations, thank you Jeff.
DR. STEINDEL: Simon, are we recommending any specific modifications or concur with? I would assume that if CHI chooses to make some editorial modifications after us that's up to them and we can just concur what with we saw?
DR. COHN: Yes, as long as the meaning is not changed and it's to us in a way that we can affix it to a letter to the full committee and to the Secretary next week.
MS. WARK: We'll send an edited version and I'll indicate where the edits are made so that you can review those.
DR. COHN: Okay, great.
DR. STEINDEL: Cynthia, are you going to attach a slightly edited version to the letter that we forward to the Secretary, is that what you're saying?
MS. WARK: I will send the subcommittee an edited version for your use.
DR. STEINDEL: Okay, then what I'm going to say in the letter, I'm just looking for wording in the letter, what I will say is that we concur with the CHI recommendations as modified.
MS. WARK: Okay.
DR. COHN: Or as enclosed.
DR. STEINDEL: Well, the specific wording we use is recommendations for the immunization domain as modified in the attached document, that's our specific wording.
DR. COHN: Okay, well with that why don't we move into the remainder of the preliminary reports and then at the end of that we can go back and look at the specific recommendations to make sure that we're comfortable with them for the CHI letter.
MS. WARK: Okay, we have four preliminary reports left to present and they are encounters, billing/financial, text based reports and history and physical, and as we agreed previously with the subcommittee we are using a somewhat streamlined approach to these preliminary reports. Because it is an earlier version of our deliberations and reflecting early discussions we have streamlined presentations by members of the Project Management Office and I'd like now to introduce Alicia Bradford who joined the Project Management Office back in August and she is a nurse informaticist having received her master's degree from the University of Maryland Nursing Informatics Program and she is a prior Navy nurse and also practiced nursing at the National Institutes of Health, and we're very pleased to have Alicia as a part of the project management team. I will be giving the first two preliminary reports on encounters and billing/financial and Alicia will follow with the text based reports and history and physical.
Agenda Item: CHI Preliminary Reports - Encounters - Ms. Wark
Okay, the Clinical Encounters Workgroup started work last spring and it became apparent very quickly that the term clinical encounter encompassed many of the other domains that we would have subsequent teams working on, therefore we put the team on hold over the summer and then we redeployed the team this fall. The team has now defined, they've used the ASTM definition of clinical encounter as an instance of direct provider practitioner to patient interaction regardless of the setting between a patient and a practitioner vested with primarily responsibility for diagnosing, evaluating, or treating the patients' condition or both or providing social worker services. And secondly that a contact between a patient and a practitioner who has primary responsibility for assessing and treating the patient at a given contact exercising independent judgment. The encounter serves as a focal point linking clinical, administrative and financial information, so if you look at the scope table below there are several sub-domains that were considered in scope by the Clinical Encounter Workgroup, and several areas that were considered out of scope. The areas that were out of scope are essentially those that are being covered by other workgroups, so allergy information is something that we would expect to be covered by other domains, the demographics of course, you've received a final presentation on, and we have workgroups under way for the remainder of the items here that are listed out of scope, diagnosis/problem list, financial payment, insurance information, and interventions and procedures.
I'm going to jump forward a little bit and talk about how the workgroup went about defining the elements that re considered in scope and those that are out of scope, and on the second page about halfway down there's a methodology that's outlined that is being used by the workgroup to identify the scope.
The workgroup started with HL7, basic encounter message, because we have already adopted HL7 messaging as a standard and it also served as a good model for looking at clinical encounters. The HL7 standard consists of 25 message segments comprising 612 data fields. The workgroup eliminated 517 data fields that fell within the scope of other CHI domains, so that would be for those domains that were noted as out of scope or being handled by other workgroups. And after doing that there were 95 data fields remaining, 57 of those were determined to be date fields, times, yes/no responses, names or other text or identifiers that did not need standards adopted. This left the workgroup with 38 data elements as candidates for terminology recommendations. So what they have done is to put together a spreadsheet listing those 38 data elements and then they did an analysis of the alternatives available for those data elements and now moving back up to the top of the page where it says alternatives identified, they've identified the ASTM standards, the HL7 version 2.4 admission/discharge transfer messages, X-12N 837 health care claim message, SNOMED CT, and HL7 version three. So what they're doing is an analysis where they look at each of the alternatives for those 38 data elements that were remaining on the list as being in scope and noting the presence or absence and then an analysis of the content or the coded values.
The workgroup further clarified its scope after it started looking at these data elements and discussing how the scope would be defined further, for example, by internal use. And it was noted by the workgroup that HL7 encounter messages are used both within a facility and between facilities, and just to give everyone a flavor for what they're dealing with, there's a lot of information exchange that goes on, admission, and transfer within a facility that gets really down into detailed levels of say moving from one unit to another or one bed to another, and the workgroup thought that if they went down to that level of detail that they'd be working on this for quite a while and that it would be very difficult to have consistency from one federal system to another. So they determined that to contain the scope that they would declare any data elements that only support internal operations within a facility as out of scope. And the action here is to determine which if any data elements are clearly for internal use only. Also for encounter life cycle the workgroup has noted that HL7 encounter messages can report information before an encounter starts, while an encounter is active, and after an encounter is completed, and the workgroup has agreed that they would not limit the scope based on the phase of life cycle.
For minimum data set the HL7 standard declares almost all data elements as optional although actual implementations impose tighter constraints, and the workgroup agreed that they would not propose which data elements should be mandatory, they would propose the terminology code sets to use for encounter data elements. And what the workgroup got into here was that some federal facilities would use some data elements within an encounter message and not others and therefore they didn't want to indicate all of the encounter data elements as being mandatory, they wanted to leave it open to the business functions and processes to make those determinations.
Encounter sub-domains they agreed to clarify the analysis by identifying sub-domains within encounter, for example, provider information, admission information, discharge information, accident information, death and autopsy information, and we are here today asking the advice of the subcommittee regarding including all of these sub-domains within our report, so the workgroup would certainly like to hear any advice in terms of the sub-domains.
MR. BLAIR: Cynthia, for my benefit would you just read through those sub-domains one more time?
MS. WARK: The sub-domains that are being included as in scope are admission information, transfer information, discharge information, patient movement information, provider information, accident information, and death and autopsy information.
MR. BLAIR: Admission, discharge and transfer, I'm sorry, let me get closer, but the admission, discharge, transfer information is not within a facility, it's only between facilities, did I hear you correct based on what you said earlier?
MS. WARK: Yes, the workgroup was determined that the within facility transfers would not be evaluated for this go round. There was also another issue they've called mix and match, the HL7 encounter messages include a large number of diverse data elements but the HL7 standard offers at best starter lists of concepts with an expectation that users will expand these lists to meet their own needs. We will attempt to recommend a single terminology system for all data elements and sub-domains within encounter. This had quite a bit of discussion within the workgroup and the option that they discuss was selecting different standards for different data elements, so for example if HL7 did not define a value set or a particular data element but there was a very robust set of codes in X-12 would we want to recommend a standard where we were taking value sets from different standards. And the workgroup is leaning toward not having that as the final solution. It was through this discussion that they added the HL7 version three tables that appear to be more populated then in the 2.4 version of HL7 and that's where the workgroup is coming out on that issue but we would certainly appreciate getting any feedback from the subcommittee on that.
And lastly government reporting, the National Center for Health Statistics collects information on inpatient discharges and ambulatory care, this information overlaps the encounter domain and the workgroup has determined that they will not attempt to harmonize the clinical encounter recommendations with these other reports. So that's the preliminary thoughts of the workgroup.
DR. COHN: Well maybe I'll start with just a question or two. First of all obviously when Jeff asked you to talk about domains and sub-domains I think you actually missed the provider information --
MS. WARK: Oh, I'm sorry, yes.
DR. COHN: -- piece of all of this stuff, which was the one I was most interested in, and was sort of curious what you were thinking about in relationship to that knowing that we have a national provider identifier and I think and a file that should be coming down some point soon as a final rule, how does all that relate to this, does that begin to, you're obviously talking about a single terminology but obviously there's all this other stuff going on.
MS. WARK: I am not intimately familiar with all of the values within that area for HL7 but I think the analogy is that when we talk about demographics for a patient we did not look at a unique patient identifier, we looked at other variables that could be used to correctly identify that you had the right person that you were administering medication to. I think it's the same here for provider information, that it isn't encompassing a provider identifier per se, that it would be other elements that were used to describe characteristics of the provider.
DR. COHN: Obviously I'm listening but obviously we're sort of aware and I would say I don't know what exactly is specified in the file that goes along with the number but that's something you certainly should attempt to harmonize with.
MS. WARK: Could we ask --
MR. BLAIR: Simon, international provider file.
DR. MCDONALD: I don't think it's ideal in the long run, I don't know what's possible in the short run to be accepting the same kind of master file mechanism to figure out who the provider is. Firstly it's in the law that we're supposed, HIPAA is supposed to make up a file, or only four years later or something like that, so there is an opportunity for a number. Secondly it's very important to a lot of things you want to do between institutions, if you want to get a record to a provider it's a heck of a job, you have to go around and kind of create this list of who's really who and where and all, life would be much nicer if you knew who ahead of time they really were. And then that becomes important for privacy because you send it to the wrong office you're maybe doing some hard. So I'd like to argue that at least in the long run you don't just, don't use the same technique as for patients where we have some push back from having a patient identifier, there isn't any known push back from to having a provider identifier, depending on how it comes out. Now there's also a provider transfer specification I think, that you send around what information you do have, which I think should be some more thought about, I don't know if it's in the scope of this activity.
And finally in terms of the questions you're saying, it is true that a whole lot of the tables are defined as user defined but those actually historically were often not that way and I read that coming from push back from the reason why they're converted from HL7 specified to user identified was push back from some vendors who didn't want to fuss with it. These are, you see these in a lot of tables and messages these actually as they are coming across and I'm looking at the PD(?) in one of the business segments, so that I wouldn't throw those away because they're defined as user identified, they actually tend to be seen in a lot of messages and we should still be pushing toward a real standard.
I guess finally it would be nice to know what segments you ended up looking at hard and what was excluded by other domains and others in scope/out of scope because it's hard to know in specifics what you really ended up with looking at. That's a lot, I'm sorry.
MS. WARK: Well, on the last point there is a spreadsheet and similar to what you have for demographics, in the final report will would provide the spreadsheet with all of the segment headers. I don't see any reason why we can't send that along to you as an appendix to this preliminary report so that you can have that level of detail available.
DR. MCDONALD: We could provide better comments back I think. And in some cases version three has fewer, the tables are less populated, I think on average they're probably better populated and they're related, so it's not --
DR. COHN: Karen?
MS. TRUDEL: I think the comment that was made about the provider identifier and the data in the National Provider File is a very fair one and since both CHI and HIPAA are within my domain so to speak we'll agree to take that back and make sure that we look at the elements of the National Provider File which may include things like a specialty taxonomy code, things like that, and make sure that whatever we're doing is harmonized across the two.
MR. BLAIR: I think the methodology that you used where you started with the general topic of encounters and then you looked at all those that were covered by other CHI activities and you sort of wound up with the subsets that result, I think that that may have been a rational way to proceed. But now that you've gotten there the ones that remain I think have a lot of issues that are not tightly similar to each other and I think a lot of the recommendations you may come up with with registration, ADT, and whether it's really appropriate to limit that to only external to a facility rather then internal, and other topics such as providers, I just sort of question whether you should, at this stage whether it's useful to keep this bundled under the domain of encounters anymore.
MS. GRAHAM: Could I ask for some clarification, this is Gail Graham from VA, could you clarify that, what your recommendation would be to transition this to, to just take the component to the other domains or --
MR. BLAIR: I don't remember all of the sub-domains but I think that if you examine those that may have similar issues or similar applicability then maybe you come up with a useful subgroup, but it looks to me that what's left now have such topics that are so diverse that the issues of how you adopt them are going to be so divergent, to do anything I think you're going to wind up having totally different recommendations for each one, or maybe there may be two or three that you could group. Did I answer your question?
MS. GRAHAM: Yes, sir, thank you.
DR. COHN: Jeff, interesting, I actually sort of, I'm about 180 degrees away from you on this one. Actually I think this is a very valuable exercise, to me really the question, and this actually captures a number of elements that wouldn't ordinarily caught with any other approach except maybe through billing or something like that. I think the question that I have had really was more the distinction, I think Jeff you did mention this about the difference between external, internal use and encounter use, external use, and I did observe that this is sort of the first discussion with CHI where you have made that sort of a distinction as you have come in to talk about what's inside versus what's outside. Now I agree with you there are some things that are so unique to an individual facility that you clearly don't want to deal with standards about them but I think Jeff's comment about well geez, does ADT only apply if you actually have to transfer hospitals as opposed to move from an ICU to a med surge floor or whatever, it just moves you into sort of some funny use cases. So I think that maybe there's, I think there is a dividing line but I'm just not sure about external versus internal as the dividing line. Clem?
DR. MCDONALD: I'd support that, for example you might say that the outside place doesn't care about beds, but what if it's the VA and you've got all these different hospitals and you want to analyze infections and you want to see if the roommates caught the same infection, you'd want to know that. Now I think what the issue you're going to get into is that you're not trying to specify the names of the rooms necessarily but you could still make it a standard that they would have to be unique identified with the additional key of the institution, so one could roll things up and use that information. It's the same with the provider, and maybe clinic types, there's some things that you don't want to specify how within their world they name their children, but you'd like to make sure their children will be mixed up with other children. So it wouldn't be just specifying the list of codes but it would be specifying what information was transmitted that would keep them from being ambiguous.
MR. BLAIR: The other thought I would have on that is, and I understand that the constituencies here may have concern, because maybe their registration ADT processes are different at this time and they don't want to be compelled to comply with a standard at this time, but I think that's the whole purpose of what we're going through, so even if they may not want to do that right away I would think the value of what we're doing is that there be at least a statement of direction that they would move towards compliance with our ADT standards, I'm just thinking of that one right now, within their facilities as well as when they exchange. And also try to eliminate as much optionality in what they agree to, and maybe it could be two years out or three years out, but I would think that that's the value of the process we're going through.
MS. WARK: This is very valuable feedback and it really reflects some of the discussion that the workgroup had amongst the members, because they all thought that there was some value to attempting to standardize as much as possible even internally. As the discussion ensued, and it's written this way in the paper to contain scope, and I think they were really looking at this as what can we accomplish in the timeframe that we have and the resources that we have and then what can we say about future work. And so perhaps by adding some of the discussion and the points that you all have made to the paper to indicate that there should be an attempt made to identify internal elements where it's possible to standardize now and then leave the remainder for the future, or for the next phase of work, perhaps that would be a way to address the concern.
DR. COHN: Obviously our intent is to provide helpful input. Kepa and then Carol Bickford.
DR. ZUBELDIA: You asked us for the domains and sub-domains, there's a couple of sub-domains that I think that could be beneficial. One would be for outpatient encounters, I'm not sure if it's here hidden somewhere else, but I don't see it and I think that would be a very important one. And perhaps some less direct type of encounters, such as diagnostic labs, which may have different characteristics from the outpatient encounters.
DR. COHN: Kepa, I think we're going to move to e-encounters and things like this but --
DR. ZUBELDIA: E-encounters, telemedicine.
DR. COHN: Telemedicine, yeah, Carol?
DR. BICKFORD: I have three questions, one of which is when we have a discussion between clinicians related to a patient where does that consultation, that component fit? The definition of encounter is specifically provider practitioner to patient, what about provider to provider? Where will that fit in the discussion as you move forward because it's an important piece of the clinical practice. Another thing that I have a concern about and it's related to the location of the patient to help us appreciate the context, recognizing that the location is not necessarily as critical, but it is and it might accommodate the outpatient concept versus the bed patient versus the home health person, I don't, does that patient location fit in the sub-domain of patient movement information? For instance when you're tracking how they're moving through the admission process, they're here, they're there, they're there, I'm just asking that as a question. And the third piece is in relation to our work in our clinical encounter of doing assessment, maybe it's part of the history taking but it's not, does that fit under interventions and procedures?
MS. WARK: Okay, there were three points there and I just want to make sure that I capture all of these. You were raising the question about provider to provider discussions as consultations. I don't recall the workgroup discussing that specific issue and I don't recall it being a part of the elements that they were looking at but I can certainly bring that back. The patient location, again I think, I'm just not familiar enough with the HL7 data elements to be able to respond to that now but when we provide the spreadsheet that has the actual data elements listed that should be clear where it's covered and how it's covered. And then lastly, assessments, that is not being covered by this domain, that will be covered by other domains, history and physical.
DR. COHN: Other questions or comments? Actually Cynthia, I know that Kepa and I were sort of joking back and forth about the issue of telemedicine and e-encounters is something that needs to be --
MS. WARK: The issue of?
DR. COHN: Telemedicine and e-encounters, electronic encounters is an issue that probably should not, needs to be addressed in some way or other, I don't think that it's actually necessarily a separate domain, maybe it's just part of the, it becomes just another type of encounter along with ambulatory encounters and others but it's --
MS. WARK: And I was just trying to sort out here the ASTM definition, which is being used as a starting point by the Encounters Workgroup, it says that ASTM contrast encounter with an ancillary service visit, which it defines as the appearance of an outpatient in a unit of a hospital or outpatient facility. So even though encounters can take place in many different settings, including telemedicine, it then goes on to say that the ancillary service visit is different from an encounter. This gets into some of the discussions that the workgroup has had about the domain of clinical encounters and how different standards kind of define encounter differently and specify data elements differently, so it's actually not an easy task to try to crosswalk from one to the other because they're not the same, so if we started with ASTM it would be very different to work backwards and look at HL7 and see how it compared. The reason they decided to start with HL7 is because we've adopted that as a messaging standard and it made sense but it doesn't neatly crosswalk to the other standards like ASTM. If we started there it might be very different. So I appreciate all of the feedback and I will take these issues back to the workgroup and we will be back to see you in a few weeks.
DR. ZUBELDIA: Cynthia, since you're starting with the ASTM definition is there another domain for the ancillary service visits?
MS. WARK: For the ancillary services? I will have to go back and look at that to answer that question, I can't answer that honestly at the moment where or how that's covered.
DR. MCDONALD: Just to follow-up on that, because an awful lot of the visit baggage is the same, it may not be, I mean what you have to send around in general, it may not, it's not a clinical encounter literally but when you start getting to what you ship around in a relationship they don't look that much different.
DR. COHN: Cynthia, and I guess, obviously we're all sort of throwing things in on this one about encounters, I notice that you, obviously you have many good alternatives identified, I would obviously just want to make sure that you use as one of your sources though information that may be actually in CMS regulations and advisories that already exist knowing that CMS has obviously spent a lot of time over many years answering questions and identifying issues around what an encounter is, they're probably mostly concerned with what's a reimbursable encounter but we would not want to necessarily have you completely ignore those issues --
MS. WARK: We looked at the Medicaid information system because it does have some data elements for billing for encounters, and it wasn't very helpful. It was very limited in the information but it's a good point, we will use the other sources available at CMS.
MR. BURKE: I would just point out that I would, that an encounter is defined differently for each type of provider and it may be an ominous process to go through and look at all those.
MS. WARK: Could I just clarify what you said? To look at the different encounters for different types of providers, this harkens back to why we put this workgroup on hold and because there are so many complicated issues and if we start looking at all the different provider types in all of the settings it gets pretty confusing pretty fast and the workgroup has trouble coming up with much that they agree upon, so I agree that there's, if we look at, if we start with ASTM or X-12 or a provider model this might come out differently, and the workgroup decided to make it manageable for now, that they would start with the HL7 as the model and then look at the different data elements within that and provider information is certainly one of those areas. So to the extent that we can have them take a look at that I'll take that comment back. It's actually a very complicated domain that interfaces with many others.
Agenda Item: CHI Preliminary Reports - Billing/Financial - Ms. Wark
MS. WARK: Okay, moving on to billing and financial, this domain was identified early on in the CHI process as one where it would be more of a validation effort then a full analysis effort of all the alternatives because of the existing HIPAA standards. So the scope has been defined as the standards that are used to implement electronic exchange of health related information needed to perform administrative functions in the federal health care enterprise, and it is assumed that the HIPAA transaction and code sets will serve as the basis for these standards.
The alternatives identified are those code sets that have been adopted under HIPAA and the workgroup added one more as an alternative, ICD-10-CM.
As the workgroup began to talk about what the sub-domains might be we began with a list of sub-domains identified under HIPAA being claim submission for reimbursement, health care claim payment/advice, eligibility determination, prior authorization referral, enrollment/disenrollment, coordination of benefits, and claims status inquiry. Also identified under HIPAA but out of scope because there are no standards adopted in these two sub-domains yet are claims attachments and report of injury. The workgroup then went on to identify additional areas where we want consider alternatives for standardization within the federal government and the areas that were discussed is a quite long list, many of which we determined would be out of scope, including electronic funds transfer, purchasing of supplies, provider identifiers, unique patient identifiers, plan employer identifiers, advance beneficiary notification, and electronic signatures which is being addressed by the Text Based Reports Workgroup. The two that the workgroup decided would be in scope are appeals and certificate of medical necessity, those are in addition to the code sets already adopted under HIPAA.
The initial thoughts of the workgroup are that health plans and health care providers who transmit any of the designated HIPAA transactions electronically within the federal government, were external to and are considered HIPAA covered entities, and were required to be compliant with the HIPAA transactions and code sets as of October 16th, 2006, therefore the HIPAA transactions and code sets are assumed to be the minimum standards for the CHI billing administrative domain. And it was too late to add this to the paper but we did make contact with the Department of Justice and asked specifically about Bureau of Prisons and those are not considered Medicare covered entities, they do not exchange medical claims outside of their system and therefore they were not listed here but we did inquire as to their status.
PARTICIPANT: [Comment off microphone.]
MS. WARK: Well, the information that we received, the comment was made that they pay for services and the information we received is that they don't exchange billing information outside of their system.
Claims attachments are considered out of scope due to the scheduled publication of the attachment NPRM in 2004, work is underway between HL7 attachment special interest group and CHI staff to map and align CHI clinical standards with the proposed HL7 claims attachment standards, so basically we will follow that work as it evolves and we will consider it out of scope for the time being. So that's the initial thoughts of the workgroup.
DR. COHN: Questions or comments from the subcommittee? It looks like we're fine with that, good luck. Carol?
DR. BICKFORD: Carol Bickford, American Nurses Association, was there any discussion of the complementary and alternative therapies being coded under the ABC codes?
MS. WARK: I'm sorry, could you repeat that?
DR. BICKFORD: Was there any discussion about the complementary and alternative therapies being coded under the ABC codes?
MS. WARK: We did not discuss that in the workgroup. Can you perhaps elaborate as to any further thoughts on that?
DR. BICKFORD: The ABC codes are considered complementary to the existing CPT-4 codes for procedures and activities and there's a financial component attached to that for the billing pieces, and so I was just asking if that needed to be included as one of those that are considered in your discussion, recognizing that it's not a complete all inclusive data set but many of the others are missing those important components for the complementary and alternative therapies, which are considered in many cases to be effective, sometimes even more effective interventions then our traditional procedures, interventions and pharmaceuticals.
MS. WARK: Thank you.
DR. COHN: Okay, next one please.
Agenda Item: CHI Preliminary Reports - Text Based Reports - Ms. Bradford
MS. BRADFORD: I'll be presenting both text based reports and history and physical, the co-leads are from the VA. Both of these workgroups had originally considered merging the reports but once they started to define their scope realized that that was too difficult and have separated them out. They're very early in their process, have only met within the last few weeks, so come to you with very preliminary information seeking any guidance and advice you can provide.
The Text Based Reports Workgroup, their scope has been defined as a domain determined by the group to include the standards and terminologies used to define the messaging architecture and syntax of clinical text documents. Initially all clinical document types were considered as possible sub-domains. Additional sub-domains were further delineated from initial analysis of content of clinical document types, including section headings and data types. The group reached consensus that inclusion of these sub-domains result in scope that was much too broad to be completed in the short time frame and resources allocated. Document components and data domains contained in text based documents overlap broadly with areas already covered by the other CHI domain workgroups.
Their domain and sub-domains for in scope, they've determined text document structure and syntax, electronic signature, document section headings, clinical document types and titles. The alternatives they've identified is the clinical document architecture, ASN1, which is abstract syntax, Notation one, HTML, XML, Rich Text Format, Portable Format, Clinical LOINC, CEN, 13606, Public Key Digital Signature and Electronic Signature.
Their initial thoughts to this point has really just been to identify these alternatives and to briefly describe what they are, and as I said this workgroup has just kicked off in the last few weeks so would appreciate any feedback in their direction.
DR. COHN: Comments? Jeff?
MR. BLAIR: Can I suggest you add one additional alternative to the list? And it's the continuity of care record, it's an initiative from ASTM along with HIMSS, along with the Massachusetts General Society, Massachusetts Medical Society, and I think that's, it involves documentation records for radiology and for discharge summaries and also for possibly for claim attachments. And they are working with HL7's clinical document architecture so it's both complementary to HL7 as well as can be used independently.
MS. BRADFORD: Thank you.
DR. COHN: Carol Bickford and then I have a comment.
DR. BICKFORD: Carol Bickford, American Nurses Association, ASTM's Committee E31 had some workgroups that were working on the format and structure of clinical records and I don't know where that process is in the ballot structure, but they were looking at standardization of headings such as allergies, what needs to be on the clinical documentation for example, the patient name and some of those pieces, so you might check with them as to the status of that one, I don't know the number I'm sorry.
MS. BRADFORD: Okay, thank you.
DR. COHN: I just want to comment and this is actually more related to this then it is a comment on your draft. But in December at our next hearings, you probably just need to know that we're going to be having some discussions about clinical data architectures, relationship both to claims attachments and I think beginning to bridge into that issue of text based reports, parts of the medical record and all of that. I don't know that we're going to complete the discussion there, only knowing how much else is going on at that session already, but I think the intent was to begin to get into that. Carol, did you have another comment?
DR. BICKFORD: Carol Bickford, American Nurses Association, another area to examine, I don't see it listed here, is the significant reports that come out of radiology for the imaging studies and the results reporting from that arena. You have identified laboratory findings but it might also be something like a procedure note, an operative report also, so it's the resulting component as a bigger picture.
MS. BRADFORD: Laboratory findings was determined to be not in scope and rather then the workgroup trying to look at the numerous types of text documents there are and standardizing those they thought they would standardize the format of the document so that it could be exchanged between different platforms and not the actual content of the document. And we do have a multi-media workgroup also that will be looking at radiology.
DR. COHN: Kepa?
DR. ZUBELDIA: One sub-domain that I think is quite important and is missing is the text encoding and alphabets, specifically the use and the encoding of Spanish and French characters and some Arabic characters and so on that may be important, and if you're going to exchange information within Puerto Rico and south Florida, south Texas, south California, you'll find a lot of Hispanic names with accents and so on that get completely mungled when you, that's a very technical term but when you transfer them from one system to another and that can create problems.
Also I would like to see, you haven't identified what the scope of the electronic signature is, but I would like to see the work on electronic signature advance if at all possible to be adopted as a standard under HIPAA for electronic signatures, I think that would be great.
MS. BRADFORD: Somehow I think that's not within the scope of this workgroup.
DR. COHN: Though I obviously would comment that, along with Kepa, I was looking for the electronic signature. There were hearings by this committee, it's been about a year and a half ago now, related to electronic signature and it was really a view, I think it had more to do with HL7 and ASTM and others and you might in terms of not necessarily the alternative but you might touch back with those groups to see exactly what's going on with them. Steve, you have a comment?
DR. STEINDEL: Simon, we actually had a discussion about electronic signature on a CHI conference call and there is a little bit of a complication from our point of view in that CHI is a federal eGov project and there is a federal eGov project on electronic signatures and e-authentification, which, or within federal use we may have to use.
MS. GRAHAM: We added that as item ten on your report.
DR. ZUBELDIA: That would provide an idea bridge between the eGov electronic signature and the HIPAA electronic signature through CHI.
DR. STEINDEL: The intent of the federal eGov projects in this area are such that those recommendations can be applicable outside of the federal system, however they're not in place yet so we can't comment on them, we have no influence on them.
DR. ZUBELDIA: But as you're looking into electronic signature, don't look just electronic signature for human readable text but also look at electronic signature for transactions, whether it be an X-12 syntax or HL7 syntax or whatever the syntax, so the same electronic signature standards and technology and mechanisms can be used for text or transactions of any kind. And that would provide enough flexibility and maybe we can one day find something that can be used for HIPAA.
MS. BRADFORD: Any other feedback?
DR. COHN: So when will this report be done, two weeks, three weeks?
MS. BRADFORD: Well they are on an expedited timeframe but not that short.
DR. COHN: Thank you. As I said you might also want to work with us around that December hearing to make sure that, it also may meet your needs in some fashion or other. Next one?
Agenda Item: CHI Preliminary Reports - History and Physical - Ms. Bradford
MS. BRADFORD: The last preliminary report, same workgroup, history and physical. The scope has been defined as the terminology used to identify, classify and name the components incorporated into a patent's medical history and the physical exam process performed by a physician. Because this process is not standardized and is dependent upon the clinical judgment of the examiner it is extremely difficult to incorporate every possible component that might be included. Therefore the group has taken a broad brush approach to this effort keeping in mind the JCAHO accreditation requirements for a complete history and physical exam.
The group elaborated a list of the major components or sections, which may be included in a history and physical exam document. Each component was compared to the list of domains addressed by other CHI groups. Where the other CHI groups have focused on a particular domain it was agreed that the component would be left as out of scope. The remaining components listed here were left in scope, and they are history of present illness, review of systems, past medical/surgical history, family history, social history, non-medical allergies, physical exam observations, and physical exam findings.
The list of alternatives listed that they came up with initially was SNOMED CT, ICD-10-CM, ICD-10-PCS, MEDCIN, DSM-VI, MEDRA, Clinical LOINC, and CPT-4 codes. Once again their initial thoughts were only to describe the different terminologies and to ask for any feedback.
DR. MCDONALD: At some risk of mentioning a word, and that's why, there's really three different apples and oranges in that list that you've mentioned already, so to over simplify, LOINC is just, the headers or sections or questions in that context. SNOMED would be principally the answers or findings or diagnoses, and MEDCIN is a little bit more complicated because it ties, it includes sort of a logic to follow in some sense within a structure. Just be aware that they're not, they're not necessarily mutually exclusive.
DR. COHN: I guess I would make a comment here on, and it seems to me there's two issues, I mean one is the structure and the other is the terminology, and I'm not sure that we've settled the structure issue yet, and it's always very helpful to have a structure to figure out what terminology to stick into it. I'm actually wondering if CDA and CDA like pieces, which I think were related to your text based reports is really, and even though these are separate areas I know you're working on both but this is sort of where they apply and I'm just not sure how much you can, how far you can get here without some sort of a rigorous evaluation to make sure your domain and sub-domain list is the right list or is the comprehensive list or is the list, and that's sort of, it almost seems to me to be where you need to start.
MS. BRADFORD: I think the group struggled with not wanting to get too granular with all the different components of an HMP knowing that if a patient was seen in an outpatient clinic for an ear infection they might not go through an exhaustive list, and if a patient was being admitted for surgery, so they didn't know how granular to get with the different sub-domains. I think they took this list from the HL7 CDA draft of what would be included in an HMP.
DR. COHN: Carol and then Steve.
DR. BICKFORD: Carol Bickford, American Nurses Association, I would invite you to revise your language of your definition, it should be history and physical performed by practitioners, others of us are doing these besides physicians. You also, I'm concerned that we're focusing on pathology and it doesn't accommodate wellness and our health promotion initiatives. For example you're doing a well child exam and you don't have any pathology piece so how does that positive spin to the history and physical get included. And another resource for you is ASTM, the standards that you addressed in the past one, the clinical record components, electronic health record has lots of already defined components that would be supportive of your activities.
MS. BRADFORD: Thank you.
DR. COHN: Steve.
DR. STEINDEL: Simon, at the risk of getting my CHI hat taken away from me entirely I'd like to ask the subcommittee to consider the idea of asking CHI to drop this as a domain from the point of view of like we were talking earlier about the text based reports, how many text based reports there are and how difficult it is to define what's involved in each one of those text based reports. I think we've heard several comments already about the difficulty of defining a history and physical and what goes into it, and this may be best left to a second stage of CHI when we do have some clinical domains better defined and terminology for those domains better defined and now we can start in the next stage looking at how to structure information, which history and physical is really a form of structuring the information.
DR. COHN: Comments from the rest of the subcommittee? Jeff?
MR. BLAIR: I think that was prudent advice.
DR. COHN: I guess my own concern and view on this one has to do with my own, that I'm somewhat ignorant of exactly all the timeframe aspects of the CHI activity. On one hand I understand that you wanted to do everything yesterday, or probably last month in reality. On the other hand I don't expect that we'll see every final report the same day, at least I hope. So I guess on the one hand I would say if stage two is a year from now I'd be sad that this would have to wait that long, on the other hand one might hope that it would come a month after the other one, which was the text based reports. Maybe you can clarify for me your timing for both the text based reports and the history and physical, when was your expected due dates on that?
MS. WARK: The schedule is for all of the workgroups to complete their initial analysis and recommendations by the end of this calendar year, and everyone is working very hard to try to finish that. Now as reflected in many of these reports the workgroup looks at the timeframe and the resources allocated and makes decisions to leave things out of scope that they don't think they can adequately cover within that timeframe. We have not to date had a whole domain put up for holding off until next year, so I'm not quite sure how to respond to that. I guess I would tend to go with some initial work that while we have the workgroup convened and they've started their discussions, and they've at least identified some of the structure and syntax alternatives and for example electronic signatures, I mean we'd have to have a discussion but why we wouldn't have them continue to see what they could at least uncover and what they are able to accomplish within the timeframe if it can serve to inform future work in those areas, they may come back and say we weren't able to come forward with a recommendation or something at this time, and maybe that's not what Steve was suggesting --
DR. STEINDEL: Cynthia, Jeff just made a comment which maybe the group misunderstood what I said, I was talking about history and physical and not text based reports. Text based reports is where electronic signature occurred and I believe text based reports should go forward, because text based reports I look at as the structure on which these more detailed reports like history and physical might be encompassed.
MS. WARK: I got it just the other way around.
DR. STEINDEL: And Jeff pointed that out to me and I want to clarify that, it's the history and physical because to me that feels like a lot of things that the text based report committee ruled out of scope, the detailed level text based report, like a report for radiology or a report for pathology, which I do not look at as much different then a history and physical.
MR. BLAIR: Now that Steve has made his clarification let me make mine on my comment and reaction of prudent. I think it was prudent from the standpoint, I think you wound up saying that you cancelled or stopped this and I certainly don't want you to stop the work on text based standards, I just felt as if the timeframe that it would take to be able to pull it together and make all the assessments of the things that are emerging out there because there's just a lot of work in process, if you could allow yourself the discretion of allowing some of these things to mature --
DR. STEINDEL: And considering the same workgroup is working on text based reports as history and physical it might be a better use of resources to focus them on the text based reports and you just make a note that history and physical is one of these other text based report type things that require sub-domain headings I think they called it in their words, section headings, etc., that need standardization and that might be best left to information theory models, information models.
DR. COHN: Gail?
MS. GRAHAM: I just, since both of these people work for me, I think that they would welcome the chance to spend, to do a more in depth analysis on text based reports and continue to in their deliberations there contribute to the other but have that domain kind of sitting in the background and being able to concentrate on text based reports.
MS. WARK: Well, I guess I'll have to say we can take that back for discussion.
DR. COHN: And I guess from my view I'm a lot less certain about these statements and putting this on hold, but I do I think as everyone has sort of commented think that this, the work needs to be informed by the work of the text based reports since that's I think fundamentally the first step here, so it once again gets down to are you still going, when do you close up shop on this first phase versus and when do you go to the next phase. So I think we'd have to defer to you on that.
MS. WARK: Technically our schedule is through the end of the calendar year for the first 24 domains to be covered and there's certainly much more work to go on in the future and the extent to which say something can't be accomplished within this timeframe and needs to wait for other processes to evolve or other work to be completed is something that we bring back to the Council.
DR. COHN: Great. Carol?
DR. BICKFORD: Carol Bickford, American Nurses Association. Let me be the devil's advocate for a moment and identify that the text based documents is sort of like structuring but the discussion about the history and physical pieces actually gets to some of the concept work and the assurance that we have the appropriate data elements in place, so it's consideration in the decision making.
DR. COHN: Any other comments on this stuff, I think you've been given about an hour of good advice hopefully. Any other comments about the preliminary before we go back to the letter on the final? Okay. Steve, did you want to run us through the recommendations and make sure that we were comfortable with the recommendations on the finals?
DR. STEINDEL: The recommendation letter itself, and I can go into the exact changes in the areas, of course if we remember our recommendation letter it consists of a standard opening and closing paragraph, and I'm going to pass over those. The actual recommendation section I now have worded as follows: The NCVHS has the following comments on the attached set of CHI domain area recommendations. The first one, the NCVHS concurs with the CHI recommendations for interventions and procedures, part B, laboratory test order names. We had no modification or anything on that section.
The next one was the NCVHS concurs with the CHI recommendations for medications as modified in the attached document. We further note the need for additional funding for the Food and Drug Administration to expedite the publication of a publicly available version of their unique ingredient identifier UNII codes and to provide continued funding to maintain the standard. That the dosage and administration sub-domains be investigated in the next phase of the CHI process. And the FDA NDC process be investigated and improvements identified be expeditiously pursued.
DR. COHN: Good.
DR. STEINDEL: Okay, the last one we have is the NCVHS concurs with the CHI recommendations for the immunization domain as modified in the attached document, and Cynthia is going to provide us with some wording modifications there. If you wish me to call back the medication domain and go over the exact modifications we're presenting they were actually all basically wording changes.
DR. COHN: Correct. Comments? Are you moving this letter?
DR. MCDONALD: I will move we accept the just dictated recommendations.
DR. COHN: Second?
DR. ZUBELDIA: Second.
DR. COHN: Okay, comments, questions, discussion? All in favor, aye --
SUBCOMMITTEE: Aye.
DR. COHN: Opposed? Abstentions? Okay, passed. Well, thank you, that's one item. Well believe it or not we're actually running on schedule at this point, Karen I think we now turn it over to you for a HIPAA report. Actually obviously Cynthia and Linda I want to thank you both very much, very good presentations. Obviously we'll be looking forward to seeing hopefully some additional discussion and final reports next week.
MS. WARK: Yes, we have three final reports that are scheduled for presentation next week, and thank you to the committee, it was very helpful.
DR. COHN: Thank you. Karen, welcome.
MS. TRUDEL: Thank you.
Agenda Item: HIPAA Update - Ms. Trudel
MS. TRUDEL: I'm going to do a very brief update on just a variety of HIPAA issues. Obviously the October 16th compliance data has come and gone and we're not hearing at this point any indications of widespread cash flow or operations problems, although I must admit that on the 16th my press office contacts did present me with a CD of what they called train wreck music for my enjoyment.
Just in terms of general readiness we are hearing that compliance is increasing, that people are testing, that they're successfully testing, they're getting into production, we're hearing that there's a sharper increase in the percentages of compliant claims. You can see that Medicare's last set of statistics, which was actually October 10th showed that I think we were at about 28 percent at that point and so my guess would be at this point we are somewhere above 30 percent, maybe up to a third of fully compliant claims in production. And we're hearing similar stories from Blue Cross/Blue Shield plans and others. I spoke to the state Medicaid directors this morning and one of the directors there shared with me that her testing numbers were almost exactly the same as ours, that it was roughly 33 percent of the non-pharmacy claims were coming in compliant, in production, no problems, and the pharmacy obviously were much higher.
Good news in terms of regulations, we have moved the provider ID final rule forward to OMB for its review, and the tentative publication date for that now is December. The Department is continuing to work very hard on an enforcement rule and I think Stephanie Kaminsky and I probably talked to you about this earlier, we are working on a substantive rule that is both substantive and procedural, it will augment the procedural rule that was published earlier in the spring, and it will cover all aspects of HIPAA Admin Simp, not just privacy, not just transactions and code sets, so obviously there are many issues on that and we will be doing a proposed rule and expect to get quite a bit of comment on it.
DR. FITZMAURICE: Do you know when that will be out? Is that also a December one or --
MS. TRUDEL: No, I think it's somewhat later then that because it's really a huge undertaking.
DR. MCDONALD: Just to clarify, the provider final rule, that means the data will be available, or we'll just know how it might be structured when it is available?
MS. TRUDEL: What it will say is what the identifier will be, how the enumeration will take place, it will address the issue of are providers going to have to pay for an identifier or not, again the kinds of data that will be in the file and what will and won't be in the file --
DR. MCDONALD: The database doesn't exist now.
MS. TRUDEL: And the ability for people to access and use the data, all of that will be discussed. No, the database will not be available and I believe that we will be adjusting the effective date of the regulation, or at leas the ultimate compliance date to take account of the fact that we want to give people a full two years after the database is available and enumeration can begin. So there would be some consideration of the fact that you won't be able to get an identifier on day one, so the clock won't start ticking on day one either.
DR. MCDONALD: I'm actually eager to get this database, I'm not trying to put it off, but I guess it sounds like there's really not a database yet and it's not that there's one there and you got to get it cleaned up, it's going to start being created after --
MS. TRUDEL: There's some system development has been underway for a while but unless, until we had final decisions from the Secretary and OMB as to the data elements and what the identifier will look like you really can't, you really can't go past a certain point.
DR. MCDONALD: And this is not an NPRM, this is a final rule.
MS. TRUDEL: This is a final rule. Okay, in terms of enforcement, we do have a system in place to accept complaints and we have received two complaints on transactions and code sets to date. That's I think about all I'm going to say about HIPAA in general. I want to talk a little bit about the issue of DSM-4 and ICD-9, which was an issue that had come before the subcommittee for some time, some time ago, and there had been a request for an OGC opinion on certain aspects of the issue, and we basically have managed to work this all out at the Department's level and I just want to give you an update.
We've developed a number of frequently asked questions that address the issues that were raised by the American Psychiatric Association, we've posted them on our website and the APA appears to find them, that answers their questions. We clarify that the diagnostic criteria, that there are no diagnostic criteria adopted under HIPAA, whether they're DSM or any other kind, and that indeed diagnostic criteria are outside the scope of HIPAA, which means that it's perfectly okay to use DSM-4 or indeed any other diagnostic criteria, and so we've specified that. We've also been very clear that while the ICD-9 codes need to go on the HIPAA transactions it's again quite acceptable to crosswalk between the short descriptors that are in ICD-9 and the short descriptors that are in DSM-4. We have further worked out with the APA that they will have a period of time to request changes to short descriptors in ICD-9 that appear to be especially outdated or inappropriate and they have indicated that they're going to take advantage of that.
DR. COHN: And that will be through the ICD-10 coding --
MS. TRUDEL: Nine.
DR. COHN: ICD-9-CM --
MS. TRUDEL: Right, through NCVHS.
DR. COHN: Coordination and Maintenance Committee, okay, great.
MS. TRUDEL: So those FAQ's are up on the CMS HIPAA website, which is www.cms.hhs.gov/hipaa/hipaa2.
DR. FITZMAURICE: Karen, does that mean that a person can use either the ICD-9 print descriptors or the APA's desired descriptors in assigning a diagnostic code?
MS. TRUDEL: What's really important is what is on the claim and what is on the claim on the transaction is the ICD-9 code, and as long as the person assigns the ICD-9 code with the understanding that it means what the ICD-9 short descriptor says it means, there's no reason that those descriptors in DSM descriptors or any other kind of short descriptors can't be cross mapped.
DR. FITZMAURICE: I think I understand, thank you.
MS. TRUDEL: Marjorie you can, do you want to --
MS. GREENBERG: It's the ICD-9-CM Coordination and Maintenance Committee, I think I called it the Coding and Maintenance Committee.
MS. TRUDEL: Thank you.
DR. COHN: Oh, is that it?
MS. TRUDEL: Yep, that's it. Any questions?
DR. COHN: Well actually want to thank you for helping assure the successful implementation of the administrative and financial transactions.
MS. TRUDEL: I don't think I can take credit for that but thanks anyway.
DR. COHN: Any questions or comments, other, yes?
DR. FITZMAURICE: Do we assume that things like the health plan identifier, the NPRM for that would come out sometime after then the National Provider Identifier?
MS. TRUDEL: Yes.
MR. ARGUS: A question for Karen on the NPI, once the NPI is fully implemented does that mean previous identifiers for providers will no longer be used?
MS. TRUDEL: I believe the way the implementation guides are structured the NPI would be used and other for instance UPINS, etc., would not on the transactions but again there's an understanding that for some period of time, maybe permanently, people are going to need to do behind the scenes cross mapping to existing identifiers.
MR. ARGUS: Right, and I assume that would be the two year period for which, between the final rule and the implementation.
MS. TRUDEL: Let's assume we're talking about the period post implementation so we have a compliant date and I will correct if it turns out that I'm wrong, but I believe that the implementation guide says use the NPI when the NPI goes into effect.
MR. ARGUS: So it will be exclusive at that point in time.
MR. BLAIR: From our experience with the financial administrative transactions we had of course a two year implementation period before compliance was required and we modified that slightly when we had ASCA because we wound up asking people to articulate what their testing plans were. But my understanding is that WEDI had asked us if we could create some type of coordinated plan for the industry as a whole, in other words for example to the, payers need to be in compliance first, or certain folks needs to be in compliance, something where it's segmented and it doesn't lead to one single date but we could create milestones for an orderly transition for the industry. Is there any thought being given to something like that with the Nation Provider Identifier so that, it may even create a little bit confidence if there's these milestone dates during the period.
MS. TRUDEL: I'd have to talk that over with general counsel but I don't believe that there's anything in the law that permits us to assign different compliance dates to different kinds of covered entities. I think that we can do it on a voluntary industry basis and I think that's something that the WEDI-SNIP group could have a big influence over. I think we learned this time that when you put out things like the WEDI-SNIP sequencing paper, the testing paper, with the testing levels, that people do pay attention to them and that indeed you can get while not 100 percent agreement, you can get enough of a consensus that things run fairly smoothly. I think we found that people who sequenced their transactions almost all did it the way WEDI suggested doing it. So I think your point's well taken, I'm not sure there's a legal way to do it but there's certainly a practical way to go about it.
DR. COHN: Clem and then Kepa.
DR. MCDONALD: Will the provider number include an institution as a provider? At least that was discussed in your recent discussion. As I'm a hospital so I get a number or is it just for persons?
MS. TRUDEL: No, it includes both institutional providers, organizations and individuals.
DR. MCDONALD: Was there a thought about a built in crosswalk of DEA numbers?
MS. TRUDEL: We've talked to DEA about that, I'm not sure that we've raised that discussion any time in the recent past but I'll raise it with the folks who are actually doing the, building the system and creating some of the crosswalking.
DR. MCDONALD: It'd be probably the most useful loading file, some of the other, UPIN wasn't so good.
DR. ZUBELDIA: Jeff, concerning the transition to the NPI, implementation guides require the use of the NPI as the primary identifier if it's available. But they do no preclude from using the other secondary identifiers, so I would easily see how the providers that are using the NPI as primary identifier will continue moving onto a secondary identifier what they're currently using as primary and continue providing the tax ID and the UPIN or whatever other identifiers they need to provide as secondary identifiers because they can provide along with a primary NPI a variety of secondary identifiers with the transactions. So it provides for a very easy transition, but I agree with Karen that WEDI can make recommendations in that area that will probably followed by most of the industry.
MR. BLAIR: I guess one of the things that, and I'm not sure if this is correct, but the financial administrative transactions, that was a massive change for the health industry, there were so many different parts of it and so many different points of coordination, but I think if the National Provider Identifier and it's one identifier and it seems as if there would be a potential, if there's something we could do to make it easy to move to it where the industry might be able to move to it earlier in the two year cycle rather then later, and I was just trying to see if we could do something. Is my perception correct? We really should be able to move to an NPI faster and there's probably less resistance to making a transition, is that not true?
DR. MCDONALD: Well there's going to problems with the file allocations in a lot of systems probably --
DR. ZUBELDIA: It's a very flexible mechanism for the transition. If the providers start sending the NPI as soon as they get it, implementation guide also requires to send a tax ID as a secondary identifier when you send the NPI as primary. And a payer that is expecting the tax ID would just have to look for the tax ID in the secondary identifier rather then primary identifier element. And it provides for a flexible transition where a provider by sending both can actually accommodate both kinds of payers, the ones who are ready to receive the NIP and the ones that are not. It's much more flexible then what we saw with the transaction transition.
DR. FITZMAURICE: Making the assumption that the NPI would come out on December 31st, Karen do you have any estimate of when enumeration would begin?
MS. TRUDEL: No, I don't.
DR. COHN: Other questions, comments?
DR. ZUBELDIA: Karen, there was an expectation set about a month ago about a potential for a modification to the transaction rules once a year and that perhaps it would be an NPRM in the spring or sometime in early 2004. Do you have a date on that?
MS. TRUDEL: We are working on a first modification, what we do not have at this point is a real good sense of what the content of it will be. We know that we have a few things that we need to raise because the Secretary's Regulatory Reform Committee has asked us to clarify some definitions and clarify some issues that relate to direct data entry, so we'll definitely be addressing those. What else is going to be in the regulation is still being discussed within the Department at this time.
DR. COHN: Other questions or comments?
DR. ZUBELDIA: I have one more. I know there is probably no change but is there any update on the individual identifier?
MS. TRUDEL: As I understand it the same language that instructs us not to allocate funds to develop an identifier is being and has been provided with our appropriation each year.
DR. FITZMAURICE: I guess we'll find out when we get our appropriations.
DR. COHN: When the bill comes through. Now what the agenda provides now is a 15 minute break and we'll spend some time when we get back from the break hopefully talking if there's any implementation issues that we need to be aware of or begin to keep on the radar at this point and then what we'll is to move into our ICD-10 discussion. So why don't we take about a 15 minute break, we'll get back together at 3:00. Thank you.
[Brief break.]
Agenda Item: Discussion of HIPAA Implementation Issues - Dr. Cohn
DR. COHN: Okay, this session we're going to basically start out for just a minute or two and just sort of ask if there are HIPAA implementation issues that we need to be aware of, put on the radar, be following obviously now and into later December and into January. Obviously Karen I think did a very nice job giving us a current status of the implementation. Do people have any comments or anything that we should be watching as we move forward? Kepa?
DR. ZUBELDIA: There is an issue that is on my radar screen and a lot of other radar screens and I don't know if NCVHS can do anything about it other then perhaps find out more, and that's the coding system for outpatient procedures in an institutional setting. That's a very big issue and HIPAA adopted the ICD-9 for inpatient procedures and not for outpatient procedures and that's a big issue for the industry right now. I don't know, maybe Karen has some ideas of what can be done about it, other then for us to hear about it and your suggestions.
DR. COHN: Well, Kepa, I guess I'm surprised because I had understood always that, and Marjorie help me with this one, but that normally Volume 3 of ICD is used for inpatient procedures but that outpatient procedures, hospital outpatient procedures as well as ambulatory visits are typically CPT and HICPICs, and then DAPC is based on that. Karen, please.
MS. TRUDEL: Okay, we're still doing some research to find out the extent of the issue but what it appears to be is that at least some health plans are requiring the use of the ICD at the claim level, not the line level but at the claim level, to explain a, if the, I may not have this right, but as a further explanation of what is happening to the patient to back-up CPT charges.
DR. COHN: Now are you talking about procedures?
MS. TRUDEL: Yes, ICD procedures at the claim level, CPT procedures at the line level.
MS. GRAHAM: Actually if I could, this is Gail from the VA. We have certain carriers that require us on the claim to crosswalk CPT procedure codes to ICD-9 out of Volume 3 procedure codes. This is specific carriers and we had actually hoped that the HIPAA implementation date would resolve the issue because we're having to manually pull these and put the ICD-9 codes on them because we code all the outpatients in CPT.
MR. ARGUS: And this is one area where we've heard from some of our hospitals, and some of it came about because the health plan is requiring the provider to report ICD-9 --
MR. BLAIR: Who's speaking?
MR. ARGUS: George Argus of the American Hospital Association. Some of the health plans are requiring providers to report ICD-9-CM procedure codes at the claim level. But there's another important point to keep in mind here, too, that many providers, hospital providers will still code internally the ICD-9 procedure code for their own purposes, and one area that you will lose functionality if you don't allow I guess some of this to continue is the ability to assign a principle procedure code to a diagnosis. If at the service line, which is HICPICs, you report the HICPIC code you don't have the ability to identify the principle procedure that coincides with the principal diagnosis. And many providers like to have the ability to compare similar inpatient services against outpatient services by using ICD-9-CM at the claim level, the procedure codes, as a means of looking at the two environments as a means of comparison. Now hospitals will probably continue to do this for their own internal purposes but I do think that precluding that does take away a functionality, that's something to consider.
MS. TRUDEL: We're still looking into exactly what the business needs are for the data and I don't feel as though I've got a good handle on that right yet, and it's interesting this came up in one of our roundtable calls, the one before last, and the person who raised it was very up front about the fact that this was just something that everybody missed, it was not that the regulation was unclear, the regulation was quite clear but the people who were using it at the claim level never thought that oh, I'm going to have to stop doing this, so it's just something that we missed it.
DR. MCDONALD: If I could just get a clarification of what we're really saying, we're saying the diagnostic code ICD-9-CM had to be added --
DR. COHN: Procedure.
DR. MCDONALD: In addition to, and that's the hospitals do it too sometimes, or was that the procedure code so that they have some crosswalk?
MR. ARGUS: Well hospitals do it for their own internal purposes as part of their own means of comparing the inpatient environment and the outpatient environment, so they'll basically develop a procedure code using an ICD-9-CM procedure code and give it a principle procedure code classification to coincide with the principle diagnosis. Simply listing the HICPIC code at the service line does not give you the principle procedure, that's something that will be lost in the transaction.
DR. COHN: I suspect that this, I suspect that there's a fair amount of variation, in fact I actually have never heard of this prior to George having brought it up or Kepa you bringing it up, but I think it's something we probably need to understand a little better and obviously Karen it sounds like you're endeavoring to understand the issue better as well as see what the frequency in the industry is.
DR. ZUBELDIA: And I think that's where NCVHS can hold some hearings and gather the information necessary to make a decision on it or to help CMS in their enforcement or whatever regulation changes, whatever needs to happen, I don't know what needs to happen, but at least we need to identify the extent of the problem.
MR. BLAIR: Is that an issue that also needs to be considered within an ICD-10 NPRM?
DR. ZUBELDIA: No, I don't think so.
DR. COHN: This seems like a very, my sense is this seems like a very different issue, interesting. Have you heard anything else, Kepa?
DR. ZUBELDIA: Sure.
DR. COHN: Any other problems?
DR. ZUBELDIA: Anesthesia, the coding of anesthesia either with minutes or units or both. There is no consistency among the payers on how they want the anesthesia coded and there are some payers that insist on having minutes, some payers that insist on having units, and some payers that will take both, some payers that will want both. And this makes coordination of benefits of very difficult if not impossible in some cases.
DR. COHN: Anything else?
DR. MCDONALD: We're just general talk or is this specific --
DR. COHN: Well, I don't have a solution to any of these things, I was just trying to get a sense of what sort of issues as HIPAA gets implemented we're seeing to know whether or not we need to jump in and do something immediately, whether there's something we need to be following in December, something even more urgent then that. I'm not hearing anything that, I mean some of these things, the anesthesia or something that's been an irritation for the industry for years and I think it's really a discussion to have with the DSMOs about the nature of their standards, since really one would be expecting that the standards should be specifying what goes into that and then the question is is the industry satisfied with it.
DR. MCDONALD: Don't you think it's too early? I mean no one gets paid in two days, it's only been 16 days, so if there's a problem it won't emerge quite right away.
DR. COHN: Well, sure, I mean this is not the final discussion on HIPAA, you mean you're talking about HIPAA --
DR. MCDONALD: I thought that's what --
DR. COHN: Yes, it is, I agree, it's really more a question of are there issues that are already arising, and actually I'm not hearing anything that's earthshaking --
DR. MCDONALD: A lot of places have 60 day delays in payments so we may not even, so I don't know where those delays are but there might be, the rule may not have hit.
DR. FITZMAURICE: I wonder if we're seeing any rise in the use of companion guides rather then having the health plans who want to put out a companion guide to allocate data elements or maybe additional data elements, rather then having them going to X-12 and bargaining for changes in the implementation guides, is that happening or is it again too soon to tell?
DR. ZUBELDIA: I can tell you our experience at Claredi with companion guides, we have corrected over 300 different companion guides and implemented edits for over 160 so far. Some of the requirements in the companion guides are clearly against the implementation guides, the HIPAA implementation guides, that doesn't seem to bother anybody, they just put the requirement out and everybody else has to comply with it. Most of the companion guides are not, most of the companion guides are pretty clean, there are a few with odd requirements but most of them are pretty clean. But there's clearly a very wide spectrum of requirements in the companion guides, I mean you'll get companion guides that are three pages that have very simple requirements for the communications and that's it. You get companion guides that are over 200 pages with very specific requirements. One of the frustrations that we found is that the companion guides in general are incomplete, they tell you about 30 to 40 percent of what the payers requirements are and then the payers will reject transactions for requirements that are not in the companion guide and that's disturbing as far as interoperability is concerned because it's very difficult to predict where the target is going to be. And they specify a set of requirements and then they have multiple additional requirements that were not specified or companion guides that change, and there's already been several iterations of certain companion guides. So that's kind of friction points. I would say that none of that is much different from what happened before and none of the precluded transactions from being paid before, so that's not preventing the transactions but it's also reducing the advantages of going to the HIPAA transactions because you continue carrying the same friction that you had before with the new transactions.
DR. COHN: Other comments about all this stuff? I mean I don't, there's not an action that we need necessarily to do about any of this in the moment, it was really more that it seemed appropriate given that we are ten days after the implementation just to look at, see if there was any major issues that we need to jump in on, and I'm certainly, I mean I think these are, Kepa I think has brought up some issues which I think are things that we should look at but probably don't have quite the urgency that we have to do them in the next month to six weeks or anything. Clem?
DR. MCDONALD: Well, I think maybe, except I wanted it earlier, or maybe bad luck, but I think this has come up, it looks like it's come off a lot better then many of us might have hoped for and I think we need to credit sort of wise government people who were responsive to issues and maybe more flexible if that's the case. I wasn't going to be surprised to hear a disaster on the 16th, but we're not done yet, but I'd kind of like to make I guess an applause for the federal agencies and groups and people who so far have kept this, looks like it's working.
DR. COHN: Hear, hear. I think what we will try to do is to have the ongoing conversation and questions being asked as we move forward into future hearings and certainly, I don't know that we have time in November knowing how tight that schedule is for our break-out but certainly in December we'll try to use, have a couple of minutes to sort of reflect, I think, Clem you were probably right when you said we're only ten days into it and that there's going to be lots of other things we discover. And certainly by January I'm expecting that we'll have a meeting with the DSMOs and others to begin to sort of talk about where we are knowing that at this point some of those implementation guides are getting a little long on the tooth and how in the heck do we move forward from here.
DR. ZUBELDIA: What process are we going to have to gather this kind of feedback? For instance, I've been getting emails from people that want to come and testify to the NCVHS, and I've sent you what I got. Is there going to be a formal process for people to tell us their complaints or their successes? Because we could find both, successes and failures, we just need to be balanced in doing this.
DR. MCDONALD: It almost sounds like a suggestion you're making that we ought to have some testimony because it would give an opportunity to recorrect if there are problems, but also could give an opportunity to brag or feel good about it and maybe get more support for further standardization if we good success stories.
DR. ZUBELDIA: I've heard from a substantially large payer that their auto adjudication rate on HIPAA claims has gone tremendously high where it was in the low 60's before HIPAA, and with the HIPAA transactions is in the high 80's, and that's a tremendous impact on the bottom line for those payers. I mean there's real successes out there and they're starting to be measurable right now.
DR. HUFF: Is it the time and is it our purview to ask what should happen next as well or is that WEDI and those guys looking at, I mean within our own institution they sort of held off on, they said we want to get this working and get it done and in part of doing they said now I see some new opportunities that maybe ought to be addressed next. Is that us or is that WEDI or somebody else that should come up with sort of --
DR. COHN: Well, actually it's partly us, I mean first of all, and Kepa I agree, I think that certainly from my view one of the things we're looking at next year is I had described it as lessons learned from HIPAA but it's also successes and failures and lessons learned from HIPAA. Certainly there is the intent, we are obviously supposed to be working with the DSMOs to identify changes, extensions, and everything else to the standards and I think we have a responsibility to do that and basically try to prepare on the early basis recommendations for sort of where to go next. Now as I say I don't think it's necessarily our job to without their input try to discover which direction to go but certainly working with them I think there is the intent to expand this. Now as I say that we are only ten days into the implementation so I think some of the industry might take some umbrage with us already talking about expanding what's going on but I think there's time next year for us to begin to have some of that conversation. Does that sound like a reasonable piece and we should have it on our sort of to do list as we move into next year?
DR. ZUBELDIA: Are we going to have a process for people that want to tell us about their success or failure or should we just an informal process and put together something? I'm concerned that we need to, that whatever the process is if there's one it has to be balanced, because this could end up being just HIPAA grip sessions, and that probably would not be appropriate.
DR. COHN: Agree. George Argus?
MR. ARGUS: This is George Argus again from the AHA. I guess what I'd like to see and what I've heard from some providers is while the focus has been mostly on claims many providers still are looking to get the bang for the buck on some of the other transactions, the eligibility, claims status, the inquiry and response, and even to some extent maybe remittance. So I guess I would be curious to find out really the progress of each of the transaction standards as far as utilization by the industry as a whole, that to me I think would be important to find out how far along everyone is with the other transactions.
DR. COHN: Good suggestion. Other comments at this point? I don't know that we need, to my view this is probably either a January or maybe even a March/April discussion or maybe beginning to do both, so it doesn't mean we have to decide this tomorrow but I also just wanted to make sure we didn't have any emergent issues. Steve?
DR. STEINDEL: Kepa brings up a very good point, there's a lot of information that's being passed around and it has to be classified right now as anecdotal information and I'm just aware that the plural of antidote is antidotes, not data, and I think we need to find a way to find data on this, not just what's happening in one place but is this a universal, and getting data is much harder. Maybe that's something we should spend some time all flying talking about with CMS and our colleagues in the industry and if we're focusing some hearings on this issue seeing if we can get some data in the area.
DR. COHN: Marjorie?
MS. GREENBERG: Well, that certainly is what's required in the annual reports to Congress, I mean we've been now for a number of years reporting on sort of progress, now it's going to be time to start reporting on what happened with the standards adoption. So I know there's been some discussions in the past about using other groups who routinely collect this information but I would agree with Steve, that I think we need to be thinking about if there's some way to gather systematic information. I don't have a solution.
MR. BLAIR: Can I piggyback on the idea? There's two areas of information that I think would really be nice for us to have, especially in our reports to Congress, is if there's some way for us to find out what the transition cost was for providers and payers and clearinghouses, the covered entities, as well as the perceived or measured savings during the first three months, six months, or year after October 16th, it might be very nice to wind up being able to show a return on the investment for HIPAA. Maybe that would have to be a year away but that would be nice data to have if we could get it.
DR. COHN: Well certainly the costs and benefits is I think a very important thing, I've not been involved in many implementations where within three months of implementation we're seeing benefit, so you're probably right we need to take a slightly longer look at all of that, but I mean very good point. I know, I'm sure CMS is also interested in those issues and I think we can sort of try to, we'll be continuing to discuss this and trying to come up with some plans for early 2004 to get into this deeper.
I'd ask all of you if there are sort of a more emergent issues that are coming forward that we need to somehow get involved with, and obviously my hope is that CMS will take care of most of those emergent issues but if there are things we obviously want to make sure that we're providing time and this has been something that we have been sort of moving along and want to stay pretty close to. Any other comments for the moment or should we move into our next agenda item?
DR. COHN: Okay, well this begins our conversation on ICD-10, what we're going to do this afternoon and I've asked Donna Pickett to sort of help us go through the information that we have. The first part is to sort or review where we have been, both the timeline as well as comments from many in the industry knowing that as I think as we looked at this we realized that we have received a tremendous amount of input from the industry and many times we sort of forget about it because it's been over such a, it's been a relatively long time period, so I wanted Donna to sort of go through our compilation of information and then talk for a couple of minutes about the summary of implementation issues that have been identified through much of the testimony.
Now obviously tomorrow, after that we'll spend a little time just talking about things and I especially want to get Clem McDonald's thoughts since he won't be here tomorrow. But then tomorrow morning we move into sort of testimony on the topic and then hopefully tomorrow afternoon we will have an opportunity to sort of talk about next steps, consider the nature of a letter that we may want to come forward with if it's appropriate and other issues from there. Anybody have an better ideas on how to spend the next day and a half? Okay, with that Donna do you want to sort of walk us through some of what we have here and sort of reacquaint us with where we've been the last couple of years.
DR. PICKET: Based on the comments that we've heard at previous meetings and for staff I think most notably the August subcommittee meetings and hearings and the September full committee meetings we needed to kind of bring things together. There's been a lot of information that's been presented and discussed over the last 12 months and for a lot of the deliberations regarding ICD-10-PCS and CM, some of this information actually dates back a lot farther then just the hearings that we've had within the last year. What we've attempted to do is provide information in discreet segments so that if you're interested specifically in the development of ICD-10-CM and ICD-10-PCS we've created a unique document that has that timeline. Some of this may seem familiar to you because in previous meetings we've had a timeline that actually incorporated not only the development of 10-CM and 10-PCS but it also included the national committee meetings, hearings and letters that have been sent to either the full committee or the subcommittee or to the Secretary, and what we've tried to do is separate those things out to make it easier to identify which things happened when and how things relate.
So the first document, development of ICD-10-CM and ICD-10-PCS, lays out the history of the development of each of those code sets. I won't walk everyone through this specifically because it's been seen so many times that it would be sort of not a good use of our time to go through it once again, and there really haven't been any changes to those timelines relative to the development of PCS and 10-CM, except for the additional information gained from AHIMA's presentation, the joint presentation rather of AHA and AHIMA about their pilot testing, so that information has been incorporated as well.
DR. COHN: Which document are you referring to now?
MS. PICKETT: That's the AHIMA/AHA information is actually included in the NCVHS timeline. The first document on the development of 10-CM/PCS again just contains the complete history of the development of each of the respective classifications.
DR. COHN: Okay, and I guess I'm confused, were you going to walk us through this or were you saying that you didn't need to walk us through this?
MS. PICKETT: Well I can if that's what everyone would like but given that this document hasn't changed that much we can through this in a very summary fashion just to cover everything.
The first page of the document basically provides an overview of where we are with ICD-9-CM, the fact that it was implemented in 1979 and it's been in use obviously for more then 20 years. It also provides a basic history of how ICD-9-CM came to be, that the diagnosis portion is a clinical modification of the WHO version of ICD-9 and that procedures classification was also developed as part of ICD-9-CM. And then we also reference the fact that CPT was developed and is maintained by the AMA and is used for the coding of professional services on claims, on physician claims and also in other non-inpatient settings. So again, it just gives the history of the classification and the fact that there was an ICD-9-CM coordination and maintenance update process for trying to keep ICD-9-CM clinically relevant to our current health care environment, however it's reached a point where there are various parts of the classification that cannot be updated because of its limited space and that information is again referenced in the background information on ICD-9-CM.
What follows are two separate timelines on the development of each of the respective classifications. The first timeline deals with ICD-10-CM and discusses how ICD as published by the World Health Organization is in its tenth revision, and that the U.S. now is using ICD-10 for mortality information. But it also goes into a lot more background about when WHO finalized ICD-10 and also when it was first released, which was in 1993. In 1994 NCHS awarded a contract to evaluate ICD-10 for morbidity purposes within the United States and a prototype of ICD-10-CM was developed following a thorough evaluation of ICD-10 by a technical advisory panel that consisted of private and public sector stakeholders. Among the membership of that technical advisory panel was the AHIMA, we had a hospital representative, what was then called HCFA, AHCPR, and some other hospital representatives. The TAP concluded there were compelling reasons for recommending an improved clinical modification of ICD-10 and that it would overcome most limitations of ICD-10 for morbidity applications. The TAP also strongly recommended that NCHS proceed with implementation of a revised version of ICD-10-CM.
Between 1995 and 1996 further work on enhancements to ICD-10-CM were undertaken by NCHS staff, including a thorough review of ICD-10-CM Coordination and Maintenance Committee proposals for modifications that could not be incorporated into ICD-9-CM. We also worked with a number of specialty groups and among those groups were the American Association of Dermatology, American Academy of Neurology, American Association of Oral and Maxillofacial Surgeons, the American Academy of Orthopedic Surgeons, and a number of other groups.
In 1997 a draft of the tabular list of ICD-10-CM and a preliminary crosswalk were posted on the NCHS website for public comment. Comments received during the three month period between December 1997 and 1998, those proposals were received in writing. At the conclusion of the open comment period the draft version of ICD-10-CM and the draft crosswalk were removed from the NCHS website. More then 1200 comments were received from 22 individuals and organizations representing a variety of groups including government agencies, two research institutions, some system developers, and other professional organizations. The comments ranged from general observations to very specific and detailed analyses. Of the comments that were received and analyzed 480 were rejected, 238 required no response or necessitated no further action, 180 merited direct incorporation into ICD-10-CM, and 268 were included as recommended. 77 were also found to have merit but required further review.
In 1999 ICD-10 was implemented in the U.S. for mortality reporting, final death statistics and leading causes of death for data years 1999 and 2000, using ICD-10 had been published or available on the NCHS website. An overview of the comments received during the ICD-10-CM comment period were also posted on the NCHS website in 1999 and a summary of those comments were also presented at the ICD-9-CM Coordination and Maintenance Committee and posted on the NCHS website.
Between 2000, 20001 development of ICD-10-CM continued with changes being made in response to the open comment period as well as additional input from physician specialty groups. May 29, 2002 NCHS posted on its website a pre-released version of ICD-10-CM. Consistent with ICD-10 the classification is alphanumeric, the codes are either a five or six digit alphanumeric. Descriptions are consistent with current medical terminology and findings. Some of the modifications features in ICD-10-CM have already been incorporated into clinical modifications that are in use in other countries, most notably the Australian modification ICD-10-AM, and the Canadian modification ICD-10-CA. NCHS will make available tools to facilitate implementation of ICD-10-CM and that would include database versions, crosswalk, educational materials, official coding guidelines, etc.
In 2003 we posted an updated pre-released version of ICD-10-CM, those files are still available on the NCHS website. August of this year AHIMA and the American Hospital Association, AHA, jointly conducted a pilot test of ICD-10-CM during June and July. The study involved dual coding of records in ICD-10-CM. More then 6100 records from a broad cross section of the health care community were dual coded by approximately 180 participants. The initial results indicate that there's general support for adoption of ICD-10-CM. ICD-10-CM is seen as an improvement over ICD-9-CM and that ICD-10-CM is more applicable to non-hospital settings then is ICD-9-CM.
The next portion of the timeline deals with the development of ICD-10-PCS. The Centers for Medicare and Medicaid Services, formerly Health Care Financing Administration, began developing a replacement for ICD-9-CM Volume 3 in 1990 in order to address a number of limitations with the current procedure classification. There are only ten codes available within each category, once a category is full one must either combine types of procedures under one code or find room in another section of the code book to place a new code. The benefit of such a system is that one could easily collapse the codes into categories when doing research to capture a wide range of similar procedures. When one puts a similar code in a separate part of the book coders and researchers may not easily find, there can be gross errors when trying to identify particular types of cases or make projections of numbers of procedures performed when codes are not carefully placed within a system such as the current ICD-9-CM. The background information goes on to talk about the generic problems of ICD-9-CM and that the updating of the system has been quite difficult and involves making compromises.
The development of ICD-10-PCS adhered to the criteria established by the national committee for procedure classification, and that is referenced in appendix one, it's all the criteria that had been identified. ICD-10-PCS is made up of seven alphanumeric characters, it should be noted that ICD-10-PCS is not based on an international classification because currently none exist. The system provides exponentially greater code capacity where all substantially different procedures can have a unique code, whereas ICD-9-CM procedure classification contains fewer then 4,000 codes, the current draft of ICD-10-PCS contains approximately 197,000 codes. There is also significant room for easy expansion as technologies are developed. The substantial increase in codes means that a coder can quite easily find the appropriate code that describes a particular procedure rather then having to use a less specific or clinically detailed ICD-9-CM Volume 3 code.
ICD-10-PCS was developed using an open process. A technical advisory panel provided review and comments throughout the development. The technical advisory panel included representatives of AHA, CPRI, ANSI, HIS, AHIMA, AMA, the American College of Surgeons, a representative from the managed care industry, medical informatics, American Association of Medical Transcriptionists, and several federal agencies.
And that kind of concludes the background for the timeline of the development of ICD-10-CM and ICD-10-PCS.
The next document is the NCVHS timeline and provides a chronicle of the long history that the NCVHS has had in relation to patient classification systems. Classification systems play an important role in nearly all uses of health data including outcomes research, reimbursement, program evaluation and policy decision making. The timeline details the highlights and major milestones of proceedings of the national committee and its subcommittees, especially the Subcommittee on Standards and Security. These proceedings specifically relate to the continued discussions to move forward with implementation of the clinical modification of ICD-10-CM to be used for morbidity diagnosis classification in the U.S. and the ICD-10 procedure coding system, ICD-10-PCS, to be used for inpatient procedure classification. In addition the timeline includes items occurring out of the realm of the national committee, external reports, letters, and other external action taken related to the need to implement these two classifications.
Testimony obtained during the April 1997 hearings on the initial candidates for adoption, the medical code sets under HIPAA acknowledge the need to move to newer versions of the ICD-9-CM at a later time but recommended that the initial HIPAA code sets be those that were currently in use. Comments received in response to the NPRM for standards for electronic transactions published May 1998 supported the need to replace ICD-9-CM with ICD-10-CM and ICD-10-PCS. The United States implemented the tenth revision of ICD-10 for coding mortality data in 1999, and ICD-10, or clinical modifications of it, are already in use for morbidity reporting in most other G7 countries.
The timeline starts with 1990. The Subcommittee on Medical Classification Systems of the national committee reviewed chapter proposals from the World Health Organization and preliminary international implementation plans for ICD-10. NCVHS noted at that time that while ICD-9-CM was responsive to the changing technologies and identifying new diseases that impact heavily on the health care community, that there was concern that the ICD classification system may be stressed to a point where the quality of the system may soon be compromised. This was noted in the NCVHS 1990 Annual Report.
The November 1990 report of the national committee concerning issues relating to the coding and classification systems included the following information. The subcommittee review found structural problems with both CPT-4 and Volume 3 of ICD-9-CM. An ongoing study and evaluation of the feasibility of a uniform procedure code was felt to be necessary and that such an evaluation should address HCFA's responsibility as a catalyst in determining the efficacy of a single procedure code set.
1993 the subcommittee continued to address issues related to ICD-10 focusing on the status and implementation of ICD-10 in the United States with particular regard to morbidity applications. The subcommittee initiated a letter from the NCVHS to the assistant secretary for health, an administrator of the Health Care Financing Administration, recommending that the Department dedicate resources to determine the feasibility of implementing ICD-10 for morbidity applications in the United States.
During 1993 the subcommittee held three meetings and three working sessions dedicating a substantial portions of the meetings to developing and reviewing its report to recommend that steps be taken to create a single procedure classification system for multiple purposes in the United States. The NCVHS issued a report on the desirability of a single procedure coding system containing the following conclusions. The single procedure classification system should possess utility as a statistical classification and as an administrative tools, characteristics of such a system are defined. There was general resistance it is noted to altering existing systems except where changes are considered necessary to reflect current medical trends. Current systems are badly in need of overall and consolidation. Pressures for change derive not only from end users who must contend with deficiencies of current systems but also from political forces that must address major health care reform. The committee notes that data sets currently do not permit the ability to track patients through the system as they enter and leave various care settings over the course of an illness or over a long period of time.
The committee realizes that recognition of the necessity for the development and implementation of a single procedure classification system is only the first step in a difficult and time consuming process. Public and private sector resources will be required to achieve a successfully and timely solution to the issues addressed in the report. The background material goes on to cite commonly noted flaws of ICD-9-CM and CPT-4 and noted that both classifications suffered from a lack of space for expansion, overlapping and duplicative codes, inconsistent and non-current use of terminology, and lack of codes for preventive services.
For ICD-9-CM Volume 3 specifically it was noted that there was insufficient specificity and detail and insufficient structure to capture new technology. For CPT-4 it was noted that it was a non-hierarchical structure, that it was physician service oriented and not multidisciplinary oriented, and that there were poorly non-discreet coding categories with variable detail.
In May of 1994 the Department, OIG issues a report on coding of physician services that describe vulnerabilities in the maintenance, use, and management of CPT-4 as they related to Medicare reimbursement. The report identified several flaws in the CPT-4 codes, guidelines, and index that can lead to improper coding. The report also noted that the guidelines on hospital outpatient services appeared to be a particular problem.
In November 1988 HCFA informed the CPT editorial panel of its concern in applying CPT-4 to outpatient services. In a December 1992 position statement AHIMA stated attempts to effectively use CPT-4 for the hospital setting have resulted in the inconsistent application of CPT conventions and general guidelines. The report also contained several recommendations for HCFA and the AMA that would correct the deficiencies noted. Prior to publication HCFA and AMA reviewed the draft report and their respective comments were included.
In 1995 it was noted in the proceedings of the 45th Anniversary Symposium of the national committee that in 1983 and again in 1986 that the NCVHS had called for strong efforts to develop a single procedure coding system for the U.S. to replace Volume 3 of ICD-9-CM in hospitals and the AMA CPT-4 in ambulatory settings, and that was noted in the Annual Report dated 1983, 1985, and again in the Annual Report dated 1993.
In May 1995 the subcommittee convened hearings whose primary purpose was to discuss data needs of managed care organizations using the proposed criteria for a unified procedure classification. Four different models of managed care participated in the discussions. There appeared to be consensus for a unified system given the varied needs of managed care organizations. The consistent message in written and oral testimony before the subcommittee was that existing coding systems were not meeting their needs.
In April of 1997 the subcommittee convened hearings on initial candidates for adoption of medical code sets under HIPAA. Approximately 30 organizations provided testimony. Testimony obtained during the April '97 hearings acknowledged the need to move to newer versions of ICD-9-CM at a later time but recommended that the initial HIPAA code sets be those currently in use.
In 1998 comments received in response to the Notice for Proposed Rule Making for standards for electronic transactions published in May 1998 supported the need to replace ICD-9-CM with ICD-10-CM and ICD-10-PCS.
In 1999 the Institute of Medicine's Committee on Injury Prevention and Control recommended a high priority be directed at ensuring uniform and reliable coding of both the external cause and the nature of injury using the ICD on all health data systems, particularly on hospital and emergency department discharge records. The committee noted that special efforts should be directed at training to ensure optimal use of the tenth revision of ICD.
May 17, 2001, a public meeting on whether ICD-10-PCS should be named as a replacement for Volume 3 of ICD-9-CM, was held at the ICD-9-CM Coordination and Maintenance Committee. 11 organizations, the AHA, AHIMA, AMA, the Federation of American Hospitals, the American Speech Language Hearing Association, NGENEX(?), AdvaMed, the Princeton Reimbursement Group, Medical Technology Partners, McKesson, HBOC, and DRG Review, requested the opportunity to present comments on whether ICD-10-PCS should replace ICD-9-CM Volume 3. It noted timing and resources were of concern to presenters. Of the 11 organizations presenting only one, the American Medical Association, did not support going forward with the process that would lead toward the replacement of Volume 3 with ICD-10-PCS.
In September 2001 the Benefits Improvement and Protections Act of 2000, BIPA, Congress addressed requirements for incorporation of new medical services and technologies into the Medicare inpatient prospective payment system. Some of the requirements addressed the lack of detail and shortage of available coders in the current coding system, current codes in the current coding system. In the September 7th, 2001 Federal Register CMS noted the limitation of ICD-9-CM regarding the ability to expeditiously incorporate new medical services and technologies into the classification. A number of approaches and techniques used for procedures such as lasers, minimally invasive techniques and the use of scopes cannot be readily captured by the current structure of Volume 3. Short term solutions undertaken by CMS to address limitations to Volume 3 included opening a series of codes that can be used for new procedures and technologies. This series of 200 codes provides some additional expansion, however, it should be noted that this approach captures a diverse group of procedures potentially effecting all body systems and that the approach taken was inconsistent with the structure of ICD-9-CM.
April 9th, 2002, CMS provided an overview and a need for ICD-10-PCS to the Subcommittee on Standards and Security. The subcommittee heard testimony from six organizations, the AHA, ADVAMED, Blue Cross/Blue Shield Association, the AMA, AHIMA, and the Federation of American Hospitals. Blue Cross/Blue Shield Association stated that the NCVHS should thoroughly evaluate the impact of migration to ICD-10 on all aspects of the industry by assembling a multidisciplinary team to provide input.
May 29th, 2002, several organizations provided testimony to the national committee regarding migration to ICD-10-CM, among those presenting included the American Academy of Neurology, the American Academy of Obstetricians and Gynecologists, the American Psychiatric Association, McKesson, HPOC, John Hopkins Bloomberg School of Public Health, NACRE(?), NCHS, the American Academy of Procedural Coders, were among some of the testifiers. With the exception of Blue Cross/Blue Shield Association testifiers supported migration to ICD-10-CM from ICD-9-CM. Blue Cross Association submitted written testimony urging the subcommittee to wait until the industry had successfully implemented the initial HIPAA standards and that the national committee thoroughly evaluate the impact of migration on all aspects of the industry by assembling a multidisciplinary team to provide input before making a recommendation to the Secretary.
A second panel convened on that day comprised of IT representatives from three vendors and two health care systems provided additional testimony regarding systems issues related to the transition to new code sets. Several of the presenters have had experience in migration to ICD-10 in other countries. The view was that migration was a challenge but could be accomplished as long as there was sufficient lead time, two to three years, to make systems changes.
In August 2002 the GAO issued a report, the HIPAA standards dual code sets are acceptable for reporting medical procedures. The GAO stated in its concluding observations considering the adequacy of ICD-9-CM Volume 3 and CPT in meeting all of the criteria recommended for HIPAA standards the practical challenges of implementing a single procedure code set and a lack of empirical evidence to either support or disprove the merits of doing so, we believe that dual code sets for reporting medical procedures are acceptable under HIPAA. In addition we concur with those representatives of the health care industry who contend that more study is needed to examine the possible benefits of adopting a single procedure code set for medical procedures before its implementation is considered.
In August of 2002 HIAA provided a letter to the chair, Dr. Cohn, the letter urged NCVHS to take into account all effects and entities involved before making a decision on the migration and the timeframe for implementation of 10-CM and 10-PCS. They also recommended that a thorough impact study with participation from industry groups, sufficient time for comments on any decision in proposed rule, and sufficient time for industry to implement ICD-10 if that decision is made.
The American Hospital Association submitted a letter to Secretary Thompson in August of 2002, the letter summarized the AHA position that ICD-9-CM had outlived its usefulness and supported replacing 10-CM for diagnosis and 10-PCS for inpatient procedures. ICD-10-PCS testing shows it is a vast improvement over Volume 3 ICD-9, and that both ICD-10-CM and ICD-10-PCS should be implemented at the same time. Lastly, the AHA noted that it did not support adoption of a single procedure classification for all services, rather as suggested by the GAO report the AHA supports a dual approach to procedure coding with ICD-10-PCS for inpatient services and continued use of CPT-4 for hospital outpatient and physician services.
Another letter received in August 2002 was jointly signed by the Blue Cross/Blue Shield Association, Blue Cross/Blue Shield of South Carolina, HIAA and Aetna American Association of Health Plans regarding migration to 10-CM and 10-PCS. The recommendation from the letter was that we should wait until HIPAA mandates were successfully implemented and that a thorough impact analysis including cost/benefit of migrating and not migrating is completed, and that other alternatives be assessed.
September 23rd the AMA forwarded a letter to Secretary Thompson summarizing their position, that it was unnecessary and potentially detrimental to replace Volume 3, which had served its purpose well and contained only 4,000, codes with 10-PCS, with the code set containing nearly 200,000 codes, and that it was unproven in any setting. They also cited a lack of involvement of organized medicine and the leadership of allied health professionals in the development and the maintenance of PCS. Lastly they urged the Department to consider adoption of CPT as a viable workable alternative to ICD-10-PCS. This letter was also co-signed by the American Academy of Family Physicians, the American College of Physicians, ASIM, and the American College of Surgeons.
An identical letter to the one that I just described was also sent to Secretary Thompson, however this letter was signed by 46 medical organizations.
September 24th, 2002, a letter signed by Blue Cross/Blue Shield Association, American Association of Health Plans, the Health Insurance Association of America, and the American Public Human Services Association, National Association of State Medicaid Directors, submitted a letter to Dr. Lumpkin regarding transition to ICD-10-CM and ICD-10-PCS. The letter urged a detailed analysis of the impact of replacement of 9-CM on the health care industry be conducted prior to the national committee making a recommendation.
Without going through each and every additional letter there are several more in October of 2002, November 2002, December 2002, all advising the committee and the subcommittee to undertake a cost/benefit analysis.
In December 2002 the subcommittee discussed the scope of work for initiating a contract for a cost/benefit analysis of ICD-10-PCS prior to making a recommendation to the Secretary.
January 2003 the contract was awarded to RAND's Science and Technology Policy Institute to conduct an impact analysis. The analysis included assessing the full range of implications on current users, including costs and benefits, those that are quantifiable and non-quantifiable, and the scope of work specifically included identifying costs, including opportunity costs associated with transition including impact on information systems, rate negotiation, reimbursement methodologies, training and changes in forms. Timing of transition including impact of timing options and costs and benefits were also part of the scope of work.
RAND presents their study plan and preliminary results at the May and August 2003 meetings respectively. July 23rd, 2003, the American Health Information Management Association submitted a letter to Secretary Thompson urging the Department to take quick and decisive action to adopt ICD-10-PCS and ICD-10-CM to replace the obsolete ICD-9-CM. Among the risks of further delay is placing the U.S. out of step with all other developed countries that have already adopted ICD-10. The letter urged the Secretary to promulgate an NPRM to adopt 10-CM and 10-PCS and offers to work with the Department to make this important change happen soon.
July 31st, 2003, AHIMA sends a letter to the HCFA administrator supporting the adoption of ICD-10-CM and 10-PCS and urging them to make CMS and the industry's case known at the upcoming September national committee meeting.
August 2003 AHIMA wrote to Tom Grissom and to Tom Gustafson at CMS urging them to make their case known in the upcoming September national committee meeting.
August 20th the Medical Group Management Association forwarded a letter to the chair of the subcommittee expressing concerns over the potential adoption of 10-CM and 10-PCS. Their concerns included significant modification to business processes, increased time spent documenting a patient encounter, potential loss of research and benchmarking data, training difficulties, significant increase in cost and delayed reimbursement. They stressed the importance of great deliberation, coordinated national strategy, and allocation of resources.
August 2003 the AHA and AHIMA conducted pilot test of ICD-10-CM, I won't read that again because it was covered in one of the previous timelines.
At the September 23rd meeting the national committee met and heard testimony by RAND of the results of the cost/benefit analysis. Other groups also were performing impact studies, testified on their results, and additionally testimony regarding reaction to these analyses were heard. Reaction testimony was heard from AHA, AHIMA, AMA, HIAA, and CMS. The subcommittee agreed to evaluate these testimony statements at their October meeting.
September 23rd, 2003, a letter co-signed by 46 medical organizations, I'm sorry, that's a repeat of something else.
This timeline has to be updated, we've received as of today two additional letters, one from AAHP/HIAA that I believe they distributed today, and also another letter from the American Association of Procedural Coders, that was also distributed today at the tables. So we know that this document will be updated forthwith.
The next document is labeled appendix one, and basically what it tries to do in a table format is synthesize everything that you just heard in the timeline. The first column identifies the stakeholder, the second column identifies the type of organization, the third column contains the dates of the testimony and/or letters that have been submitted by that organizations, the next column identifies whether the organization is in support of implementation of ICD-10, ICD-10-PCS, the next column no support, and then the last column for undeclared. This table contains approximately 83 organizations that have provided the testimony that you've just heard me review in the timeline. Particularly for those letters that were jointly signed by a number of organizations we have listed all those organizations out separately in this table. What we've also tried to do is to show not only the dates of the testimony and/or letters but also to show where those may have been joint letters, so for instance for one of the organizations it says JL4, that means that it was a joint letter signed four organizations.
DR. FITZMAURICE: Donna, what does a check mean?
MS. PICKETT: The check means that that organization is undeclared in terms of their position on ICD-10-CM/ICD-10-PCS. We were trying to look for a very visual way to show that, it was not as easy to do that for support and implementation column because some letters or testimony were specific to either ICD-10-CM or ICD-10-PCS and not necessarily both, so we weren't able to use the check-off format in that column. And in those instances what we've done where an organization supported one of the code sets but not the other we make it clear that they only provided input on that one particular code set. You'll also see for some organizations based on one of the last letters, joint letters we received where there's support for ICD-10-CM and ICD-10-PCS but for the inpatient segment only. So we tried to be very detailed within the structure of the table to synthesize everything that appears in the timelines.
DR. FITZMAURICE: Can I follow-up, I see a check under support implementation of ICD-10-CM and ICD-10-PCS per se, for example Tenant Health Care under support implementation, I don't see a check on undeclared, so does the check mean undeclared?
MS. PICKETT: In that case it means it supported both. And if there's a way that you think maybe we can better annotate the table to address any of the concerns about the representation of the information staff would be pleased to hear it and try to incorporate those comments.
DR. COHN: I'm presuming that this thing says support implementation of ICD-10-CM/ICD-10-PCS, these are for current uses right?
MS. PICKETT: That's correct.
DR. ZUBELDIA: Donna, where it says no support for implementation you actually mean that it opposes implementation?
MS. PICKETT: That means they do not support implementation.
DR. ZUBELDIA: That means they oppose implementation?
MS. PICKETT: Yes.
DR. ZUBELDIA: Because there are some that are just blank, there's no support and no position --
MS. PICKETT: Some raised issues relative to the need to do cost/benefit analyses first, with no declaration as to whether they did support or did not support, so we left that blank.
DR. ZUBELDIA: I'm trying to make the distinction between lack of support and opposing implementation. There are some that have no expressed support clearly, pending whatever, there are some that have expressed that they do not --
MS. PICKETT: If they have expressed they do not support there would be a check mark in the column labeled no support. If they have not taken a position they are in undeclared.
DR. MCDONALD: Can I get some clarification, I see the American College of Physicians is listed as supported it and I don't, I'm on the regents board and I wasn't on it when this might have happened but the last meeting which was Saturday I heard from some regent members some fear and trepidation about it. So I just was wondering if you could actually point me to the actual support letters so I could be clear that was --
MS. PICKETT: The dates are in here and we do have copies of those letters. And that's one of the reasons we wanted to include the date so that if we needed to refer directly back to a letter we have that information.
DR. MCDONALD: -- that one to make sure that I understand what their message is --
PARTICIPANT: Clem, was it the Family Physicians?
DR. MCDONALD: American College of Physicians, I thought I saw ACP, and AMA's on there too and I didn't hear their testimony sounding that way, I'm just reading the bullets, they didn't say they opposed it, they didn't seem very supportive in their testimony, the AMA.
MS. PICKETT: The AMA position, which has now been jointly signed by 55 organizations, their stated position currently is that they're focusing only on 10-PCS in the inpatient setting.
DR. MCDONALD: But they have a letter supporting that --
MS. PICKETT: We have a letter from them.
DR. MCDONALD: Was the ACP one of the co-signers on that?
MS. PICKETT: Yes.
DR. MCDONALD: If I could get a copy of that I could help make sure the channels are clear in the organization.
MS. PICKETT: We'll make that available to you.
DR. COHN: Carol, did you have a, I'm sorry, I was trying to get that one thing straightened around.
DR. BICKFORD: Carol Bickford, American Nurses Association. I stepped to the microphone because I wanted to make sure I was clear on how the inquiry had gone to the various organizations at some point and that was ICD-10-CM and PCS in all settings, and then they made their statements one way or the other. Is that correct?
MS. PICKETT: All this is synthesized from the testimonies that have actually been provided and there were two separate hearings for, one for ICD-10-CM and one for ICD-10-PCS.
DR. BICKFORD: And not being privy to the graphic because you've got it and we don't have it, it's difficult to appreciate the year that these testimonies and so on may have taken place, so I guess I'm wondering about the currency of the decisions, sort of like Clem is querying, have we had some changes in leadership and positions, so do we truly have a current listing on who's on board and who's not.
MS. PICKETT: The information that's in here is current up to the letters that we've received and if there is a change in position that is noted in the table. For the others, there was no further outreach to have them clarify what their previous written position was.
DR. COHN: I guess I would suggest that we may or may not choose to send this particular appendix forward. As I'm looking at this one, having just listened to the timelines and all of that I'm wondering if maybe there's a better way to express this. Actually I'm sitting here, I don't know whether there is a better way to express this. I mean we could choose within the timelines to put asterisks and the letters that were, I mean since the timeline has all the letters we could just note all of the people who signed a letter at a particular point, and that would also share the same amount of information. I think the intent here was to more give you a sense, I mean first of all we wanted to make sure everybody sat through it so that none of you could say that you didn't have a view of the history of what had gone on, but it was also just to give you a sense that we have heard from a lot of players in the health care industry and this is not a conversation that has somehow missed entities. Now it may not be that there's consensus or universal agreement or anything but it has not been for lack of talking to the industry about it. Donna I think you're expressing that very well and I think Clem does bring up a very good point that, as well as Carol, that a position taken two years ago may not be the same as the position held today. I mean all of us know that we're, we all live in organizations where positions change and certainly sometimes it's hard, for example a letter we just received from AAHP/HIAA, even though I think it's probably still an undeclared letter, depends on how you read it and so it becomes a very hard in how you put a check box in a particular place.
MS. PICKET: Other staff and I had noted that trying to come up with a discreet way of codifying what that position is has been difficult.
DR. COHN: Exactly, but I think this is great work and I would certainly share, everybody had a sense of looking at all of this, and I don't want, as I said we can decide whether the appendix should go beyond our meeting today, which I think is one piece, and certainly if we have questions about it or feel that it is, doesn't well represent everyone's thoughts, I think there are ways to deal with the timeline to more adequately represent the various stakeholders that commented. So, anyway, we can talk about that.
Donna, did you want to go through, I know that we're beginning to run a little short on time, we're not doing too badly. I notice you had a summary of implementation issues and I thought, did you want to run through some of those?
MS. PICKETT: Well, for the organizations that actually supported implementation and for those who had concerns about it, what we wanted to try to bring forward was a summary of all the implementation issues that had been raised in all of the testimonies and the letters that had been forward, either to the subcommittee or the full committee. This slide will seem very familiar to you because it's actually something that was covered in the timelines relative to previous national committee reports, so I won't go into that one but it talks about the pressure for change derived not only from the end users but those who are contending with all the deficiencies. This was referenced in the NCVHS timeline because it was the bullet points on one of the pages.
This slide just provides a summary of the disadvantages of remaining with ICD-9-CM, again culled from all of the testimony that we've either heard or read over time.
DR. COHN: Be aware, we don't have copies of this presentation.
MS. PICKETT: We'll make copies available tomorrow.
But it talks about the limited ability to expand the classifications to incorporate new diseases and technology, the lack of comparability with U.S. mortality data and international data, limited ability to adjust payment methodologies because of the lack of specificity and detail, and limited ability to evaluate health care outcomes.
And the possible benefits were some of those that were noted within the RAND report and in the testimony that has been provided previously, those include better understanding of health conditions and health care outcomes, improved disease management, better understanding of the value of new procedures, the potential for fewer miscoded, rejected and improper reimbursement claims, and more accurate payments for new procedures.
Throughout all of the various letters and testimony there's been a lot of issues raised related to timing. What I've tried to do is put onto a few pages some of the major issues that have been recurring themes as it relates to timing, clearly there was a sense that we shouldn't implement any new code sets until after the initial HIPAA standards have been implemented and that we should look to lessons learned from HIPAA implementation. That there should be a well defined implementation and maintenance process. On the subject of timing you get variable views that includes either implement in two year, implement in three years, no later then 2005, or the earlier possible timeframes, so we've kind of run the gambit here in terms of thoughts on timing. There was also mention of quality controls that need to be in place to monitor the systems during transition and that the crosswalks should be made available prior to implementation.
The issue of resources came up in several of the testimonies and there was a suggestion that perhaps payment methodologies be adjusted to be sensitive to increased regulatory costs of implementing the new code sets.
Issues related to training included ample time for education and training, that it should be broad based training and retraining, using the train the trainer and having identified tools for the training. And that the training should be coordinated across sites, regions, and stakeholders.
Lots of issues related to systems changes. There were issues raised about product redesign, testing, systems integration between various applications and foreign systems, upgrading or total revamping of hardware/software, software conversion, changes to optical scanning systems, changes to decision support systems, field lengths, coding edits, changes to order entry, lab, radiology, pharmacy, medical record, patient accounting systems, revisions of product and user manuals, and the need for coordination between vendors. Also cited for the need to run parallel systems, increase storage capacity, and that there would also be changes on state reporting systems, cost containment and discharge reporting systems, and quality reporting systems.
Other changes that were noted were changes related to policy and fee schedules, fee schedule calculations and structures, authorization streams, DRG assignment and pricing interfaces, changes to groupers and end coders and the need to renegotiate contracts. Also changes to databases that are used for modeling, adjustment to adjust care management policies, redesign of paper forms, data abstracts, report layout and format, and the need for new claim forms and other treatment forms. Again, this was just to try to bring together everything that has been provided to the subcommittee and the full committee over the course of the last 12 to 18 months to get it in one document.
And that's the last page.
Agenda Item: Subcommittee Discussion - Dr. Cohn
DR. COHN: Donna, thank you. Comments, questions? Clem, anything?
DR. MCDONALD: I just want that we get a chance to talk at all about the proposed letter.
DR. COHN: Actually we're probably not going to talk about any letters until tomorrow but I'm curious about if you have thoughts about what a letter might contain.
DR. MCDONALD: Well, I think the strongest thing I feel is the statement about the scope is is almost buried here, there's one sentence at the very end, as a replacement code for institutional procedures is the only hint that there's a scope limitation. And I made some proposed wording that we'd say this does not apply to professional billing of procedures, very explicitly especially since that's the only thing anybody agreed to in that whole table, I think it was down the whole line of it even American Hospital, even the strong proponents only were supporting, support for this in the institutional setting. I don't think that the average reader or necessarily the Secretary is going to understand there's any constraint on this.
DR. COHN: So you want to make very clear that whatever letter we may come up with makes very, very clear that we're talking about ICD-10-CM for diagnoses and ICD-10-PCS for inpatient procedures, it's actually really institutional --
DR. MCDONALD: Well, that wording is in there, I think to make it clear you say this is not, what it doesn't apply to. And I gave you a sentence but I don't care about what sentence, and I think Kepa you thought the same thing or no?
DR. ZUBELDIA: Yeah, I sent an email about that.
DR. MCDONALD: The other thing, I think the letter has gotten progressively, even with Mike's changes it's really more and more and more sort of this is what we got to, it's got more to, when I first read the sample letter it was one that was sort of really, really very meager support into something that was completely supporting, and I voted for the in-between one. This one is very, very completely supporting, all it says we should just do it. I'm not going to take a position against that, it's just that we're doing an NPRM and part of that is to get the responses, and I'd like to ask, plead with the group to ask for one other bullet and the question is are there any other alternatives. Because the thing that never got discussed and I know the politics is hopelessly complicated and there's two powerful organizations that have, I don't know what it is but they have like, they don't like each other, enough talk about this, but it seems to me the easiest thing to do to solve the ICD-9-CM problems would make a deal with CPT some way that would make everybody happy and add the other thousand codes you need to handle the special things inpatient and then we'd be having one code and we'd be done with it. But that I don't think is ever going to happen because of the politics and whatever, I don't understand, there's something there but it can't be discussed. So I at least bring it up because that's the reason I'm frustrated by this particular direction we're going.
MS. PICKETT: Dr. McDonald, I'm not sure that it's exactly that, I think if you refer back to the national committee's annual report back in 1993 it specifically stated that both classifications had major flaws and I think that's part of the issue in terms of looking at one versus the other, they're both flawed.
DR. MCDONALD: Well, I think you can talk, life is flawed, my other, I object to making this huge expensive change and we still got a structured code that we're going to change again in ten years, so I don't, the flaws we can find, he who does not, throw the first stone and all that kind of stuff, but all this stuff is flawed including 10-PCS, and the other one is, I'm kind of in a railing, the other thing that gets me upset every time I see it, it says we will stop fraud, well, firstly that's a pejorative statement and it begs the question about the difference between error and fraud, which some physicians think is a very, very small, the government calls it fraud and we a lot of them think is errors, but when you get 200,000 codes versus 9,000 codes they're going to pick the wrong one more and they'll be more opportunities to cry fraud in the choices people make and when they send it in then less, so I think actually the fraud story for the physicians will get more painful rather then less painful.
DR. FITZMAURICE: Let me make an important point and it's one that strikes me hard, too. I don't see why we need to go through this every ten years when countries can negotiate with WHO to have a continual process so that it's 10.1, 10.2, 10.3, maybe we never get to 11 but they're continual changes, not these sharp edges that create all these lumping of costs into one or two years.
DR. MCDONALD: Well, it's more then that, when you do the sharp edges people forget the history and they go we're going to make it really clean and it's their imperfections versus the last imperfections and you create this huge upheaval where things don't map perfectly, whereas you're forced to, now I know I sense that the one is limited in terms of the space but ICD-9, plain old ICD-9 ain't that bad, you can't throw the same criticisms at it, it's just that we have this contract, and we have this agreement with WHO, etc., etc. But starting over is a bad principle in the coding world, people live a 100 years and we'd like to be able to compare care for longer then 10 years or 15 as it really lasts.
DR. FITZMAURICE: Are you also suggesting an addition to the letter that says we ought to have continual change rather then lumpy change?
DR. MCDONALD: We ought to have a system that permits continuous improvement rather then have to throw away the old codes. But my main thing is that if we could have a statement, what other options are there, I think this has gotten stronger and stronger as though there's no doubt in anybody's mind and I want to have a uniform agreements, I don't want to vote against it, but there's some doubt in my mind and let the NPRM decide.
DR. COHN: Karen?
MS. TRUDEL: I'm not arguing against putting that in the letter but if it puts your mind at rest at all the regulatory process will require that we go through analysis, we can't just say we think this is a good idea, there is a requirement to say we looked at alternatives and this is why we picked the one that we picked, and that we're clearly going to get comments back on that. We don't constrain the kind of comments that we get and in fact we can invite comments on certain issues and perhaps that's what you're saying, that we should be very clear and up front about inviting comments on that issue. Because we'll get them anyway.
DR. MCDONALD: One you can't ask for it, can't we make do with some of the codes, one of which is not mentioned in here? Five letter word, four letter word --
MS. TRUDEL: Doing nothing is always an alternative.
DR. HUFF: -- universe, you can do things different, but even you say some of these things we just can't do given, what I've heard you say is this is hard and it's ugly and nothing's perfect, but I haven't had you say is a clearly stated different decision you would make, what you would really want us to do.
DR. MCDONALD: If I was king, well I'm not, no, I'd put the parties that I have this biggest discomfort together and figure out if there's, at least try to find out. The gap in the codes could be solved just by fiat or whatever, the question is its control and who is going to be comfortable with what, and you really need to have two separate treads and there's lots of other things that are hard to get your hands on to know what the real issues are. And the charges, could you make a deal, okay --
DR. HUFF: So is this specifically with ICD PCS, so you don't have any trouble --
DR. MCDONALD: No, I don't have trouble with it, it's just too bad that we have to do these upheavals because I don't think, truth is there's any problems in the U.S. that are being caused by ICD CM, it's just that we aren't --
DR. HUFF: Can you clarify why you think this current scheme will lead to upheaval? I mean if PCS were adopted or CM, 10-CM were adopted, why does that necessarily lead to upheaval?
DR. MCDONALD: I thought we just saw the list on all those slides.
DR. HUFF: No, it causes upheaval this time, why does it cause the next upheaval, what in this decision is locking this into anarchy ten years from now?
DR. MCDONALD: Structured, well, maybe I'm wrong about CM --
DR. HUFF: You can ignore, if they put structure in the code you can always ignore that, that is at some point in time you can just say that's an identifier, it no longer has meaning as a hierarchical code, it's like planes, there's a strategy for assigning codes to planes that includes who bought the plane first. But then, if that company goes out of business, that identifier stays the same and it no longer means that it was bought by Delta or United or whoever.
DR. MCDONALD: You just suggested something else then, I'm glad you brought that up, that is that's what we should have done with ICD-9, we should just have called them codes and extended it the way we wanted to. But I think we ought to say something about that so that we don't anticipate that they'll be another assumption about structured code in ten years, some kind that we don't have to do it again and maybe this time around we'll just do what you said, that would be, I think that would be a major gift.
MS. PICKETT: I'd just like to make one comment. We haven't updated, we didn't revise every ten years, 9-CM has been in place for 24 years so it hasn't just been a ten year cycle.
DR. MCDONALD: No, I know, it's supposed to be every ten years, we've resisted long enough.
DR. COHN: Well, actually my understanding was is that there are processes related to support of ICD-10 that are different then ICD-9 that allow for that updating. Am I mistaken about that? I'd understood within the new ICD-10 process that there were ways to update the actual ICD and the problem that you have with ICD was is that you could update CM however you wanted but the actual base ICD from WHO never got changed, isn't that, am I, that is a correct statement isn't it?
MS. GREENBERG: There is now an updating process for ICD-10, which is ongoing. Just regarding Clem's concerns, and I think this really gets back to the issue of single procedure classification, which of course this committee has wrestled over, it's been almost, well the entire time that I've been involved with it, which is since early ‘80's, and I think there's been generally an agreement that this is probably not currently feasible, whether it was a GSA, or the GAO report or some other letters that have been received. But I think what you're not appreciating sort of is that although yes, there are obviously political and institutional and other reasons I think we have heard from both CMS and the hospital industry that CPT as currently constituted does not meet the needs for hospital statistics and hospital reporting. So I think it isn't that it reflects physician time, physician input, and it doesn't capture everything that's needed in the hospital environment. So I think that is testimony that's been received from the part of the industry that would be asked to start using 10-PCS, so I think that has to be, it's not just the political and institutional issues, I mean we're talking --
DR. MCDONALD: If they could add the codes that they needed --
MS. GREENBERG: Well, it's not just a question of adding codes I don't think, I mean one possibility is to create a procedure classification that would suitable for all settings, and in fact most other countries that's what they have, I think we're the only country that is active user of procedure classifications that has more then one in use. But I think generally that would take a lot longer time to get that kind of convergence and that's why we've really backed away from that.
DR. STEINDEL: Marjorie, I think the question that Clem is asking is phrased the way you phrased it at the end of this, is that we haven't sat down and studied the ability to create a single procedure coding system that would work for this country, we've said that a single procedure coding system will not work the way we're handling things today, we haven't investigated whether we could create a single procedure coding system that would work. And I think what Clem is suggesting is that maybe it is better to invest the time to answer that question and have a single procedure coding system going on in the future then revisiting the issue of dual coding systems every N period of time, whether that be every decade, every five years, every 20 years. So he's asking if that question can be included in the NPRM as something that people can comment on.
MS. GREENBERG: That's not what I heard actually.
DR. MCDONALD: Well, I wasn't actually asking that but you're right, you're right --
MS. GREENBERG: It's an interesting point but I didn't think --
DR. MCDONALD: How I heard it in the testimony, the dimensions, and since it's, this is only dealing with, 99 percent of what we're talking about is physicians procedures in the hospital so there's a strong connection. I heard this doesn't work, so you get another 50 codes, another 1,000 codes, you only have 3,000 right now in the whole gizmo and you've got 50 or 60,000 in CPT, so it seemed like with another couple thousand codes you could make them work together. But that's why I attribute it to politics, that someone else wants control and they're fighting over control and so we're not going that way. And I'm willing to accept reality.
MR. ARGUS: This is George Argus from the American Hospital Association. I just wanted to make a couple comments. I appreciate Clem's suggestions in terms of trying to work out the politics but there is an upheaval with your proposal as well and there is a timeline that would be extended as well, taking that route. And I was struck really by the timeline that Donna presented over here, I think we're really at a point or this committee is at a point where it needs to make a decision to move forward with an NPRM and let the industry flesh out some of the remaining issues. But I think at this point in time there is benefit in moving forward with ICD-10-CM and ICD-10-PCS for the inpatient institutional setting.
DR. COHN: George, thank you very much. Carol?
DR. BICKFORD: Carol Bickford, American Nurses Association. I think we need to consider the fact that we now have new technologies and we have an evolving health care system, which is no longer just inpatient and then minimal outpatient, the majority of our care is being conducted in the outpatient setting and it's not being done just by physicians and we have multiple technologies that we don't have in our current coding systems. From my perspective I see this context, the place where these activities are occurring is another element, it has nothing to do necessarily with the procedure or the diagnosis, it just is part of the setting piece that was sort of addressed in the domains that CHI was trying to grapple with. So do we have the technologies that allow us to do things setting neutral? And then add on that extra piece which then we can put the financials attached to whatever setting gets it from the other queue, so I'm thinking of trying to help us move through this, it's inpatient or outpatient.
DR. COHN: Thank you. As we have this discussion I'm once again reminded about the fact that there really is no national consensus on any of this and I think we have observed that, I mean I thin we are far despite a couple of years of testimony, hearing from 80 witness or certainly with many more potentially writing letters, I mean this is an area where there's a lot of feeling and clearly we're not, I think it's going to be difficult to come to the consensus solution to solve all needs and all agree.
DR. MCDONALD: I'm supportive of a letter moving forward with NPRM, and I just, the changes that would make it --
DR. COHN: And I think what you were hearing is a letter that reflects the fact that there is not a single industry view on how to move forward or necessarily even when to move forward as Donna sort of commented. But I think it's important that whatever we do as a next step, and I agree with one of the previous speakers who said that we really have had a long history on this and there's a certain point at which an advisory body has to really sort of move something along so that the whole industry can sort of comment, but I think I'm also hearing from you that we need to be sort of honest with this diversity of what we're hearing and all of this recognizing that we could have 100 more testifiers and we still would not have developed a single point of view. And we need to be relatively subdued in terms of what we say recognizing that diversity in the industry, I don't mean, I'm not putting words in your mouth but is that --
DR. MCDONALD: That's exactly what I said.
DR. COHN: Okay, I guess I'm quoting Clem and I just didn't realize that, I apologize for quoting you. And certainly as I said this is something we'll be discussing more tomorrow now. Other thoughts or comments before we thinking about breaking for the day?
DR. ZUBELDIA: I think in the, I know we're going to be talking about the letter tomorrow but I want to go one step further from what Clem is going, I think that in the letter to the Secretary about an NPRM it should be with a recommendation from NCVHS to adopt the ICD-10, not just an NPRM to see if it needs to looked for adoption. You know what I'm saying? We've been looking to see if it needs to be adopted for the last few years and it gets to a point where yes, we need to have feedback from the rest of the industry but I think that the NCVHS needs to say well, at this point we're recommending it for adoption, or we're not.
DR. COHN: I think that if we actually can't come to a, I mean if the word is the Secretary just needs to further evaluate, I think that's more of a NOI almost rather then an NPRM, it's hard to do an NPRM and not recommend anything.
DR. ZUBELDIA: That would be more like saying well we don't know what to do, we're advising the Secretary to look elsewhere because we can't tell one way or the other, and I think we need to be more assertive then that and say okay, we recommend it for adoption but we still want to get feedback from the rest of the industry, but this is what we're recommending.
DR. MCDONALD: I think that's what the letter says now.
DR. COHN: Other comments? I mean once again we will have a chance to, this is not really a discussion of the letter because there really is no letter in front of the subcommittee today but there hopefully will be some work that we can do tomorrow afternoon to consider various alternatives and come to a decision that we can live with.
DR. MCDONALD: Well, I appreciate the chance to articulate.
DR. COHN: Sure, Clem, thank you. Okay, well now we're going to be starting tomorrow morning at 8:30, not at 11:00, and as I said we'll be starting out with some testimony on ICD-10, both from Tom Grissom and from Blue Cross/Blue Shield Association, and then after the break some more conversation plus an open mic and then we'll get into discussing the next steps. Okay, thank you all, the meeting is adjourned.
[Whereupon the meeting was recessed at 5:02 p.m. to reconvene the following day, Wednesday, October 29, 2003.]