[THIS TRANSCRIPT IS UNEDITED]
Hubert H. Humphrey Building
Room 505A
200 Independence Avenue, SW
Washington, DC 20201
Call to Order and Introductions
Second Panel - GCPR Framework Project Management Team:
Convergence of Ideas on Medical Terminology
Review and Comment on HIPAA Guiding Principles
Review of Focus Areas from Work Plan
Review Hearing Agenda for the Rest of 1999
Compile Definitions of PMRI from Testimony
MR. BLAIR: For those of you who have been listening on the Internet, waiting for the National Committee on Vital and Health Statistics, the Computer-Based Patient Record Work Group to begin, thank you for your patience. We are technically live now.
I'm Jeff Blair, the chair of the CPR work group. The first order of business will be for us to introduce our committee members and the other attendees here in the room.
[Introductions were made.]
I'm aware that those of you on the Internet may not have been able to hear all of the introductions. However, you will be able to pick up most of them, because many of the folks are folks that are waiting to testify, and during the morning you will be hearing them again, introduce themselves in front of a microphone.
For those who are listening for the first time, the National Committee on Vital and Health Statistics was given the mandate from the Health Insurance Portability and Accountability Act to the administration simplification portion of that act, to proceed with studying different aspects of standards, and make recommendations to the secretary of the Department of Health and Human Services.
One aspect of that mandate was a portion of the administrative simplification provisions which indicated that the committee had the responsibility to study union data standards for patient medical record information, and the electronic exchange of that information, and report those findings to the secretary of the Department of Health and Human Services by August 2000.
That particular mission is the charge for this CPR work group. In order to execute that charge we have developed a work plan, which we will be reviewing later this afternoon, a calendar of events that we also will be reviewing later this afternoon. On December 8-9 of this last year we wound up soliciting comments and feedback as to whether the work plan to pursue our mission was on target. And then we also had hearings in March 29-30 on the focus areas.
I just might reiterate for you, just to put it quickly in perspective, the work plan basically identified six areas of focus that we would pursue to study uniform data standards for patient medical record information. Those six included understanding issues related to message format standards. That's number one.
Number two is understanding issues related to medical terminology. Number three is understanding issues related to business case issues for developing standards. Number four is understanding issues related to standards that may affect the health care information infrastructure. There is another work group in the NCVHS that is working on health care information infrastructure. The role in this committee would be the standards aspects of that.
Number five, in terms of the focus areas that this work group is working on, studying issues related to data quality, accountability, and data integrity. And lastly, studying issues related to how different state laws may affect the ability for us to have uniform data standards for patient medical record information. So that gives you a perspective on the focus areas that this work group is pursuing.
In March of this year we had our first hearings on message format standards, and just a little bit on data quality. In May we had two days of hearings on medical terminology, and we're going to be reviewing some of the results of those findings. Today we are picking up one of the folks that we had hoped to testify for us and wasn't able to before, and that is Claudia Tessier, on the use of XML syntax with respect to medical transcription and representing ASTM.
We will be having hearings in September and October, this work group will be. That I think gives you a little bit of a background on the work of our work group.
Before I proceed further, Dr. Simon Cohn is just to my left. He is chair of the NCVHS Subcommittee on Standards and Security, so before we wind up going to our testifiers, Simon, do you have anything that you would like to add?
DR. COHN: No. Jeff, I think you have recapped the issues and opportunities here very well. Obviously, the subcommittee, which is the current body of this work group is obviously looking very much forward to recommendations as they come out of the work group, and hopefully being able to forward them to the committee.
We are also hoping, I think as you have commented, to have a letter to the secretary for consideration by the full committee some time in the September time frame just to begin to lay out the issues.
DR. COHN: Thank you, Simon.
The last thing I'll mention is just a very quick run down. For those folks that are here, we have a copy of the agenda. For those of you that are listening on the Internet, the first thing we'll do is be receiving testimony from Claudia Tessier. Then at around ten o'clock, we may be able to move it up a little bit, because our second person is going to testify, Dr. John Mattison was not able to join us today. So we hope to be able to move up testimony from the government Computer-based record group. So you're here, and please be prepared to begin a little bit early, maybe 10 or 15 minutes early on those hearings. We would like to hear from you.
After lunch we are going to focus on the guiding principles that were developed for the review of standards for HIPAA administrative simplifications. We're going to take a look at our work plan. We're going to take a look at the calendar. We're going to look at the budget for the work group. We're going to also wind up receiving definitions of patient medical record information. And finally, as Simon pointed out, all this leads to our ability to wind up drafting an outline for an interim letter to the secretary of the Department of Health and Human Services with respect to the progress of our work group.
On that note, as many of you are at least aware here that I'm blind, so let me just do a quick poll before we wind up going to our first testifier. Is there anything else that anybody on the committee would like to say or point out at this juncture before we begin? Hearing nothing, we would like to thank Claudia Tessier.
Claudia, could you please introduce yourself to the committee and to our Internet audience?
MS. TESSIER: Certainly, thank you. I'm Claudia Tessier, executive director of the American Association for Medical Transcription, chair of the ASTM E31 Committee on Healthcare Informatics, and co-chair of the ASTM E31.22 Formats Task Group.
Thank you for the opportunity to provide testimony to assist the National Committee on Vital and Health Statistics in developing and understanding the issues related to patient medical record information.
The American Association for Medical Transcription, AAMT, is a not-for-profit, national membership association representing the profession of medical transcription. Medical transcriptionists trans reports for clinician dictation that communicate patient medical record information. Transcribed patient information includes data related to demographics, genetics, health, illness, injury, family history, social history, diagnostic procedures, medications, treatment, and invasive and noninvasive procedures in relationship to the patient's past, current, or anticipated future condition, as well as interpretations regarding that information.
AAMT promotes the continuing education of medical transcriptionists so that they will prepare transcripts that communicate patient medical record information clearly, consistently, completely, and correctly.
ASTM is an ANSI-accredited standards development organization. It celebrated its 100th anniversary last year, and its E31 Committee on Healthcare Informatics will celebrate its 30th anniversary next year. The E31 committee promotes knowledge and development of standard classifications, guides, specifications, practices, and terminology for the architecture, content, storage, and communication of information used within health care including patient-specific information and medical knowledge. It also addresses policies for integrity and confidentiality, and computer procedures that support the uses of data in health care decision-making.
ASTM's E31.22 Transcription Subcommittee's efforts toward the development of standard formats for patient care documentation have brought me to this hearing, and are the basis of my testimony. AAMT representatives participate actively in this subcommittee's task group on formats, which is jointly chaired by myself and Paul Schyve, M.D., of the Joint Commission on Accreditation of Healthcare Organizations.
On behalf of AAMT and ASTM I thank you for including this topic of standard formats for patient care documents on your agenda today.
Patient data recorded through the processes of dictation and transcription is primarily for the purpose of documenting the patient's history, current status, treatment and prognosis. This documentation is likely to be the basis on which future care is determined. Thus, ASTM's efforts toward developing standard formats for patient care documents are for the primary purpose of enhancing patient care.
In the fall of 1998, the ASTM task group on formats conducted an informal survey among inpatient and ambulatory care settings in order to assess the volume of transcription, and to identify the most frequently transcribed reports among respondents, as well as to gather sample report formats. The 69 respondents reported a total annual volume of 9 million reports.
This reinforces our estimate, based on admission discharge, surgical procedure, and outpatient that the annual volume of inpatient and ambulatory transcripts is well over a billion. Of note as well is that medical transcription is a multi-billion dollar business, with estimates as high as $18 billion annually. Transcription, thus, is a significant part of the health care industry, yet it is one that until recently has not drawn the attention of standards developers.
Given the impressive volume of dictation and transcription, consistency in formatting transcribed patient care documents is becoming increasingly important, primarily to improve the quality and efficiency of patient care delivery. Other benefits include enhancing communication of patient information among providers, access to data for research and statistical purposes, performance measurement in health care, the reimbursement process, efficiency and cost savings for document preparation, and development of the electronic patient record.
The ASTM survey also found that the respondents' most frequently transcribed reports are discharge summaries, operative and procedure reports, histories and physicals, consultations, emergency department reports, progress notes, clinic notes, and imaging reports. Many survey participants included sample copies of their report formats.
Variations among these sample reports include titles of reports, section headings and subheadings, section sequencing, and highlighting of specific sections. For example, when recording a patient's allergies, some formats create a specific section titled Allergies, and some highlight the word allergies or the drug name by presenting it in all caps or in bold type. Others include allergy information in the narrative of the report without drawing attention to it, thus diminishing the likelihood that the information will be readily accessible to future users.
This example demonstrates the benefits that could be gained by a standard for representing allergies in a patient record. In some instances it could be a matter of life and death, as indeed some physicians have reported. In perhaps less dramatic ways, consistency in other formatting areas would greatly enhance ready access to patient information.
This effort toward standardizing formats for transcribed patient documents would contribute to the collection, storage, and communication of comparable patient medical record information. Its strengths include its emphasis on involving key stakeholders and seeking widespread input and reaction. Meetings, for example, have been held with representatives of medical specialty societies and of nursing organizations, and we will continue to reach out to document originators as well as users.
Its weaknesses include the difficulties that are likely to be encountered in adoption of the standards. Traditionally, clinicians prefer to use their own formats, and we anticipate resistance to efforts toward conformity. To some degree, this resistance is being addressed and reduced by the proposal that medical transcriptionists be responsible for formatting the reports according to the standard. This is frequently the practice now, except that medical transcriptionists must learn and apply multiple types of formats. Standard formats would greatly facilitate the transcription process.
Of note too is that at the urging of physicians and others attending our stakeholder meetings, we have widened our scope to include the development of standard formats for all forms of data entry by clinicians into the patient record. Thus, the standards we develop will not be limited to transcribed reports, but will also include handwritten notes, as well as those electronically entered by keyboard, structured entry, touched screens, speech recognition, or any other form of data entry.
Comparable patient medical record information is desirable, because it allows the past, current, and future status of a patient to be assessed regarding that specific patient, as well as in relationship to other patients, and aggregate data from multiple patients contributes to statistical and other research studies. Additionally, such data may be used for other purposes, including reimbursement and peer review.
Interoperability among dictation and transcription systems, as well as other documentation systems is not common, resulting in difficulties in accessing and merging patient information in systems other than those used to create and store such data. Such difficulties occur between institutions and across departments within the same institution, as well as well between provider institutions and those who dictate, and those who transcribe.
Efforts are underway among SDOs to address the difficulties of incompatible systems. The efforts of ASTM E31.22 related to formats are being coordinated with the work of ASTM E31.25 Subcommittee on XML. Thus, as standard formats are developed, they will be handed off to the XML subcommittee so that is can prepare DTDs, document type definitions, for these standard formats. It should be noted as well that as standard formats and their applications through such means as XML are developed and adopted, we anticipate that vendors will in turn incorporate them into templates within their systems.
Furtherance of this formats effort and the work of other standards development relevant to patient medical record information would be enhanced through the federal government's development of legislation or regulations that define the minimum requirements for such information. Additionally, it could use its position to involve more professional organizations in the standards process, both in development and implementation. Finally, its continuing support of ANSI Health Informatics Standards Board's efforts could accelerate the development and coordination of standards for the electronic patient record.
Thank you for the opportunity to present this information to you. I will be glad to answer any questions or expand the presentation.
MR. BLAIR: Thank you, Claudia. Are there any questions from anyone on the committee?
DR. COHN: So you mind if I start out? Claudia, thank you very much for your very informative presentation. I actually had a question regarding your testimony, and wanted you to please explain to me a little more what you meant by this comment where you said we have widened our scope to include the development of standard formats for all forms of data entry by clinicians into the patient record. What exactly do you mean by that?
MS. TESSIER: Well, at the time that we met with the medical specialty society representatives, there was a range of reaction from resistance to disinterest to enthusiasm at the beginning of the meeting. As the meeting evolved there was increasing interest in proceeding with developing standard formats for transcription, and it actually went further to if we are going to do this for transcription, let's look at other forms of data entry as well, because that isn't the only form. It may be the most voluminous form of entry into the record, but it's not the only form.
There are handwritten notes. There is more and more structured entry, touch screen entry. There is increasing use of speech recognition of course. And the physicians that were present at that meeting encouraged us not to limit our efforts to developing format standards that would be applicable only to transcribed reports, but to expect those standards to be applied whether the information was entered directly by hand, by keyboard by the physician in whatever form of entry.
So that for example with the issue of allergies, at any time that allergies would be entered, the standard that was developed for identifying that section with allergies would apply no matter what form the entry took.
DR. FITZMAURICE: Claudia, again, I appreciate the testimony. You covered a lot of salient points, and it was a very brief form, a very terse form. You talk about the benefits of format consistency -- improvements in quality, efficiency, performance measurement, quality measurement, which is important to my Agency for Health Care Policy and Research.
This is a relatively new effort just getting underway. Given the benefits that you describe, where do you expect the greatest support for the development of these format standards? That is, whom do you think would benefit the most, because they probably should be participating very actively in this. I'm thinking of computer systems, consumers, government. Where do you think the strongest support is going to be?
MS. TESSIER: The strongest support I think will come from medical transcriptionists, and from the clinicians who recognize the value of that consistency in formats. There was one physician at the meeting with medical specialty societies who spoke from a very personal point of view that she saw a benefit to this not just between physicians and institutions, but within her own practice.
She found that as she went back to past documents for the same patients that she was seeing a decade later, that she was unable to find some of the data she wanted, because her own system of entry had changed. So she was very eager, and had developed for herself a standard format, and wanted to see not that particular standard, but the issue of standard formats applied across the board.
I think like so many things, one of the pushes for this I think will come from reimbursement. I think to a great extent the increase in volume in transcription and dictation and transcription has been fired by the reimbursement issues. That you can retrieve information so much more easily when it is in a keyboarded format rather than a face sheet or handwritten entry. And I think that over time the reimbursement system will recognize that standard formats will also facilitate that process.
But I think clinicians for future patient care for consistency, in order to be able to retrieve information, there will be a strong support from that arena.
DR. FITZMAURICE: I would follow that up by saying too that people who are responsible for large populations, and the care of those populations need to have consistent information in order to make good decisions to make available in different areas, and to different populations. I would think that they too, the managers of these large service delivery organizations, whether they be managed care, whether they be government, would also be very strong beneficiaries of this, because they could then pull data together to make those decisions.
It's the lack of information, whether it's at the clinician's level, or at the high level management and policy level that needs to drive health care in the future.
MS. TESSIER: Yes.
DR. BRAITHWAITE: Claudia, thank you very much for your testimony, and for your work in this area. It's very important I think. But coming off of Mike's question, and from your testimony, the physician that talked to you oh, it's wonderful, I invented my own format. And it's great and wonderful. And yes, every physician would love to do that. As you know, physicians are a very independent lot, and don't like other people to tell them what to do.
Is there any experience that would indicate that if the financial folks say to all the doctors who are sending them or working for them or whatever, that you must follow this format that we have dictated to you, that they are actually going to do it? What's their reaction going to be when the thing they dictate comes back in a form that they don't recognize?
MS. TESSIER: I think that first of all, the must do it is the turn off. I think that what we are trying very hard to do is to enlist the interest and support of physicians and other clinicians in the development of these standards so that they are a part of the process, and not just impacted by the outcome of the process.
There certainly was within the meeting with the medical society representatives a very clear expression that if this comes out from HCFA, from the AMA, at least some of them don't want any part of it, and will be very resistant to it.
Over the course of that day's meeting what we saw was an awareness and an acknowledgement among the participants that this is probably going to happen, so let's be a part of it. Let's design it. I think that is part of the impetus behind their suggesting that we go beyond transcription, and not limit it just to that area.
Quite frankly, I think some of the reason for going beyond transcription was that they didn't want transcriptionists to be the ones designing it either. They want very much to be a part of it. And we acknowledge that. I think transcriptionists have an opportunity for input, because they are so experienced with this. On the other hand, they admit that they are not clinicians, and want to work with clinicians to design this so that is works for all parties to the best advantage.
MR. BLAIR: Bill Braithwaite, did you have a follow-up question?
DR. BRAITHWAITE: No.
MR. BLAIR: Claudia, my question gets to coordination, cooperation with your activities with other standards development activities. You indicated that the syntax that you were looking at to be able to help to format that you capture through medical transcription, the syntax would be XML, which tends to be consistent with activities within the special interest group in HL7 on patient record architecture that is looking at XML. It is consistent with a number of other standard activities that are beginning to look at XML as a syntax, including some others in ASTM.
So the question that I'm getting to here is what do you do when you have voice recognition and when you have handwritten notes? Are they also going to somehow be transcribed into an XML format? How do you wind up pulling those into a syntax that is consistent with other standards?
MS. TESSIER: I think it's going to be a long time before we have got everything consistent. The reality in the transcription world, although I hate to admit it, is that there indeed are instances out there where typewriters are still being used, and not electronic entry systems. So that not only is the issue of how to integrate handwritten notes there, but also what do you do with the more traditional typed entry that is still going on. I don't think that it's the majority, but it exists, and exists primarily in doctors' offices and to some degree in clinics. I would hate to think it exists in acute care facilities, but I would hesitate to say absolutely that it did not.
So the whole integration issue I think is a very big one, and one that we don't have all the answers for yet. Certainly XML is something that ASTM and HL7 are looking at. The current assessment is that that's the direction it is going in. Ask again in three to five years, and who knows whether XML will still be considered to be the answer.
MR. MAYES: Claudia, thanks for your informative testimony. Certainly, coming from an agency that deals with an awful lot of actual material, standardization would be a real benefit. I sort of hesitate to bring up issues of privacy and confidentiality, but I think it might be useful. As you pointed out, one of the biggest beneficiaries for this standardization, or at least self-perceived beneficiaries would be the transcription industry itself.
Could you maybe give just an idea of whether or not currently contractually transcription agencies are limited in terms of reselling information that they capture through transcriptions? I'm thinking along the terms of the kinds of issues that have come up more publicly with pharmacy benefit management companies and those sorts, who actually are sort of third party processors of information, but have recently been getting into the business of repackaging that information and marketing it for other uses. Is that going on in the transcription industry?
MS. TESSIER: Yes, at least from anecdotal information that we have, indeed it is going on. In some instances we understand that transcription businesses are storing back-up copies of the transcripts that are prepared. In the past that used to be primarily for billing purposes, to back up the charges, but more and more we hear of it being done, and the discussion is going on of whose data is it.
In some instances that storage is on a contractual basis, where the institution specifically asks the business to do it. We have been told as well that pharmaceutical companies are investing in transcription businesses, and that a good deal of data mining is going on. We have even heard comments to the extent that there is more money to be made in the data mining from the transcription industry than there is from the actual transcription itself.
MR. MAYES: So it would be fair to suggest that we might want to look at the relationship of the confidentiality and privacy types of standards?
MS. TESSIER: I would absolutely look at it. In the reviewing the draft confidentiality bills from Jeffords and others, the issues of storage are among those that we are looking at, because I don't think that there is as widespread awareness of the extent to which storage of back-up copies, data retrieval copies.
There is the also the issue if you are looking at security and confidentiality, there is the issue of transcription crossing not only state lines, but the international transcription that is being done. There is a good deal of offshore transcription being done. I think that the whole concept of confidentiality and security needs to be looked at in terms of crossing state lines and country lines as well.
MR. BLAIR: We have had two additional work group committee members join us here. So before I get to some of the other questions that are queued up here, Dr. McDonald and Richard, could you introduce yourselves, please?
[Introductions were made.]
MR. BLAIR: Kathleen? Kathleen Fyffe, please.
MS. FYFFE: Claudia, thanks for your very informative presentation. I have two questions. In your testimony you talk about multiple types of formats that medical transcriptionists must learn. Can you tell us a little bit about those multiple types? Are we talking about scores of formats, or hundreds of formats? And do they tend to be geographic, people in certain parts of the country use it, or do the orthopeds use one thing and the I people another?
MS. TESSIER: Transcription business have either in hard copy or on-line, the formats that come from the institutional clients that they work with. So that if you are doing a discharge summary from Hospital A, this is the format that that hospital would like you to use. If it's Hospital B, there is this format. And there are as many formats as there are institutions. That's the beginning of it.
Then you have always use this format except for Dr. Smith, and with Dr. Smith you change to this format. For Dr. James it's a different format. So the variations are variations among institutions, departments within the institutions, and individual physicians within departments. It is voluminous.
MS. FYFFE: My second question is has there been any discussion or thought given to having a training class or some sort of educational unit in medical schools?
MS. TESSIER: Yes.
MS. FYFFE: Can you tell us a little bit about that?
MS. TESSIER: I think that ultimately that is likely to happen. We would like to involve the medical schools sooner rather than later. But quite frankly, there is little attention given in medical schools to dictation to begin with. So coming in with this particular concept may or may not be received well, but we do intend to do so. I think that the number of stakeholders in this is quite widespread.
Of course in medical schools you've got a population that is much more familiar with electronic data entry, and more likely to be responsive I think to how to use it in order to standardize. But we do anticipate getting medical schools involved as well.
DR. FERRANS: Yes, I want to thank you for your testimony. I want to thank Bob for raising that very important question, and you for providing such valuable information. That particular issue is very concerning to me.
With regard to development of XML, it is interesting -- and in the hospital I serve on the medical records committee at a very large hospital. Just trying to get changes on electronic forms is an arduous and sometimes political process. One important driver for standardization of electronic forms and transcription I think is Medicare documentation requirements.
I think especially with academic institutions it's the sort Damocles that everyone is afraid of. That they are going to get audited, and find out that they have not sufficiently documented care. And having template-based, sort of collapsible XML based systems I think is a very useful method to insure that. I certainly logic systems can be built on to that to kick back a document if one has not tried to do this. Actually, this is something that we're looking at doing.
I'm curious about number one, the time lines for some of the draft DTDs to go out. Specifically, I was looking, they have had a broken link for a draft radiology DTD in the XML SIG of HL7 for months now. So I'm curious as to what program you all are making on that.
I'm also curious as to anyone has also brought up the idea of just standardized hospital forms, and the sort of blurring of the distinction between paper forms and electronic forms?
MS. TESSIER: I first brought it up about 30 years ago when I was doing transcription myself. It's been brought up repeatedly. There has been tremendous resistance to it. Hospitals in the past wanted their own letterhead, their own format. In the distant past I think a good deal of the resistance came from the institutions themselves.
The movement toward XML I think indeed is going to facilitate the process. And we are continuing to look at developing these standard formats whether or not XML is used. Quite frankly, the institutions are increasingly interested in it. They are finding it much more attractive than they have in the past.
I think that part of what we are dealing with, you mentioned Medicare documentation requirements. On the one hand I think that can be a driver. On the other hand it's that initial resistance, don't tell us what we have got to do that we hear from the clinicians. Indeed, in terms of suggesting that transcriptionists apply the standard format whether they use XML or just the electronic cut and paste, there are physicians very resistant to transcriptionists doing that.
Transcriptionists, they believe, should not have that responsibility. The reality is transcriptionists have had that responsibility for decades. It used to be that it had to be done by re-entering the report, because you didn't have the electronic ability to cut and paste. So the electronic system has greatly facilitated already the efforts toward that standardization that more recently comes from the institutions saying we want all of our reports to look alike, except for Dr. Smith's.
And so transcriptionists are doing this already. There just hasn't been as much of an awareness on the part of clinicians that they were doing it. And I don't think we have to wait for XML. Where are the efforts in XML in terms of doing this? It is kind of the chicken and the egg. They are more ready to move with this than the formats group, because we haven't yet developed the standard formats.
In fact, I should alert you that at our most recent meeting of the task group the conclusion was reached that if we said what we were going to do is take document by document and develop a standard format, we would never finish in my lifetime, or that of anybody else on the task group, because there are so many different reports; not just your general medicine, but then you get into the specialties.
So what we have decided instead is to start in a different way. That is we are going to look at three things. We're going to look at the names of parts of the report. So for example, if you are doing a physical examination, it will always be called physical examination in whatever report it appears. It will not be called physical exam. It will not be called PE. It will always be called physical examination, assuming that's what we come up with.
Secondly, that if there is a physical examination and a review of systems, the review of systems will always precede or follow the physical examination, whatever the decision happens to be. So the sequencing of the sections is something we are going to look at. So that when one picks up a report with a review of systems in a physical exam, you will always know that one comes before the other.
Thirdly, we are going to look at topography. TYP, typography. That is, for example with allergies we might decide that allergies will always be titled allergies, and will always be in bold print, so that one can readily identify it. Those are the three characteristics we are going to begin with, so that what we come up with can be applied to any report in which that particular section might appear. We are about to start drafting the first standard for that.
DR. KOLODNER: Thank you. I very much appreciate your presentation today, and the number of the issues that we have touched on. In the discussion about forms and about XML, I am confused. The reason that I'm confused is I thought that the power of XML is the data can be tagged with XML, and then it can be poured into any format we might like.
Although there is benefit from the standardized format as somebody is looking at it, that we might, without forcing the standardization at first, might be able to take any format that is provided, and begin to parse out the data that is there in XML. So that as we develop formats, we can give Dr. X his or her format that they like, whether the hospital wants a standardized format. They could get the data in that way.
In some ways if that distinction is an accurate one, then the battles that we could get into about format become actually less important than can we make sure that whatever format we are given, we can parse out the entry into the XML, leading to something in the future, but capturing valuable data now.
I'm wondering why the emphasis is on the form, rather than on really doing the parsing part?
MS. TESSIER: Well, I think part of the reason the emphasis is where it is, is because we are still dealing in the world of the paper record. While XML ultimately may solve some of these issues, meanwhile we've got over a billion transcripts being prepared annually, of which I don't know what the percent is, but in discussions with our membership of the professional association, there is very little that is being used electronically.
It may be transcribed within an electronic environment, but it is used in the paper form. So it is printed out and used in a way that the eye has to make those adjustments to the format, rather than being able to have the system do it for you. So we really are starting with the paper record.
DR. KOLODNER: Are there any of your members that you know of who have done even pilots, where they have taken a format and moved it into XML?
MS. TESSIER: No, transcriptionists for the most part are not that developed with the technology, and the application of XML to transcription is going to come through our working with IS people. There may be some of that being done, but it's minute.
MR. BLAIR: In order for us to stay on schedule we have Clem McDonald's question and Dr. Cohn's question, and then at that point we will take our break.
DR. MC DONALD: I have a number of comments and then some questions. I think you're actually with the XML. There is a real challenge with dictation systems, and to start to define the pieces. I think that's the strength of this effort. The XML group got all the publicity, but this is the hard work.
MS. TESSIER: This is the real work.
DR. MC DONALD: I compliment the work. I only would worry about getting into bolder, not bold, because it's the same. Actually, HTML dealt with that by saying emphasized, because maybe it's going to be red on a color printer rather than bold.
But the other question was the biggest problem we had with what I call kind of naked transcription, just coming across as a file is knowing who it's about, what the date, and what it is. Is part of the standardization of the raw text version going to be where you put the hospital number or the identifying number, and where you put some of the other things so we can parse those out?
The other side is when you have actual systems. There are already a lot of HL7 communication with transcription systems to give those pieces of information. How is that going to fit together?
MS. TESSIER: Well, I can't say exactly, because the outcome isn't determined yet. But yes, at our last meeting we acknowledged that part of the difficulty is in just where does the information appear. In some instances patient name is upper right. In others it is lower left, whatever. So we are looking at placement of the patient demographic information, as well as at least what the minimum information needs to be.
DR. COHN: Actually, I do know of a number of transcription groups that are using XML. The problem is that there don't seem to be any standards for how this is all used.
Actually, as I was looking at your testimony, I was surprised by the discussions around what I would describe as the business case, and you have it becoming an industry standard. I guess as I think about this and your discussions around trying to interest physicians and others to adopt the standard, perhaps from my simple view I would imagine that if there were a business case around this, and it was actually cheaper to do, and transcription services said, gee, it will be X dollars per or X cents per word if you adopt our standard format, X plus 3 if you use your favorite format, it might become an industry standard very quickly.
Yet I don't see that in any way referenced that there is a business. Is there a business case like that?
MS. TESSIER: Absolutely, and in fact that's the example I have used. That if I had a transcription business and standards were developed, or even are in the process of being developed, I would offer to my clients you let us do it with this standard, and you get a discount.
DR. COHN: Can I just follow-up on that one then? Wouldn't that change the nature of the discussion?
MS. TESSIER: Theoretically it does.
DR. COHN: It moves from a standards discussion to a business discussion.
MS. TESSIER: I would have thought so, but the transcription businesses are not giving us that response, at least to date. I think again it's each transcription business wants to do its business its own way, just as clinicians want to enter data their own way. So we're in the process of selling this not only to clinicians, but quite frankly even to some transcriptionists; certainly to transcription businesses.
I will say among the greatest enthusiasm that we have seen in going out and meeting stakeholders was from the nurses, because they are users of the record. And they started giving us input on how we could address this in terms of how the users will benefit from it. As a result of a very small meeting we had with a number of nursing organization representatives, 30 or 40 nurses signed up for the work group in order to participate and have input so that the user will benefit from it as well.
MR. BLAIR: Claudia, thank you very much for your testimony. In the tradition of a former chair of the NCVHS, Don Detmer, he felt that it was important to have agenda integrity. So may I ask you all to take your break, and please be back. We will convene promptly at 9:55 a.m.
Thank you.
[Brief recess.]
MR. BLAIR: Our second panel are representatives from the Government Computer-based Patient Record project. I believe we have four individuals that are here to testify. My I invite you please to introduce yourselves?
MR. GROEN: I'm Peter Groen and I'm from the Department of Veterans Affairs. I'm the GCPR project manager. To my right is the project management team. Let me have each one of them introduce themselves to you.
DR. MARTINO: I'm Dr. Janet Martino. I'm a physician with DOD. I'm acting as the deputy project manager and essentially COO for GCPR.
CMDR. MC CAIN: I'm Jim McCain with the Indian Health Service. I'm the chief technical officer for the project.
DR. KENTSMITH: Good morning. I'm Dr. David Kentsmith. I've been chief of staff at two VA hospitals, and I'm on loan from the VA to the GCPR as the chief medical officer.
MR. BLAIR: I assume that you have already determined the sequence of your presentations. So please go ahead.
MR. GROEN: What we've done is given each of you a copy of a handout, sort of an executive briefing slide show that we use wherever we go and speak. We're going to work our way through that, with each of us taking maybe five minutes to discuss the GCPR project from our own perspective. Then we'll open it up for questions and answers, and let you direct where we go and what the discussion is going to be about.
What I'm going to do here as we walk through this is review the background of GCPR very quickly, the goals and objectives, the organizational model, the governance model that we're using, the phasing of the project so you get a better feel for that, and then we'll address some of the clinical and technical aspects.
The speaking order is basically myself as the project manager. Dr. Martino will then follow-up with the phasing. Then Dr. Kentsmith on the clinical aspects, followed by Cmdr. McCain on the technical issues.
So looking at our executive briefing handout slide show, very briefly the GCPR history, we began back in January 1998. At that point I was not the project manager. There was Ms. Maureen Coil, who was initially selected to head up the effort. The initial partners were Indian Health Service, the Department of Defense, the Department of Veterans Affairs, and the Louisiana State University Medical Center.
That formed the basis for the partnership. Unfortunately, as the project progressed, a year into the effort there were some congressional issues that general counsel raised, legal issues, and as a result of that Louisiana State University is not a part of the activity at this point, although I believe there is some legislative interest in this.
In January 1998, we started the effort, and then started moving forward with the initial procurement activities, and ultimately that culminated in the selection of PRC Litton as the prime contractor for this effort to deliver the solution for us. That contract was awarded in September 1998.
After September 1998, we have a period of time where memos of understanding, interagency agreements, and other procedural issues had to be addressed. Budgeted funds had to be obtained. So the actual delivery order to PRC Litton to begin work was not awarded until April 1999, just a couple of months ago.
The GCPR project in essence gets it impetus from the presidential advisory committee, PRD Presidential Directive Number Five, in which the president directed the VA and the Department of Defense to work together on the comprehensive life-long medical record. And in addition to that you will see in the following slide that aside from the presidential executive directive, there has always been congressional interest or direction for our agencies to work together.
We do have a long history of 10-15 years of partnering with Indian Health and DOD in our health care systems. In fact, the CHCS system from DOD, the Vista system from the VA, and the RPMS hospital information system on the Indian Health sides all have common roots.
The effort also, in addition to presidential and congressional interest, basically was also moved forward by mutual agreement by the senior executives within the agencies. And it just made a lot of sense for the GCPR project to move forward.
The GCPR vision was established by the board of directors and the coordinating committee is basically to improve the public and individual health status by sharing clinical information. And our goals are to create a collaborative effort to appropriately share clinical information via a standards-based, comprehensive, lifelong medical record. Where no standards exist, the partnership will seek to advance the development, establishment, and adherence to standards.
The GCPR effort that was initiated, in essence a program, then spawned off very specifically something called the GCPR framework project, which is what we are focused on. The GCPR framework project is the first project initiated by the interagency GCPR coordinating committee. This is where we are focusing all of our efforts. There may be additional projects that emerge over time, but this is our primary focus at this point, and that's what we are going to discuss here today.
The GCPR framework project will provide the means to share information between our organizations, between our disparate information systems with the VA, DOD, and Indian Health system. That's basically what this is all about. And then in our slide presentation you will see a couple of graphics there where what is the GCPR framework?
Well, the bottom line it's as an individual moves from private life into the military, and then perhaps gets married and is deployed, and services his country and then leaves. Maybe moves back and is cared for by the VA, and perhaps in some instances will be a Native American then is cared for the Indian Health Service. And in turn there may be private organizations that we subcontract with to provide health.
And as these individuals who are treated by VA, DOD, and Indian Health then proceed through their lifetime, maybe they are recalled because they are in the Guard, or they are in the reserves, there is a need to keep that lifelong medical record and move it back and forth, and have it always available when needed.
The GCPR program's organizational structure, basically the way we have set this up is there is a board of directors that oversees this effort. It is composed of Adm. Fisher from the Department of Defense, Adm. Church from Indian Health, and Deputy Undersecretary Garthwaite(?) from the Department of Veterans Affairs.
Immediately below them is the executive committee. The executive committee is composed of two representatives from each of the organizations. These are senior individuals. You happen to have three of them right here. Dr. Kolodner is one of the members on the executive committee from the VA, and Col. Ray is a member from the Department of Defense, and Capt. Garvie is from the Indian Health Service. We have another member from each of those organizations.
That committee actually guides us, and we report as the project management team to them. Any major policy decisions, any major resource issues, any new direction or strategy, change in strategies is provided to us by the executive committee.
The GCPR project management team is composed of myself as the project manager from the VA, and Dr. Martino from the Department of Defense as the deputy. Then we have two major components. We have our chief technical officer. All the technical work falls under Cmdr. McCain from Indian Health, and all of the clinical activities come under the chief clinical officer, Dr. Kentsmith.
Below them we have a series of working groups and what we call integrated product teams. The working groups are government controlled groups, acquisitions, legal, marketing, architecture, and others. And then working with our prime contractor, PRC Litton and the variety of companies that have teamed with them like Unisys, 3M, AT&T, Sun and others.
We have integrated product teams such as the security team, external interfaces, data management infrastructure, and others. And that's where we work very closely with the prime vendor on the solution that we hope will be prototyped here later this year.
At this point we're going to go into the project goals and the project phases. Let me turn that over to Dr. Martino.
DR. MARTINO: Thank you, Peter. Just to reiterate some of the goals that Peter mentioned earlier, our primary objective is to create an infrastructure, sort of middleware piece or layer, technical environment that would support sharing of patient information across the agency-specific systems.
Each agency has a budding or complete, in various stages of complete CPR system. We're not trying to build another CPR system. What we are trying to do is build the infrastructure to allow the sharing of information between those existing systems. The end in mind of course is to create that patient focused sort of interoperability architecture that can hopefully, through our work with standards development organizations, and by virtue of this being a government and federal project, hopefully will form the basis of or feed into the movement towards the national information architecture for the medical domain.
A related, I don't even want to say secondary goal, but a related goal to that is the emphasis we are trying to put onto our modeling efforts. We have or are developing what we call the GCPR reference model, which would include both information model components and terminology model components.
Again, what we are trying to do is act as a catalyst for existing activities that are actually going on worldwide; not only nationwide, but worldwide. So we're certainly not starting from scratch in any sense. We are trying to leverage agency assets and other activities that are in the civilian, and again act more as a catalyst and forum for bringing those activities together and coordinating them.
We have broken the framework project into several phases. Peter has already mentioned the initiation phase, which lasted about 12 or 13 months. And the activities in the initiation phase are listed in this slide. On bullet that actually is missing is one of the first things we had to do is establish more formally the sharing agreements between the agencies, and the processes by which the agencies would partner.
We had had a variety of partnership arrangements on a more limited level, and GCPR really was the first time we tried to bring those all together between the VA, DOD, and Indian Health Service.
MR. BLAIR: Dr. Martino, for my benefit, since I can't see, if there certain items on the slide that you want in particular for me to be aware of, could you just briefly mentioned them. I don't want to slow you up.
DR. MARTINO: That's quite all right. As I said, most of these things were already mentioned by Peter Groen. Once we established the interagency sharing procedures, the bulk of the first year was really spent in articulating the objectives of the GCPR, which culminated in the statement of objectives that formed the basis of the acquisition of the prime vendor, and basically finalizing project planning documents. Those were the activities of the initiation phase.
The next phase, Phase I is the current phase, so that's where I'll give the most details. Basically, since it is the starting point of our development work, there was great emphasis on beginning the development of the GCPR reference models. This we expect to be an ongoing effort. As you may be aware in modeling this is what we call the long hole in the tent. It does take a long time, but it is a crucial underpinning of everything we do, so we're really trying to emphasize that.
Another government activity is the development of the interfaces between our heritage or agency-specific CPR systems to the framework of the middleware piece. That also would be ongoing as we enhance functionality. Actually, as part of the modeling effort we are also trying to model and build into our models the roles for access privileges as the first step in beginning to create a security infrastructure.
Then we're designing and developing the technical environment in IPTs with our prime vendor. Out of that work, what we expect to happen is by late fall, we expect to have a proof of concept for testing, culminating in early next year, probably in the March time frame, of a working prototype. So that basically outlines the activities of Phase I.
The modeling work that we're doing, as I pointed out before, we're trying to leverage existing models that are derived from a variety of standards development organizations such as HL7 and CORBA MED. We are leveraging whatever sort of current assets the partners have. Each of the partners has done some work in this area.
And trying also to expand our horizon to include international modeling efforts such as those models that are available from the UK, the European community, Australia. For terminology we are also looking at again the usual standards that have sort of floated to the top such as SNOMED and LOINC, and are also very interested in the work being done by Kaiser Permanente with Mayo Clinic in extending the clinical utility of SNOMED terms.
Interestingly, as a follow on to this morning's testimony, our model development is taking place in five basic partitions. Four of those are sort of core infrastructure areas, in areas such as person demographics and identification. One of our areas is medical record architecture. And there is a clear area of collaboration here with the AAMT as they try to define record formats. So we'll be talking later.
We're also looking at continuity of care in a shared environment, and that would be across space and time, again, the longitudinal patient record. And as I mentioned previously, role-based access privileges to support security.
As part of our first iteration and for the prototype we would like to also begin to share data on clinical observations. We have chosen laboratory information as our first area for shared data.
Possible next domains we have been discussing and expect in the next few months to settle on, we have talked a little bit about pharmacy and medication profile, patient summary of care information such as allergies, problem lists, medication profiles, and transcribed/dictated narrative summaries. A third possible area is the area of consult tracking and consult information flow.
In parallel with the government activities that I have just outlined, there are also the Phase I prime contractor activities, primarily the establishment and the prototyping of the key enabling technologies that we feel are necessary to provide the longitudinal patient record.
These encompass such things as a master patient index and record location service, the implementation of role-based security, and what we have entitled information interchange infrastructure, which depends upon such things as the common reference models, and their implementation through the use of what we call semantic mediation, a lexicon to help implement the common terminology that we would like to use to share data.
We also will be focusing on messaging support for such things as HL7 messages, and extending into the area of XML-based messages. So the goal of all of this of course is to support the lifelong longitudinal patient record, both on an individual basis, and also to allow the initiation of population-based studies.
The primary activities or status, where we are right now, in the early spring we have stood up the program management office, our officers named, the support staff acquired. In setting up a collaborative work space to enhance communications since we are geographically separated was a major problem for us, which we feel that we have begun to solve.
We have set up certain program management processes such as government and agency deliverables review from the prime contractor. We're starting to move into requirements management, configuration management, and risk management activities. And we have started to put into place the software development processes using the object oriented unified process. And that helps us bridge the gap between the government requirements elicitation and analysis phase, moving into the design phase that is being done by the prime contractor.
Quickly to just touch base a little bit with Phase II, as we move into pilot tests, what we would like to do is move out into an alpha test facility, a small limited test, but in a live environment, so we can insure adequate user feedback. Our status there is that we have started to draft the site selection criteria. We expect those criteria to be finalized within the next month. And probably within the next 60-90 days, we will determine through the executive committee, the sites for the initial live environment testing, to happen mid next year.
Phase III is our deployment phase, and we expect a multi-year sort of phased development, also coordinated with an incremental increase in functionality as we broaden the framework to handle other domains.
So in conclusion, we think of the GCPR as a utility service. Col. Ray has heard me say very many times the agencies have the lamps in the form of their CPRS. We are trying to build some house wiring to enable the communication between them.
With that, I would like to pass the torch to Dr. Kentsmith. I would suggest since we did a little rearranging of our slides, I believe your copies you should jump to page 15, which will bring you to Dr. Kentsmith's first slide.
DR. KENTSMITH: The two areas that I'm going to talk about have to do with the benefits, both patient benefits and government benefits. Those are covered on the slides on page 15.
I think when you step back and understand how the DOD, Indian Health Service, and VA have been working together to prepare for patients over many years, it suddenly becomes quite apparent why this is a very necessary and important part of continuity of care to have a tool such as a framework, that will help unite the three electronic medical records between the three agencies.
This is the basis for what we are doing. And that is that it is a quality improvement initiative, and it provides a tool for continuity of care for the three agencies. We share many patients and many resources between the three agencies, and basically being able to do this will significantly affect the care that is provided to the patients, and more specifically be able to insure that there is continuity of care.
This is a rather unique event also in that it brings together three very large health care agencies in the government who have a wealth of experience and wealth of patient data and information, and a common purpose, and that is to improve the health of individuals. In doing this, we are bringing the individuals who care for patients, who provide care together so that they collaborate and are able to essentially in the first phases of this process, define and identify what their needs are for tools to help support them in caring for those patients.
Those tools are designed and are established in a way that really support the practice patterns of the physicians, the health care providers in these three large managed care organizations.
We also look at the providing of better management tools when we are looking these days at scarce medical resources, and sort of determining where those resources will be applied. The allocation of those resources certainly are very important, especially when we are looking at limited resources, or high volume, high cost areas where patient care can be improved if we can share better. Sharing is very dependent on the timeliness of the data is collected, or what has accumulated on the patient, being able to get to where the patient is being cared for.
We have a number of good examples of how connecting up the three systems can certainly improve in cost sharing. We look at telemedicine as an example of that, being able to share for example the reading of radiographs between agencies. Having accumulated a panel of experts in certain areas of radiography for example. X-rays can be forwarded to them using the framework to do that, and share that intellectual resource.
Or telepathology, being able to consult directly with the Armed Forces Institute of Pathology by sharing a slide with someone there immediately, using the framework to be able to do that. So these are things we look at in terms of the sharing of resources.
As I said, the real benefit is improving the quality of the care that is provided, especially the continuity of care. Fragmentation of care between the three agencies has been apparent to anyone who has worked in the system, and certainly to myself, having cared for patients for many years, of the patients going between the three different agencies, and attempts to try to get the paper record or some report of what happened in that encounter in the other agency with a patient.
Or just simply having a lab test done for say a Department of Defense facility in a VA facility, and getting that information back in a timely and useful manner to the agency is quite apparent, and something that we are focused on and want to see.
There is another aspect to all of this, and we're pretty aware of that, and that is accumulating a longitudinal lifelong record in a format that the patient can actually read and understand, and be able to be part and collaborate in their own care. Simply being able to take that electronic medical record, and read it as a patient, myself, and correct some errors in transcription, or errors in just recording of the data has ramifications in terms of the quality of care that can be provided to someone.
Other benefits to this in terms of connecting up the three agencies is just in case management. Being able to follow through with the care of a patient once that care has been established in one of the agencies, for example an active duty individual who on active duty develops an illness that is going to be a chronic illness, that will need case management as they then move from that setting of active duty on into the VA, and from the VA on into for example if they happen to live in an Indian reservation, being able to follow that care on with them, and make sure that the preventive measures and other aspects of insuring the quality of their life is preserved.
Other benefits obviously include just simply the possibility of sharing innovations or treatment approaches. For example, looking at the treatment pathways, looking at treatment guidelines that have been established, or have been demonstrated to be effective in the different agencies, and being able to immediately share those between the three agencies has a lot of aspects of benefit to the patient.
Communication between the three agencies through the framework is also a benefit for providers, who in managing a patient, and sharing the care of that patient, often times find this a problem, of being able to share back and forth what it is that they have understood and learned about the patient, and what care that they provided, and what recommendations they have in this framework. This allows us to be able to do this more effectively.
Then as sort of a final comment, the use of the framework to do research, and just taking advantage of some of the agencies that have done research through population studies across their population, now gives the whole three agencies the possibility of being able to do cross-agency population studies that can result in some very significant changes in the way care is provided, and what may be discovered for treatment.
Those are the benefits I think that seem to be pretty obvious, and what is driving this whole approach, and that is our patients, and keeping our patients as the center of the improvements and the tools that we are developing to support that improvement.
At this point I will turn this over to Capt. McCain for his comments.
CMDR. MC CAIN: Thank you, Dr. Kentsmith.
My portion of this panel presentation is the tactical review of the GCPR project. What I want to do is try and make that fairly brief in terms of just my dialogue with the panel here. But then we go into questions, inform this in more detail later on. So just to lay the foundation, this portion of the technical presentation won't be an in-depth discussion of the architecture or anything like that.
But just kind of going back into the concept of what the GCPR framework project does for us. What it does for our agencies is it allows our agencies to adopt a new technology base. One of the things that each of our agencies has had the capacity to do is to basically collect information from other sites within our own organizations, maybe share it with another organization, another facility within each of our organizations.
But we have many, many different ways to accomplish it. I'll start off with maybe sharing the pharmacy data is one of the historical aspects within the VA system. And you have certain methods and so forth to accomplish that. You could share that information within the VA sites, but you really couldn't share that information beyond the VA sites. So each of the organizations has the capability to share information already, but we do it in many different ways, fashions, and at very, very high costs to accomplish that.
So the framework project basically allowed us a new technology, a chance to investigate and choose a new technology, go out to the vendors, the private sector community and say here is our problem. What technologies should we utilize to solve our problems? How do those technologies help us? How do those technologies give us the adaptability and the scalability that we need for the future?
It is very interesting to me that we had roughly seven or eight fairly major vendors out there within the Department of Defense on the DCITMS(?) contract who proposed to the government's problem I guess is the way you would say it, saying how do we share this information, what technologies should we use.
We had seven companies go ahead and submit proposals. What I found very interesting as part of the review team on the proposals was that in basically all of them they shared some common things. It really kind of set the focus for what our solution was.
Number one it was very much standards-based. The solution had to be very much standards based, whether you were talking about a messaging standards or distributed architecture standards, security standards. Whatever, they said you must be standards-based.
Now again, not necessarily should you be standards-based upon standards that you create, but standards that the international community or the private or health care industry in itself goes ahead and does. So that was one of the fundamental tenets that came across in all of the proposals.
Another aspect of things was a distributed object architecture. They came across and said, we want to propose CORBA as your backbone for your technological architecture, CORBA being the Common Object Request Architecture proposed by OMG. They felt it was the best architecture to form the backbone to provide the services for the transportation of the messages and the information in content. So that was another fundamental thing that came across through all of them, was well, we want to use CORBA and many of the CORBA services and distributed object environment.
Another one was security. You have to deal with security in a very, very confidential way, and in a very secure way. So in the proposals there is a very heavy component related to security and provisions of security in this.
The other thing was in communications they said, well, you're in a distributed object architecture. You are talking about sharing information across the globe, across the world to wherever, a number of sites and so forth. You have to have good communications, and good communications once again, security.
Now this is where the government said we will insure adequate band width for you. So one of the things that we have to face with the framework project on the government side is making sure that we have the adequate communication network infrastructure in place to support the things we are trying to do.
So given that as kind of the common theme in all this, I want to go ahead and just talk a little bit about the security component. In the security component what we have done is we have in our modeling efforts, we have a security partition. Now to emphasize the role of security even within the project itself, we have a security integrated product team is responsible for implementation and design of security within our framework.
So our modeling for security is based upon the activities that have been done by the CORBA group in terms of their secure services layer, in terms of what formerly was called the HRAC(?) from the CORBA-MED domains area. We also used a foundation the HL7 security recommendations that have come forward from the security group within the HL7 group. And then we looked at the security models that are being proposed through the European security type things.
We also went and looked back onto this group's reports on security, and said okay, within our framework we need to be able to accommodate whatever security requirements come forth from this group as recommendations for the HIPAA legislation things on that.
So we have gone ahead and we have identified many, many different aspects and characteristics of the security that we want to put into place. We're in the process of going ahead and modeling that activity.
The other aspect of security though that I think is fascinating is in anticipation of legislation that may be coming regarding the patient's rights to control their data, things like that, in our security model we are modeling for the capabilities of a patient being able to specific whether or not what information can be shared with what entities and things.
So our modeling effort is already anticipating as much as we can anticipate at this point in time, but the primary thing is that we put into place the process that allows the flexibility and adaptability to change those models, and change the processes as they are further defined by your group and other legislative groups.
Talking about the common information models areas there, as Dr. Martino talked about, we have information modeling going on primarily in the five partitions that she mentioned which once again were the patient or person; the security; the patient record architecture; the continuity of care; and the clinical observations. So those are the five primary things.
I think once we started with a very, very good set of standards. We went to the HL7 groups and said, give us your use cases, and give us your models and things that you have regarding these five partition areas. We went to the Australian group, the European group, and the to the UK group saying okay, we're doing work in these areas. Tell us what you have in these areas. How can we adopt what you have? How can we add to what you have?
So again, in all our modeling efforts there is founded and rooted upon a base set of recognized standards, either that were formally adopted or looked like they were going to be pretty good forerunners for making adoption into the realm.
So we have done that. Now the other aspect with modeling though is we have gone ahead and said, okay, well, the government wants to help promote standards in both the international realm and in the national and health care industry. How do we accomplish that?
Well, again the primary ways to accomplish that is by active participation in the standards group, the standard SDOs out there, like the HL7 group, like the SENTI(?) 251 groups, like the ASTM groups and things. Our organizations have good representation on those standards bodies.
But one of the things that we also said is well, in those areas that you may be a little bit shallow on yet, because most of the SDOs tend to be voluntary-based, where you have work going on in between quarterly meetings or something like that. Or there is fair amount of work on quarterly meetings, the standards development sometimes can be a little slow.
So we said in those areas where you may or may not have sufficient resources, or when we look upon the areas and it doesn't quite have the depth and breadth that we need, what we want to do is we want to feed back what we had to do to help meet the informational exchange methods that we adopted within our project.
So one of the first things that has happened of course is in the HL7 standards group there is now what is called a government projects special interest group. This special interest group is a special interest group that is comprised of not only government entities, health care entities like CDC, VA, IHS, DOD, but it also has membership from the private sector, Kaiser Permanente, others in there, and it provides a forum for articulation of common issues and common challenges that we all have in trying to put our brains together and say how do we accomplish that.
So in relationship to the modeling activities we did then is say we know that there are going to be certain areas that we are going to want to contribute to the HL7 models in certain areas. And the HL7 group basically said, the group project SIG will be your primary forum to accomplish that.
You, as government agencies, will go ahead and get together in the government projects SIG, define the modeling efforts, discuss those. Then come to some sort of concurrence among those. Then once you reach concurrence, you go ahead and move those through the HL7 technical committee, the standard for model adaptation and adoption of things. So those are kind of some of the areas and some of the ways that we're dealing with the modeling itself.
With this I kind of just want to go ahead and finish up the technical and go on to wrap up.
MR. GROEN: I think we're sitting here with sort of a wealth of experience and information that we can share with you. We are more than happy to sort of draw to a very quick close here, and let you actually start asking us any questions. Maybe we can help you out.
The person sitting behind me, who has been driving the slide show, I want to make sure that she is properly introduced. This Sylvia Endicott Sullivan. She and I both served as the national project managers for the highly successful roll out of the VA's computerized patient record system that is now basically 90 percent or better complete. So Sylvia is just joining me here on the GCPR activities.
So let's stop.
DR. KOLODNER: Peter, there is one other slide that you have in the handout here, and for those who are on the Internet and for Jeff, it may be worth just summarizing, and that is the one that says GCPR supports other major initiatives. I guess the implications of that. I think that might be worth another few minutes. It is on the top of page 16.
MR. GROEN: One of the things to remember too is this slide show and the GCPR Website are available to those of you who would like to go see that at www.gcpr.gov. And then also on the VA's Website at www.va.gov, you'll also find the live testimony from this organization.
On the other initiatives and Dr. Kolodner and others can jump in here, there are a number of areas where this project is going to have an impact on other activities that are simultaneously taking place. Certainly within the VA we have efforts underway to insure that discharge physicals or the exchange of information from patients from the DOD system and the VA system take place in a more orderly manner. That's always been a problem for us, and this GCPR framework is going to provide us the solution in that area.
There is a variety of other technologies or activities that we are actively involved with like the smart card activities, and this framework, and a lot of the standards that we are going to hopefully employ within the GCPR project will facilitate our efforts in those arenas.
Certainly the Web-based solutions as we move to providing on-line patient access within the VA to the veterans so that they can gain access to their benefits and medical record information. The GCPR framework will again facilitate that activity. Those are projects that are already underway, and at various stages of development and implementation that will benefit from this project.
The MPI arena, we touched on that, but we are all struggling with a variety of solutions. Again, the framework will facilitate and bring I think our efforts together in those arenas. Certainly the introduction of the PKI technology. That too is a VA-wide initiative that we have underway.
These are a series of funded projects that exist not only within the VA, but within the partnering agencies that we're all going to benefit as a result of the GCPR project, and we'll see some significant advances there.
Are there other things that you had in mind?
DR. KOLODNER: I just was looking at the bullet there.
MR. GROEN: So why don't we just stop here. As a project team we have the expertise from the clinical to the technical to management. We'll try to answer whatever questions you have.
MR. BLAIR: Thank you much for this very informative briefing that you have given to us. I think it's important for our committee to be able to hear this information, because the GCPR project is an example of a project that is trying to take advantage of the standards that are developed and are evolving both nationally and internationally, and trying to see how they can use these standards to be able to develop a fully functional, multi-enterprise electronic health record that would be a core of an information infrastructure.
You could see from the testimony the things that have been necessary for them to do, because there are gaps in these standards that exist. There are difficulties in getting some of these standards to coordinate, and to be interoperable. So this is an outstanding opportunity for our committee to wind up seeing what needs be done, and to wind up trying to pursue this vision.
With that comment, let me solicit questions from our committee. After we get questions from our committee, I would like to expand it to other attendees that might be here as well.
DR. COHN: Maybe I'll start. I think we still be at the stage were we are sort of seeking to understand. Certainly I'm hoping that that will not take of all our session though, because the reason why I thought we needed to have the GCPR here had to do with the discussion around what sort of collaboration exists. So I don't want to just have us spend the whole time trying to really understand what you are doing.
Having said that, I am perhaps lacking a piece of vision or understanding, and maybe you can just articulate that for me. It may be a change in vision as you move forward. But as you talk about moving information back and forth from systems, and between health care institutions, I do recognize that there currently is a standard. It is called the fax machine generally.
I was trying to figure out the endpoint, as well the interim steps that you are taking to show functionality that is greater than a fax machine. For example, in the piece that you talked about, I was trying to figure out in my own mind whether you were going to be sending information from the DOD system to the VA system, and have it be of sufficient quality that it could be run through automatic support algorithms to identify preventive health care alerts and all of that. Which is a whole level of sophistication much, much deeper than a fax machine, or even sending things in a format that can be read.
Where is the vision of the GCPR in relationship to that? What are the requirements?
DR. MARTINO: I think the vision is to move information electronically, because clearly the paper record doesn't really support our business needs to be able to quickly access and utilize and save for research purposes or larger studies the kinds of information that we need to perform those.
The presidential directive -- I don't know if it said outright -- but is looking at such things that we used to call the service member life cycle. And to be able to understand what happens and how it happens to the patient in the force, and is coming in the civilian sector as a reservist, entering military service, being deployed, all sorts of things going on there. What we did in terms of immunizations. What experiences has he had there, exposures. Then coming back to the states and returning to civilian life.
Being able to create a longitudinal record across those activities, and have the information available for study and understanding of what happened to that patient as an individual, or as a population group I think was one of the primary drivers. I think that that need is not unique to the federal agencies. There are analogous requirements I think in all sectors of health care. So it's more than just faxing.
I think again, we are focused on the enabling technologies so that the information as it does move between agencies is truly meaningful. It carries the context in which it was captured, and the true moving also moves with it and is saved. So that you do have a longitudinal picture.
DR. KENTSMITH: I was just going to add a comment in terms of just as I had emphasized, the continuity of care. And having a longitudinal, if you, as a physician, have had the experience as I have had of trying to do a review on a patient, and having to depend totally on the patient's recall, or trying to figure out where the record might have been, the paper record, if you have it faxed to you, to determine what part of that record to have faxed to you is sort of what this is all about.
If you can accumulate over time, all the places that the patient has been cared for, and have that then be available to you in longitudinal, in a chronological manner, that certainly changes totally what your care is going to be for that patient, and may significantly affect your ability to diagnose and treat that patient, and as I mentioned, be able to provide case management and ongoing care as that patient is handed off to other places.
DR. KOLODNER: Jeff, one other thing. Because those of us who are staff to the committee were actually there at the beginning, I think it may be helpful to also answer Simon's question.
The intent as we moved forward, was to find the common ground. Part of that was to separate the CPR, the concept of the record and the data from the systems that delivered it and that the clinicians touched. The kind of decision support you are talking about is vital for improving the health care, but we decided to do, because we had different systems and couldn't just slam in an application level to those that we knew would operate in all of the environments here, and especially when LSU was there initially, and we hope they and others will be there in the future.
That we felt that we had to rely on the CPRSs, the systems that were there in each of the institutions. But we wanted to feed them the information, so that they would be fully capable of anything that is inherently built in. So for example if Reagan's tree for LDS were to ultimately be able to have an API, a link into GCPR, that the data that comes across would be translated into a form that would be able to be used to the full capability of those systems, whatever they might be.
And the fact that it comes from a system that may not have those decision supports built in wouldn't limit the use of that information. So although we knew that many of us have been working for 20 years knowing that we wanted the information, we wanted the decision support, we want that fine interface that doesn't get in the way of the clinician-patient interaction, but that is fully integrated with the interaction, so it gives us the decision support at the time that we are collecting the information.
Although most of us don't have that system in front of us right now, we knew that having the data there as data, rather than as a report, would allow it to be integrated into the systems to the degree that they are capable.
MR. BLAIR: Let me just acknowledge one thing. When you finished with your testimony and I made my observations of the benefit of the information that you are sharing with us, Dr. Cohn pointed out an additional purpose of your being here is also for us to begin to explore how we could collaborate and begin to also do things that may complement or support your activities.
DR. FITZMAURICE: Let Jeff and Simon, I also look to emphasize the synergy that exists between the GCPR and the Computer-based Patient Record Working Group. We're both interested in patient medical information. We're interested in the standards for that information versus in using the information to achieve a functionality within the capability of the systems that receive it.
I'd like to be able to say that I understood everything that you said, but I didn't. I think there is a learning process that has to go along, at least in my mind. But I have some specific questions. They don't require long answers, but short answers.
The first one I would like to ask is of Janet Martino. You talked about looking at reference models, CORBA-MED, HL7. You currently have a contract with HL7 to use its reference information model for GCPR?
DR. MARTINO: I think Jim is more aware of those positions.
CMDR. MC CAIN: With that the HL7 group does not require any licensing to utilize their information model.
DR. FITZMAURICE: So you can use it freely without payment?
CMDR. MC CAIN: Correct.
DR. FITZMAURICE: That's great. And is that the major source of your own information model?
CMDR. MC CAIN: No, it is not the major source. It is one of the major sources that we are working with and collaborating very closely with. But we also are deriving a lot from the European standards too, the SEN251 groups and things.
DR. MARTINO: I would think most important in our collaboration with HL7 is our intention to utilize their established processes for creating the standards. We can feed in what we are doing, but there needs to be buy in and acceptance from a wider community.
DR. FITZMAURICE: That leads to my next question of David Kentsmith. As you identify needs for standards for patient care that need to be developed that aren't already there, is it possible you would support existing standards development organizations to develop those standards for you, who go through the process, even though it might take a little bit longer?
DR. KENTSMITH: That's very interesting that you should bring that up, because I talked about collaboration between the three agencies and bringing the subject matter experts together, physicians, nurse practitioners, registry record administrators, a number of people who are focused on patient care and providing for the patients.
In that collaborative workshop, the modeling workshop, what grew out of that was a recognition of the importance of standards, and a recommendation that standards be used. This comes from the physicians and the clinical users. So there is an awareness that there is a tremendously wonderful opportunity at this point to be able to do that. Certainly the large number of our clinical subject matter experts really were in support of that.
DR. FITZMAURICE: I see the opportunity as well.
CMDR. MC CAIN: Can I just offer a real concrete follow-up on that? For example, within the CORBA-MED domain there is a service called COAS, Clinical Observation Access Service. One of the things that they do is they work in conjunction with the HL7 group, particularly the vocabulary group and the patient record architecture group on trying to specify our term templates for queries of clinical nature.
We have started discussions with Dr. Stan Huff to go ahead and utilize some of our subject matter experts from our three agencies to help populate those templates, and fill the COAS templates a little more in depth.
MR. BLAIR: Could you tell us what the COAS acronym stands for?
CMDR. MC CAIN: Yes, Clinical Observation Access Service.
DR. MARTINO: Clinical observation being a very broad, generic term to account for anything one can touch, feel, measure, or obtain from a patient.
DR. FITZMAURICE: My next question is for Peter Groen. Since we're interested in patient medical record information that roughly extends to the sites, you mentioned the smart card technology being used in GCPR. There is some feeling that smart card technology might be best used for identifying who this patient information is about, authenticating if this is the person who should have access to the information.
But not to carry the information around so much, because often times when you see a physician, the information you want on the card doesn't pertain the physician's office, lab tests for example; although the emergency care, blood type and some of the information as you have spoken.
What are the effective uses you see for the smart card in the US in the private sector or the military sector? I don't want a long answer. But what do you think is the best use of the smart card with the current technology?
MR. GROEN: I think if we got into any major discussion of smart cards, I'm probably the wrong person to talk to. We do have a number of specialists, such as Mr. Dan Maloney, who is in the VA, who is our representative to the G8 community on smart card technology. He can probably answer that type of a question a heck of a lot better than me.
I know that the GCPR framework certainly is going to have an impact on the VA and the DOD's use of smart cards, and certain is going to make us step back and think things through perhaps a little more thoroughly or differently. One of the benefits that we are anticipating from the GCPR framework is that like you said, if you need information and you need complete information, and you need it up-to-date, hopefully the framework will give us the tools to do that to those very small, remote, local points of care, whether they are out on the battle field or in an Indian Health reservation.
In that instance it may lead us to step away from trying to populate the smart card with the level of detail that you mentioned.
DR. FITZMAURICE: I assume since you mentioned Dan Maloney at the VA is the smart card specialist, that when we get involved in that, we could probably call him up and ask him for his expertise on that subject?
MR. GROEN: Yes, he sits on some --
DR. KOLODNER: He'll be here Thursday.
DR. MARTINO: For DOD the one place where smart card might be complementary to the GCPR is as Jim pointed out, the GCPR does require a fairly robust communications infrastructure. We know that there are a lot of parallels between DOD and the small, remote Indian Health Service sites, where the communications are less than optimal. And I think there is a role of smart card for back filling.
DR. FITZMAURICE: One last question for Cmdr. McCain. You talked about the security system that you wanted to put into effect, and yet elsewhere in the presentation you have a patient taking charge of their health care. What are the aspects of the security system that would allow patients to gain access to the patient medical record information? And will they have access in the right time, the right place for them to make the decision, as well as the physician having access to the information to make a decision?
CMDR. MC CAIN: Let me clarify that you basically enter two different levels there. One, you talked about where patients have access to the information and viewing of the information, things on that order. The primary aspect I was referring to was where the patients could control what aspects that would be made available to the other health care providers. So that's an important distinction.
I don't want to say that currently within our model are we developing anything that says the patient is going to be the one pulling it up, although technically and architecturally that could be accomplished.
DR. FITZMAURICE: Particularly in the military, people carry their own medical records around. So they are responsible for their own medical records. If they have access to it, they could look at it any time they want to. I guess it remains for the future to determine how that functionality would carry over into the GCPR project.
CMDR. MC CAIN: That would be correct.
MR. MAYES: Yes, I think that I have rightly characterized the GPCR project feels that information models and common data representation structures is pretty core and central to the entire framework project. What I would like to get is just an opinion based on your activities so far as to one, would it be desirable, and two, would it be feasible to begin to develop the processes, not a specific model or specific data standards, but the processes for achieving this type of activity on a national basis?
You are very large organizations, and I would think that some of the experiences that you have had to date might be applicable. Do you think that we could in fact have a national health information model? And that we could have national data initiatives? I would assume that there might be multiple types of initiatives to look at doing this type of work on a national level?
MR. GROEN: I would say that our initial work that we are doing with the HL7 group is our initial foray into trying to establish whether or not we can work in some sort of national modeling type activity role from the government, along with the private industry. Where we are with that is that we are just now getting through the first iteration of our information model on the GCPR side. So we still have a little bit of time yet that we need to work through those.
Once those are through, we'll have done the gap analysis between the HL7 models and ours, and then we'll take that back through the process that we have articulated earlier through the GCPR government SIG process. And that will be kind of our first test of whether or not we are putting into place a process that works for large national data modeling type activities.
DR. MARTINO: We're protoyping processes as much as we are technical approaches.
MR. MAYES: Right. The focus of my question was not so much will you come up with a national model. There are lots of models for different purposes. But rather are you discovering any interesting intricacies in the process of trying to do things at a national level?
CMDR. MC CAIN: One thing we are discovering is that in many ways we are finding that the other national modeling groups like the UK and so forth, in many ways would rather work with us initially than work with some of the other SDOs and try to kind of harmonize through us and then to on to the other SDOs.
DR. MC DONALD: I think I should point out that the SEN models, there is not a common model. It's a bunch of separate models. I think if you explored the industry in Europe, you wouldn't find very heavy use of them.
I think similarly, these things are developed in absence of the real developers, such as some governmental models are often of no interest to anyone. So they are trying to sell them to you guys, instead of the people that should using them over there. It's possible. I think you have to be a little bit cautious about that.
The question I really wanted to ask is about security. That is you didn't mentioned IPSEC(?) or IP6 or the things that are coming there, and whether that's a deep part of your planning. IPSEC, the IP security standard, which is actually in IP4 as well as IP6. It's a beautiful thing. I think if you haven't, you should. It's quite glorious. It's supported by many of the routers already. It will talk to certificate servers for HTTP and it's really quite nice.
MR. BLAIR: One of the activities that you have described or alluded to is in developing this framework, to be able to knit together all of these standards so that you wind up having an information infrastructure where you can actually build a practice solution is the fact that you have not been able to just pick standards off the shelf, and have them work together.
And you have evolved to activities where you have taken a look at an information model which is in the process of being developed at HL7, the reference information model, which is an event-driven model to be able to facilitate the development of message format standards. And you have also taken a look at terminology models, in particular SNOMED-RT, but also including other terminologies.
And in an effort to see how these models will work together, I think it would be helpful for many of the committee members to understand why it was necessary for you to include in your project the coordination of terminology models and event-driven reference information models. Is there someone that could give us a concise explanation of why you felt that this was necessary? I think that would be helpful for the committee members.
DR. MARTINO: Well, I think in the first case, in the information model case, what you are trying to describe is the structure and relationships between the information elements or data elements. It's sort of like horse and carriage. The other part of that equation then is the terms that you use to fill those information slots. Those two things need to be tied fairly closely together.
If you have an information slot for family history, and another one for diagnoses, then your terminology can just be basically a list of medical diagnoses. If you don't have that distinction made in your information model, you can compensate for that in the terminology model by adding terms for family history of, as opposed to current diagnoses. So they very closely aligned with each other, and one impacts the other.
MR. BLAIR: Thank you for that description. Let me just follow-up with that. That is an activity that is probably very core to having widespread development of electronic health records by independent vendors. You're a large project, and you have the resources to be able to do that. Clearly, it hasn't been done up until now, so that kind of says that many small vendors don't have the resources to be able to do that work.
I would imagine that many of these vendors are delighted that you are doing that work, and will want to benefit by the modeling that you are doing. Could you give us some idea of how long it might be before you have results from this modeling effort that can be proliferated to the private sector?
DR. MARTINO: Basically, as part of our development paradigm we are trying to do this in an iterative way, so that we start first at a fairly high level. I have to say that part of what we are trying to do is not arrive at a lowest common denominator, but rather arrive at areas of commonality, and that can be generalized to many different business needs and sectors, which is very different from the lowest common denominator, which tends to be very limiting.
One of the reasons why we try to look broadly at a variety of models in the international community is that you can't arrive at the appropriate generalizations without having a wide base of viewpoints taken into account. So that's sort of the first stage. Then with each iteration, what we hope to do is not only bring that to the standards development organizations, but then reiterate and keep refining sequentially.
Also, I think one of the most powerful things we offer to a standards development organization is the fact that we are actually using these standards to implement. That's, to my mind, the acid test of how good the standard is. If it's not implemented or implementable, then it becomes a shelfware standard. So we are really trying to help by our prototyping efforts, to actually show that yes, this standard is working.
I didn't answer your question.
MR. BLAIR: You did, because what you said is it's going to be iterative. It's going to evolve, and you really have difficulty winding up saying what necessarily might be available, one, two, three, four, five years from now. But you have got to prioritize in terms of your information needs. So I understand your response.
MR. GROEN: We're driven by a real need. We've got to have a solution. We've got to have something in place. If this project doesn't help us to better treat patients, and doesn't come on line pretty darn quick, then there is no sense in having this project, and I probably wouldn't be associated with it.
We did roll out the computerized patient record system nationwide in the VA in a fairly well organized and very rapid clip, and that's the intent here. We're moving forward to prototype in the March time frame of the year 2000. At that point we will know whether the solution or the track we are on is working. We certainly won't have the final solution.
It will be a series of bills, an iterative process that keeps on growing. But we then move to Phase II, and Phase II is the alpha-beta testing. That is through that next fiscal year into the year 2001. That's where the rubber hits the road. That's when we find out whether this is of use, and whether it really works, and whether right down at the clinical work station level, the physician is truly benefitting, the patient is benefitting.
If that is achieved, which is what we have set out to do, then very rapidly after that we move towards releasing the first version of this. Basically that's fiscal year 2002 becomes the target to begin the phased implementation. That's what we are working on right now, is all of the issues related to phased implementation. And yes, that might be a multi-year effort. It depends on the strategy. It depends on how we go about doing this.
But our goal is to get that out so that we, at least the partners within this project, achieve our objectives and reap the benefits of our work. And certainly as soon as we have something that is workable, our intent right from the beginning is to share this with anyone and everyone that we can.
MR. BLAIR: Dr. Cohn is next. We have about 12 minutes.
DR. COHN: First of all, I want to thank you all for being here. I think it's been a wonderful presentation. I had a concern that we weren't going to have enough time, and I'm certainly reminded of that as I have 12 minutes to solve this problem.
First of all a comment that there may be a number of questions that we send to you in writing to ask for your comments on. I will hope we can get response to them. In some ways I'm concerned, because I've asked for things and I get 90 page reports back. I know that this is a big project and all of that.
MR. GROEN: We have literally thousands of pages of information we'd be happy to share with you on functional designs.
DR. COHN: Those on the Internet I'm sure understand my concern.
The committee, as you know, is working in a number of different areas. The work group is focused on message standards, on terminology, on issues related to data quality, accountability, and integrity, which seem to be key areas that you are also all concerned about.
I guess we have talked about two different areas of focus. We have sort of segmented our activities. One is sort of near-term issues, and then overall, the total solution to everything. And I think one of the things that might be very helpful for the committee is to understand as a litmus test, what are the areas that you are having rough times with?
You are obviously working on things. Some things are probably slam dunks. Other things, probably even the standards development organizations can't help you with. Do you have any comments about particularly vexsome issues at this point from your views?
DR. MARTINO: That's an easy one, security. Security policies. This is the first time that the agencies will be sharing information, or attempting to share information across agencies. If it grows, and we start getting larger with other federal agencies and potentially even moving out to the civilian sector. To my mind it's not the technology, it's the need for the senior leadership of these agencies to determine what will their policies be for sharing data that is lean? And who should get to see what, and when? And remaining cognizant of the patient's right to control who gets to see what and when.
So if there was one thing that would keep me awake at night, it's how are we going to nail down those policies, or at least come up with a reasonable set of recommended policies?
CMDR. MC CAIN: Another area that I feel we've been trying to address, but recognize obstacles is in the area of terminology; trying to map the terminologies to our information model, and which terminologies to use. It has been recognized especially strongly over the last four or five years that the lack of a common terminology through information systems and for the medical field has been a barrier to sending the information with the proper context, what it really means.
Well, there are a number of agencies that are trying to deal with that. I believe your work group is trying to deal with that. I think it's very timely that for example, when the subject matter experts came together and said, well, which terminology are we going to use for laboratory results, the answer was we're going to use LOINC for the test name. Great. That's good public domain type of vocabulary.
But then they said well, we're going to use SNOMED for the results. So now we're right back into the old licensing issues well, okay we're going to use LOINC for the test name, but we're going to have to have SNOMED for the results. And the subject matter experts are going forward again and saying in the whole vocabulary effort, which vocabulary for which piece of data and things like that.
So we are kind of right back in my mind to the same point of where you have got all these different vocabularies out there, with all different licensing issues, with the proper context and meaning, along with the data structure and data information itself.
DR. COHN: Other comments? Richard, do you have a question or a comment?
MR. GARVIE: When we saying bringing the subject matter experts together, I wouldn't want to give you the impression that that's an easy thing to do, as you well know. So maybe hidden in your question there are some real process issues. It's quite a task for our respective organizations to get the appropriate people together to engage in the process from which these would be the desirable outcomes, more to the point of the question.
And in any discussion of this it isn't just our organization, but it's the private sector as broadly as possible. Engaging those people who are most involved in delivering care, let alone patients, and bringing the right people to the table to engage in these processes so that we are assured that they are closer to the outcomes is really a very formidable logistical, organizational kind of task. So perhaps hidden in your question there are some real issues of logistics.
DR. MARTINO: I would just like to add, and this may be heretical to this organization, I think we need to be careful that we don't use standards to lock ourselves in our business practices to such an extent that we lose the amount of flexibility and ability to respond to unique requirements that exist because they are unique.
The language that we use has an enormous amount of flexibility for a reason. It meets human needs, communication needs. So one of the things we are really trying to look at is how we can utilize technology to allow the flexibility that is necessary to meet our needs and our not only individual, but specialty and institutional unique requirements, especially as concerns language and terminology.
Yet, still be able to interoperate with each other through the use of translation services, so that we are not trying to look everybody into using the exact same term for the exact same thing every single time, but rather allow the human need for flexibility and subtly, and yet still be able to transfer information and combine information.
DR. COHN: Janet, I appreciate your comment. I just wanted to make a comment about this. There seem to be a couple of things going on here at the same time. I think when we started out this session I was a little confused about really the role of the GCPR. I think I may have misunderstood, thinking of them more as almost standards developers.
I have come out of this session personally feeling that you are sort of on very on my list as standard users. I really personally think that one of the great values -- I know this work group is going to be over the next several months, starting in the fall, having a number of hearings where we really are trying to pull in users of terminologies, users of standards to get insights as to where the troubled areas are, what we can do to try to help.
I really see that the value here may be a beginning discussion, recognizing that you are going to be going through all the bumps that the rest of us are, and will be able to provide us some insights on that. So I personally really want to thank you all for luminating for me.
Jeff, did you have other comments that you wanted to make?
DR. FERRANS: I have a quick comment and a couple of questions. I guess the comment was about what you said about the GCPR being both developers and users in some sense. I think bridging that gap, looking back over some of the earliest testimony that the work group took, that was one of the issues that people raised about how to bridge the gap between the two groups. I think that this is really a perfect example of how it can be done, and how it's been successful to date, and I think we'll continue to do so.
I did want them to clarify a couple of things that I think will be important to work group and to people listening. One thing is you mentioned that the information model is not the HL7 reference information model. My understanding was if someone were to use the HL7 reference information model, that given the architecture -- and this question is for Jim McCain -- that it would be fairly to map.
CMDR. MC CAIN: I think one of areas of discovery for our project, and that touches on Clem's earlier responses is that when people talk about modeling and information modeling there are very different levels of modeling that can occur. One of the things we had to do was look at what the level of modeling in the HL7 realm was compared to what our needs were. It depends at what level of modeling activities that you want to work with.
For example, in the sense of the HL7 RIM, in trying to gather the use cases for the HL7 RIM which would help us drive down to kind of making sure our models fit with theirs, found the HL7 organization did not have a central repository for the use cases, and they were unable to fairly quickly provide us the use cases for us to go ahead and compare our models against and stuff.
So the short answer there is I do not believe at this point in time that there are really any models out there that can you could just go ahead and just take and run through. You have to kind of compare what your level needs are, and where you go from there.
DR. FERRANS: The other question I had was about licensing. I just looked up some licensing prices last week, much to my dismay. I think this is an issue in the context of terminology. There has been discussion in the work group about terminology available at a reasonable price to help, because that may be a perceived barrier. Do you perceive price -- I know you mentioned, well, there are licensing issues, but specifically looking at the sizes of your organization, do you see price as a significant barrier? Just some general comments on that.
DR. MARTINO: We certainly have a significant bargaining power I would think.
MR. BLAIR: May I do this at this point, because it's a little past 11:30 a.m. I think the point that you are raising right now with respect to licensing agreement, whether in terms of the complexity of dealing with it, or the price burden, especially small vendors, I would like to indicate that the testimony from the GCPR project is extremely valuable. I think maybe three or six months from now it might be helpful for you to come back and give us additional progress on how you are doing in terms of issues that relate to standards.
Agenda Item: Convergence of Ideas on Medical Terminology
MR. BLAIR: At this time I think we need to move on to our next activity, which is the review of the information that we gleaned from our testimony on May 17-18. Should we just take about 30 seconds? I believe that that document has been distributed. Is that what you all have?
Let me just explain to you just briefly. We had I think it was over 30 folks that testified to us on May 17 and 18. Interestingly enough, this was on medical terminology. As we began to wind up identifying the folks to give us testimony, we found there were so many different developers of medical terminology that it virtually took us two days. We had originally hoped to be able to wind up having users, meaning vendors and message standard format developers as well. We found that we couldn't hear from the diversity of developers in less than two days.
So in the fall and later on today we are going to wind up getting to our calendar, and we will have another session where we will have vendors and message format standard developers and other users testify to us on their experience with uniform standards for patient medical record information as well.
Given that, just with the 30 folks that testified to us, there was a tremendous amount of information. So Michael and Simon and I and Bob made an attempt here to pull together a working document to just try to help us convert some of the ideas just in the interim. This is not anything that is final. This is not going to be a report or anything else like that. It's just to help the committee members begin to converge ideas from those two days.
Simon had made the observation that this might be a good thing for us to do for message format standards, and for the information that we got in December as well. So if you find this of value, we may wind up doing thought convergence documents like this in additional sessions. Give us some feedback on that.
The real gist of this particular document is the last three pages. That's really where I would like to drive your focus and attention. Let me give you some idea of why I gave you all of the other pages before that, which was it was an attempt to consolidate. I went through the notes that Simon and Michael had taken during those two days. I tried to pull that together with some of the documents and information, and then tried to make a stab at converging that information. We'll wind up seeing whether you feel like this is close to the mark, or divergent, or whether we missed some things.
But essentially the first thing in there is just a preference. It's kind of a disclaimer, just simply saying this is work in process, and it will change.
That next page there are the ideas that Jim Samino(?) presented to us. Jim Samino essentially wound up saying if we are going to wind up having clinically specific medical terminologies that facilitate comparability and interoperability, then what are the characteristics those terminologies need to have? That is very similar to the work that Dr. Cohn and Chris Chute(?) came up with in terms of a framework for the convergence of medical terminologies.
The next one was the testimony from Keith Campbell where I think it was called "Strategic Imperatives." Again, these are the types of things that we look for in terms of medical terminologies that could help us with the goals of comparability and clinical specificity.
The next two or three pages is an imperfect stab at just simply trying to capture and collect the major observations that we heard during those two days, again from Simon and Michael's notes.
Then following that is the issues that we believe we heard. That's another couple of pages there.
Then after that, another just attempt to wind up saying was this homogeneous or heterogeneous in terms of the medical domains? So it was just a glance at saying are there certain segments or domain areas where there are different types of problems with medical terminology, or different types of progress with medical terminology. So you wind up seeing just a stab at seeing if they cluster in certain areas. Again, that's imperfect. That's only an exercise for us to help us to begin to think through this process.
I'm going to skip the next page or two. Let's go to the last three pages. I think that starts on page 15. In a sense we tried to assimilate all these in terms of saying what do we learn from this in terms of what is the environment for medical terminology and the trends. There is I think like five or six key trends that we observed. We did begin to see that there is significant movement among medical terminology developers to try to become clinically specific and comparable.
We did wind up seeing that in at least certain domain areas there are attempts at trying to coordinate and converge or provide some type of framework for convergence of medical terminologies. That's both in terms of among the different terminologies and in terms of being able to marry or harmonize those with the message format standards. I think Simon used the term both vertical integration and horizontal integration that's going on in process there.
A third key observation is that even as clinically specific medical terminologies begin to evolve, there will continue to be an important need for mapping terminologies back to mandated terminologies, like ICD and CPT codes, which are required for reimbursement purposes. That will continue for quite a long time, and that is not going to go away.
The other things that we observed in terms of important observations on trends and the environment is that there are medical terminology developers that are beginning to come up with new ideas a la David Rothwell and Health Language Center, where he's coming up with some new approaches, and is working on some prototypes. XML is new as a syntax that is being used in terms of medical terminologies.
SNOMED-RT is still fairly new in terms of its idea of being able to use ontological principles to be able to wind up coming up with a way to converge terminologies to basic concepts. So there are a lot of new things that are evolving and emerging right now. It's not a "mature" marketplace yet.
The other piece is that we still see a significant amount of user influence in driving what medical terminologies should be in terms of their usage, their scope, their specificity, their interoperability. The vendors and users of course are still very influential in driving what these things should be. Then finally of course it gets down to the question of what should be the government role to be a facilitator or a catalyst to promote the good works and good efforts that are going on out there.
If you were to go to the last two pages, we made an attempt to wind up saying what were the major questions as to what should be the role of the government to able to facilitate medical terminology development that is more clinically specific, comparable, and interoperable. And I think we have got about five or six of those. We tried to group these.
A number of folks wound up saying we should consider the possibility of subsidizing some of these standard developers, because many of them are small, and the task is large. And if we go down that path, then it raises certainly other questions such as if we do decide that it's a proper government role to do some subsidizing, then what should be the criteria or models that would guide us in the subsidizing.
Another question that arose was should we wind up facilitating coordination? If so, do we do that across the board? Do we do that in terms of specific domain areas? Again, what is our criteria for doing that? Other questions that come up in terms of coordination are should we go through a process of identifying gaps, limitations, overlaps in the process of determining what coordination efforts need to be done?
The next area, Simon or Bob, could you help me with the next one? My memory isn't perfect.
DR. COHN: Should the government recommend we adopt certain TMI?
MR. BLAIR: Here there were some folks that testified that they were really looking to us to do recommendations similar to what we did with the financial/administrative transactions for the first part of HIPAA with the ASC X.12 and ACPDP message format standards and the code sets that related to those, and security standards that related to those. Should we be doing that with respect to standards for patient medical record information as well? So that's another key question.
Then the other area is many people began to suggest that there might be different incentives that we should consider recommending. Incentives either to promote better data quality, or better documentation of clinical information, or maybe on a broader level should we have incentives for better quality of care? Maybe that gets at it.
Or should we have incentives to try to pull through the use of standards in terms of encouraging trade associations or professional associations to adopt, enhance, or work on the development of these standards? So these are other aspects of these questions, and there may be more.
After incentives I think that -- what is our next one that we had there? Many of the terminology developers have limited resources to be able to do the work of coordination. And this isn't quite the same as subsidizing, but it is basically saying should the government do something to facilitate the participation of subject matter experts either at these terminology developer meetings, or to facilitate the coordination covering the expense of key individuals to be able to move these standards forward a little bit more expeditiously.
And then this last one, which was points of view. Should the government adopt certain viewpoints or points of view in what we recommend to the secretary? A la should we wind up saying that we need to have revolutionary change, or should it be evolutionary? Should it wind up endorsing or adopting ideas such as reference terminologies and interface terminologies or mappings?
In short, this is the working document that we have come forward with. I have sort of stepped you through it really rather quickly. If this is of some help to us, I think that what I would like to invite you all to do is critique it. Is this converging? Is it helpful? What did we miss? Where are we off the mark? What do we need to change?
So I'm opening it up for critiques and comments and improvements, or you could wind up saying this is off the mark.
DR. COHN: I was actually going to make sort of an introductory comment. I think overall -- I think this is almost like the working set of issues, is what I would start with, and what we heard at the last meeting, without saying that we necessarily have solutions. I know I was sort of hoping at our September that we could spend more time going through all 13 pages of it, and probably also since I see Ms. Humphreys at the table with us, I was hoping that we could have her make come critiques and comments based on her understanding of things, and hopefully to come to perhaps a more common understanding.
I think it's a terrific document. I can't say yea or nay on any of the issues per se, but I personally think these things are very helpful. Do others have comments? Betsy?
MS. HUMPHREYS: I think it's a great document. I think it does raise the major issues which I gather were raised by yourselves and others at the meeting about what some of the options would be in terms of your eventual recommendations about what the federal government should or shouldn't do.
I mean I think it's a very interesting job, and I'm awfully glad you did this, because I was terribly sorry I couldn't either appear or listen. Now I've read the abstract; maybe I'll skip the paper.
DR. MC DONALD: I think there are a number of minor comments that I won't make now. There are corrections like the statement about NDC being not scalable. Well, it does cover every darn thing. So I don't know if that's exactly true. I know where it came from.
I think there is another dimension that isn't addressed in this. I think there are some subtle negative effects of government processes on this whole vocabulary thing. We recently have been reviewing how we can go from medical record statements of problems to codes. And it all works as long as we can live in parallel universes, and not have to fuss too deeply with the CPT.
It also works if we can go from a more specific term to a less specific ICD-9 term. But the subtle thing about government processes is -- and I can't find the track, I can't find the complete chain, but I am told that because the AHA and some other groups have said that one must code to the most specific detail, that then locks into the regulatory effect on the billing and fraud.
So therefore, if you convert something that is very specific like -- there are a zillion of them -- but the one specific one like melanoma. You can't code melanoma specifically into ICD-9. You've got to code skin tumor. But then you must code at four levels that aren't part of our consciousness, like is it remission? Is it a severe case? There are a bunch of funny categories that if you followed the regulation, you would have to do, which I don't think regular physicians think that way.
Or if I record depression, like I record on 20 percent of general medicine patients, that's all I know. I know they are depressed, and that's all I know. But they don't like you to code just depression, because there is remittent, unrelating, severe, sometimes intermittent depression, and there is always intermittent, and often not so bad depression.
There are a bunch of what I consider completely nonsensical categories, but maybe if we had some psychiatric helpers we would understand where they came from. But we are going to get bound into some nonsense I think if these links aren't broken with some of the current specs in ICD-9. I think we really have to be careful, because they are very subtle.
Everybody in the house says we go to fraud if we don't -- I can't find the regulation that says that, but they believe that chain of evidence is there, that we have to do these crazy things that don't make any sense in clinical context.
So whatever we do, I think we're saying blindly that we should say that government should not interfere with the proper practice of care in terms of the coding we do in managed care by some of the odd things that have evolved in some of these coding systems.
DR. COHN: I certainly am hopeful, and recognize that we have only heard from terminology developers so far. I would be hopeful that we would be hearing more comments of what is really going on from the user perspective in the future.
DR. MC DONALD: Most of us are completely unaware of this. I stumbled into this because we are dealing with the computer in this new ferocity about fraud. Mistakes are fraud now, so you go to jail. I think the penalty in the books for this stuff is higher than the average actual jail sentence for manslaughter in this country. So it's kind of gotten really weird I think. I have been stumbling into this in the last four weeks, these linkages in the codes are really crazy in some areas.
DR. FITZMAURICE: Has somebody ever been fined for a coding error?
DR. MC DONALD: No, but I think that there have been some big, severe fines charged against university hospitals for some of which were really just mistakes, $30 million.
DR. FITZMAURICE: Even in admission to recovery of excessive reimbursement?
DR. MC DONALD: Big time.
DR. COHN: Dr. McDonald, are you then supporting the move towards less specific terminology?
DR. MC DONALD: No, I'm saying that there is an evolution, and all of it has reasons. You see how it evolved for the various purposes of reporting for mortality and morbidity, and all the sudden now -- and it's all been fine, because we had our problems lists, and then someone else does the coding. That's when the hospital spends $10,000 on admission(?) and you spend $150 on coding.
Now the lock that closes on ambulatory care, where you've got two minutes to see a patient and 20 minutes to do the documentation, and on top of it if you kill a patient you go to jail. It's just really getting crazy. It is subtle, because where the lock is just closing now, and the specificity of saying that you must code to the most specific, which is a recommendation in some formal document that I'm trying to argue against, because it creates this lock. It changes the whole dimension of it all.
DR. FERRANS: Is there a question in here about the committee somehow addressing some of the problems having to use billing and reimbursement codes with regards to clinical content versus more clinical code sets? I don't know if we can make a clean distinction, but the issues have been raised. Certainly Chris Chute has raised it in his testimony to the work group.
DR. MC DONALD: This thing has a number of hints in it that the clinical coding should define -- the problem is when the specificity goes sort of orthogonal to the specificity that you might do naturally in care is where you might get the problem. As I say, you have this funny orthogonality, and you have some specificity that just doesn't really make sense.
Whereas you can't say most lab tests have one code. You can't say those. That's okay, and as long as we're allowed to leave ICD-9 not at it's most specific, and let us be specific to the level of care that is necessary, we would be better off.
MR. BLAIR: Could I try to paraphrase your thought, because I would like to capture it if possible as we go forward. I think what you are winding up saying is that while we have had many problems in health care up until now because we were not able to capture things with clinical specificity into computing systems, or it was difficult or limited to do so, and we have driven forward to improve clinical specificity, what you are saying is that as we go forward, we need to come up with a balance. That we don't wind up making clinical specificity a god or a mantra, and wind up going to an extreme. Is that correct?
DR. MC DONALD: Well, more than that. We shouldn't let non-medical definitions of specificity control the medical definitions. And we should be pragmatic. Let me give you two other examples of crazy stuff in ICD-9 which I have just become aware of.
One is other one is if you record diabetic necropathy as a diagnosis, you must also record diabetes as a diagnosis, which is like saying pizza pizza for Domino's pizza, because you have already said it in the semantics of the definition. That's also a fraud and a misbilling, all kinds of other implications of that. That's somewhat nutty.
The other one is that you must record -- and I think this is true, I may be as sure of this -- you cannot just say they have diabetes. You have to describe the level of control. That makes sense on the surface, but I have a hemoglobin A1C in my chart, which all I need. I don't need to do discretizing on the basis of no good rule to describe the level of control. So that if they want me to control the level, let me send them the hemoglobin A1C.
There is a mismatch now. The artifact that created for other purposes, and it is now dominating all of the decision processes.
MR. MAYES: Because it's HCFA that you are talking about. And I think that you've got it somewhat confused. I think a lot of the drive to specifying coding to specific levels is being driven by usually the most specific code pays more. So I don't think that's HCFA telling you to specify that. What they are saying is if you code to that, you better darn be able to document it.
DR. MC DONALD: I accept that.
MR. MAYES: It's not just the government.
DR. MC DONALD: It's our medical record librarian stands on a table with a gavel and says we're going to all go to jail if we don't do this. Now it may be confusion on the part of the industry. I don't say it was HCFA. You can track through it. The government says you must follow the rules of the industry, the American Hospital Association, et cetera. They say you must do specific coding. So I think it all gets locked together in that.
DR. LUMPKIN: I think I'm a little bit disturbed by the discussion, because one of the things that we haven't really had a discussion about is what are the principles involved in the process. I believe that one of the principles that needs to be involved is that the purpose of a medical record, the primary purpose is to convey medical information from one practitioner to themselves somewhere down the road in the stream of time, or to another practitioner in some other location who are distracted in time.
That should drive the level of specificity, because actually that's going to be a much higher specificity. And if we get concerned about the fact that the whole finance system is built upon, and we build it for the finance system, but it doesn't meet its primary, principal goal, then I think we are going to come up with a system that will be nonfunctional when the primary financial system changes, or if it changes. And we're certainly seeing that that's in transition.
So I think that one of the things that we may want to use as we go through the list of questions is some basic principles on how we decide the answer to the question.
MS. HUMPHREYS: Without this particular case, I know exactly what Clem is talking about, having looked at this from a variety of angles in terms of mapping terminologies and concepts in the Medsource(?). Where suddenly you know this concept is very clear to you. It means this specific thing. And somebody says to you, well, does it belong into category A or category B, which is some higher level construct, which is totally meaningless to you.
So even though you know what this is, since you don't know what non-A, non-B elsewhere classified, not otherwise specified, XYZ disease is, you don't know what category that is. Even though what you know this one is, you don't know whether it falls into this category or the other category of that, which has a similarly gross organizational name. You really have to say it's going to take an hour and a half for us to figure out whether this is logically in this category or that category.
And even though those type of examples don't come up that frequently, what you are saying is people have gathered all the information they need to practice care to another physician, a patient, a nurse, whatever. We could all understand this perhaps. But that is insufficient. Even though you know everything about the episode of care, it is insufficient to make the correct selection automatically or in some easy way of this other code, because that has been organized by some other means.
I think that the long term goal should be that if you correctly represent the care using some relatively standard approach to doing that, that you should be able to generate the categories that are needed for statistical or billing processes without another intellectual activity beyond that, which is to describe what really happened to the patient.
We are not there yet, because we haven't got the detailed terminology, and we haven't used that sufficiently to know where the disconnect between it and the existing categories are.
DR. MC DONALD: I think the model is broken though. I think it's deeply broken. I basically would like to pray every night, don't make me use ICD-9 to say what is wrong with my patients as the principal mechanism, because of all these funny situation. We really shouldn't be deciding that the whole chart should be coded as discrete elements because we have numbers we can send that give that information much better. And that's part of the model broken. They are trying to say everything in the code.
MS. HUMPHREYS: Yes, and I think that the issue here is, Clem, that until you have an appreciable number and experience with computer-based patient record systems that have some standard approach to a more detailed level of clinical information, which we do not have today, then we can't say here is the better approach. I'll send you this type of standard data. You will be able to figure out what is wrong with the patient, and you'll be able to figure out whether the care was appropriate.
I agree with you that what we have doesn't do it, but we don't have the replacement system.
MR. BLAIR: We have two other folks waiting to ask a question. Simon and then Marjorie.
DR. COHN: I'm not sure I'm asking a question. I'm just sort of making a response. I think you are bringing up a really important issue. I'm not sure that it's -- it feels to me a different issue though than what this work group has been discussing. I would also think of it as more of an issue around payment policy, and as you commented, fraud and abuse legislation, and having to do a lot more with what I would describe as reimbursement codes.
As we were looking at the actual work activities of the overall subcommittee, I think one of our jobs is to monitor and hold hearings about implementation issues having to do with the HIPAA administrative transaction standards. I think you are describing things that fit fully within that. We may want to reflect that as we look at the subcommittee activities tomorrow in the work plan.
So I think that there is probably an issue. I'm not exactly sure what it is still, except these other comments. But I think that we may have an alternative mechanism to begin to vet the issue. That's just a suggestion, and we should make sure to bring this up again as an issue tomorrow.
Marjorie, you had a final comment?
MS. GREENBERG: I agree with you on that. I think some of these specific issues may require more in depth discussion with the people who are responsible for the different policies and the coding guidelines and everything else, the identify the extent to which these are really systematic problems that could be addressed in some way.
I do think you have to differentiate maybe as Bob was saying too between what people think they should be doing, and what in fact people are told to do. Because the coding guidelines for outpatient care very specifically say to code to the level of specificity that you know, not to code to the highest level of specificity that you can find in the code book; but to code to the level of specificity that you know.
Those are the guidelines. Now as Bob said, maybe people know that at that level of specificity you won't get as high a reimbursement as you would at another level, so it's affecting things. But that is the guideline on specificity.
MR. BLAIR: Let me close this discussion. I wish we had more time. I think I'm going to extend our lunch break to 1:10 p.m., then we'll begin prompt at 1:10 p.m.
Thank you everyone.
[Whereupon, the meeting was recessed for lunch at 12:05 p.m., to reconvene at 1:10 p.m.]
MR. BLAIR: We're about to reconvene. We have two segments this afternoon. These are essentially business issues and administrative issues related to the CPR work group. Our agenda we'll be stepping right now, and we tried to organize this is in some semblance of order and sequence, but we may modify that if some additional information is available to us, or we want to be a little bit flexible here.
But generally our tentative plan is to cover the principles. Michael, what did we call them? Guiding principles. The guiding principles were developed when HIPAA legislation was new. It was almost a year and a half ago I think when they were put in place. That was to help us as we were going through the recommendations for the standards which be adopted for financial and administrative transactions and the code sets and identifiers related to that.
So we'll hit that first. As a matter of fact, for those committee members that are here, you'll find that there is a hand out. All of these things you should have a copy of. Once we have gone through the guiding principles, to sort of refresh our memory in terms of what we are trying to stay focused on within the NCVHS -- this was broader than this work group. This was basically NCVHS guiding principles.
The second thing is our CPR work plan. Bill Braithwaite had made a couple of observations on our last version of that. We want to be able to cover that. And the next item after we have gone through that work plan and looked at the focus areas, is to take a glance at the remaining calendar days and come to an agreement as to what things we would cover with the days that we have left during this year for either hearings or other activities.
That ought to bring us then to our three o'clock break. After the three o'clock break, if I recall correctly, we should have on our agenda a discussion of whether or not some of those issues in terms of the focus areas in the calendar bring up some budget issues, or whether there are other budget issues that we need to consider or discuss.
That would be followed by I believe it is Richard who pulled together definitions of patient medical record information gleaned from many of the folks that testified during the first half of this year.
Then finally that hopefully will put us in a position where we could then begin to consider what are the things we want to include in an interim letter to the secretary of the Department of Health and Human Services. Actually, I might ask Michael at that point maybe to take the lead. We have a brief outline on that, but I think Michael will take the lead on pulling that letter together. So maybe you would be the best person to lead us through that portion of our discussion here.
We don't have hard and fast time lines here. We have about an hour and 45 minutes now for the first section. So if you were to take a look at the guiding principles, Michael and Bob, both of you pulled that together. I think, Michael, you took the lead on pulling that together. Michael, would you like to kind of give us a background on those guiding principles and the thinking that was there?
DR. FITZMAURICE: I'd love to, Jeff. The guiding principles arose as HIPAA was passed, and we started saying well, what standards are there, and how do we make choices where there are competing standards? We needed some principles to guide our choices. And so we brainstormed. I put down a bunch of them. I must have had about 15 or so. I came back and people quarreled with them, and so I adopted what they said.
We synthesized it, sent it out to places like American Medical Informatics Association, people on NCVHS. We got feedback from the community, and began to narrow it down on the criteria by which we wanted to make the HIPAA transaction standards a success.
I'll give you an example of some of the quarreling that went on. Number 8, for example, I had originally written it as be as simple as possible, but be precise and unambiguous. Somebody said, no, you don't want simplicity first. You want to be precise and unambiguous and then be simple, but you want to know what you mean. Yes, you're right.
So all of them have benefitted from a lot of vetting. And when they went over to OMB, along with an initial HIPAA package, OMB said, yes, we like this. In fact, we like it so much we want you to evaluate yourselves on the basis of these guiding principles. So they may come back to haunt us more than one time.
What we are looking for at this point is how the guiding principles help guide us in producing a report on standards for patient medical record information and their electronic exchange. The first paragraph of the guiding principles talks about the goal of HIPAA, and then the standards show us how to achieve those goals.
Would it be helpful if I went through each one of them?
MR. BLAIR: Yes.
DR. FITZMAURICE: The second paragraph starts, all standards reviewed by evaluation teams for designation as a required national standard should be evaluated based upon the following HIPAA guiding principles. Each HIPAA standard should:
(1) Improve the efficiency and effectiveness of the health care system by leading to cost reductions for, or improvements in benefits from electronic HIPAA health care transactions. Obviously, if you don't do that, you haven't met the goal of HIPAA. It's got to improve effectiveness and efficiency.
DR. COHN: Are we discussing each one?
DR. FITZMAURICE: Let's discuss each one as we go along, if that is appropriate.
MR. BLAIR: Simon, could you let him go first, and then we'll come back.
DR. COHN: I just wasn't sure what the process was.
DR. FITZMAURICE: I was going to do all of them.
(2) Each of the standards should meet the needs of the health care standards user community, particular providers, health plans, and clearinghouses. After all, they are the targeted people in the HIPAA.
(3) They should be consistent and uniform with the other HIPAA standards. That is, their data element definitions and codes, and their privacy and security requirements. And secondarily, be consistent with other private and public sector health data standards.
(4) They should have low additional development and implementation costs relative to the benefits of using the standard.
(5) They should be supported an ANSI accredited standards developing organization. Remember, that's required by law. Or other private or public organizations that will assure continuity and efficient updating of the standard over time. So it's not just sufficient to be supported by an ANSI SDO, but we want continue. That's the reason for this principle.
(6) They should have timely development, testing, implementation, and updating procedures to achieve administrative simplification benefits faster.
(7) These standards should be technologically independent of the computer platforms and transmission protocols used in the HIPAA health transactions, except when it is explicitly part of the standard.
(8) They should be precise and unambiguous, but as simple as possible. I like that one.
(9) These standards should keep additional data collection and paperwork burden on users as low as is feasible. After all, the purpose of having electronic transmissions is minimize the burden of data collection and distribution.
(10) These standards should incorporate flexibility to more easily adopt to changes in the health care infrastructure such as new services, organizations or provider types, and to changes in information technology. So we want flexibility and easy adoption when it comes to a choice among standards.
Those are the ten guiding principles that we adopted for the HIPAA transactions. You will find those I think in every notice of proposed rulemaking that goes out.
Now what is before us, and I don't know if you want me to lead this or you want to take over Jeff and/or Simon, is the application of these principles to standards for patient medical record information and their electronic transfer.
MR. BLAIR: Simon, you had some comments or questions.
DR. COHN: I had one in reference to one of them, but I think that you are asking the more general, so maybe we should just explore that.
DR. FITZMAURICE: All right. What I'm exploring is the standards for patient medical record information, should these also guide us? Secondly, are there some that are less applicable than others. Are there principles that may guide us that are not in these principles?
Let me run through them again, maybe less quickly. Improving the efficiency and effectiveness of the health system. We probably, I would guess, want a principle that says that standards for patient medical record information ought to do that.
MR. GARVIE: I would wonder whether it's necessary to have the remainder of that after the comma.
DR. FITZMAURICE: The remainder is by leading to cost reduction for, or improvements and benefits from electronic HIPAA health care transactions. I suppose that's a good point, because we are looking beyond the electronic health care HIPAA transactions.
SPEAKER: I've lost where you are.
DR. FITZMAURICE: Number one, the first part of it is improve the efficiency and effectiveness of the health care system. The second part Jim was saying well, it may not be directly applicable to us, because we are looking at clinical transactions predominantly, not the administrative health transactions. So if we were to modify it for our purposes, he sees just truncating the second half of that sentence.
MR. BLAIR: Michael, are you able to capture these modifications as we go along, or do you need some help? Does somebody else need to help you with that.
DR. FITZMAURICE: I can do it so far. This is just a deletion. Does everyone agree with that?
MS. HUMPHREYS: Another option would just be to strike "HIPAA."
DR. COHN: That probably is better.
DR. FITZMAURICE: All right, we will strike HIPAA from this.
DR. KOLODNER: Except that one of the issues for the standards is that they may contribute to all sorts of other things, such as roll up of medical knowledge if they are done right.
MS. HUMPHREYS: Either than just transactions.
DR. KOLODNER: So I think it's broader.
MS. HUMPHREYS: All right, striking the whole --
DR. MC DONALD: This quote at the top I thought is our overall guiding light, right? This comes out of the legislation. By encouraging development, et cetera, standards for electronic transmission of health information. Isn't that the constraint about which these standards are talking?
In that case, what I was going to go on to say was whether we should exclude some things from this, because a good definition says what it is not.
MS. HUMPHREYS: Except it does seem to me that the HIPAA requirement that this group is working on is broader than just the electronic transactions. It's advice to the secretary and potentially to Congress on electronic medical records, right?
DR. FITZMAURICE: Certainly that is broader than just the HIPAA transactions. I agree with Betsy on that.
MS. HUMPHREYS: It's even broader than transactions I think though.
DR. MC DONALD: When we had the discussion about creating this though, it came coming back to it's this abstract thing that is not the medical record that goes between them is what I thought we more or less talked about. Because the legislation -- these are the words out of the legislation. That doesn't reduce the fact that it's medical record information. I think we get focused on the physics of the health record, and we start building the wrong kind of standard.
DR. FERRANS: We have a paper that discusses that a little later. You are talking about what's the definition of patient medical record information.
DR. FITZMAURICE: Is this best resolved by putting a period at the end of "improving the efficiency and effectiveness of the health care system"? Anyone object to that?
DR. COHN: The only issue is it's a little redundant with what we said in the first sentence. Having said that, is there anything else that we want to put as a qualifier?
DR. KOLODNER: Clem, although the phrase at the top says data standards for electronic transmission of certain health information. When you set those standards in place, there are both the intended consequences, and the unintended. And because once you put that standard in, it has implications for other things.
DR. MC DONALD: Wonderful implications for other things.
DR. KOLODNER: Right. We hope wonderful, if they are done right. So the benefits really are broader.
DR. MC DONALD: Okay, I get you. So put the period after effectiveness of health care system.
DR. FITZMAURICE: Let's move on to the next one if there are no objections. Meet the needs of the health data standards users community, particularly providers, health plans, and clearinghouses.
DR. FERRANS: There is nothing on public health there, public health agencies or some reference to public health. Because certainly they're important users.
DR. FITZMAURICE: Good point.
MS. FRAWLEY: The other thing that I was think is we don't talk about the patients at all. The patients were the other thing, and we've got to get the patient in there. Really, we can make the system as efficient as we want, but ultimately the person that really should be benefitting is the patient.
MS. HUMPHREYS: Well, particularly because at this level of electronic medical record -- well, no, we succeed in turning the whole thing over, then we will send pieces of this data, this specific to health plans. So we better leave them in.
MR. MAYES: I would just put user communities.
DR. FITZMAURICE: These are the health care standards user community.
DR. KOLODNER: Well, it leaves it ambiguous, as opposed to explicitly stating that talking about health insurance and all this other stuff, and the person at the center of that is the consumer or the person, the patient. And it would be nice to recognize them, because sometimes we do forget them.
DR. COHN: A suggestion on this one. I'm not sure whether it is really meeting the needs of the health data standard users community or the health data users community, which does include the patient.
DR. FITZMAURICE: And you would put a period after community, or would you extend it to include the providers, health plans, clearinghouses, patients, and public health? Two choices here at least.
DR. MC DONALD: The implication of always putting patients in I resent a little bit, because it suggests that we providers don't care about patients in some way. So at least I would like to get that on the record.
DR. LUMPKIN: Well, thank you for that. I think the point that Kathleen made is an important one, and I think we need to think about putting it in there somewhere. We are not doing this just because it's going to make it easier for the provider. We imply the patient when we say to improve the quality of care. But if we really believe that, we ought to say it, because there are a lot more patients out there than there are providers and standards developers.
DR. FITZMAURICE: So it may read something like meet the needs of the health data user community, particularly providers, health plans, clearinghouses, public health organizations, and patients.
DR. MC DONALD: The truth is patients are not now in the circuit of data communications. Whether they can or should be is still something that hasn't been fixed or solved. I wouldn't put them first.
DR. KOLODNER: The other thing is if we keep simplifying and it's meeting the needs of the health community, and it happens to be data this is there, but the idea of a health data user as opposed to the health community.
DR. FITZMAURICE: That's what we are talking about. We are talking about data. We are talking about patient medical record information. It's the people who use that data or that information. I guess there is a presumption that it is to be used for good things like improving the quality of care or efficiency. But if that's not explicit enough, then we can modify it by taking out data and user, and saying meet the needs of the health community, which is what Rob is suggesting, particularly patients, health plans, clearinghouses, and public health organizations.
DR. MC DONALD: I think leaving data in there is appropriate, because that's what we are talking about. We're not going to make them happier. We're not going to meet all their needs. If they are hungry or whatever.
DR. FITZMAURICE: We are hearing testimony from people who use the data, and who modify the data.
Is there another thought on this or a concluding thought? Rob has introduced a concept. Clem has supported a different concept. I have kind of leaned over toward one side, and don't want to say everybody.
DR. KOLODNER: The reason that I think that we need to be thinking more broadly, Clem, and I go back to examples from the early days of developing the hospital information systems. That when we developed our pharmacy system, we focused on the needs of the pharmacists. Instead of focusing on the needs of all the people in the health community, we optimized for that data user, which then precluded its use. We actually had to go back and restructure it for the broader needs.
So that when we are looking at standards --
DR. MC DONALD: I think what I'm arguing for is leave the word "data needs," and put all the kind of individuals in there you want.
DR. KOLODNER: Right, but I'm saying that it isn't just the data needs. It's not just the people who need the data, but the implications for those who may not be needing the data, but who are affected by how you encode that data.
DR. LUMPKIN: I think we have to be careful, because if we keep on modifying number two, it's going to end up looking like number one. But I would suggest that if we were keeping in the word "patient," but eliminate the word "standards." I think that may convey the meaning that we want to have.
MR. GARVIE: I would also recommend substituting information for data.
DR. MC DONALD: What does it say now?
DR. FITZMAURICE: With that suggestion, replace data with information, because we are looking at standards for patient medical record information.
MS. HUMPHREYS: It sounds like Medline users now, so I'm not so sure you want to make that change.
DR. FITZMAURICE: Meet the needs of the health information user community, particularly patients, providers, health plans, clearinghouses, and public health organizations. We could also put in researchers.
MR. MAYES: Sort of a follow-up on where I think Rob was going. That is it is going to be interesting to see whether or not we can resist suboptimization in this effort, because you meet the needs of each of those listed groups individually, and come up with quite a different solution than meet the needs of the broader user community, or the broader community involved in that. I think that's along the same lines of what Rob was talking about.
DR. COHN: Can you re-read this thing? It seems like it has gone through a number of changes.
DR. FITZMAURICE: It would say, replacing data with information, it says meet the needs of the health information user community, particularly patients, providers, health plans, clearinghouses, and the public health organizations.
MR. BLAIR: Maybe if you said including, then if there is something we don't have in there, we haven't excluded it.
DR. MC DONALD: Information/data, I know the distinction, but I think it's more provocative of what we are speaking of, because we are talking about sending stuff over wires that systems can understand. And it may not be information always. It might not be a high level thing. It's going to cause confusion with library information.
DR. FITZMAURICE: And your point might also be that data can always be aggregated into information if you have the logic there. But you don't always get back to data if you are talking about information.
DR. MC DONALD: So I would say meet the data needs, rather than all their needs, because that's a big, tall order.
DR. FITZMAURICE: All right, any objections to replacing information with data?
DR. MC DONALD: No, it's a different change. It's where the word data or information goes.
DR. COHN: You're saying data?
DR. MC DONALD: I'm mostly arguing about is it was meet the needs of the health -- information was after the word needs.
DR. FITZMAURICE: And you would rather have needs of the health data?
DR. MC DONALD: No, I want to say meets the data needs. Arms could be twisted on information needs.
DR. FITZMAURICE: So whichever one it is, move it in front of needs?
DR. MC DONALD: Yes. Then the rest of it I think will work.
DR. FITZMAURICE: And do people prefer data or information?
[Several people say "data."]
DR. FITZMAURICE: All right, it reads now,"Meet the data needs of the health user community, particularly patients, providers, health plans, clearinghouses, and public health organizations."
MR. RAY: If I need a certain type of data, it doesn't come out of a standard. It might be easier to obtain.
DR. FITZMAURICE: The PMRI standards should meet the data needs of the health reporting requirement. I guess should facilitate the data needs. Do you prefer facilitate or meet the data needs?
DR. COHN: Can I just make a comment? I was actually the one who suggested that we sort of look through these. It was not an intent to rewrite them completely. But it was I think to anchor us in relationship to the overall work that has happened already. And so at least my view of this, I would ask for other input, is not see how extensively we can rewrite, but how much we can keep them pretty much the same with minor modifications, and show our linkage with what has already gone before.
DR. FITZMAURICE: All right, that would argue for meet instead of facilitate. All right, it will be, "Meet the data needs of the health data user community, particularly patients, providers, health plans, clearinghouses, and public health organizations."
MR. BLAIR: Can I offer one consideration? The guiding principles were there to try to help us at the beginning of the administrative simplification process that we were engaging in, namely, where we were to make recommendations for standards and code sets and terminologies for the first dates that we had to do this with.
My thought -- see if you agree with this -- is that if we decide in our CPR work group that we would recommend the adoption of standards, then these guiding principles would then be invoked for us to use, to help us with the guidance in those recommendations. But we may never decide to do that. We may decide to facilitate or do incentives or do other things, in which case these guiding principles are sort of an interesting background for us, but they may not be relevant.
So my vote is to leave them the way they are. They are standing the way they are as guiding principles for us if we do decide to recommend the adoption of specific standards.
DR. FITZMAURICE: So you would suggest tabling this discussion until such time as we make recommendations?
DR. COHN: I was actually going to take a slightly different view on this one, because I guess I had looked at this, and I had thought -- I guess we need to come to some consensus on this -- I wasn't seeing this so much as our guiding principles for choosing standards, but it's more a guiding principles for the work of the work group, be it whether we recommend standards or not.
I had expected the first paragraph and other things to have significant revisions, because it was unclear whether you are recommending standards. It was more the way that recommended things. Are we recommending this work to move something forward otherwise? And how are we deciding to do that, under what basis?
DR. MC DONALD: Actually, I was going to suggest we leave two alone, except to add two other items there. Because saying health data standards users community does qualify it, because it is guiding principles for choosing standards is the name of this. So I think I have to go against what you are saying.
So once we say the guiding principle of choosing standards, it's nice to qualify it, to make it good for those who are going to use these standards, and not for all goods and for all needs for all of the universe. So if you just add in patients and public health into this list, I think we can satisfy most of the requirements.
MR. BLAIR: Could I get just a sense of the group? Essentially, I think Simon is suggesting that we review and modify these guiding principles to make them appropriate guiding principles for our work group. And is it the sense of the group as a whole that that's appropriate, and that's what we should do?
DR. FITZMAURICE: Now, Jeff, are you asking this of the Computer-based Patient Record Working Group, or of most people in the room? Who gets to vote on what you are asking?
MR. BLAIR: To be honest with you, we are a small, but feisty little committee work group, and I wouldn't exclude anyone's views.
DR. FERRANS: I was wondering, do you think that this is guiding work for the work group in a broad sense, or do you mean that it should be like this is the desiderata of standards, that we should measure standards that come before the group?
MR. BLAIR: Who are you asking the question to?
DR. FERRANS: I guess to Simon, because he put forth the notion. If you put forth the notion that this should sort of guide the work group, are you saying in sort of the broad context? Are you saying if someone talks about a standard before the work group, that this is the ten principles that we should use as how to judge a standard. I think sort of to guide the work group is kind of broad. Is that what you meant?
DR. COHN: Let me just step back for a second, because we've got many different levels of misunderstanding. If you will notice, this document is actually at the bottom version 2.5, written by Mike Fitzmaurice and Stan Knudsen(?) in April 3, 1997. This was a document that was specifically meant to help guide the overall first activities around --
MR. MAYES: This was the federal IT infrastructure team document. The federal work team's recommendation.
DR. COHN: And I guess in my simpleminded way -- I'm obviously running back from this direction quickly -- I had thought that there were a number of very good items in here that did not have to do with necessarily choosing standards, but that had to do with the overall guiding principles that we should be adopting as a work group.
Now I didn't see them as tied to standards, but I saw them as tied to the general principles of what are we doing anyway. And certainly one of them I thought that we were trying to do, had to do with improving the efficiency and effectiveness of the health care system. I also thought that we were trying to meet the needs of the health data user community.
I thought that we needed to have things potentially be consistent and uniform with the other standards that we have already adopted. That was at least my sort of view of it. Now I didn't expect that we would be taking this as a standards document, because we haven't decided as Jeff has commented, whether we are actually going to be even recommending these standards.
It seemed once again from my own view, that it would be useful to have a set of guiding principles that we had actually articulated, that we could then look at all the other things that we're doing, and saying yes, is that consistent with it, or against it.
MR. BLAIR: Given what you have said, would you feel as if it is appropriate to go through and edit those words, to tailor it to our use? Or should we accept it as is, as guiding principles for the NCVHS overall?
DR. COHN: Before I make a comment, I want to see what other people think.
DR. LUMPKIN: I think we're almost done. The difference is we're not talking about standards for transactions. With patients, as long as the bill is paid, they don't care about the transactions. They don't care if the doctor has to send in a paragraph or an encyclopedia. They just want the bill paid, or at least they don't want to be charged for it.
I think it's different when you get down to the medical record. So the first two are really where the essence of the differences are. If you skim down the list of the other ones, they are I think more generic to the subject at hand.
DR. MC DONALD: I've got a couple of different thoughts. This document is very specific, and it has a clear anchor against which we can argue or discuss. The subject you proposed, that is general areas to guide our group, I think is jello. We'll never get done with the discussion.
This document, it was maybe focused more on the business transactions, but we have the job of defining standards and other activities related to medical record information. I think that's how it is stated in the law. This comes pretty close, and it's very consistent. There is no ambiguity. You've got a title that says what it is for, and a paragraph that says what it is for. So that once we finish the top paragraph, you are screwed more or less. We all went through that, that is, in terms of changing the direction.
We all agreed that this all reads good, the first paragraph was good. Okay, now what are the points? Then you have accepted because it says in the big paragraph that the purpose is to guide choices for standards to be adopted by the secretary of health and human services. So I would suggest that we go through this thinking about standards for the health record information transactions, and we're almost done, like John said.
DR. FITZMAURICE: All right, with regard to the second one, Clem has suggested meet the needs of the health data standards user committee. Going back to the first and adding just patient and public health organizations. But what is on the table is meet the data needs of the health user community, particularly patients, providers, health plans, clearinghouses, and public health organizations. How much change do you want?
DR. COHN: It's good enough.
DR. FITZMAURICE: Go with what I read secondly?
DR. MC DONALD: I don't think in the context of reading the early paragraph and the title of this carefully, my guess is adding patients and public health would satisfy it.
DR. FITZMAURICE: All right, there is a general agreement on that.
DR. FERRANS: I think the remaining question on the table is should it be health user community, health data standards user community, or health data user community?
DR. MC DONALD: Since this is a choice of choosing standards I think it becomes easier.
DR. FITZMAURICE: Well, it's not just standards. It's to make recommendations for standards.
DR. MC DONALD: I'm trying to be guided by what said at the top.
MR. BLAIR: If you have data needs of the user community, then don't you have the data in there already, if you are putting the data in front of needs?
DR. FITZMAURICE: What it reads now is meet the data needs of the health user community, particularly patients, providers, health plans, clearinghouses, and the public health organizations.
DR. MC DONALD: Your original words ties to the title and the purpose of this document, which is to pick standards. So you want to pick standards that will help those that use the standards.
DR. FITZMAURICE: See, we may not be picking standards though, or making recommendations to the secretary about specific things.
MS. HUMPHREYS: I was just going to say that I just want to announce that we do not want to do this as a committee. But actually, the opening paragraph has to be rewritten if you are going to apply it to this. Then it is still, however, about standards, because I believe the law says that what you are supposed to do is recommend what the federal government should or should not do to help promote standards for electronic medical record information.
So I think Clem's point that these things do all relate to standards is still a good point, but the first paragraph doesn't actually cover the scope of this working group that is very specific to the initial HIPAA standards work of the government.
DR. LUMPKIN: I think we are getting close, although there is a difference that is not just nitpicking. I tend to think that are goal is not to meet the needs of the standards user community, because you can develop standards that will make things easier for the standards user community and harder for the providers and consumers and so forth in the system.
We need to keep in mind foremost what we are about. I think that's what number one. It talks about the efficiency and effectiveness.
DR. MC DONALD: Just a clarification. Who uses standards, if not them? That is, you've got a model of standards users that is just maybe standards groups for their own sake. I would just like to clarify the model we are talking about, who uses standards, because we're actually going to list those specifically.
MR. BLAIR: Can I make an interjection here? Simon has just said, and I really very much agree it was not our intent to rewrite the guiding principles here, because we really didn't allow the time to go ahead and do that. I think that maybe it sounds like the group feels like a rewrite is appropriate, and Michael, do you think that you could draft a modification and send it to the rest of us for review and edit? I think that would be a more efficient way for us to deal with it, rather than for us to send the next hour going through this.
DR. FITZMAURICE: Sure Jeff, it would be my pleasure to do that.
MR. BLAIR: Simon?
DR. COHN: And we can hopefully reach agreement in September, and that would be part of whatever communication is sent out. See we couldn't even agree what the purpose of it.
DR. MC DONALD: I don't think most of the people around the table disagree that much with it. We just have our little favorite twists on the words. But I worry that now we have opened it up to a total rewrite. I think this is a pretty good document. I think it's internally consistent. The disagreement is, is this the document you want to talk about today or a completely different one that hasn't been written yet.
MR. BLAIR: Well, Clem, would you want to be part of that rewrite there?
DR. MC DONALD: What I hear is a response to a typical -- we're all committee nigglers. And you're kind of panicking now because we're burning all this time. I didn't hear that people are rejecting this document out of hand in any sense. Simon has got this notion that this document should be a different one. I would like to challenge that one. I think this document is a pretty good one just like it is, with some minor to modest modifications.
We may need a different one too to say what the committee is going to do, but I thought we did that before we started. I thought we had another one that said what the committee is supposed to do.
DR. FITZMAURICE: What is the pleasure of the group for moving ahead? I'd be happy to rewrite this, making minimal changes, accepted all penciled or penned changes you want to make on the document you have in hand, and incorporating those that I choose, and then we'll review it in September. Emailing it around, and letting you make additional comments. If your favorite comment didn't make it in, make it again and let everybody see what your comment was. Then we'll review it again in September.
MR. BLAIR: Visually I look out and see total consensus. Onto our next topic?
If you recall, during this year or so we created a CPR Work Group work plan. From June through December we had seven or eight revisions of the work plan. In December we had hearings to have outside folks come in and critique whether or not our focus areas were correct. You should have a copy of version 8 of our CPR Work Group work plan.
There were a few things after we sent that out, I guess like in February of this year, version 8, Bill Braithwaite you had an email that you sent where there were a couple of concerns and questions that you had. To expedite things, maybe I can step through just a tad on that. You also should have a copy of the email from Bill.
Just to refresh your memory real quick, when we talk about the focus areas, there are six focus areas in the work plan. The first is studying and looking at message format standards. The second is focus on medical terminologies. The third area is business case issues related to standards. The fourth was how we relate to the work going on in the health care information infrastructure work group of the NCVHS. The next I think was data quality, accountability, and integrity. Then the last was the issues of whether different state laws may make it difficult for us to create national standards related to patient medical record information.
So those were the six areas of focus. Bill, are you here?
DR. BRAITHWAITE: Yes, I'm here.
MR. BLAIR: In your email you wound up observing that I guess even before we got to that, there is a statement in our work plan that referenced that it would go to Congress. That was absolutely my mistake. I had observed the word that we would study and make recommendations and legislative proposals. So I stuck the word "legislative proposals" in there. You observed that it never was in there, should not be in there. It goes to the secretary of the Department of Health and Human Services. We can correct that.
Is there any other thing, Bill, on that item that you wanted to point out?
DR. FITZMAURICE: Jeff, let me interject that I think you are right, this is a congressionally mandated report that we are required to write, directed toward the secretary of HHS.
MR. BLAIR: I think that the next observation you had, was that with respect to the health care information infrastructure?
DR. BRAITHWAITE: Yes, and standards supporting the NHII vision.
MR. BLAIR: Right. And my read on that was that our focus in our CPR work group, Uniform Data Standards For Patient Medical Record Information, was that while the health care information infrastructure group had a little bit more of a broad mission and a charge, that they were maybe a little less focused on what implications there were for standards for patient medical record information in that group, the standards piece.
So that was the rationale for thinking that we had a piece of work that we do to make sure that we were in sync with recommendations coming out of the Health Care Information Infrastructure Work Group. That we needed to look at those and be in sync with those. Does that still not change your thoughts, that work item doesn't belong in here?
DR. BRAITHWAITE: I agree in principle. It just seemed to me like that focus area was of smaller importance, and weaker than the others, who have major focus areas.
MR. BLAIR: Do you think we should remove it?
DR. BRAITHWAITE: My recommendation would be to take it out.
MR. BLAIR: The recommendation is to take it. Is there a general feeling that that should be removed from our six items?
DR. MC DONALD: We've got a whole other work group doing that, right?
DR. FITZMAURICE: Except they are not focused on the patient medical record information. It's more a much broader focus. So the purpose of item number four is to show a link of NCHVS, and to make sure that we also have a vision. I would hope it's a shared vision with that work group that we're not inconsistent with what they want to do. We are consistent with what they want to do.
DR. FERRANS: It's interesting, because I have a document right in front of me that says that HI neither refers to a set of technologies, standards, and applications that support communication and information. It's at least the paper that was submitted, that was written by the work group or whatever that kicked it off. That does reference standards. So I agree there is a common thread.
DR. MC DONALD: I would take it out. Either that or we say a general one that will coordinate with all the committees.
DR. COHN: Let me actually ask the chair of that work group, since he is here with us. Is this an activity that NHII work group -- we should just lateral this whole issue to them?
DR. LUMPKIN: Well, actually, I have a different view. I think that all of the work groups of the national committee should conceptualize where they fit within regards to the NHII, because that really is an overarching vision for the national committee. I don't have a problem with the way it is in here. It could just be discussed as somewhere in there, but I think that it's important for all the work groups and the subcommittees to have as part of their work plan, thinking about how what they are working on will fit within that structure.
DR. FITZMAURICE: Could I make a comment too? I tend to agree with Bill that it is maybe of lesser importance to strike forward strongly in this area. But I think it is also important to show the shared vision that we have with the other work groups. I would recommend leaving it, but giving it less weight, less focus, and fewer words in the final report.
MR. BLAIR: Any other comments? Do you feel comfortable that we resolved this by having a show of hands for whether it remains as a focus area, or whether we delete it?
DR. MC DONALD: No one wants to delete it now. They were persuaded by the chairman.
MR. BLAIR: Okay, so it will be retained? Should we go on to our next item? I think the next item that you mentioned, Bill, was that with respect to medical terminologies?
DR. BRAITHWAITE: This was when the work plan discussed information gathering activities in sort of a generic form.
MR. BLAIR: Oh. Why don't you articulate what you were thinking of there?
DR. BRAITHWAITE: I just felt that we ought to be a little more specific about are we going to hold hearings on this, or are we going to commission papers or hold workshops. Just a little more specific about what we mean by information gathering activities. We are not just going to clip the newspapers here.
MR. BLAIR: That probably reflects a lack of clarity. I don't think we knew the answer to that. But that was five or six months ago. I think you may have noticed that at this point we have sort of looked at our time frame, and looked at our resources, and I think that we have considered that the most expeditious thing that we are able to do -- this may not be the right answer -- was that if we had hearing, gathered information at that level, and then in the first quarter of this next year, pull that together into recommendations. If that is not the right way to go on one or more of these focus areas, then you know.
DR. BRAITHWAITE: I think the working group has in fact, since this was drafted, come to some agreement about what we mean by data gathering activities. We have started to implement those. So it's kind of a moot question now. Back in March this comment made sense, but now it doesn't.
DR. MC DONALD: We have had hearings.
DR. BRAITHWAITE: Then the fourth one was about medical terminology. I just was looking for a more explicit statement that we're going to work on the development of an overall framework in which existing and developing terminology efforts can be integrated and coordinated, because we have discussed that as one of the activities that this group would work on. At least in my reading of it, I didn't see that as one of the major activities that we were going to be working on.
MR. BLAIR: I agree with you that that the things that we wanted to do. I picked some words in there I think like coordination and convergence and some of those things. In my mind I had thought that, but I have no problem with modifying our wording on that, so that it more clearly includes the concept of looking at a potential framework for converging medical terminologies.
DR. MC DONALD: I like your words very much like they are, but it would be a way to assuage those in instead of or in addition to. You are not saying converging. You are really saying make something whole. That's what I read, by whatever technique it takes. That's a little different. That's not quite the same as saying somehow we'll get them all the same.
DR. BRAITHWAITE: It's not the same as make it all into one. It's a methodology, if you will, of taking the existing and future works in this area, and building a common framework in which they can do their job, on which then we can publish to the outside world in a way that everybody can then use the integrated product.
MR. BLAIR: Bill, would you like to introduce a modified phrasing for that?
DR. BRAITHWAITE: I guess the phrasing in my comment is sufficient. We just have to find the right place to put it. I would be happy to work with Mike in getting that into the document.
DR. COHN: Actually, what we may want to do is say in the work plan where we talk about the area of focus, include issues related to convergent medical terminologies. We might say frameworks for medical terminologies.
DR. MC DONALD: I would suggest this is another good one to do off-line.
MR. BLAIR: We just took a look at some of Bill's thought with respect to whether or not the focus areas are articulated in a way that is giving us a direction that we need. Does anybody else have any thoughts or concerns with respect to the focus areas, given that we look to them for guidance on exactly how we are pursuing our hearings for the rest of the year?
DR. MC DONALD: This is a very good document. It gives direction. I like it.
MR. BLAIR: Any other observations, comments, critiques?
MR. BLAIR: Then let's go on to our next topic, which is our calendar for the rest of the year. There is a handout for that, which turns out to be not what I intended to give you.
Here is one that I would hope that we could come to a general agreement on the focus areas we need to target for additional information. So I might even suggest these, and just scribble them down and we can discuss them.
Michael, we have two days in September. I think that there is September 16 and 17? So we have the 16 and 17 of September. We have October 12 and 13 for hearings for our work group. That's four days. And there are two days in December, but we are trying to reserve those for the Subcommittee on Standards and Security as a whole. So we are trying to see if we could do our work group activities essentially in those four days of September and October.
Simon and Michael and I and Bob on a conference had the following ideas. We felt that we needed at least one more day on medical terminologies, and possibly a day and a half. That would be so we could hear from the folks that are using medical terminologies, a la vendors that are building it into their systems, providers that may be building it into their systems, and the message format standards organizations that are trying to incorporate medical terminologies into their message format standards.
Again, we started looking at the folks there that we would need to have, and we also in particular -- Betsy, are you here with us? She is not here. Well, we also really wanted to give Betsy a good deal of time to enlighten us on her perspective, since she has worked quite a bit in trying to converge medical terminologies. I shouldn't say converge them, but map them with the Unified Medical Language System.
So we wanted to have her on the agenda. There was an additional individual that has been requested by Henry Hefernin(?) on otologies as a diverse approach that we should at least hear about in terms of medical terminologies.
So we are looking at a day, day and a half for that. We were looking of at least a half a day, if not a full day on data quality, accountability, and integrity issues. And among the folks that we were looking at to hear from in that area was Gary Dickinson. Gary, by the way, is one of these voices that has been out there for some time, who has been winding up indicating that we are moving beyond one-to-one connections with sending health care information with messages.
Now that we have integrated delivery networks and managed care organizations, you may have two or three or four or five links. Right now many of the standards don't have structures for accountability and auditability as it goes through multiple links; that the data retains its data integrity. And he has some thoughts that I think we ought to hear about.
There are other folks that have dealt with data quality, who are in the business of analyzing data and sharing it back to health care providers in terms of performance information, outcomes information. They often have to deal with scrubbing the data and cleaning up the data, and they understand what it takes to do that, and the kinds of problems they have with data quality.
And so there probably needs to be like I said, a half a day or a day for that. The third area which comes out of our focus area is the area of business case issues for standards. And I have gotten emails from a number of solo care and small group practice physicians that are very, very concerned that we are heading down the path where they will be forced to accept computer-based patient record systems.
They are very concerned, and want to be heard in our committee so that they could articulate some alternatives to that, that they feel may exist out there other than those that we have considered so far, and that's a broad view of a business case issue.
The other is a number of small standard developers and terminology developers have indicated that there are individual subject matter experts who can't afford to go to all of these standard development meetings. Or if they can go to those, they can't go to the coordination meetings that exist. So there are a number of those that we wanted to hear about in terms of business case issues.
Kathleen Fyffe, I think you had raised this as an issue early on. We really want to make sure that there may be some other folks that you may want to make sure we hear from.
MS. FYFFE: Hospitals.
MR. BLAIR: Small hospitals. Then the last area is the area of the impact of having diverse state regulations and laws. What impact that has on the standardization of patient medical record information. That's in terms of record retention, authentication, permanence.
I'm not mentioning privacy and confidentiality, because that is being handled in another group. So exclusive from privacy and confidentiality, these issues. So if you notice, as I went down this list, that basically gives us a day, day and a day on medical terminology, a half a day to a day on the data quality, right now one day on the business case issues, and one day on various state laws, and how that affects everything.
So that basically is the five days. Is that all right?
DR. COHN: That's fine.
MR. BLAIR: Comments, critiques, thoughts?
DR. MC DONALD: You gave a really long litany there, and I couldn't classify them all clearly under these headings. The one I was having trouble with was you talked about small standard developers who couldn't come to any meetings.
MR. BLAIR: The business case issue?
DR. MC DONALD: For what? The business case for having standards, or business case for them participating in standards?
MR. BLAIR: What it was is --
DR. MC DONALD: Give me an example. I can't imagine anybody in business who can't come to a meeting some time.
MR. BLAIR: Sometimes you have small vendors that have folks that have difficulty attending HL7 and ASTM and CORBA-MED. They have to pick, because they can't be in all the places. Then you have other situations where you have small terminology developers.
DR. MC DONALD: I'm still trying to clarify. Yes, life is hard, so they pick. That's how things get started out in the marketplace. Why is this an issue?
MR. BLAIR: The issue is that many of them have indicated that they feel as they have knowledge and issues to share in the standard development and terminology development process, but they lack the funds to be able to participate in the process. And they were looking for the government to facilitate in some way, or subsidize, or pay for, or fund some of these folks so that they can accelerate the standards development process.
DR. COHN: First of all, I'm not sure where to start on this one, but I would probably say that I'm not sure -- I think that may be a piece of business case. But it's at least my view the business case was like why would anybody do whatever we said anyway, was really to meet the business case issue around we said there should be a standard for X. What would make anybody want to implement that? That to me is really more the business case, to help drive a case to the secretary. Hopefully, we don't lose that in the process. I think Kathleen Fyffe -- that's more what she thinks is the business case too.
Now I wanted to just reference back a bit, because I'm actually feeling right at this point as a work group, and I think that we're still talking about five days. Let me say that as I looked at what we needed to do, and we'll talk about this tomorrow, the overall subcommittee probably needs a day anyway, probably around December, maybe a little before, but of course it's the same group, so surprise. We'll somehow work it altogether.
But there is a generic need that we have in this work group to begin to hear more strongly from the many users, be they hospitals, be they vendors, be they managed care organizations, probably be they patients, since we have just identified in our previous discussions that maybe the patients are one of the customers of the work that we are doing.
That we really need to begin to hear and solicit their thoughts. Now when we solicit their thoughts, we probably need to take it in a slightly different way than what we have done before, which is that I don't think it's reasonable to ask a vendor to come once to testify on how they use terminology, and then another time how they use standards, and yet another time on state regulations, and another time on something else.
Maybe we need to put the questions together in a way where if we have people come to talk to us, we can ask them on all the questions that we would have of any user of all of this about what sort of bumps they are having, what works for them, what doesn't. How in their view, the government can help.
DR. FITZMAURICE: What does fit together.
DR. COHN: What problems are they having. And that that's really I think a lot of the work we have in the fall, at least my view.
DR. FERRANS: I guess I would add to that I agree that we need to get more users in here. But when you ask them about the business case, what exactly are you asking them? I had an opportunity to review all the testimony from the first hearings, and we actually asked people about the business case for standards. And implied in the question was the business case for computer-based patient records, because if we are going to put standards on, people are going to have to use computer-based records to enter the data -- I'm talking about the end users or the generators, whatever you want to call them.
I was surprised by the lack of hard data that was given. I did want to say that when we do ask people to come, that we ask them to cite specific data or something, because we need something to hang our hat on. Everyone will come in here and say that it's going to improve clinical care and research and education, and it seems like everybody agrees on that, but I haven't heard any data to support it.
Not that I disbelieve it, but it would be really nice for us to have that, and to build that business case. But from the users, I think part of the question of the users is what is the business case for you using computer-based patient records also? Not just what is the business case for standards.
DR. MC DONALD: I don't think those things are conjoined the way you said it. I think the regulation says medical record information transfer. There are lots and lots of needs for that, short of doctors typing in coded information into their system. I think in fact the answer you get from guys who say I don't want to type coded information.
So I worry about conjoining. Clearly if you had that, you're going to have a lot of needs. But the data is going that way. The guys want to get lab data in their office computer, I'll tell you that, and they want to get those pharmacy records in their office computer if they could. They would like to get the discharge summary from the hospital in their office. So it doesn't have from then outward.
DR. FERRANS: I know, but then they are probably not going to need SNOMED-RT if they are just going to be getting some of --
DR. MC DONALD: Like the pathology reports.
DR. FERRANS: Yes, so I mean when we start talking about the overall framework, we're talking about getting granular data. Again, we get back, and I see the distinction between patient medical record information and computer-based patient records.
But when you talk about the users, there is an implication if standards start coming around, and they are expected to use it, they are actually going to have to purchase the systems to implement some of that. There has to be some discussion of the business case there.
DR. COHN: Richard, I share your concern that we need to be very careful to craft any sort of entry into the business case issue. I think the discussion we have just had is what happened. We asked them what business case, and they said computer-based patient records. That's the business case. I think we need to be careful. We want to find out about that. We need to look at that.
I think the other piece we need to look at is the piece that Clem was referencing, which is really sort of the soul and the heart of what it is we are trying to do. Now I'm actually sort of looking at Kathleen Fyffe, who I'm hoping is going to help us with some of these issues in terms of crafting the appropriate questions. Am I looking at the right place?
MS. FYFFE: I hope so. The concern I continue to have is whether the average hospital in this country -- some of those average hospitals are actually in the red, and it's only going to get redder for them -- are going to be able to enter into this type of environment at all, to participate in the electronic exchange of medical record information.
Now I'm pretty well convinced, and this is just my instinct, that there is a certain percentage of hospitals in this country that can and will afford to do this. There are some more that will try, and then there are others that it will just be out of the question.
So as one of the testifiers pointed out early on, I think back in February, you have to look for the lowest common denominator, and the lowest hanging fruit. I hope that the recommendations that we come up with are not so elaborate and so costly that the average hospital in this country could not participate.
Let me say again, and I'm going to sound like a broken record, reimbursement to US hospitals is declining, and it's going to continue to get worse. It's sort of almost contradictory for the secretary to -- I'm not saying she is directly responsible for reducing reimbursement to hospitals, but to come up with a recommendation that might be costly, and then at the same time the reimbursement is getting squeezed out. That's my view on it.
MR. BLAIR: I feel as if this is a very important issue for our work group. If we allow our scope to increase where we are considered, whether inferred or any other way, to be mandating the use of computer-based patient records, we're in a lot of trouble. That is not the mission that we are given in the law.
We are given a very different mission. The words are study and recommend uniform data standards for patient medical record information and the electronic exchange of that information. I think we have to stay tightly to our mission, because the uniform data standards are facilitating the evolution of an information infrastructure. That's different saying somebody has to install a computer-based patient record.
What it is doing is it facilitating the information infrastructure where over time people will be able to -- over time if you have an information infrastructure, just like you have a transportation infrastructure, then you can build cars inexpensively. People can choose whether or not they want to use them. You could wind up using the telephone analogy. You could wind up using many other analogies. We should not in any way, shape, or form mandate the use of computer-based patient records.
But if we build the information infrastructure intelligently and wisely, we will facilitate low cost use and productive use of data, whether it is called a computer-based patient record, or whether it is a subcomponent, or whether it is a practice management system, or whatever it is. We are facilitating and enabling the growth this information infrastructure. I think we need to stay closely to that.
DR. FERRANS: Jeff, I would just say that in the previous business case testimony that was given, because again, I had an opportunity to review it recently, a lot of the "return on investment," people talked about decision support systems, which implies this. So I think when we have asked the question before, there has been the sort of mission creep, or the perception. So if that's the clear intention, then I agree we need to phrase what we are asking, and what the business case is very carefully.
MR. MAYES: I want to put a little bit different spin on this. It may not be this committee's charge to dictate these electronic health records. It is happening. HCFA is moving ever rapidly towards requiring legislatively, electronic reporting of ever increasingly complex clinically-oriented data; other payers are.
Whether or not hospital can currently meet those demands is going to be the question of whether they stay in business. Frankly, I think if they can meet those demands, they should be in business, because I would question how they can really provide adequate health care in today's information intense world.
So while it may not be this committee's charge to dictate the use of that, I think it would be the peril of this committee to ignore the fact that there are many other business drivers that are in fact dictating the move rapidly -- much more rapidly than some might like, but rapidly towards that kind of environment. So the kinds of recommendations and standards should at least recognize that that's the real world and where it is going.
DR. MC DONALD: The goal is to enable, rather than require.
MR. MAYES: Others will declare. Let someone else get the heat. But they will declare, and they are declaring. So if we sort of pretend like that's not going to happen, I think would be a disservice.
DR. FITZMAURICE: I would also add that it's one thing to require that somebody submit you data in electronic form. It's another thing to require that they have a system on site that does it. In many cases there are services that spring up that will translate paper information into electronic form so that you can meet the requirements of payers, insurance companies, and of HCFA.
Now in the longer-term, as we get patient medical record information standards and transmission standards polished, it may be more efficient for them to buy a system, but that's their own business case decision.
DR. COHN: I think I agree. I was just going to actually add one more other item to our list for the fall. Is that in order that this point? Just in terms of our fall activity, and I think Jeff had commented about this also, there is actually a fair amount of international activity in relationship to standards which Jeff and I have both talked about previously.
I thought a briefing and probably a quarter day or something, a half day related to the international standards activity, and what exactly the government ought to be doing in relationship to that. Is there an issue, and is there a need will probably be well within the scope of the work group. I just wanted to throw that out.
MR. BLAIR: Can I elaborate on that just a tad?
DR. COHN: Please.
MR. BLAIR: A tad of elaboration. It isn't just to be informed about what is going on internationally. It's that both Simon and I and I think others that have been involved in things like the ISO US TAG are beginning to realize that the world is working on standards for patient medical record information or health information. They are looking on their criteria for what technologies to choose, what standards to choose, what priorities to use.
It may be only a few years before international standards activities can begin to influence what standards we choose in the US. And therefore I think the committee needs to understand what is happening here, because if one of the thing that we are really looking at is what is the role of the government with respect to uniform standards for patient medical record information, one of the roles may be to make sure that US interests are properly represented on the international scene, because that is becoming much more active.
DR. FERRANS: I just wanted to add one more comment, getting back to the previous discussion. Certainly, I don't think anyone here is advocating that we mandate computer-based patient records. However, I still would suggest there is a strong link between exchange of clinical information and electronic data standards and computer-based patient records.
And if what Kathleen is saying is true, it is very helpful to hear people talk, end users talk about costly implementations that have not given them return on investment. I can name several other types of health care technology systems that have these business models that have actually not been proven out in the literature. It says you pay millions and you save X, and it pays for itself in three years, and that's really not been the case. That's one thing.
The other thing is if we are very worried about that, our name is still the work group on computer-based patient records. I think that would lead people to suspect that we are working on standards for computer-based patient records. That's just an observation.
MR. BLAIR: Any other comments or observations?
DR. MC DONALD: Just a comment, maybe a retort about the ISO activities. I just want to remind people the ISO standards cost up to $40 for a three page standard, and the Internet standards cost nothing. There was ISO OSI thing that was going to sweep the world. The cheap, fast one -- we all know how that came out. So that I think that it's a worthy activity, but I think we should be cautious about how and when we support it, and try to encourage ISO to be more like Internet.
DR. COHN: When we get ready to hear from them, can we arrange an appropriately balanced presentation to explore all sides of those issues?
DR. MC DONALD: ECME(?) has become a real competitor for ISO, as has Internet in reality, although they don't talk that way. But the costs are ridiculous. It's $40 for four pages.
MR. MAYES: Being involved in two of them, we are pushing drafts, because we can publish drafts for free. Basically, the drafts are pretty darn close to the finals. People who are actually involved in some of these things, particularly in the health care side seem to be well aware of that problem, and are trying to make as broadly available, the actual, true information as possible.
DR. COHN: Is it time for a break?
MR. BLAIR: I was going to make a proposal for how we deal with the calendar.
DR. COHN: Okay, fine, and I'll just talk about something else after that.
MR. BLAIR: Good, because we do have time. We're running believe it or not, early.
DR. COHN: In that case, let me make a comment. This just had to do with another issue, which I actually had hoped was going to be brought up today in the GCPR presentation, but we did, because we ran out of time. I would just remind you that there is a tension in the committee probably in some way reflected in our discussion around business cases and all of that, about whether the committee is going to come up with a framework and overall solutions for the long-term versus sort of the short-term what is it that really -- there is a big bang for the buck in the short-term.
I don't mean to resolve that issue, but we certainly need to reflect it in all the work that we do from here to fore.
DR. MC DONALD: Is it either/or?
DR. COHN: Actually, it probably isn't. It's a tension. I think we probably need to do both, but we need to be listening to some of the near-term things that have strong business cases, that are these low hanging fruit things, and begin to catalogue them, and begin to put them together in whatever future testimony we have.
MR. BLAIR: Okay, that's good. Actually, we will proceed with this calendar. My thought is that if Michael, Bob, Simon, and I on a conference call, were to lay out which of these focus areas we have on those five days, if anybody else wanted to join, that's fine. We have time. We have 20 minutes here. Do you want to pick which days are which, or would you prefer to leave it to our small, but feisty group here to lay it out?
DR. MC DONALD: I would be happy to have you lay it out if you would try to whittle it down. When I hear some of the subject matter which I think is more peripheral, like some guys wish they had more money. Or some guys have trouble choosing between what meetings to go to. Who doesn't? I don't know if we need to hear from those, given the amount of hearing time we are investing in this, and how hard it is to get to all these meetings.
I don't think that all ideas that have come up should be accepted. I think there should be some selection process, and pick the most important.
MR. BLAIR: Michael, could you capture that point? We need to whittle down and a selection process.
DR. FITZMAURICE: Yes, Jeff, I have that.
DR. KOLODNER: Jeff, you said five days. I have four days I'm aware of.
MR. BLAIR: I think we're feeling that hearing from users probably would be more than a day. We also are adding in international.
DR. KOLODNER: We have four dates.
MR. BLAIR: Simon I think was granting us some of the time that had been set aside for the subcommittee itself in December.
DR. COHN: Do you need the actual dates? We're talking about our first fall set of hearings on September 16 and 17, followed by the next set of hearings on the 14th and 15th of October.
DR. MC DONALD: We're not going to whittle any days then?
DR. COHN: We're not going to whittle out any days. For some reason I have the 14th and 15th of October. Actually, I would defer to you, Mike.
MS. FYFFE: Monday, October 11 is a federal holiday, Columbus Day. Was that part of the reason?
DR. COHN: I had understood the 14th and 15th had been called because I think it had to do something about Thursday and Fridays being -- we will clarify that after the break. Then the final dates were December 9 and 10, except that I see its 7 and 8 on one of these things. I think we are going to have to clarify with Jackie which dates these are.
DR. MC DONALD: I have six days I'm counting now. You said there are five days of hearings and we might whittle down some or no?
DR. COHN: Tomorrow we will talk about the work plan for the full subcommittee. If you will remember, there were actually some HIPAA and other related things.
DR. MC DONALD: If you only had four days, would you choose differently who came? We had international participation at these meetings. We had the British. We had the Canadians. We had the Australians. I really think that let's see the least we can do, rather than the most we can do. Is that wrong?
MR. BLAIR: Clem, you're probably at a little bit of a disadvantage, because you are so involved in the standards community that these things are basically familiar to you. But I think that there are many people on the committee where they don't have that breadth of exposure.
DR. MC DONALD: I'm only saying that in most things in life we've got limits. If we set limited days to how much we spend on this -- it just keeps growing. I mean they said 14 days would be required for this. I think I'm going on 29 this year. So someone has got to push back. We've got limits in life, and in those limits, what are the most important things that are presented. Not that we can keep adding days, no matter whatever comes up.
DR. COHN: This is a difficult decision, only because we haven't had the discussion on what the subcommittee is doing, which we plan to take probably one day this fall on.
DR. MC DONALD: That's worthy.
DR. COHN: So if we have four days for this committee, and we have one day for the full subcommittee, then we're talking about five days.
DR. MC DONALD: I think two half days; if we were able to discuss what the priorities were, we might whittle and get a whole day off. If I'm the only one feeling this way, I'll just shut up.
MS. FYFFE: I was just going to say, Clem, you're not alone. We were just concerned also about -- I guess my concern is we could spend a lot more time doing this activity. But I know we are scheduled to also talk this afternoon about the report that we eventually have to come up with. So I'm just starting to get a little concerned, because we could keep holding hearings and inviting people, and it would be the quest for more knowledge.
I think at some point we just have to say, okay, we've heard enough. There are some seminal reports out there already that we can incorporate by reference. And I think the more important thing is the next part of the committee's work plan. That also is a concern, because as you start to realize how much time we have already spent this year on activities, we still have an awful lot of work yet to do in order to produce that report, which has to go to the secretary.
So I think I would support your recommendation that if we can get everything done in four days, then we should do four days. Then what we need to think about is how we're going to get the rest of our work plan.
MR. BLAIR: Let me add one thing. Clem and Kathleen, I am also tired of traveling and going to meetings. So on the one hand, I agree with you that I really don't want wasted time. I don't want redundancy. I don't want unnecessary testimony.
On the other hand, while many of us are involved in certain standards activities, there is kind of this call from folks that are peripheral or diverse in one manner or another, that have different views. If we wind up limiting the time, then we either scrunch things together, or we wind up cutting out some of the divergency.
Now that doesn't mean that we wind up needing to have ten days for hearings, or whatever the right day -- it's a balance.
DR. MC DONALD: I really think you could cut a day out by subject matter. You just piled on a lot of subject matter, and we haven't been forced to decide what is the most important. I'll just leave that lay.
DR. COHN: Can I make a comment? I first of all have the dates of the meetings that we're holding, and I have a suggestion on how we might approach this. First of all, September is indeed the 16th and 17th. October the dates we have held are the 14th and 15th. Then we have also held December 9 and 10.
Now my belief, probably like Kathleen Frawley's is there seems to be a lot of work which we'll reflect on after our break as we talk about what else we need to do into the year 2000. I think we should definitely plan for the September and October, and then if we need, we can do the December.
DR. MC DONALD: That would mean we would have discussion in December, and October not just hearings.
MR. BLAIR: That might be.
DR. MC DONALD: I still say we have taken all comers. I haven't heard anything rejected. You always have to winnow. Somehow you've got get the best use of time.
DR. COHN: I guess I would suggest that probably when you are talking about the users, it's probably not the appropriate time for us to be winnowing.
DR. MC DONALD: I don't know how much time you had suggested for people saying that they wanted to talk about the business case of how they can be participants when their companies -- I think I would argue about the international. This is an American standards group activity, and we're doing American standards for the US government. And we can get reports about the international stuff. We can see those in writing.
DR. FERRANS: The point I wanted to make was the summary on the terminology issues that we received earlier today. We could talk about that for weeks. I agree with both sides in that obviously we need some kind of balance. But I would certainly weigh not cutting down the time, but redistributing time so we get going on the analysis part. Because I think that that is going to take us a lot longer than we think it's going to, because the issues are incredibly complex. We'll never have enough testimony that we totally understand all the issues, and that we're just perfectly fluent in them and can make decisions.
MS. FRAWLEY: I agree. That was my concern. Because I think at some point we're just going to have to stop with hearings. We may have to do some work outside of the formal committee process, because there may be some follow-up that has to take place as we begin to work on our report.
People may have to contact people or follow-up to ascertain more information. But I think that just the terminology piece alone is so huge, that even the list that we just read off, we could spend more time just on that, and never get to the business case, which I'm concerned about. I don't know that we're going to get the information we need, because I don't think a lot of people have good data out there.
I think that was the point, that again, we could be hearing a lot of anecdotal things, but not really getting our hands around something. That's really what we would need for our report. So I guess I think what we need to do is probably cull that list a little bit, decide what are the most critical things, and then probably figure out the next piece, which is the more difficult piece of how are we going to put this all together, analyze it, and make sense of it for our report. I think that's going to be the tough one.
DR. COHN: If I can make a rejoinder on that. That was actually why I requested the discussion after the break, which is really to look into the year 2000. My need is that we come up with budgetary items and things like this, to assure that we get to our finish line successfully. But I think it's absolutely the right time for us to begin to talk about that.
MR. BLAIR: Are there any other comments on this? Otherwise, my inclination is for us to take our break about seven or eight minutes early --
DR. FITZMAURICE: Could I just ask a question, Jeff, about how did we resolve this? Did we resolve it by the four of us will come up with a plan to fill up at least the first four days of hearings and see if we can get it all in those four days? Then we'll expose the schedule to the group, and let the group respond to that schedule?
MR. BLAIR: Could I just modify your words just slightly? Instead of being our objective to fill up the first four days, is that we know we have some time that is on the calendar that is available to us. But I think what we want to do is we want to look at each of those focus areas, and say what do we need to do to gather information to help the committee decide? And do it in the most expeditious time efficient manner.
If we can do it in less time, we will. But, yes, the four of us trying to figure out -- lay out times on those four days. Clem, did my comments reflect the right attitude?
DR. MC DONALD: Yes.
MR. BLAIR: Are there any other comments or feeling about this, or have we concluded this discussion with respect to the calendar for the rest of the year? Then afterwards, Simon, when we reconvene in about 15 minutes, I think you have some good thoughts with respect to how we might approach the budget issues. Maybe you could lead the discussion on that.
So let's see, back at 3:15 p.m.
[Brief recess.]
MR. BLAIR: Our agenda for the next couple of hours here is three topics. Simon really corrected my characterization of saying our first topic is budget. It really is the tasks and things that we need to do that might wind up reflecting possibly budget items or other issues or resources.
If we have two hours between now and adjourn time, just rough and dirty, unless we change this here, we might be looking at 45 minutes a piece for each topic. The first one would be the tasks that we need to do to be able to produce the final report. The second one is your presentation Richard on the definitions of patient medical record information, and to give time for people to comment on that. Is 45 minutes enough time, Richard, for that?
DR. FERRANS: More than enough time.
MR. BLAIR: Tell me how much time you'd like.
DR. FERRANS: I think we can do this in 15 minutes, because we're just going to discuss it very quickly. It's a very short list of comments that people have made directly from testimony. And we are either going to come to consensus, or it is going to go back for a rewrite.
MR. BLAIR: All right, let me ask you this, and Simon, you may be thinking the same thing here. What do you think of maybe having Richard go first on our agenda here? The third item, by the way, is the interim letter to the secretary. What if we do that?
DR. COHN: Yes, that's fine. I want to clarify that I think most people were expecting a 5:00 p.m.
MR. BLAIR: Richard?
DR. FERRANS: I'll hand these out. After the guiding principles went through various iterations, I was glad I kept this short. This draft report issue paper one, I would emphasize the word "draft" really starts out in the first few paragraphs by talking about HIPAA and what the charge of the committee is. Interesting enough when I kept looking at it, it's uniform data standards for patient medical record information, and electronic exchange of such information, patient medical record information.
So we submitted written questions and invited panelists. Actually, it was in the first hearings. The second group of hearings we submitted different questions I believe to panelists, where we said how do you define patient medical record information?
Reviewing the testimony, a lot of people didn't answer the question necessarily. They did talk all about patient medical record information, and its virtues and standards for such, but a few people did define it, and in the interest of trying to be objective I took off any references to who said what. We had a lot of good thoughts on this though.
I also put down the Institute of Medicine definition of a patient record. There were several definitions. The first one is actually of the patient record. Which is the patient record, "is the repository of information about a single patient. The information is generated by health care professionals as a direct results of interaction with the patient, or with individuals who have personal knowledge of the patient, or with both."
Then in Item 2 they talk about how they distinguish between a primary patient record and a secondary patient record. The first being used for health care professionals in the provision of patient care services, reviewing patient data, documentation actions or instructions. The secondary record derived from the primary record, and containing selected data elements to aid non-clinical users in supporting, evaluating, or advancing patient care. So again, this is all about patient care. That's very useful.
The IOM also gave a definition for the computer-based patient record, which Jim Garvie was able to get me, as "an electronic patient record that resides in the system specifically designed to support users through availability of complete and accurate data, practitioner reminder and alerts, clinical decision support system linked to bodies of medical knowledge and other aids."
"This definition encompasses a broader view of the patient records to serve the basic function of storing patient data for retrievable by users involved with providing patient care." It's a long sentence. "Even this classic function must be broader in the future, however, especially with respect to a key feature of flexibility. Different health care professionals will require different modes of information retrieval and display." That's all from the IOM book.
We had one testifier who said patient medical record information is data about the events and findings that characterize an individual health status and health care. That was a very useful definition. It's used at all levels, private agencies, local, state, and federal levels for the process of overseeing and managing patient care and patient care processes, as well as for monitoring public health. This is a good, broad definition of what it is, who is using it, and why.
Another testifier said that the patient medical record is a collection of information that contains narrative text from providers, and basically was more descriptive. It contains other types of information that might be in other formats such as x-rays. That information today is document-centric.
We had another definition that patient medical record information is basically any information needed by a clinician to care for a patient under his or her professional responsibility. And it went on to say in addition to clinical and administrative areas, this includes enough financial and insurance related information so that a determination can be made as to what services or medications or procedures a clinician is able to order for that patient. So this gave an expansive view of what types of data are included in patient medical record information.
Another definition, this is number seven, and there were ten that I listed, computer-based patient medical record information, and now in this definition we are having the prefix about it being computer-based, otherwise called CPR, is electronically maintained information about an individual's lifetime health status and health care.
It replaces the paper medical record as the primary source of information for health care meeting all the clinical, legal, and administrative requirements. It necessarily encompasses all data related to the person's health care, including demographic, clinical, and financial. So again, this took an expansive view from the content standpoint.
The patient medical record information encompasses the entire content of today's ubiquitous patient chart and all of its various permutations. This includes literally all health care information from the patient's medical history and initial assessment to the record of conversations with and/or about the patient, to the typically extensive document associated with an encounter, visit, or episode. It will also include the full gamut of administrative, financial, and clinical data representing a patient. It sounds like a chart.
Another definition was data required to deliver appropriate, effective, and efficient care at the appropriate place in time to achieve optimal health outcome. That's a nice definition.
MR. BLAIR: Which number was that?
DR. FERRANS: This was number nine. As another part of that, bringing in prevention aspects of the care mandate, and implicitly the population focus, which is something I think that we have discussed a number of times in our discussion.
We believe it becomes necessary to make the distinction between patient medical record information and what they call person health information. The former having acute care and high acuity, probably a disease connotation, and the latter being more concordant with NCVHS defined computer-based health record.
Number ten, computer-based information on health care and health status can facilitate coordination, research, and assessment for both clinical care and public health, and permit individuals to participate more closely in their own health care.
I made a few comments, and I kept my comments brief on this. The original IOM definition of the patient record, that being a repository of information about a single patient generated by health care professionals as a direct result of interaction with the patient with individuals who have personal knowledge of the patient or with both, that this while useful in that provides a patient-centered view, does not necessarily address the comparability, interchangability or population-based needs for the information that is certainly part of the charge of this work group as we are looking for a definition.
The data could meet the IOM definition, but not serve the broader needs achievable through data interchange and aggregation. However, PMRI is not necessarily comparable. It is what it is. So should comparability be part of the definition, or should it be a goal derived from standardization of? It's a question.
This definition does not take into account the difference between the paper-based and the computer-based information. There is no distinction in PMRI if it is paper or computer. So should we include that in our definition or not? Are we talking about computer-based patient medical record information? Or are we just talking about patient medical record information? I think looking back at what they said in the law, they talk about it and the electronic exchange of such, so I don't think necessarily it needs to be added in as a modifier.
Associating PMRI with health care and health status I think is useful as a definition, since wellness information is as important as information on illnesses. And this method adapts to both individual and population-based care. The distinction between PMRI and personal health information that was mentioned before I think is an artificial one, and we don't really need to deal with that.
The definition when I talked about uniform data standards for PMRI, and electronic exchange of such must take into account the concept of uniform data standards. It doesn't seem to preclude paper-based records, but it must have the capability of being electronically exchanged somehow.
So putting all this together, I include just a couple of definitions that were in here, and put them together into sentences. I put down as a preface I think it is useful to start with the Institute of Medicine definition, so I'll read it again. Defines a patient record. So we define the patient record, the repository of information about a single patient. This information is generated by health care professionals as a direct result of interaction with the patient, or with individuals who have personal knowledge of the patient, or with both.
So this is nice, because here this says what the patient record is, and it at the source, is very direct. Then my definition that I came up as a draft as the starting point in this discussion is patient medical record information, PMRI, is information about an individual's lifetime health status and health care needed to deliver appropriate, effective, and efficient care at the appropriate place in time to achieve optimal health outcome. It is used at multiple levels for the process of overseeing and managing patient care and patient care processes, as well as for monitoring public health.
So that can be the starting point. I would just add looking at it again, it's kind of long.
DR. MC DONALD: I worry that you have defined in operational terms. You're saying that data needed for this purpose, which is effective and efficient care. So now you have to say is an S4 gallop(?), which we record on the chart, does it qualify? So you discuss this operational definition on it, and I don't think -- you also want it provide all this care, but it's defined in terms of that.
MR. BLAIR: Other comments?
DR. COHN: I have a question for Kathleen Fyffe. I think since Kathleen had earlier indicated that she wasn't clear what PMRI is, does this help you?
MS. FYFFE: No, it doesn't. Sorry.
MR. BLAIR: What areas of ambiguity does it leave?
MS. FYFFE: Well, first of all it mixes up an individual's health status, and the health care needed to deliver appropriate, effective, blah, blah, blah. I mean, you could say that everybody in this country needs cadillac health care to optimize them individually, but that shouldn't be part of the record. It's too health care needed. How about health care received? I have several problems with it, but I'm going to have to take a little time to think about it.
DR. MC DONALD: If you go back up to the Institute of Medicine one, does that define it?
DR. FERRANS: I guess it's focus on a single patient.
DR. MC DONALD: All we're trying to do now is just characterize this class of information, which we describe with this name. The disadvantage of yours, you are saying things that are true and we all support, but they aren't central to the definition of it.
MS. FYFFE: I agree with what you are saying Clem. It also talks about overseeing and managing patient care. Well, is that actually in the medical record information, or is that something that is outside of it? Or is that a decision that is made about the information in the record?
DR. FERRANS: I guess that it is used for that purpose. It's part of the definition.
DR. MC DONALD: It's a true statement about it, but again, what you need is to just define it.
MS. FYFFE: Yes, it's too broad.
MR. BLAIR: If I'm hearing correctly, it sounds like there is pretty much agreement on that Institute of Medicine definition. And I think when you started to go beyond that, one of the areas I think that you felt was lacking in the IOM definition was some allusion to wellness or continuity of care. Is that correct?
DR. FERRANS: Two things, wellness, and I guess public health, and not just about the care of the individual patient.
MR. BLAIR: Are we able to sharply focus on adding those two concepts to that Institute of Medicine definition?
MR. MAYES: I don't think that -- I mean public health monitoring is served by aggregating data from individual patient records. It is not in and of itself, part of an individual's patient record.
DR. FERRANS: Let me change the term "public health" to population-based.
MR. MAYES: The same thing; it wouldn't make any difference to my comment. Actually, I like them both. I could go with either one. I think the shorter Institute of Medicine one is actually pretty complete. It gets away from implying the various uses of it. But as Clem says, that's really not part of the definition, or a necessary part of the definition. It is more of a modifier.
DR. FERRANS: The only other thing about the IOM one, not to nitpick with them, but if it's generated as a direct result of the interaction with the patient, what do we call laboratory data, x-rays, things like that?
MR. MAYES: You have to take a sample from a patient it's a direct interaction.
MS. FYFFE: Or if you have a pathologist who is reading a specimen, he has personal knowledge of that patient, of the specimen.
DR. MC DONALD: I guess the question is, is there information that we would usually categorize that doesn't fall into this category?
MS. FYFFE: Well, let me make some comment about that. Wouldn't it depend upon the definition of health care professionals? Because you have kids with attention deficit disorder that are evaluated by school psychologists, so there is a record there. You have got people who are injured on the job and so there is a worker's compensation file held by a human resources manager. Now is that part of the patient record? So those are under the definition of health care professionals, regardless?
DR. MC DONALD: It sounds like they worked this for a long time. If some examples don't fit in, then we have to modify it.
DR. FERRANS: What about administrative and financial data?
DR. MC DONALD: That's a good question. I almost don't think of that, even though some of that is added in there as part of the patient health care information.
DR. FERRANS: This is a debate. Obviously everybody wants to keep the focus on clinical care. At times it is helpful to have that other data in conjunction with it to look at past pregnancies, outcomes, whatever.
MR. MAYES: I don't see where it is restricted to that. It's a repository of information about a single patient. Fair enough. The information is either generated by a health care professional, or as a result of an interaction with the patient, or with individuals who have personal knowledge. It doesn't say those individuals are health care professionals by the way, of the patient or both, and it doesn't restrict. It just says it's information about the patient. It could be about their income if that were relevant. It doesn't say anywhere in here that it's clinical information. It is just generated by a contact with the health care system.
DR. MC DONALD: I think we want to take out the first three words, because this is talking about the record, and we're talking about the information within it.
DR. FERRANS: So you want to take out the repository of. You want to say patient medical record information is information about a single patient.
DR. MC DONALD: The other advantage is it has some panache because it came from another body.
DR. COHN: Actually, I like the suggestion with the IOM and removing those first couple of words. I do think we need to think a little bit --
DR. MC DONALD: Do you want to give examples, such as?
DR. COHN: I guess that's almost what the next paragraph begins to get into.
DR. MC DONALD: No, I mean specific examples such as findings, symptoms, diagnoses, results of diagnostic studies, history.
MR. MAYES: That's where you start to get into the issue of the structure of the medical record.
DR. FERRANS: In one way the first paragraph is about what the definition will be. The second paragraph is not to get into terminology again, but what we all understand what it is we are supposed to be talking about. I guess that's what I was sort of trying to make explicit, because we dance around this.
DR. LUMPKIN: I've got problems with two words. I kind of like staying with the IOM definition. I think it gives us some credibility. The first is health care. I'm trying to think it through some scenarios, because it talks about a direct interaction with the patient or individuals who have personal knowledge of the patient or with both.
Now a child who lives in a house with lead on the wall, and that information becomes part of their health record. It was collected by a sanitarian. That person would not be a health care professional. They would be a health professional. So that's my first one.
The second one is professional.
MR. MAYES: You could argue that that individual does have personal knowledge of the patient in that he knows the patient lives there and reported it.
DR. LUMPKIN: But I think that the second part is really dealing with the fact that you get historical information from --
DR. COHN: I have a suggestion here. John, we're looking for examples to understand this one. Having just read and MPRM recently, I like first of all quoting the IOM. I think we all like that. I think that maybe what we want to do is to do that, and then maybe add a sentence where we would expand the definition to include maybe some other pieces of things that, because it both aligns us with the IOM. It doesn't mean we have to go around rewriting their definition.
DR. FERRANS: So something along the lines that we adopt this definition with the understanding that the committee felt that the term "health care professional" may have been too restrictive, and that we would broaden what would be considered a health professional to anyone who delivers health services to that individual, or something like that.
MR. BLAIR: Could we have health professional and care givers, or is that still not broad enough?
DR. MC DONALD: The other way to fix this is to interact with a patient or their environment. Would that fix it?
DR. LUMPKIN: Or we could give examples. Included in this would the following kinds of individuals. That's where we could say care givers, sanitarians.
MS. FRAWLEY: Why don't you just say health personnel instead of health care professionals?
DR. LUMPKIN: Because we're trying not to change the IOM wording.
DR. KOLODNER: The IOM wording doesn't allow for the person himself to add any information in there. The way this sentence is structured is that the information is generated by the health professional either as a direct result of interaction with the patient, or interaction with other individuals who have knowledge of the patient. It is not in the way it is structure --
MR. MAYES: It doesn't say other individuals, it says with individuals. And the patient is such an individual.
DR. KOLODNER: Right, it is the information generated by what whatever we label X, by their interaction with the patient or the individuals who have personal knowledge of the patient.
DR. MC DONALD: I think they meant that. We can argue about whether we mean that. I think they meant that, because if someone has got a private record at home that has no bona fide on it, I don't know if that necessarily counts. If the patient says I really am very well. My mother says I'm very well. I've never been sick in my life. Does that work when you come in to do work in a kitchen?
DR. FERRANS: If he records his own fasting blood sugars at home, and writes them down.
DR. LUMPKIN: It's patient entered data, but it's generated by an interaction with the health provider.
MR. MAYES: It's actually not patient entered data. It's patient captured data, but it's entered into the record by the health professional.
DR. LUMPKIN: Because I was thinking what about the patient who comes to the office, and in many offices now they give them a questionnaire to fill out. But that's an interaction.
DR. KOLODNER: So we are maintaining the distinction between the PMRI and the personal health record?
DR. MC DONALD: That's another discussion.
DR. LUMPKIN: Let's take another example. Let's take the individual who reads their own personal medical record and has a disagreement with what is in there. Then they write in a note. That's still an interaction. They have now interacted with that health care provider. They disagree with what's in there. They send into that health care provider, whoever the keeper is, to add that to the record.
DR. FERRANS: I'll just say that it talked about the person, the computer records, and the community health records, whatever the three types of records. You could also add on saying it's all the information meeting within or that it constitutes. That would somehow broaden the definition to include all those other things. That's another way of looking at it.
MR. MAYES: I guess Rob's point actually goes to the fundamental issue of who owns the data. The motto that we have is that basically it's the health care provider. It's the health system that owns these records and the data. It maintains them and manages what goes in. It could have input from a whole variety of people.
Another model that is gaining some interest and currency is you actually own your own information, and you allow others to add to that information, but you still retain the fundamental ownership.
DR. MC DONALD: You don't have to get into the ownership issue. What the model we have today really has is an independent player besides the patient just declaring. The individual saying I have no TB, so now I can work in a nursing home. There are a whole bunch of notions of the secondary professional, whose professional eyes by society, which is a big deal, your license and all, and that has a different meaning.
But I think the whole rest of this is to explore, and that is what we are really going to do with patient recorded information. Do they write it down right? All the other kind of things. And there is the traceability you have to have for all the other kind of records? So I think it's a great explorer, but I don't know that we want to build the whole mansion on it.
DR. KOLODNER: Our charge may be limited to -- and I think that there is reason to maintain the distinction between a personal health record and the patient medical record. Our charge is in the area of patient medical record, and the interactions, and the exchange of that for monetary reimbursement, and whatever else. So we can stay focused on that.
I think it would just be presumptive to present that as if it is everything, because it isn't. And I think there is a certain chance that that will be superseded by something that the patient feels meets their needs better if it's their personal health record that encompasses all of their record. But what we can standardize or what we can work on is that interactive thing. The one that results from the interactions with the professionals.
DR. COHN: This is an interesting discussion, because if we remember looking through the work plan where we talked about standards for national health information infrastructure, and here we are talking about personal information as opposed to patient medical record information. I think that we probably actually want to be inclusive rather than exclusive on this particular definition, I would think just taking cognition of the NHII activity.
Just to try to put things together, I'm hearing a couple of things. I'm hearing people generally like the IOM recommendation. I think there are a couple of areas where people feel it needs to be expanded, or there is some argument about whether it includes it or not, depending on how you define certain things.
I think if we're as authoritative as this group, and can't figure out whether it is included or not in the definition, we might want to then take the liberty to clarify or expand the definition to explicitly state that it does include things. And John mentioned one where he talked about professionals who deliver health services, and give a couple of examples as one group this would now need to be expanded to.
The other I think is probably the patient directly entering information in. That we would probably not want to necessarily exclude that from our definition. Anyway, that's a proposal of two things I heard.
DR. FITZMAURICE: I came in late into the discussion, so it gives me license to run over anything anybody may have said. But I'm thinking of an empirical definition. We haven't had any trouble in talking about these. We've gotten through hearings and nobody has said that is or isn't as part of the patient record. So we've all got the same kind of vision.
I would state it as patient medical record information is that information generally found in the medical record of a person. You could say health record of a person. Then you could go on to qualify it however you want to. You could talk about who maintains it, who stores it, who enters the information, who uses the record, and what kind of information it is. But it is basically what you find generally in a medical record.
It's circular. It's an empirical definition. It's not a definition that doesn't use the same words, it uses the words of what it's called. But it's what is in the medical record. And to really define it, you need to have somebody go through a billion medical records, list out all the things that are in there, and that's what it is.
DR. MC DONALD: What the previous thing was that we accepted this definition in -- we're politicized. There is one species as a patient and another species as providers. The provider species doesn't really care about the patient, and all this other kind of stuff which I said I resent.
I think let's work with this a little longer. What if the patient, all by themselves wrote down in a letter, I'm healthy today, is that part of the medical record?
DR. KOLODNER: Clem, you are picking up on entries that imply a judgment, as opposed to I have a pain in my left ankle. There is nobody else who knows better whether or not I have that pain.
DR. MC DONALD: I'm not disputing that. But there is a difference. Like when a tree falls in the forest and no one is there to hear it. I think there is maybe sort of a philosophic, metaphysical thing here. I have my great-great-grandfather wrote a diary every day for 14 years. Almost every day he says something about how sick he was. I don't think anyone considered that a medical record. But he said things about his health.
DR. FERRANS: It could be considered classically a symptom.
DR. MC DONALD: Well, still and all, I think there is a phenomenon. People are reporting to their nurse practitioner, their podiatrist, whatever. There is that interaction that has something to do with making this an official document of a symptom. If they do it by themselves and they tell their whomever, then it becomes this more official thing. If they are just thinking it, gee, I'm not feeling good today, it isn't part of the record.
MR. BLAIR: Could we kind of have a process so that we don't lose the benefit of where we are going in keeping us together? Richard, it does appear as if we focused in on the beginning sentence out of that IOM piece, and Dr. Lumpkin has indicated a modification with respect to eliminating the word "care" and "professional." Could you kind of after each person makes their suggestion --
DR. LUMPKIN: If I could suggest that what I came to was that we would adopt the IOM definition verbatim, and then add some explanatory or illustrative language a couple of sentences afterwards that would make sure it's clear what we mean by the terms that the IOM used.
MR. BLAIR: Could we go down that path then? Are you able to suggest a sentence or two that are good examples?
DR. LUMPKIN: I would say including reports by sanitarians on housing conditions, something like that, that sort of gets to the broader definition of a health provider.
DR. FERRANS: In some ways, and I don't mean to split hairs, but if we are going to talk about people who are not classically considered health care professionals, but we want to keep the IOM definition, we ought to add to the IOM definition and but the modifier in addition. Information generated by other people, other methods are also considered part of patient medical record information. Then we can put all the qualifiers we want as to things that fall technically outside, but are true to the definition.
DR. LUMPKIN: I think the essence of what the IOM says, at least my interpretation now, having listened to this discussion, and really trying to look at their definition, my assumption, having seen some of their processes is that they didn't come to this definition lightly. That there is a process that occurs. How we got into medical records is by a patient going to a clinician, a care giver.
Now there are numerical factors and objective data that is just transcribed, but there is other data that is interpretative. The patient has a respiratory rate of 15. Some of that data is very important because it is interpretive. The patient looks cyanotic. You don't have a picture of a blue patient. You just have to make that judgment, does this person look cyanotic? Well, if I say that, it conveys something to every other clinician.
That's what the medical record it, an interpretative process. That's what the definition is getting to. If we define it in that way, it's maybe more restrictive than it could be, but I think we're going to benefit in this process in having a more restrictive definition, because the smaller the thing we try to get our arms around, the more likely we are to actually achieve something.
MR. GARVIE: I guess I agree. I think that we can account for -- this has been said before -- we can account for those areas where we feel we need to elaborate or whatever in subsequent wording. For example, just as another example, today the term "repository" might have to be qualified if we are pulling medical information -- I was going back to the idea that we simply adopt this.
DR. MC DONALD: We didn't agree to adopt it, except that we are going to throw out the first three words, because we're not talking about the definition. We're talking about the information.
MR. GARVIE: Okay, all right.
DR. COHN: But otherwise leaving the quotes in place.
MR. GARVIE: Okay, so you'll dot dot dot into it.
DR. COHN: Let me just make sure where we are at, because I think we have reached consensus, which is that we are keeping the quote. We are adding some additional explanatory language as a way of being illustrative and whatever afterwards that would potentially further --
MS. FRAWLEY: I came up with a sentence to follow on the IOM definition, so I'm just going to throw it out. It's to try to capture the fact that there are things that happen outside of health care encounters. The other thing that we saw in our testimony is that we also have administrative and financial information that we need to incorporate.
So what I thought we could add is the next sentence after the IOM definition is in addition to the clinical information captured by health care professionals or health professionals, whatever we want to say there, patient medical record information is information collected in encounters outside the health care setting, as well as administrative and financial information.
Health care setting reflects your sanitarian, or it could be something in school, where a school aide notices something. It may have to be tweaked in terms of some of the words, but that was what I was trying to get to, was that there are going to be these kinds of situations occurring.
MR. BLAIR: Can we declare victory on this?
DR. MC DONALD: No. The beauty of the current definition is that is a definition. I can operationalize that definition. I can look at license requirements. I can tell whether it is or is not. With your definition I cannot tell what's in it or not. Any encounter, well, I went to the movie. Sure it's the whole library of all information ends up sliding in.
So the challenge to those who want to loosen it is to put some boundary on it when you loosen it. It can't be just anything in the universe that people decide because we'll never get this thing done if it's everything.
DR. FITZMAURICE: But can it be anything that a doctor decides should be in it?
DR. MC DONALD: Well, this first one would cover that. We're not trying to focus on doctors, because it's truly an encounter because somehow they decided they had some kind of interaction.
DR. FITZMAURICE: But my point is that a doctor could decide that a patient going to the movie is important enough to put in the medical record.
DR. MC DONALD: The challenge is to broaden it in a way that it doesn't open the whole universe until it becomes a black hole.
DR. FERRANS: I think in some ways you were talking about outside of the traditional health care setting. Then that sort of evokes images of then we're trying to add in alternative medicine or whatever, which a lot of people are accepting as health care professionals anyway, so they already fall into the definition.
MS. FRAWLEY: It was just an effort.
DR. MC DONALD: If we could say something about outside, that just says everything else. So it certainly is a constraint. Everything gets in.
MR. BLAIR: Can you think of a way to constrain it? Because once you start opening up beyond the health care professional.
DR. MC DONALD: I think this works for now, for the purpose we have, and for the purposes of what they are talking about. We're not really dealing with -- all the special regulations don't apply to what people keep at home in their own private drawers.
DR. FERRANS: I think I'm getting to where Clem is.
DR. COHN: Fine, leave it the way it is?
DR. FERRANS: Because by and large if there is a report from the sanitarian, it is going to go someplace into the health professional system. So are we going to define it as PMRI is information about a single patient, and from there on?
DR. COHN: This information is generated by a health care professional as a direct result of interaction with the patient, or with individuals who have person knowledge of the patient, or with both?
Let me just make a suggestion on this one, because we are going to easily go to 5:30 p.m. or 6:00 p.m. debating the perfection of it. If you have suggestions and want to write them down, I think we can consider them at another time to add on as additional pieces if there is something that you can put down that makes sense. Why don't we leave this for the moment, and then in September if we can come up with another sentence, that's fine also.
MR. BLAIR: Richard, thank you. That was a very easy job. Thank you for enabling us to come to this point. So you've pretty much got that phrase? Could you communicate that in a form to Michael so that it is captured in our record here?
MR. BLAIR: My inclination right now is to go on to the topic of the interim letter to the secretary, and then we'll end up talking about the tasks to pull that together, since that's one of the tasks.
Michael, I would like to ask you to maybe take us through that. There is a copy in the handouts of that notorious email that I sent sort of outlining. That's sort of a starting point.
DR. FITZMAURICE: The purpose of the letter, we need to be clear about before we look into what it contains. So what do you think the purpose of the letter should be? I would suggest that it is to inform the secretary about what to expect in the report to her that we will be sending out.
Here is what we were asked to do. So Jeff has the charge of the CPR work group. Here is the definition of what we are reporting on, the definition of the patient medical record information. Here is why you should consider it important, madam secretary. Here are the focus areas on which we're concentrating in the report. Here is a preliminary issue of potential issues that we expect to address in the report.
We then might ask for any input. But I think generally a letter to the secretary that simply reports on our activities, and that she and her staff would feel free to come back to us and say we want you to look at something else as well. And then a thank you and a closing statement.
We have the focus areas as developed in the work plan. We have the definitions of patient medical record information. We know what the charge is. It's in the law. The only thing that isn't concrete before us is why this information will be important to the secretary.
And secondly, the preliminary list of potential issues. We can glean some of those from the questions that we have been asking at the hearings -- what is the role of the government? What kind of support should the government give? Where are the gaps in standards? Do we overlap, duplicate, or conflict with standard development going on elsewhere in patient medical information, such as international. Again, I wouldn't give that a large weight. We probably wouldn't mention that in the letter. But maybe that's an issue. Maybe I would mention that as an issue.
DR. MC DONALD: I wouldn't.
DR. FITZMAURICE: Clem wouldn't.
DR. MC DONALD: It's not international business.
DR. FITZMAURICE: Then why are we concerned with the European directive on privacy if we are not exchanging patient medical record information and other information that is identifiable by the individual with Europe. It's a concern.
MR. BLAIR: If it starts to affect our ability to develop standards in the US, then it becomes an issue.
DR. FITZMAURICE: Let's get on with is this what you want the letter to say? I'm proposing this for your consideration. I'm not saying this is what it should contain. I'm saying this is the way I see it playing out. That madam secretary will look at it and say, ah, Congress charged with doing this. This is what they are looking at. This is how they define it. It's important to me because, and here are the focus areas, and then here are the issues.
MR. MAYES: I just have a quick question of clarification actually for John Lumpkin. This was a mandated report to the secretary. And obviously the focus of this work group is on these. But when you look at the other work groups, whether it's the population work group or the security and privacy work group, whether it's the infrastructure work group, many of the items that they are working on could in fact be cast within the framework of a patient medical record information, standards and the exchange thereof.
Is there any expectation that the final report going up will include sections that were developed by the other work groups? Have there been any discussions from any of you that are sitting in on the other work groups? None of the other work groups have this nice, clean mandate to actually send a report forward. I just bring it up, because if there is, we should mention it in the letter.
DR. COHN: Maybe you can help me think about this one a little bit, because you are bringing up an interesting issue. I can appreciate how the health information and infrastructure work group, there is a fair amount of overlap. I haven't been so aware of other --
MR. MAYES: I'll give you an example that comes to my mind. The security could probably be handled separately, because of all the privacy and confidentiality issues. But for instance for the population, there is an increasing interest in core data sets or minimum data sets for various types of activities. I know that that has come up in discussion for instance, like the OASIS and things like that.
So I don't know whether there has been any further discussion in the population group. But certainly if we were to recommend as one of these health data standards that there would be some sort of content standards or a move towards some nationally defined data sets, something like that, I just don't know if it had come up. I was just asking the question.
DR. LUMPKIN: The answer is no.
MR. MAYES: Then we don't have to worry about it.
MS. FRAWLEY: The question I want to raise is that we had discussed this before in one of our conference calls. I thought we had come to the decision we weren't sending a letter to the secretary. So I'm kind of confused as to why we are talking about sending a letter to the secretary. Let's just say it will go this fall.
MR. BLAIR: Could I clarify that? I think it was back like in the February meeting where I think if I recall correctly either Michael or Bill Braithwaite had suggested that it would be a good idea for us, because there is such a long time frame before our final report would be in place, for us to send a letter to the secretary indicating what we were doing, and the purpose of what we were doing, an interim report.
Then I believe in the March time frame, a little bit later than that, we wound up coming to the conclusion that we didn't have very much to tell her back in February or March, so we deferred this until June to talk about it. So this is the first time we've had a chance to try to say well, what would we put in that letter. Is that what you all recall?
DR. COHN: I was actually going to add a little bit to that, because I think it certainly has been in our work plan. It was actually for June, but we weren't really ready in June. At this work group we had expressed concern about the fact that the report was going to occur right in the middle of elections next year.
I had brought that issue up to the executive committee, to talk to them about that issue. I think in a way with the recommendations, there needs to be some communication planning. And so that meant among other things that you didn't put all the information in just a single report one time that might happen to hit right in the middle of elections. But you tried to create some awareness of what was going on early on, and then we would continue then on from the August time frame.
I think that was the recommendation to the executive committee, and I think it was a recommendation that most everyone concurred with. This is really not meant to be a 20 page letter. This was meant to be a relatively short letter of this is what we are doing. Be aware this is what we are looking at.
DR. MC DONALD: Is there a downside? I think what you are just saying is sort of what's in the document we have more or less, the work plan.
DR. COHN: You mean in terms of this piece? Yes.
DR. MC DONALD: If there is a downside, we need to hear about it.
MS. FRAWLEY: I don't know. I guess staff will be able to tell us if there is any downside. So we are talking about something that would be relatively brief, not very, very detailed. I'm just concerned, because we still have the big report facing us.
DR. COHN: Exactly. This was meant to be an interim, relatively short piece. This is not going to be a 30 page report. This is going to be a couple page letter.
DR. FITZMAURICE: It may simply list a focus area, rather than give them the whole description.
DR. FERRANS: In the bullets I thought that one thing that was missing that probably the secretary would want to know is what we have been doing. I don't think it's just explicit on this list. Certainly the charge, the definitions are widely important, and we have the work plan focus areas. Then we can say we have elicited testimony from a variety of experts from blah, blah, blah. Here are the preliminary lists of potential issues that we've gotten.
DR. MC DONALD: Or the opinions.
DR. FERRANS: But in terms of saying here is our focus, and then we took testimony; very brief on that. Then list the issues that have been pulled out of it.
DR. FITZMAURICE: Other comments about this?
MR. BLAIR: I think we're ready to move on to our last topic for the day.
DR. FITZMAURICE: But let me ask a question. Would you like for me to draft a preliminary copy of that and circulate it around?
MR. BLAIR: Yes.
DR. FITZMAURICE: But feel free to make changes in it.
MR. BLAIR: Simon and I have begun to start to look beyond the gathering of information to the analysis, the pulling together the issues, and then taking the hard step after we wound up pulling together the issues into winding up, saying what our recommendations would be. And then taking it through the process of getting the approval of the full committee, and whatever approvals we have to do.
This is really at the very loose conversation level here. There is a lot that apparently we going to have to get accomplished in order to have this letter to the secretary by August 2000. Simon, jump in at any point here, because I'm beginning to grope a little bit.
DR. COHN: Sure. If indeed we're talking about getting a letter to the secretary, we've got it pretty well planned out through the end of this year. The question is how do we go from December through August and get a successful report to the secretary? I think there was a realization or a reminder anyway that June would likely be the meeting.
June 2000, would be the meeting that we would have to have this report approved at. This is not going to be, as we know, a two page report without controversy. So realistically, it needs to be at least looked at and discussed not only at the June meeting, but probably also likely at the March meeting of the NCVHS.
MR. BLAIR: At least the issues or recommendations at the March meeting, and then maybe pulled into like a formal letter. Is that what you are thinking?
DR. COHN: Well, I'm actually thinking that they are probably going to want to see things written out, and be able to make comments and think about it. I was expecting that there will be some controversy in relationship to this. I may be wrong, but I think we should expect to have things pretty much together, and that would be a draft.
But we would want their input, and then be able to go back between March and June and do whatever revision or call whatever witnesses we need to respond to whatever comments they might have about it. So that basically leaves us between December and March.
Jeff had commented on, and I know Kathleen Frawley had also observed the fact that we have a couple of months to finish up whatever analysis, identify the issues. I think Jeff had commented that December should be a time for hopefully coming to full agreement on what all the issues are, if not beginning to get some of the solutions. But effectively that time between December and March is our solution time. And I think there is a recognition that that is going to take a fair amount of work.
Richard, it sounds like you had a comment?
DR. FERRANS: I just had a thought that that's going to end our data collection time, and then we're going to enter into the data analysis. I'm just curious. We had some people who just testified in December, six months ago, and actually the year can be a long time, especially when things are in transition like HL7 for example.
I don't know whether we want to also at the very end elicit some written updates from anyone. I don't want anyone else to testify, don't get me wrong. But it would be helpful to get updates just so that we are current. Because it may have some impact. Some people are making great strides in sort periods of time. We don't want to base it on our recommendation on information that is outdated a year, especially since the report will go in August. SNOMED and REED(?) may be almost merged by the time that the report hits.
DR. COHN: So what I'm hearing is the suggestion that we try to minimize the actual testimony after December, but try to get things in written reports, specifically as updating, which I think is a great idea.
DR. FERRANS: I don't think we need any new ground.
DR. COHN: Other comments or thoughts about how to ease our work in that? I'm actually sitting here looking at this, and I'm sort of saying geez -- I made a joke to Dr. McDonald about us meeting six days in December. That's not likely going to happen, but I think we may have to look towards January and February meetings, not for hearings. But we may have to come up with some time for meetings to see if we can come to agreement.
As we know, it's easier to identify the issues than it is to come to agreement with what the recommendations will be out of those issues. And I think that's going to be some fairly hard work. Once we come to agreement on what the issues are, then it becomes an issue of wordsmithing.
Mike, I think you are going to have to look to see what sort of help you are going to need, or whether you're going to have to take off those three months from your main job. So I think we will need to know about what sort of help you are going to need during that time frame. But it's going to be very clear that we'll need to pretty much try to fashion a document in those three months.
DR. FITZMAURICE: Actually, I can tell you some of that right now. We've had good responses already from Richard, from Bill Yasnoff. I need issue papers on things that we agree upon should be fleshed out. Then we will plan to take from the issue papers, substance and body for the report.
There may be some summaries, some reviews of past testimony. We may have to hire someone to go through and write that up, and depend upon our judgment for what should be either more emphasis or less emphasis. But to put it into text that we can comment on. I would see that happening beginning in the summer, if we could identify the person to do that, so that the work group --
MR. BLAIR: Beginning as of this summer?
DR. FITZMAURICE: Because we've had the testimony in December. We've had the testimony in March. To put it together in a way that has connected sentences and connected paragraphs. And then we look at it and say if this had more emphasis. No, we don't think that is important, because the person who would do that might not have been here. It's a problem of not being able to clone everyone who can contribute to the report.
But if we continue to have outstanding responses as we have gotten this time on the definition and on the -- we haven't looked at it yet, but the literature review of the computer-based patient records. Then we identify the issues, we farm it out to the large staff that we have in support of CPRWG, and rely upon them to give us information on the issues.
So I see that process happening not in December, but starting well before December in order to make us more comfortable with what we think the final report is going to look like. Then our work is what are the main issues? Do we like the way they are described? So that we are consolidating the January-February-March arena, rather than taking pen for the first time and beginning to write anew.
Does that sound like a reasonable plan?
MR. BLAIR: In a sense, can I build on your plan a little bit here? If I am hearing you correctly, we pretty much gathered the information on the message format standards. We've got a running start on the medical terminology issues. So at least on those two -- those are big hunks. Those were our number one and two. That may be a good piece of it.
So you are saying at least if we got a running start on those two, then the other four we could evolve during the fall and December, is that right?
DR. FITZMAURICE: Yes, and I also see a component where in the first set of hearings we went out to get buy in from people. What did you think about the charge to Congress? So I would want a section of the report to show that we went out and we sought opinions from people, and they came in and they gave us opinions.
MR. BLAIR: Documenting the process that we've gone through.
DR. FITZMAURICE: It was an open and a public process.
MR. BLAIR: Okay.
DR. FERRANS: Would it also be helpful in this final report to have -- if you look at terminology, for example, there are probably between five and ten papers that cover a lot of the high level stuff that can be cited in addition to the testimony to sort of back it up. Looking at for example how the Institute of Medicine has done some of these reports. Certainly we want it not just based on the testimony, but on the perception that the appropriate literature has been reviewed. It would certainly add weight. I think we all know a lot of those. Some people on here have written a bunch of those.
DR. COHN: Is that something we would want to put in an appendix, a couple of short pieces on things? We can certainly ask for funds to have somebody write a paper for us to affix to it sort of summarizing and reviewing certain pieces on that. Is that sort of the way you were going on this one?
DR. FERRANS: Or put it in the report itself. I mean I don't have a preference either way.
DR. COHN: I think we can ask for it, and it's really a question of you identify the task, and then you look for the money. But I think we are talking about the next year's fiscal budget, which does start in October, as I remember.
DR. FITZMAURICE: Maybe some from this year's fiscal budget if we can find somebody who can begin this summer.
MR. BLAIR: Can I make sure that I understand what you are saying, because I was a little concerned. I sort of feel as if there may be some hard hitting recommendations -- I would hope there are some hard hitting recommendations that come out of the work group where we're pushing state-of-the-art forward. We're pushing standards forward so that it makes a difference. We are not just saying the obvious.
If we are doing that, I think that our focus -- we need to really make sure we have consensus within our work group where we really are ready to support something that is aggressive. And my thought is that if we go through a lot of documentation, rather than focus on the specific issues and recommendations, it might dilute our focus.
Now maybe I misunderstood what you were saying, Richard. Maybe you were thinking of something where we had already decided on the recommendations, and we wind up having staff that could wind up doing some research and maybe a position paper or a white paper to support positions we are taking. Was that what you were thinking?
DR. FERRANS: If we are going to do hard hitting stuff, I want to make sure that we back it up with the literature, and that we cite it, and that we have a scientific and broad basis behind it, and that it's not just based on the perception of the work group, or the selected people that can say, well, all you did was invite the usual suspects to the hearings. That could easily be a critique. I just think it's helpful to have broad support. It should obviously support what our recommendations are.
DR. COHN: I think that's a great idea, and certainly it's a reason to ask for some money to make sure that we can get these things done, and done in a professional way.
Are there other thoughts that people have as we -- this is not the detailed work plan for the year 2000, but it's the beginning of thinking about how we get from here to there.
I have actually one other issue that I just want to bring up to the work group, and that has to do with you were talking about communication planning for the secretary. I think there was another issue called beginning to inform and educate the overall committee. I don't think that this is going to work very well within March. They haven't heard anything about what we have been doing, and walk in with thick documents to review and comment on.
I think we are going to have to think about what it is that we want to inform them about, either through panels or some other mechanism during the meetings. I don't think we felt the need to do that until now, but certainly it's on there. So there probably will be a reason at the November meeting to begin to do something.
MS. FRAWLEY: I was going to make the suggestion that maybe at the November there be an actual briefing to the full committee on the work group. The document that we have that has the overview about what we have been doing with the calendar attached, something like that could also be distributed.
MR. BLAIR: Are you speaking of the work plan?
MS. FRAWLEY: Yes, that document. I think that would be very helpful. But I think that Simon is correct, we can't wait until next June to say here's the report, give us your blessing. So probably the November meeting, and then whenever we have our next meeting, which will be during March probably. It will have to be incremental probably.
DR. COHN: Maybe we can also talk to them about our issues document we have begun to develop by that time.
DR. FITZMAURICE: When we send the letter up to the secretary, the letter will go to the national committee, and they will be informed by looking at the letter and approving the letter what the issues are, what the focus areas are. So that's going to be at least the first wave of educating them almost a full year before the report comes out, but probably six months before they should see a draft report.
MR. BLAIR: This is an idea that just kind of the blue here, but on the back of this medical terminology work paper -- I don't know what to call it -- we've got two pages here of issues of what actions the government might consider. And admittedly they were derived from the medical terminology hearings, but a lot of those apply just as well to message format standards as well.
I almost wonder just to see what kind of consensus or lack of consensus we have. I guess there are five or seven questions of what role the government should play.
DR. FERRANS: Are you saying take out the words "medical terminology" and substitute the word "standards?"
MR. BLAIR: I'm not even talking about changing this document. All I'm thinking of is that if during this next week everyone on the committee looked at those and just gave some feedback to the rest of us on the committee on the role the government should play. Is it right for the government to consider subsidizing? Is it right for it to facilitate coordination? Is it right to wind up getting it to the areas of incentives with respect to our entire work plan?
I'm wondering, if the committee as a whole for example winds up saying subsidizing is not going to be something we are going to do, and we're not going to mandate standards, but we are going to concentrate on incentives, that converges our role very, very quickly. And we may be able to gather a lot of that stuff during this next -- yes?
DR. COHN: I don't mean to break into what you are saying. I would personally think that that might be a little premature, since we haven't heard from any users yet on any of this stuff. There are a lot of things beyond the user discussion.
MR. BLAIR: But that's September.
DR. COHN: That's right, which isn't too late I think.
DR. LUMPKIN: I tend to agree with Simon on that. I'm a little bit concerned. It might be an interesting exercise to collect those, but not collate them and hand them out. And then see where people's positions are in December after the hearing. But I'm a little bit concerned that the committee has taken a vote on a position, and then conduct a hearing on it, because the public may believe that it is already a fait accompli.
MR. BLAIR: Well, that's a good thought.
DR. COHN: But the idea of trying to do that process, but just after we have gotten enough hearings together to feel comfortable doing that, I think makes a lot of sense.
DR. FERRANS: I think questions need to be answered individually per focus area. I can just tell you that my own personal is -- I'm not concrete on any specific area, but I know it is going to vary, because the problems and the distance from the solution and the factors are all different for focus areas. So the role of government specifically may differ greatly.
DR. LUMPKIN: Just brainstorming off of that though, we've got a list of usual suspects. It may be of some benefit to send the questions to them and get their feedback. That may guide some of the direction of the hearings that we conduct on those issues.
MR. BLAIR: And the usual suspecting being?
DR. LUMPKIN: It would be the people who we would normally think of to invite to the hearings. People who have been at the hearings before. Say here are the questions we have refined. Here are some of the questions we are pondering. What do you think?
If there is agreement on some of the questions, total agreement, which I doubt, but if there is, then we know that we shouldn't spend much time at the hearings on those questions. But if there are some where there is disagreement, then we can perhaps better frame the issues that we want to get to at the hearing based upon what we have gotten in written comments.
DR. FITZMAURICE: If I can repeat what you are saying so that I understand it. We're thinking of questions to be asking at the hearing. Send the questions out and get feedback before we send the real questions out that we want the testifiers to answer? I think the only thing in the past has been the time available, but we should have time now to do that if we get a list of questions up.
DR. LUMPKIN: We've got earlier documents, the broad policy questions that we want answered. So we've got them. But that may not be exactly the way we want to ask them at the hearings, because there will be subquestions and some refinement that we'll want to learn about.
DR. FITZMAURICE: In other words, are these the right questions? What is your opinion?
DR. FERRANS: We kind of did that. Maybe one way to do it is to ask the more refined questions now in the specific areas. I mean we asked the question what role should government play, and we got answers like government should be facilitative. Government should impose. Government should stay out of the way. It's not a one size fits all. If we ask what should government do about this specific issue. Ask them the more contentious ones.
DR. COHN: Certainly if we present the issues paper that we developed for terminology, if we had that for all the various areas that we had, we could even in September send it out to the people and say, give us your comments on these issues we are pondering. What do you think about what's going on? Unfortunately, we haven't come to agreement on what those other issue papers are.
MR. BLAIR: The only thing is in a sense that's what we have already heard them. It's a synthesis of what we have already heard from them.
DR. COHN: But that's not what we heard from them as a group. These are a synthesis of comments that they made to issues they identified, in some cases without solutions. The question is now that we are beginning to look at the solution sets, do we agree that there is an issue, and in that solution set what they think. And I think we need to figure out how to put that into the process, because I think it's a very interesting approach.
MR. BLAIR: The reason I'm pausing is I'm trying to think of how you implement that. My thought is each of the individual testifiers has focused on their area of focus, and you wind up getting a fragmented set of answers. Let's work it through and see what we can do. I just don't know how to do it yet.
DR. COHN: I was going to make a comment about next year. I think one of the things that we'll probably start doing as the summer progresses is actually beginning to query people for dates in January and February, probably in some way in relationship to how will the NCVHS actually get scheduled for their dates which haven't occurred yet either. But I think one should expect that we will probably have to have meetings.
DR. MC DONALD: I think there's actual a law that says you can't have more days in meetings than your parent committee has days of meetings.
DR. FITZMAURICE: No, that was repealed three years ago.
MR. BLAIR: We deported that individual that set up that rule.
My thinking is in hearing how we are struggling, to be honest with you, on these issues of how do we pull this together, is there seems to be a need for at least a draft extension of the work plan into next year that we can start to react to and work on.
I think we need to probably do that pretty quickly, so that we can go through a few iterations, maybe electronically, so that by September we have some degree of consensus on what our work plan is for the first six or seven months of 2000. Would you feel the same way, Simon?
DR. COHN: Yes, that sounds good.
MR. BLAIR: So we've got some work to do.
DR. MC DONALD: To come back to Mike's point, which I missed. It's probably all closed. But I thought he was maybe saying telescope some of this stuff. Let's do some reporting while we are planning. I think that's a very good idea. I actually think that there is only X solutions to many of these issues, and could be discussing some of the options. You have mentioned some in the bulleted pieces here, like do we say you must, you might think about. Do have evolutionary versus revolutionary?
There is a whole series of dichotomies you put in there. It's true that we'll have to -- in the context of the message vocabularies there might be different issues. SNOMED-RT is going to be very attractive, and what do we do? Can we enforce anything that has high cost on it? How do we deal with this cost issue? There are all kinds of deep tangles, but we can grapple with them. We've got pieces to grapple with now. We don't need to discover any more.
MR. BLAIR: Simon and I were mentioning this, we have moved pretty quickly in the work group, but I think for the most part we have tended to educate all committee members in a manner where we have been able to retain pretty good consensus so far. That's a fine line sometimes we have to walk, but I think we want to continue to move quickly, but not so quickly that we lose our consensus.
DR. MC DONALD: I don't think that's what is going to make us lose our consensus. We haven't decided much. That's why it's easy to have consensus. Once we start to deal with the tough issues -- we may as well find out where the fracture lines are early. We can put this off. Some of the stuff we'll still discover.
DR. LUMPKIN: I don't think the full committee has been adverse to making decisions that have not been unanimous.
MR. BLAIR: I missed what you said. The full committee hasn't?
DR. LUMPKIN: Been adverse to making decisions that aren't unanimous. We prefer not to, but I think that we do have to keep in mind our time frames. And there will be some issues where we'll just have to make a decision. And people will feel strongly about it, but we'll all be buddies afterwards.
DR. FITZMAURICE: So if I could suggest something. I will have a draft letter to the secretary that will have some issues that we'll talk about in smaller groups like Jeff, Simon, Bob, myself, and anybody else who wants to participate.
DR. MC DONALD: I thought that early report was just going to reflect those issues we have already defined in your document. I didn't think we were making new issues for that.
DR. FITZMAURICE: All right. Tell me again where the source of those issues were.
MS. FRAWLEY: The work plan.
DR. FITZMAURICE: Those are the focus areas. But there is also a section on what issues need to be addressed. Once you see a draft of that, you can then be thinking about recommendations. It's going to be incomplete, because we won't have heard from many of the people who are going to be testifying. But it gets us started thinking about what is the range of things that we could recommend; who feels what way about what.
DR. FERRANS: I think one other suggestion, talking about when we were going to start talking about the issues. I know we had talked about having Betsy Humphreys, I guess here it's looking like it's December. I don't know when we were planning on doing that, but rather than sort of have her testify from the NLM point of view, maybe she could sort of synthesize the testimony.
That sort of leads into discussion, and we already have just this excellent what are the issues that are raised that we could then go right into. Then she sort of establishes some sort of framework of a discussion. That can really kick off how much consensus or lack of consensus we are going to have. We can just dive right into that one.
MR. BLAIR: So if there are no other issues or considerations? Yes?
DR. MC DONALD: We've got a couple more minutes. Let's just dive into one of the things that has a fracture over. Can we recommend any code systems? Or can we only make suggestions? I don't know if we can.
MR. BLAIR: Let's pick that one. The question working on that issue is do we need to pick one in order to accelerate -- are there impediments out there where we need to break a logjam and therefore pick one, or are things converging on their own for the most part, and moving forward.
DR. MC DONALD: I was actually talking about we pick CPT4 versus -- there is a nice one.
MR. BLAIR: We have already done the financial/administrative one. That's already done. Maybe the process of these things is -- Clem, I think the issue is what do we need to do to break logjams or create level playing fields or to resolve the lack of progress? I think that's going to help us decide where we need to intervene.
So I think if we articulate the fact that there is a lack of progress because the government hasn't adopted or picked something, then there becomes a compelling reason for us to adopt or recommend one.
DR. MC DONALD: Maybe what we should do is clarify what code systems really lack now. I'm thinking in terms of the CPT4 versus -- I thought we put that off -- the other code system.
DR. COHN: That's actually a discussion that I think is on the work plan for the overall subcommittee.
DR. MC DONALD: Okay, so that's not settled yet.
MR. MAYES: But, Clem, it's the more generalized issues. The legislation charges you to make recommendations and legislative proposals. It doesn't say make a standard. There is a lot of levels that this committee could work at and still add value to it.
DR. MC DONALD: That's the kind of thing I would like to be talking about. Should we be recommending process?
DR. FERRANS: Or funding?
MR. MAYES: That needs to be worked for the six work areas, what is the appropriate level of a recommendation? Is it clear that boy that there are just two competing standards and we ought to choose one of them? Or is it gee, there is nothing out here. The code set is really a good one. We were talking about it at the break. This whole issue of funding in terms of licensing and things appears to becoming a major wall that everybody runs against. There are issues, but boy when you finally talk about implementing it, particularly broadly nationally, this whole issue around licensing becomes --
MR. BLAIR: Even if it's not mandated. Because if it turns out that the marketplace for example picks SNOMED-RT -- I'm just picking that as an example at the moment -- it picks it, but it has major problems because of licensing complications and the costs. But technically that's the terminology of choice for the areas that it serves. Then maybe the government can play something to ameliorate or help that issue. On the other hand, if the marketplace is not picking it because of technical flaws, then that's another issue.
DR. MC DONALD: What about the drug codes? Didn't we decide, isn't that NDC?
MR. BLAIR: That's another one where we have had one of the drug code vendors wind up saying, look, they tried to converge, but market forces are preventing them from coordinating.
DR. MC DONALD: That's not exactly what happened. What happened was there were four companies who were collaborating, and then one of them bought three of them. And then they didn't need to collaborate anymore.
DR. COHN: Time everyone. We are two minutes to five, as I understand it. I would just remind Dr. McDonald, Clem, that we had identified that there were going to be many different approaches to many different problems, and there were going to be near-term solutions, and others were going to be long-term. I don't think that we can come to consensus in two to three minutes about all of this stuff, as perky as you may have gotten in the afternoon here.
I think that one of the interesting things is going to be structuring the discussion so that we can identify really what the issues are, and then figure out how to put them together.
DR. MC DONALD: I agree, we'll never do it in a few minutes. But we should be spending time on these issues now to get the thinking going. When you look at all the constraints, and you stop and think about it, there is no way on earth someone could do something, or there is no way on earth --
MR. BLAIR: Clem, let me make the argument, you wound up saying well, we haven't attacked the difficult issues yet. I would argue we have attacked a number of them, but they didn't seem so difficult, because we did it in a stair step manner, where we carefully built consensus. For example, the issues that are at the back of that terminology piece right now, it appeared from the discussion that we took things in a stair step manner where people accept those -- generally accept those.
I don't think that several weeks ago that we would have said that was an easy thing to do. So what I'm saying is that I think that we can tackle the tough issues, and build consensus if we carefully prepare ourselves and do it in an organized manner.
Are we in agreement that we should adjourn for today. Thank you everybody.
[Whereupon, the meeting was recessed at 4:59 p.m.)