January 6, 1999 Public Meeting of the Workgroup on National Health Information Infrastructure, NCVHS



National Committee on Vital and Health Statistics


January 6, 1999

Hubert H. Humphrey Building
200 Independence Avenue, S.W.
Room 405-A
Washington, DC

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


Work Group:

  • Don E. Detmer, M.D., Chair
  • Jeffrey S. Blair, M.B.A.
  • Daniel Friedman, Ph.D.
  • Richard K. Harding, M.D.


Introductions and Agenda Overview - Dr. Detmer

ASMB Inventory of HHS Telehealth Activities - Ms. O'Connell

Discussion of Second Draft of "Charge Statement"

Presentations on Computer Health Records:

Discussion on Core Elements for Patient, Population, and Personal Records

P R O C E E D I N G S (10:05 a.m.)

Agenda Item: Introductions and Agenda Overview - Don Detmer, Chair

DR. DETMER: Hi, Jeff, and good morning everyone on the committee. Well, I think I have ten, so I think we may go ahead and get started.

I'm Don Detmer, and chair the Work Group on the National Health Information Infrastructure. What we wanted to do actually today, and I'll start just by sort of going over the agenda very briefly. Then we'll go around the room as usual, and introduce ourselves.

We want to start frankly, getting up to speed a little bit on what is happening both in the department and government, as well as sort of in the fields that relate to this issue of health information infrastructure. So we have a number of both presentations, as well as some discussion items.

We do have also a charge statement that we would like to review and approve to take to the full committee that is meeting in February.

So at least that's my sense. Mary Jo, what did I miss with what we're going to do?

DR. DEERING: I think that's the core of it. There was some feeling at our very brief meeting that we had back in November that one of the tasks today is to begin to refine some definitions for each of the individual health records. And then based on that, begin to see how that builds into the matrix, and then be mindful of how that might impact our original vision statement.

DR. DETMER: Great. Well, with that, first I want to express my appreciation and gratefulness for all of you being here today. Let us then just go on around the room and make our introductions.

[Introductions from were made.]

DR. DETMER: The first item that we have on the agenda is the inventory of the department's telehealth activities. Erin, I think you are prepared to lead us into this?

Agenda Item: ASMB Inventory of HHS Telehealth Activities - Erin O'Connell, ASMB

MS. O'CONNELL: Yes, I am. Thank you for the introduction. It's a pleasure to be before you, and this is just going to be a brief overview, a little bit of an introduction to why I'm here, and how I got to be here. I thank you for inviting me to come speak with you.

I work upstairs for Dr. Callahan(?), the assistant secretary for management and budget. I am a budget analyst. Last summer he had begun the process of assessing where we are at the department in budget authority and spending basically for telehealth and telemedicine. I work for the Division of Budget Policy and Management. We do the cross-cutting issues. We maintain tables for cancer spending, diabetes funding, different kinds of health sector spending.

Through my bosses, the assignment was channeled to me just to begin to assess where each of the agencies are in what kinds of spending, what kinds of systems they are running, where are the barriers to service, what is happening.

This was not my primary assignment. I've had different kinds of assignments, but it's been a very interesting process. I will be giving a more complete slide presentation in the future, but today I just thought I would come down and introduce myself, and let you know that Dr. Callahan considers this a very high priority; not only assessing where we are, but helping maintain the foundation, and provide new direction for the future.

We'll be glad to be of assistance to the committee wherever we are needed. So it is an honor to be here, and please do use me, and work through me if you wish for any kind of data that you need.

So with that said, and with gratitude expressed, I do want to say that what we are doing is beginning to compile budget authority and general spending for each of the agencies in the department. I did a presentation to Dr. Callahan on October 2, which basically outlined for him my findings so far.

A lot of this is grassroots. A lot of this is calling around and seeing who, if any, are the contact people. But what we did assess so far -- and like I said, I will share greater detail later -- but approximately in the fiscal year 2000 request, there is appropriately $101 million being requested for telemedicine or telehealth services. It is pretty broad in terms of its scope.

IHS, the Indian Health Service has been doing telemedicine for approximately 30 years now. They have a very high reliance on store and forward technology. They also do use it quite frequently. They are trying to train a lot of the tribes to use the basic otoscopies for ear, and using the connections with the western hospitals in order to provide services to their patients.

We have the CDC involved. NIH, as you all are aware, is quite involve. AHCPR is researching any kind of barrier to clinical services. HCFA has been very interested and very integral in the development of regulations as you are also aware; and SAMHSA and the FDA. So we do have a whole scope of services, and they do range across the spectrum.

Dr. Callahan had specifically asked that we begin to collect data from 1993 onward. So prior to that, I will need to speak with other people. He thought that it would be a good idea to stick with the administration, and go from there.

So far I would like to basically take an assessment on what kinds of things would you like me to talk about when I come.

DR. DEERING: I think the point being that we had arranged for her to come for our February meeting, at which point her work would be truly advanced. So this was an opportunity to find out the source of information they are gathering, and see what would be the most useful to us. And given the fact that we too were trying to come up with a matrix of what was going on in matrix, if they are doing the matrix for us, at least covering most of the areas, that would be very helpful.

MS. O'CONNELL: I'll give you a brief outline of what I had thought might be useful, and you can tell me if further items are necessary. I thought I would go into a history of how the department started, and which optives or agencies were active from the beginning.

Then I would do a brief inventory of what kinds of services they are using. Then go into the increases in budget authority and spending over the last nine years. Then I would go into what kinds of barriers to services that the evaluation has told us, and highlight the differences between the perspectives and the definitions between optives or agencies.

DR. DETMER: That would be great. I think some other issues that would be useful are what are some regulatory barriers as well. One of the things I have run into in the past in telehealth was that although the telecommunications reforms a few years ago did some things, it doesn't necessarily move transmission to actually the point of use, all the way to the person, at their home and so forth. A lot of the game that can come in telehealth is actually dealing with chronic illness, to people's homes, with low end, very inexpensive equipment.

But if you don't have, as a regulatory decision, all of the communications companies just as a matter of regulation saying, you will do this, none of them competitively will take the first step to do it. On the other hand, the cost ultimately is not that great for them if all of them had to do it. But if one of them was to take the lead, they won't do it.

So I would also like to get some sense of what you found out, not just barriers that may be out there, but also sort of the barriers that aren't necessarily costly barriers per se, but nonetheless are things that perhaps the committee could bring to the secretary's attention, and to the White House's attention.

MS. O'CONNELL: Thank you.

DR. DETMER: Others?

DR. DEERING: One other thing that occurs to me is a simple delineation of the terms that you did end up using. If it were possible for you to even get that to us in advance. I realize that that too is a work in progress, and that's sort of what we are going to be doing today. But I know that you spent some time trying to figure out what do we call these things that we're trying to track. So that would help us.

MS. O'CONNELL: In terms of a definition of telemedicine, telehealth? Yes, I do have working definition.

DR. DEERING: It's not so much as a process definition, so much as your cells, your categories. I don't know what those categories are. I'm guessing some of them might be clinical services. I'm guessing some of them might be information -- some of NLM's work might come under information management. Again, I don't know how you have described those cells.

MS. O'CONNELL: That's actually a very interesting point, because as I was compiling -- I do have a working table -- as I was compiling, each of the optives and agencies had rightfully asked, what are you calling clinical services. Or what is the cut off for what is telehealth? What is the cut off for the equipment? So I understand where you are coming from. I will be glad to do that.

DR. DETMER: I think to some extent getting into this, a lot of our first paper I think was almost a why of infrastructure, and what pieces are in there. But increasingly as we move to laying out the terms and all of this business, it becomes a how issue. So I guess the other question, or to take a different kind of cut at it, trying to help us lay out these terms and such, but to the extent that you can now, but also in the future, kind of see us as maybe a port of call for some health issues, for deliberations, testimony, and so forth, we would like to be a resource to you folks, not just the other way around.

MS. O'CONNELL: Thank you. Well, I look forward to speaking with you in February, and again in the future, and I'll be glad to be of help where I can.

DR. DETMER: Okay, questions or comments?

DR. STEIDEL: Are you going to be touching on the legal barriers across state lines, et cetera, in your talk?

MS. O'CONNELL: I would like to. I don't know very much about them. Between now and February we are going to be thoroughly immersed in the budget. If I can address that, I will be glad to.

DR. STEIDEL: That's one of the biggest things that been in a lot of areas.

MS. O'CONNELL: I have a few lawyer friends who are involved in telemedicine issues, and also in licensing. I'd like to be able to speak with them, but I understand the concerns.

DR. DETMER: Yes, those are big, both policy and political, as well as legal issues. Sharing that would be a good idea.

DR. DEERING: I don't know whether you will touch on -- another aspect of the regulatory barriers has to do with reimbursement issues. Don's statement that the greatest bang for the buck would probably be in chronic care management needs, lowering delivery to the homes, what are we doing in HCFA and IHS, and even VA. I know Claude isn't here now. He works very closely in those issues.

Then a related issue, what is FDA from a device regulation point of view doing about this, because I know they have been sort of tiptoeing around the edge and sticking their toes and feet in from time to time. But those are two agencies whose actions impact our delivery of low end technology and content to the home.

DR. DETMER: Jeff, do you have any questions or comments?

MR. BLAIR: No, I don't. Thank you.

DR. DETMER: I guess to add to this, I think when you present your materials at the next meeting, and once we have gotten through this phase of getting things up on the table, it strikes me that what this work group will need to do is to lay out its own work plan for the year. But we've got to get more on the table right now to do that, because there is a huge amount of things in this, and decide what our own priorities and expectations will be for the year.

Some of that probably will require getting this stuff on the table, and then saying maybe we move, and where do we maybe make more impact, and start trying to lay out a couple year plan, something like that. Does that sound reasonable?

MS. GREENBERG: I just also wanted to thank Erin, and to note that as the sort of in-house historian at some of these meetings, because I've been working with the committee for a long time, and I think in my experience it's the first time the committee has had interaction with staff from ASMB. I think it speaks to the benefits of working with the Data Council, and having broader staffing, and just broader exposure.

I would confirm what Dr. Detmer said too, about that the committee should be seen as a resource from the point of view of hearings and other activities they can carry out. Hopefully, it can be a mutually beneficial relationship. I think this is really great. I thank you for coming.

MS. O'CONNELL: I was invited. I do want to add just one last thing. I do want to clear something up. When we were asked to begin to compile the information on telemedicine and/or telehealth spending in the department, at that time it was made clear to me only that it was Dr. Callahan's initiative that wanted the information. As to what drove his fire behind that, I'm not certain.

At this time, this report -- I'm working on a collection of my ideas and a report of where the department is. That entire paper was not shared with OMB, or it had not gone down to the Budget Office. So OMB was driving this request for information more about health information technology. That request came after I had compiled this information, and this was just added on top, because there was a little bit of confusion I believe among the department on who initiated ASMB's involvement.

The assignment came to me from Dr. Callahan, and through my bosses, but afterwards we only shared a piece of the information with the Office of Management and Budget. So I know there was a bit of confusion before as to was it OMB that was driving the request.

DR. DEERING: This might also be a chance to give Maureen an opportunity if she wants to say anything, because Maureen is really the matchmaker here for us. So helped make this happen. I don't know if you wanted to add anything.

MS. WILLIAMS: I think I'm the newest member of the committee here. My background prior to coming back to the department level was I worked a number of years at St. Luke's Hospital, then at a federal agency, then for a few years for the DC government. I was head of financial and legal affairs over there.

I happened to meet Mary Jo in another context and we started talking. I realized that some of the activities that you all are dealing with, are things that are near and dear to my heart from a prior life, but also by virtue of the fact of working directly for Dr. Callahan. What I enjoy seeing is a process, and how activities actually come to fruition.

One of my current assignments for Dr. Callahan is an officemate and myself are writing the plan for the department on how we are moving into electronic commerce. This is something that is amazing about what you are talking about today, how much of this is actually part of what we are already analyzing in the department.

So it's amazing how we can have a number of different activities linked together, but we can be very, very productive as long as the left hand knows what the right hand is doing. I really look forward to working with you.

DR. DETMER: Okay, well, it sounds like you will be a great resource to us, so I encourage you to really weigh in.

MS. WILLIAMS: No problem.

DR. DETMER: That's good.

Richard Harding just joined us, a member of the committee. Do you have anything you want to say?

DR. HARDING: I apologize for USAir. The crew couldn't get out of Pittsburgh to get down to Charlotte, but we got here. It's good to see you. I'm glad to be here.

DR. DETMER: We have just heard from Erin O'Connell about some work that she will be presenting in greater detail at the February detail; and Maureen. This is Erin, that's Maureen.

Thank you then. We will see you later. You are welcome to stay as long as you want.

MS. O'CONNELL: I would love to stay, but the budget calls.

Agenda Item: Discussion of Second Draft of "Charge Statement"

DR. DETMER: Our next item is to discuss the charge statement, and I think you've got it in your materials. It's called, "Second Draft."

What we hope to do is to revise this to the point that we can approve this today, and take it to the full committee for review, and hopefully passing. Right? Is that your goal? Okay. So that's open. It's on the floor for discussion.

MS. GREENBERG: May I ask one preliminary question? That is, we have noticed in just in roster, minutes, agendas, whatever, sometimes the work group is referred to as the Work Group on National Health Information Infrastructure, which is what it says here, and which I had thought was the name of the group. But other times it is referred to as the Work Group on A Health Information Infrastructure.

I don't think it fundamentally changes what the group is doing, but I think we need to agree on what the name, just for starters.

DR. DETMER: All right, what's your preference? Since we're the National Committee, I guess the question is, is the "and" assumed or not, really, don't you think? I don't see us sensibly not being national in the way we are looking at it.

MS. GREENBERG: Well, that's true. Also, I think the "national" links it in maybe a little better with the NII.

DR. DETMER: Right, oh, I see what you are saying. So you see although it's not a matter of real confusion, it might help to leave it on there?

MS. GREENBERG: I think so, but on the other hand, since I have seen other materials coming out with the alternative --

DR. DETMER: Certainly when you use an abbreviation, an additional letter, it's hardly a problem, NHII, as opposed to HII.

What are other thoughts?

DR. FRIEDMAN: I'd rather not change my file folders, so I vote for NHII.

DR. DETMER: All right, just so we have some sense on that. I don't know if you want to go down through each of these by paragraph, or if you want to lead us through it, or what, but it probably is a good idea to do that. Let's at least read the paragraph. We'll kind of go through it that way, and then see if there is comment or discussion or concern.

Any suggestions on the first paragraph? Too long, too short, things not there, not clear? Obviously, it's not as though these don't still get periodic revision or use, so it's not like we're going to put this in marble downstairs somewhere.

DR. DEERING: The language comes mostly out of the vision paper. The only changes made to this paragraph after the November were the addition of line four of the separate stand alone sentence, "It is not a unitary database." I think there is strong interest in reiterating that as strongly as possible.

Then in the two lines further down, "needed to improve the public's health and health care" was added to emphasize again the collective nature of this, and the public health and individual health, as well as health care specifically.

DR. DETMER: I don't hear much, so I assume it's okay. All right, how about the next one, "The areas initially identified in the Work Group's October," -- essentially these were the headings actually of the paper. What this basically is laying I think is a strategy at this point as basically monitoring information and identifying things, not so much making recommendations. Just trying to get a lay of the land, as much as anything else, but then talking about recommendations in the next one on ways to integrate these activities and move things along.

DR. DEERING: The other thing I'm noticing is that if you jump down to number four under what we do specifically, that's where we get into international models. Would there be any desire on the part of the committee to include in this second paragraph, the fact that we'll be monitoring international activity?

DR. DETMER: Well, we probably ought to. It is an item that is not listed -- it is listed in the paper, but it's not listed in these subtitles. So, yes, I think we probably ought to add it.

DR. DEERING: In the first sentence.

DR. DETMER: That's a good idea. International collaboration. We could say relevant global activities or something like that. I think that's a good suggestion.

Other suggestions?

MS. GREENBERG: In the second sentence I think it would be clearer to me if it said the work group will monitor activities in these areas of other -- because otherwise to make the connection, or will monitor these activities. I guess what you are saying is --

DR. DETMER: One is content areas, and the other was sort of where the thing is coming from. You see what I'm saying?


DR. DETMER: That was the goal on that, sort to lay out both the topic areas, but also then what primarily are things that we would see these coming from.

MS. GREENBERG: NCVHS subcommittees are related to all of these areas, I guess is what you are saying?

DR. DETMER: Yes. Do you want to wordsmith it?

MS. GREENBERG: Maybe it may be a little wordy, what I said. The work will monitor activities in these areas of other -- or relevant activities.

DR. DETMER: That's fine. That's easy enough. Why don't you put relevant in? It relates then to that. So we'll add international, global, something like that, and relevant.

Others? Okay, "The Work Group will make recommendations to the full Committee on opportunities and barriers to integrating these activities more effectively to help achieve the goals of the NHII."

DR. HARDING: In the last sentence of the first paragraph, "The NHII as a system should seek improve and enhance privacy and confidentiality." "Should" is a funny word. Is designed to, or that's the intent? Will improve, is intended to improve? Something should is kind of like maybe it will, maybe it won't.

DR. DETMER: I think it would be tough to say it will, because we don't know that it will until you find out. So I'm not sure what you are raising.

DR. HARDING: The intent is to make a difference, something like that?

DR. DETMER: Or should seek to, or something like that?

DR. HARDING: Seek to, that might be better.


DR. FRIEDMAN: Don, since we're picking, you may want to think about moving the first sentence of the first paragraph down to the second paragraph. I think it would just read better if we just started out with a statement of what it is, and then the second paragraph is really where we start dealing with what we are going to do.

DR. DETMER: All right, that's fine.

DR. FRIEDMAN: It would take a little bit of editing, but we could probably knock out a sentence or a few words in that second paragraph then.

DR. DETMER: Any others? Okay, let's move on down then to, "the Work Group will." The idea here is that we would see actually in a way the Concept Paper being something that stays a relevant document, depending on what is happening at any given time. So obviously I would think that you would want to review that maybe at least annually, don't you think, whether we make that explicit or not?

It seems to me like we would at least want to look at it in detail each year to see where we seem to be. So maybe review the concept paper at least annually, or something like that. Produce iterative revisions. Then the first task, that's fine. Other comments on one?

The second was a sense that we really needed to just sort of lay the terrain map out on this thing in terms of what is going on, and where is it happening. Again, I don't know if this is understandable English or not, but at any rate, that's the goal of the thing. Other comments on that?

Not hearing any, three, analyze it to look at progress areas meaning refocusing areas where new work is needed. It's essentially okay, once you've got the thing, so what? I guess the question in my mind relevant to identifying and analyzing relevant models, maybe to some extent it does say we will make recommendations to the full committee, but it seems to me like what I'm seeing in conversations with colleagues around the world is really a need for us to pretty actively try -- if we are the primary part of the NCVHS that is going to be tracking what is happening internationally on some of these things, and I think we are, aren't we, this work group?


DR. DETMER: Then it would seem to me like we shouldn't just identify and analyze, but I think really sort of routinely report to the full committee on this, because things are really active right now. You look at what Australia is doing, and Europe is starting to move. I think actually HL-7 even internationally is doing more than it has been.

MS. GREENBERG: Going back to number two, when you say "develop the matrix of existing activities," is it to be activities in all of these areas that were identified in the concept paper?

DR. DETMER: Right.

MS. GREENBERG: And if so, will they fit into this model of the three types of records? I see that the three types of records are a critical aspect of it, but I don't think everything will fit into those.

DR. DETMER: No, that's a good point; it may not.

DR. DEERING: I think the tension was raised from the beginning in that there was no pretense whatsoever that we knew exactly what this matrix would look like. To kind of get there, it seemed that one useful first step, and that would drive our work later this morning, would be to try and figure out exactly what is unique about those records and their core elements, and then see whether they would lend themselves to it. Certainly, there was a great degree of modest in our assumptions about how quickly we could build this matrix.

DR. DETMER: One option might be initially assessed against the elements of or something in light of that. You are right, they may not. The sense that I think in terms of the model, just to speak to that for a moment, is that there really appeared to be three domains of databases if you -- of data, not necessarily databases -- one relating to personal health information that people want both for personal education. Information about how to keep and maintain their own health; their own just health data in a personal kind of sense.

There is also a set of data that relate to what you would call the delivery systems data as it relates to direct delivery of services to patients, and what is involved in sort of managing and monitoring that. Then a set of data that related to populations; both aggregates of individual data, but in some instances, specific population initiatives that are not coming out of either of these other two personal repositories of information, or necessarily coming out of the delivery system database itself.

But I can't think of any other kind of domain, but I'm not saying that it wouldn't possibly exist once you went down through this. I'm not arguing with you.

MS. GREENBERG: What I was thinking is that may pretty much cover the waterfront on content, but if you look at some of these areas, they are going to cut across all of them, and maybe interact two different ways.

DR. DETMER: Part of the thinking -- and again, whether we are pushing something through this that won't go -- is how do we try to get from just something that is a nice kind of lay out everything on a board, and then sort of look at it, to how do we get something into concepts and frameworks that allow you to take action off of it?

I think the sense was if you parse it into these more actionable kinds of things conceptually, they may have faster legs. Now I don't know that they will, but one of the problems I think of governments, and particularly big governments is that you can be very busy just writing things, and not necessarily have it in a way that really creates an infrastructure, and serves things, and gets things done.

So that is my concern. That is the tension on the other side of this that comes with just sort of saying, well, let's not try to give it. But I'm not saying that you're with me, or that that's where we ought to go.

DR. FRIEDMAN: What I would suggest in light of what you are saying is to try to simplify a little bit that second point. Maybe something to make it a bit more general, to the effect of develop a matrix of existing activities related to the NHII, with particular emphasis on computer health records as outlined in the Concept Paper, and whatever other areas identified in the Concept Paper. Something like that; the words aren't quite right. Rather than defining exactly what the matrix would look like.

DR. DETMER: I think that's useful.

DR. DEERING: I have a question that picks up on both three and four, and harks back to the stand alone sentence, which is paragraph three. That says, "We will make recommendations," but it doesn't say what about. Now of all of the items that we have down there, it seems that three and four are the two that really lend themselves to specific recommendations, because you have just added to number four the concept of reporting back to the committee.

Clearly, the purpose of number three is to drive recommendations. So my only question is as much procedural as editorial -- it certainly isn't substantive -- is there is a need to bring in language to items three and four, or somewhere else about specific recommendations on or to reporting on, and follow-up on what is the outcome of that action?

MS. GREENBERG: You were going to add that to three right?

DR. DETMER: We're going to add it to four. We could add it to three, or we could in fact clump those four together. I think you point is well taken. I think you are right, they are more actionable kinds of items.

MS. GREENBERG: You could have between four and five.

DR. DETMER: In terms of recommendations.

MS. GREENBERG: You could have a new five, which is analyze, identify, and make recommendations related to these areas or report to the full committee, and then go on to promote and conduct.

DR. DETMER: All right, well, I think we can wordsmith that. I guess the question of whether six needs to be there. That's just sort of implied. We also are going to be awake. Do we need that there? I guess for completeness' sake, probably.

DR. DEERING: That was a protocol issue, that if you didn't leave the door open, since this is such a formal entity, did you need to acknowledge that you have limited these points that are specified, but I don't know.

DR. DETMER: Sure. I guess the other question -- well, no, I think it's fine.

Jeff, any issues?

MR. BLAIR: No, not at this time. Thank you.

DR. DETMER: All right. Others?

DR. HARDING: I have a comment on number five, promote, conduct, or collaborate with whom?

DR. DETMER: At least we would be in a position to respond to it. That doesn't mean we would necessarily do it.

DR. HARDING: Did that mean international, or the things above?

DR. DETMER: Oh, I see what you mean. Well, that's a good point.

DR. DEERING: With any relevant parties? With any relevant partners? With any appropriate partners?

DR. DETMER: Some of that would require obviously the sort of review and approval of in a given instance, the NCVHS itself, and quite possibly the Data Council as well.

DR. HARDING: If it's open-ended for a purpose, that's fine.

DR. DETMER: I guess what we were sort of saying here is that we did see us being in an activist role relative to outside players, as well as the government on how to move the agenda forward. I think that's really the intent of the thing, but I think your point is that it isn't very precise.

MS. GREENBERG: You could just say collaborate as appropriate, to leave how that gets defined by the full committee, by the Data Council, by the department, who knows, open.

DR. DETMER: Others? We had this as a draft. It's up to you as to whether we want to act on it and carry it forward. I would like to. It seems like we kind of massaged it a fair amount at this point. If you are comfortable with that, I would accept a motion. If not, we'll either pass it or not, or won't get a second.

What is your pleasure?

DR. FRIEDMAN: I think we should pass and accept it as revised.

DR. DETMER: Is that a motion?

DR. FRIEDMAN: Yes, it's a motion.

DR. DETMER: Is there a second?

DR. HARDING: Seconded, if we could see it though before the final vote.

DR. DETMER: Okay, we'll be here tomorrow, so maybe what we could do is table this motion until tomorrow. We'll bring it back in a time specific, and talk about it with the draft prose. So we are talking about tabling to tomorrow's meeting. All in favor say aye. Opposed?

MS. GREENBERG: My only concern is who do we know is going to be here tomorrow?

DR. DEERING: Jeff will not be there tomorrow. An alternative is that I can, depending on what is going on in my office, I find this really moderate. I can run upstairs and redo it, and bring it back this afternoon.

DR. DETMER: Yes, maybe we can do it this afternoon.

MS. GREENBERG: Then we'll put it in the agenda books for the February meeting, so we can give people a chance to see it before the meeting even.

DR. DETMER: All right. So we'll do it at least by the end of tomorrow, if not today.

We next have some presentations that relate to essentially this kind of matrix model we are talking about. I think Jeff is going to give us a presentation on the dimension of computer-based health records from the perspective of patient care. Then Dan will pick up on population records, and Mary Jo will give a demonstration of probably what is I'd say, the most rapidly developing phenomenon I've almost ever seen in my life in the area of health records. It's quite remarkable what is going out there right now. So she'll be talking about that. That will take us to our lunch break.

Jeff, are you prepared to present?

DR. DEERING: I had a question, or actually it was sort of an alert to everyone as that they listened, that they must listen in as active a mode, because my understanding based on our earlier discussions of what we have already said this morning is that the purpose of these presentations is for us to try and discern what are the core elements of each of these three types of health records.

Now clearly, these are records are in some cases being defined by forces beyond our control, so it's not like we are defining what the core elements are. But we are trying to understand at least what the core elements are, because it is on those elements that we are think we may be building the matrix.

So the challenge is to listen to what you hear to be as core elements. One of the reasons we have old fashioned flip charts and grease pencils is that at one point then we will be moving on to review the table that was in the vision paper, and start adding and deleting based on what you hear in these presentations.

DR. DETMER: Yes, that's essentially the work plan for the day. That's really what we are going to be doing pretty much today.

Agenda Item: Presentations on Computer Health Records: Patient Records - Jeff Blair

MR. BLAIR: I think what I'll do, as I present each section, I will pause afterwards, so that I can hear if you have any questions. The other thing is I would say that this happens to be a presentation that was prepared for Europe a month ago. I believe the reason that Mary Jo asked me to review this with you is because it does kind of give a conceptual overview of the basic components of an electronic health record at the vision level, and shows how it might relate to health care information infrastructures.

I am going to try to focus on those points. I'm going to skip through a lot of the presentation, so that we could hopefully get through this in maybe 20 or 25 minutes, and there will be time for questions. Is that okay?


MR. BLAIR: If you look at the agenda, the agenda winds up having about five major sections in it, which is the factors that are leading to the development of a framework for electronic health records. We then wind up hitting what is the vision of electronic health records.

The next piece is what are some of the major initiatives. That's the next section. The next one after that is what are the essential elements of a global health record. That may apply whether it is global, or whether it is part of a health care information infrastructure, because the vision of an electronic health record might evolve over 15-20 years. By the same token, so might a health care information infrastructure.

We will probably be proceeding in phases, so we probably need to identify what the first or second or third phase might be. I think that is going to be consistent with some of the early discussion during this last 30 minutes, as well as the presentation that Erin made on telehealth.

Then the last thing is what do we need to do in order to try to move the ball forward a little bit?

If then try to slide three. Slide three is pretty much what I think you all are familiar with as to the forces that are driving towards the need for health care information infrastructure. Many of the major challenges that we have in health care can be helped if we did have a health care information infrastructure. I'm not going to take you through that slide. I think you have seen all those points before, and understand it pretty well.

I think the next slide, I believe that is slide four, is the divider slide for the next section, which is the vision section.

If you go to slide five, I thought it was appropriate, and I think this is consistent also with the thinking on the committee to begin the beginning of how we look at the vision of electronic health records, with a statement of values. That happens to have been the one from the World Health Organization, Jublajon(?) Charter.

Why don't we go on to the next slide? Should I pause here for a second in case there are any questions?

DR. DETMER: Yes, I don't want you to roar through this. If everybody is comfortable, fine, but make sure that people break in.

MR. BLAIR: I just wanted to make sure that I could hear you, because if I'm talking, I wasn't sure that it would come through on the phone.

Part of the next two slides are to indicate that the vision that has been emerging within the United States, and the vision that is emerging at least in England is very similar in terms of the vision of an electronic health record. The words are a little bit different, because we wind up referring to our institutions a little bit differently, and sometimes caregivers or processes a little differently.

But if you wind up just even looking at the graphic, what you can see is certain basic concepts of that vision. I would argue that both visions are patient- focused. You wind up seeing at the center is the repository of information with respect to a patient, but maybe gathered by episodes or visits or encounters, but both of them wind up having the vision that over time we will wind up having a longitudinal record, which will be multi-episodic, and maybe someday a lifetime health record.

I think we do have to be a little bit careful about using that word "lifetime" health record. That may be a little bit aggressive in the political arena.

DR. DETMER: We're having trouble reading your black boxes. Are you in the slide on the NHS?

MR. BLAIR: Actually, I'm even referring to the NHS and to the amalgam of visions within the U.S.

DR. DETMER: Somehow we have one handout that we can't see the boxes, and another handout where we can. So I don't know -- do we have the visuals that go with this?

DR. DEERING: I don't have anything. He just brought those in.

DR. STEIDEL: Yes, I brought them. They were on the Web. I pulled them off.

DR. DETMER: Do you know the URL, Jeff?

MR. BLAIR: We have it on the Medical Record Institute; nedrecinst.com.

Can I just give you the bottom line on slides seven and eight, which was that the center was the electronic health record. The primary user of that would be the provider, and then the secondary users would be clinical research, public health research, reimbursement, and administrative uses within health care institutions. So that's basically the point of that.

The next slide there is a depiction of a triangle, and I refer to this as aspects of the health record. Are we on the same slide? Okay, the next seven slides in this section are concepts that we derived several years ago when I was on the Patient Record Institute Concept Model Subcommittee. I don't think they have changed too much. This is kind of high level. The way we broke it down was just to try to articulate and get consensus on what some of these thoughts were. So I'm not trying to set forth something that can't be looked at in other ways.

But when we did this, we broke it down in terms of aspects and functions and characteristics. On the aspects, we used that little triangle there. We indicated that there were three different aspects, content, time frame, and information representation.

The next slide, which I believe is slide 10 is the content. As you can see by that slide, it pretty much includes everything that has been in the traditional medical record, but it expands it, or at least hopes to expand it over time to include wellness information and behavioral information, and maybe environmental information as appropriate. Sources would be all of those that you see listed.

Why don't we go on to the next page, which is the time frame. I have kind of alluded to this. It would be multi-episodic, eventually leading from prenatal to post-mortem.

The next one is information representations. There it isn't just that it's data and text and graphics and images and audio. I think the main thing here is that it would have the kinds of information representations that would be clinically specific, so that it really would be able to facilitate information exchange, to be able to measure quality, and improve clinical processes that we have never been able to do up until now. I think that's the major significance of what we are looking for in the vision of the electronic health record.

The next slide is going to be talking about the functionality and characteristics. We happen to have organized these simplistically into information capture, information storage, information processing, information communication, and information presentation. Underneath each of those topic areas of course are many different application functions, and they do become interdependent. This was a way of at least glancing at that.

Then in terms of the characteristics, this is where you again, begin to get somewhat different from the way we have looked at the traditional medical record, because now you have characteristics where you expect that record to be available 24 hours a day, seven days a week, especially in acute care environments.

You expect it to be far more reliable. You expect the data integrity to be much higher, because you are not going to wind up having handwritten notes. You are going to expect it to be more accurate, more secure, and in fact, we would hope more useable, because if it is not more useable, then it's a burden, and clinicians would be deterred from adopting it. So those are the basic systems characteristics.

The legal characteristics that we observed was retention of records, and confidentiality, and authentication.

Are there questions at this point on these basic areas of the aspects, the functions, and the characteristics? Any surprises here? Is this pretty much consistent with what people were envisioning?

MR. RILEY: I have a question under the legal aspect of that. Are you including the concept of evidentiality as a part of the archiving function? How do you address that? Attestability, evidentiality, is that included in some of the words that you have there? I don't have the slides in front of me right now.

MR. BLAIR: We didn't get down to the details of all of the legal issues, but whatever would be legal issues would be included. So I'm waffling on the answer.

Now clearly, while this was very simplistic, and I was able to take you through this, just like a lot of other things, you do the types of things where you can get consensus on, so it's at a high level. But clearly, there are an awful lot of functions and features to assist them to be able to do this.

This is not easy to do, especially in a dynamic environment where the concepts of how to practice medicine are changing, medical terminology is changing, drugs and medicines are changing. You have to be able to design this in a manner where it could be updated continually, and where there can be continuity, inversioning. That's why it has been so difficult for vendors to be able to achieve this vision, even though they have been pretty much working at it now for I would say 7, 8, 9 years.

Should we go on to the next section, which is the major implementations? Okay. You'll notice that the ones that I was able to gather information on for this particular presentation were Australia, Canada, the European Union, the United Kingdom, and the United States. I'm going to slide through these rather quickly.

In terms of Australia, which is the next slide, the major thing there is they have come up with a national health information model. They have based on ISO standards for meta-data. I don't know if Bob Mayes is there, but Bob Mayes has taken a look at what they have done in Australia to see if that might be applicable for data registries here in the United States within the Department of Health and Human Services. I should also indicate that when I was in Europe, a number of the European countries also feel as if this might be a good example to step forward on.

The idea of meta-data is to be able to have data definitions in terms of roles and classes and attributes and many different specifications, so you are not just having a single data element.

Are there questions on Australia, or should I go on to Canada?

DR. DETMER: I don't know how clear the meta standard concept is to folks. It's probably worth talking about that a little more. Bill, you may want to weigh in on this too. Do you have more you would want to add first, Jeff?

MR. BLAIR: Other than the way I've hit it right now, my knowledge in this area is limited. I think Bob Mayes is more knowledgeable on this. If you want to go into greater depth on this, I would kind of call on Bob to take it in more depth.

DR. DETMER: Bob is not here. Bill, do you want to add?

DR. BRAITHWAITE: Well, obviously the concept is fairly simple, but it's the higher level look at the characteristics of the data that you are collecting, but not the data itself. So you are looking at like a data dictionary which describes how a person's name is structured. That's meta data in this sense.

If you agree on exactly on how an address works, then no matter where you use an address in the database, you refer to this meta-data about addresses as a description of what addresses are, and how they are interpreted, and how they are checked, and how they are used, and all that sort of stuff, and that's all called meta-data.

DR. DETMER: Okay, Jeff.

MR. BLAIR: It just occurred to me that I kept mentioning one thing early on in the presentation, and let me just kind of get it in here to get the linkage. The linkage was that when people have looked at the health care information infrastructures, they have felt that their electronic health records are a core component of health care information infrastructures. That seems to be consistent. As we step through a number of the country initiatives, you will begin to see that relationship.

Let's go on to Canada. Richard Alvarez in Canada has set forth a strategy for health care information infrastructure there. He is with CIHI. I think that's the Canadian Institute for Health Information. It calls for pulling together some of the major different projects and initiatives in Canada, and setting them forth in a way where they can begin to utilize standards to be able to create a framework for health care information infrastructure. It also puts some emphasis on trying to create national databases to support protocols and guidelines.

That is obviously at a proposal stage, so it's kind of similar to our own. The encouragement of electronic health records is of course another major element that is there in Canada.

Shall we go on to Europe?

MR. RILEY: Is there a point of contact or Website where we can pull information down about the work that they are doing?

MR. BLAIR: Well, actually I ended up calling Richard Alvarez directly, and he faxed me a copy, because we couldn't get a clean copy. Mary Jo, is she there?

DR. DEERING: I have a copy of the original paper in a group here. It's right on top. But also at the tail end of the original vision paper there is a Website that you can go to. It doesn't have this particular paper in it, because this was a paper that was prepared at a conference, but it has other information about the Canadian initiatives.

DR. FRIEDMAN: There are a couple of different Websites. One is the CIHI, and the other is if you go to Health Canada and you do a search on the Canadian Advisory Council on Health Infostructure -- Statistics Canada -- you can get to their September/October 1998 report.

MR. BLAIR: Going on to the European Union, there are three slides that we have there on the European Union. Of course they approach health care more from a top-down viewpoint than in the United States. You see that reflected by the funding in the projects that are there.

They have -- I'm questioning from the time span here -- I think my inclination is to just let you be aware of a number of these things in order to stay within the time frame here. What you will notice on those slides is they have projects that many of the vendors and consortiums in the United States do in the private sector, but in Europe they wind up having these as funded within the European Union.

There are things like the Good European Health Record Project. There are items like synax and synopsis. Synax are like demonstration projects. The other types of things are to try to set up client server architectures to support electronic health records.

If you go on to the second slide of the three you will see the Gamen(?) Project has characteristics that are similar in some ways to the things we have been doing with SNOMED-RT here in the United States, in terms of trying to develop a convergent medical terminology, one that could be kind of like a reference. They have gone a little bit further in developing some of the tool sets to be able to support these. Of course that has been integrated into a number of their projects. They have also developed tool sets for clinical decision support. That is mentioned there.

If I were to go on to the third of the third of the three European slides there, in the standards efforts in Europe, again they were a little bit more top-down. They have done some really impressive work in trying to develop architectures and frameworks, whereas the standards efforts in the United States have been more pragmatic, where the vendors get together and say how do we make these two systems talk together, or how do we come up with a data set or code set that we could use right now?

So many of these -- it's the slide that talks about the Committee European du Normalization, CEN, with a four word group, that are focused on information models, and the terminology and knowledge bases, the data security, and I guess communications, and technology for interoperability.

For the most part, you find that these are referenced quite a bit in Europe, and they are guidelines and frameworks, as opposed to standards that are actually being implemented by vendors.

Are there questions on Europe?

DR. DETMER: That was the thing. Actually, the European Union invested a lot of money in from the top, and it took really a long time. I think for some time it was not considered that maybe it was going to be that helpful or go very far. But I think what is happening is some of these things are rolling forward, so that architecture framework is actually starting to fit kind of nicely in a complementary way to the way the U.S. has kind of gone at it.

MR. BLAIR: Don, I think that's very true.

Let's go on to the United Kingdom. The United Kingdom I think has been very proactive in setting forth a defined strategy, the national health service information strategy. As you may know, they have had a coding system in the United Kingdom called the Reed(?) System, which has some similarities to SNOMED international. I'm not talking about the reference terminology so much as SNOMED per se, which is clinically specific code sets.

They have been encouraging its usage by the GPs in ambulatory environments. They are moving forward now. I think it's like a 5 billion pound funding for this over a 5-10 year period, to be able to go forward both in an ambulatory environment, as well as acute care. The emphasis is on ambulatory environments, however.

Unless you have questions, based on time, I would move forward. Are there any other questions on these two slides for the National Health Service in England?

DR. DETMER: Back to the U.S.

MR. BLAIR: Now this, I think to put things a little more in perspective for us, because when you do look at the U.S., what you see is this very rich list of standard initiatives in the U.S. that while they still fall short of the ideal of interoperability, and the ideal of clinical specificity, and the ideal of entire data security, nevertheless, these are implementable standards.

Not only have they been implemented in the United States, but if you take a look at HL-7, I think there are 9 or 11 countries around the world that are now implementing HL-7. There are at least 13 different countries that have implemented SNOMED international, which is translated into I believe, 13 different languages.

What you see as you look at this is it covers a good part of the waterfront, at least for what might turn out to be the first phase of an information infrastructure. The reimbursement areas, clearly we have been working with that on the committee. In terms of clinical message format standards, HL-7 takes the leadership in that. In terms of communications between medical devices and bedside systems there is IEEE. You want to see NCPDP for retail pharmacies. ASTM has done a lot of work with standards that have been precursors to HL-7.

There are consortiums that have been moving forward, trying to develop object-oriented standards, and see you CORBA MED, and all the working groups, CCOW, which is now part of HL-7.

The other piece that is of interest I think to us is that as we begin to look more at the Internet, there are tool kits that have been developed for Java and Active-X which will probably lead to some health care information objects-oriented standards during these next several years. We haven't seen those yet, but now that those tool kits are available, I think that may happen.

I think the reason that's important to observe is that the traditional syntax that is used from the rest of those message format standards may be modified. They may be modified to be either more object-oriented, or they may be modified to use XML, which is a derivative of SGML, standard generalized mark-up language, which is a way of tagging different data elements and textural areas within documents, and using that as a syntax. So there are some advantages to doing that.

I don't think I need to spend time today in going into more details on that, but at least you get a feeling here for how the U.S. has something a little bit different to offer in some of the other countries. Each country or region of the world I think is offering slightly different things here.

This comes to a close of this section of the presentations. Are there questions before I go on?

DR. DETMER: Just any one liners on Japan?

MR. BLAIR: I just didn't gather the information on Japan. That is an omission -- an obvious omission. I wish I could have gotten information, but I don't have it.

DR. DETMER: I'm going over there next month, so I'll try to plug you in when I get back.

MR. BLAIR: Any other questions? Let me go on to the next section, which is the essential elements. I don't remember exactly the slide. The point I was trying to make I think on this first slide within this section is that I wasn't going to address in this presentation the underlying information infrastructure, which would involve hardware and networking standards that are not specific to health care.

I wasn't even going to address the more global issues of a health care information infrastructure. I indicated on these first few slides here that I'm going to drill down to just those elements of standards that might facilitate some basic functions for electronic health records.

The two that were in my mind as I was doing this was just the ability to share an electronic health record internationally to support global travel. Just as individuals travel from country to country, if they require emergency care, can we at least get that record communicated between different countries to different health care institutions?

The second basic element of the first phase that I envisioned was telehealth at a very preliminary level, to be able to provide care to underserved populations. So I started, as you begin to look through this section, to say if we were to look at the aspects and the functions and the characteristics that we defined in the vision section, which of those areas would be necessary for a first phase?

I identified the content area. We didn't have basic minimal standards for content, minimal standards for communications, minimal standards for concept representation or medical terminology, whatever you want to call it, minimal standards for security.

Then what I did from there is the next section in the presentation which basically says if those are the four nucleus areas that we would look at first, then what can we pull from the work around the world that would begin to fit to give us answers for that? At this point, unless you tell me differently, my inclination here is I think this begins to go beyond what you want to accomplish today. Is that correct, Don and Mary Jo?

DR. DETMER: Yes, I would think so.

MR. BLAIR: I don't think there is any need for me to then continue on here, but hopefully the portions of the presentations we have reviewed I hope would be useful.

DR. DETMER: I think it's very useful. It's obvious certainly the standards work group gets into a lot of this, so we are not really interested in getting into the nuts and bolts I think, as this work group, in some of those smaller subcategories anyway.

What other questions or comments do people have?

DR. DEERING: I think I'm going to be very honest and say I think what this challenges me immediately with is what do I take away from this that I can then go put up on that sheet there? It's not a function of just one patient's health record. It's going to be the same question that we ask ourselves; are we focusing on core content areas? Are we listing a whole series of issues?

I believe in our original table we had focused on mostly what the content of each of those records might be. But we also had a couple of bullets I think under each of them as to what their functionality or purpose might be. So again, it is thrown back in our laps to decide what do we take away from this that we might want to add now under our list of "core elements" of the electronic patient record?

MR. BLAIR: With respect to that comment, I couldn't tell who was talking there.


MR. BLAIR: One of the other things, and I'm not exactly sure how to deal with this right now, but if we start to go forward and create any scenarios, then although this simplistically states the vision in terms of aspects and characteristics and functions, you then wind up getting specific scenarios where you get into different medical domains, and you also get into different health care settings, whether it is acute care, or ambulatory, or long-term care.

I tried to stay away from getting down to that level, but just should be aware of that. Keep that in the back of your mind as you start to go forward with other discussions on this.

DR. DETMER: I think -- at least, I don't know if I want to get into it now -- but I think I've got some answers to your questions from what I've heard, so I think we can deal with that.

You'll be with us all day, won't you, Jeff?

MR. BLAIR: Yes, I will be listening in all day.

DR. DETMER: I hope you will be weighing in all day as well.

MR. BLAIR: Okay.

DR. DETMER: That's good. I think that's fine. Any other questions at this point? If not, I think we will move on to Dan's speech. Further questions or comments for Jeff?

I think this has been very useful. I see at least maybe a half a dozen of these screens of yours that are going to be very helpful to us. So thanks very much.

MR. BLAIR: Thank you.

DR. DETMER: The whole thing I like, as far as that goes, but I mean relative to our task today.

MR. BLAIR: Now what I'll be doing is I'm going to put my phone on mute, so it may take me just a moment if you ask me a question, for me to respond. This way at least there won't be any noise here that will travel to the phone.

DR. DETMER: Okay, Dan, take it away.

Agenda Item: Presentations on Computer Health Records: Population Records - Dan Friedman

DR. FRIEDMAN: When I got the agenda from Mary Jo, I was somewhere between puzzlement and panic when I saw my name next to this topic. It's not as if at least in the U.S., as far as I know, we've had any rat brain for what a population health record is, and as far as I could come up with particularly good references in other countries either.

So what I tried to for myself --

DR. DETMER: It's fascinating in and of itself really, isn't it?

DR. FRIEDMAN: It is. I may be missing something. I may be missing something in the country, and I certainly may be missing something internationally, but in any case, what I tried to do for myself was really try and clarify what I was thinking of when we were developing the concept paper in terms of what a population health record could mean, and go from there.

The first element that I'm going to start us with was trying to say to myself, hey, what are we talking about when we're talking about a population health record? At the most basic level, obviously we're talking about a record of a population with two key elements in the tautological definition.

The first being, what are we talking about when we're talking about health? This very much reflects what Jeff was saying. When we are talking about a population health record, I think it's really essential that we be very explicit that we're not talking about only an illness record, or only health events records. That we are talking about essentially a much broader definition of health similar to the old, but very useful World Health Organization definition of health as being a state of complete physical, mental, and social well being, and not merely the absence of disease or infirmity.

The second part of that tautological definition is population and population-based data. This is another one of these terms that certainly we in public health, we on the committee bandy about, but in fact when we start pushing it around a little bit, we're often talking about somewhat different things.

One of the classic definitions of population-based data -- this is from John Last(?) -- where he is referring to data pertaining to a general population defined by geo-political boundaries. The population is the denominator and/or the sampling frame. Now this definition, while standard, I think is a little bit too limiting.

When I was thinking of population-based health records and the population part of that, what I'm thinking about is a known denominator -- this is also very tautological -- representing a known population. In other words, not just a collection of folks who happened to walk into the hospital or happened to agree to a clinical trial.

Known population meaning that they have known characteristics, and any data on a sample of that known population is in fact generalizable to a larger population. The intent of this is to say that this can include not only a geo-politically defined population, i.e., a city or town or state, but can in fact also include a demographically defined population. For example, an ethic group, if you happen to have a sampling frame, and you happen to have demographic characteristics for that sampling frame.

It could also include members of a health plan, where they are enrolled members, and you happen to have demographic data on them. So in other words, you have a real denominator, and real information about that denominator.

DR. DETMER: So simply just adding up X number of individuals, and calling them a population without those other kinds of things being identifiable does not meet the definition?

DR. FRIEDMAN: Does not meet the definition in terms of differences between for example population-based data and clinical trials data. What I think of as population-based data encompasses both what we talk about as health statistics, as well as what we talk about as public health surveillance, with health statistics being another one of these terms that has in fact really remained undefined.

What I think of when I'm thinking of it is health data generalizable to a known population used for designing, monitoring, or implementing or evaluating health policy or health programs. So it's really policy and program directive data, in contrast to public health surveillance, where the emphasis is -- and this is a classic CDC definition that I'm sure at least a couple of you have committed to memory many years ago -- of ongoing systematic collection, analysis, and interpretation of health data essential to the planning, implementation, and evaluation of public health practice, with the emphasis on public health practice, closely integrated with timely dissemination of these data to those who need to know. The final link of the surveillance chain is the application of these data to prevention and control.

So when I'm thinking about population health records, I'm thinking about health statistics, i.e., population-based data useful and intended for policy and program development and evaluation, as well as surveillance data much more immediately oriented towards public health interventions.

Just to ground it for a minute, examples at the most obvious levels being health interview survey, for example, BRFSS, vital communicable disease surveillance data, classical discharge data sets, where we have them for a defined area, Medicare records, where in fact we do have a defined population by characteristics.

In thinking about what population health records could consist of, if we can think both in terms of individual records, which would be my preference, and which I'm much more comfortable with, or in terms of population area records, or some combination of those.

The Manitoba has an information system that in fact seems to be a combination, where they have got, from what I can tell "administrative" data that cuts across different types of health care interventions. I think it also includes pharmacy. I think it includes hospitalization. I think it includes ambulatory.

But from what I can tell at the individual level, it does not include at the individual level what we think of as here as public health surveillance or health status information, or census data. Those data there seem to be found in a small areal level.

But what I have been thinking of is population health records in terms of individual records, which as Jeff has said about the electronic health records, would need to be -- ideally would be longitudinal over time. This is something that the Canadian Advisory Council on the Health Infrastructure has really emphasized in their report, which is a terrific document.

What they have talked about that the importance of using longitudinal data as the only means in fact for really evaluating what is the long-term impact of health in social programs. At the most extreme, as Jeff said, it would be life long data, covering all aspects of health, all aspects of health care, and all transactions between the individual and the health care system.

One of the things that really tends to happen I think both with our administrative data and with our surveillance data is that in fact on some types of things we really lag behind what is going on out there. We lag behind not so much in terms of how long it takes to have the data reported to us, analyzed, et cetera, et cetera, but lag behind in terms of our systems are so past focused that there are major aspects of health care that we just don't necessarily pick up until we're sort of past the curve.

One of the things that struck me most recently was the JAMA articles about "alternative" medicines, where there is apparently a huge individual investment being made in people going to "alternative" practitioners, alternative medicines, et cetera, et cetera, and at the same time none of our regular public health surveillance health statistics surveys picked up any of this, as far as I can tell, because we hadn't thought to ask the question. So major things fly in the radar.

Population health records I would think need to be person-based, obviously population-based, scalable. In fact, when we are talking about population health records, we are talking about locally collected records that need to start at the smallest possible unit, and scale up.

And also obviously minimizing a data provider burden, particularly in the sense that right now we have surveillance data. We have administrative data. We have got health statistics data, all of which are reported obviously on the same people, on the same subjects, from the same providers, with just sort of a rat's nest of repetition.

MR. BLAIR: Can I ask a question?

DR. FRIEDMAN: Sure, Jeff.

MR. BLAIR: The phrase "person-based" and maybe you've got it on the graphic there, and I just am not able to see it, I'm trying to understand that a little bit better, because I have tended to think of the population health record as being derived from a large number of individual health records.

When you say something is person-based, do you mean derived from individual health records, or do you mean that they would be a population-based record at an individual level? If it's the latter, then I don't quite understand that.

DR. FRIEDMAN: To the extent that we want to gather longitudinal information that integrates information on health status, health behaviors, health events, and health care, I think that sooner or later we are going to have to grapple with trying to get all of that integrated at an individual person level, rather than less integrated at an ecological level.

Is that clearer, Jeff?

MR. BLAIR: Some of your words got cut out. I'm sorry, so I missed some of them as you were saying that. Could you get a little closer to the phone? Sorry.

DR. FRIEDMAN: No, that's okay. To the extent that we are talking about -- if we are talking about trying to obtain integrated information over time that would include health status, health events, health care, and essentially transactions between the individual and the health care system, it will be much more powerful and much more useful if in fact it is integrated at an individual person level, rather than being essentially a collection of different data sets that describe a census block group, or residents of a particular city or town.

DR. DETMER: So you are saying allow a unique identification, not necessarily being interested in Dan, but knowing that you are dealing with this person that's got a name?

DR. FRIEDMAN: Yes. Sooner or later if we really want to talk about population health record, I think we're going to need to squarely face the issue of what are the pluses and the minuses, and from my point of view, acknowledge that individual-based population health records may in fact provide much more information. We may not be able to or be willing or want to deal with that. Or there may also be alternatives.

DR. DETMER: Right. I just wanted to make sure I understood you, and I think I did.

MR. BLAIR: Thank you.

DR. FRIEDMAN: Obviously, the most basic prerequisite is a legal and regulatory framework that would guarantee confidentiality, guarantee security. Another basic prerequisite is obviously linkage enablers, and obviously standards. There's no sense to go into that.

When I think about population health records, what strikes me as a use is essentially much more integrated intervention, where sooner or later we are not differentiating health surveillance data from health statistics data, or necessarily not even differentiating from clinical care data or administrative data, but we're talking about essentially a unitary data system that can be used for a variety of purposes that will obviously have health policy uses.

One possible avenue that I think is worth thinking about that the Canadians seem to be thinking about is the development of sample-based pilot population health record systems. This could either be plan-based. It could be geographically-based. They could be based upon -- certainly include an informed consent element, but I think that if we are going to move forward on this at all, the only way in which we would be able to do it is in some of small pilot-based effort, if it is going to be individual person-based. I think if it is going to be areal-based, then there is a whole other realm of possibilities.

DR. DETMER: Questions and comments for Dan? I assume in the last one you are talking about sort of Framingham thing?


MS. GREENBERG: Can we have copies of the slides?

DR. DETMER: Yes, I think we will be wanting those this afternoon, without a question.

Other comments?

MR. BLAIR: May I ask if I could get a set of the presentations either mailed to me, or maybe if I could get them at the February meeting?

DR. DETMER: Sure. We can mail them to you.

MS. GREENBERG: We could even fax these to you, if that were feasible, if they are going to be used this afternoon.

MR. BLAIR: Oh, that would be great. Should I give you my fax number, or you may already have it.

DR. DETMER: We have it.

MR. RILEY: There may be a group of ongoing studies that already do what you are talking about over in DOD. After they got beat up on the Persian Gulf illness fiasco, one of the things that we did was institute a massive collection of data at the individual level for those in Bosnia, and for those who are now deployed to Southwest Asia.

So the data sets may already be collected at the level of granularity that you are talking about. It will just be interesting to follow longitudinally what the outcomes are of that. But there are some large populations that are apparently being followed like what you are talking about here.

DR. FRIEDMAN: I did not think of that. I think you're right, that's probably the single U.S. model that we would have.

MR. RILEY: It includes things like health risk appraisal evaluations. There is surveillance data. There is environmental data. We probably know more about the environment there in Bosnia than they ever wished to know in terms of what has been collected sample-wise.

DR. DETMER: Excuse me. You came in after we initially introduced ourselves. What is your name?

MR. RILEY: I'm Dave Riley. Col. Lynn Ray(?), over at DOD Health Affairs asked me to sit in as his representative.

DR. DETMER: I'm sorry I didn't ask you to introduce yourself earlier, but we need to do. That's Dave Riley, Jeff -- I don't know if you could hear that -- asking that question.

DR. BRAITHWAITE: I think it's interesting from Dan's presentation, it points out the thing where we sort of started. The emphasis that the population record -- that people who are interested in population records are simply looking at the same data as the people interested in clinical records for individuals. They just look at it in a different way, for different purposes.

We have to be aware of the fact that the population-based health records do depend on -- totally depend on the availability of the individual records for aggregating and linking together about other information in order to improve the health of the population. We can't think of the population health record as something independent from or separate from the individuals.

DR. DETMER: Well, partially. Things like HANES and so fort is the population in a record in a sense that is different. It is coming typically out of clinical --

DR. FRIEDMAN: It certainly is, but it's a snapshot.

DR. DETMER: I understand. I'm not arguing with you. I'm just saying that in a Venn diagram sense, those things don't totally overlap, but there is a lot more overlap than I think is generally appreciated.

DR. STEIDEL: But if you had a record like was described, you won't need HANES.

DR. DETMER: I agree. On the other hand, you also have to have a system, however, that is attracting alternative medications and such, which a clinical setting doesn't necessary track either. You think it would.

MS. GREENBERG: Actually, Steve, I think you still would need something HANES, because you are unlikely in health care, even a person's lifetime health care doesn't necessarily collect some of the information that you collect in a study like HANES on a sample of people.

DR. DETMER: Well, whether it's HANES or something else, I guess what I'm saying is you may still want to do discrete studies focused on particular things that isn't typically in there, but could have some relationship to health, or you are wanting to know if it relates to health.

DR. FRIEDMAN: But it would hopefully be integrated.

MS. GREENBERG: Linked, right.

[Technical problems with the telephone connection are being addressed.]

Agenda Item: Presentations on Computer Health Records: Personal Records, Demonstration of Online Initiatives - Mary Jo Deering

DR. DEERING: Well, let me just start talking while we are waiting for this to come up. You had you package, Homes pages from two activities talking about consumer health records. The third was a piece on Dr. Koop's dot com, and a marketing study where I have indicated in the margin the fact that they too are intending to create a personal medical record system.

By way of background, as I begin to pull some of this up, let me just say that as Don has said, this does seem to be a very rapidly growing area of interesting. Namely, to give patients their own medical record or health record, or health information collection. There are different formats for it. It is defined differently by who is offering it.

What we are going to take a look at today are a few of those models that actually do attempt or purportedly attempt to be helping you track some of the core elements of what you would give in a patient history in a record of care.

The purpose behind most of these of course has been to enable consumers to keep a comprehensive record of their individual and/or family health. The secondary purpose was to enable them to share their history in a consistent fashion with a variety of providers, and to enable them to make better decisions about personal health or medical care.

Some of the factors behind this, on the consumer side it's of course just a reflection of the broader movement toward consumer health activism and consumer health empowerment and self-care, frustrations about losing records. I don't know how many of you are parents, and even if you are M.D.s, have you really kept your kids' immunization record handy. And to go into a new provider's office and to have to fill in the form each time, and not know whether you have remembered it correctly from one day to the next as to what your grandmother died of.

On the market side, meaning the people who are putting these up, to date there hasn't been a commercial market interest behind it in that no one yet is charging, at least those that are up and running on the Web by definition are usually free, but they come with some sort of registry behind them in many cases.

There is a sense that there is this interest, so it is a service. It is a way of attracting people to what are otherwise gateway health sites, or sites that are trying to present health information, most of which are advertiser-driven by the way.

On the policy side, there is an interest in terms of portability and consistency. I guess I would point out here the comment that was raised, and I think Dan you raised it. If we are aware that one-third of the people report going to an alternative medicine provider, probably totally out of plan, paying for it out-of-pocket, there is never going to be any record of that care.

If it does represent a significant quantitatively, if not qualitatively, a proportion of health services that are being delivered and a reflection of health status, then again, no plan driven record or top-down record might necessarily be able to capture that kind of information on a consistent basis.

I originally said there were two models, and I will tell you about a third in a moment. I used to say that there is the Quicken model, which is the individual enters the information according to a standard template, and keeps it at home in one form or another. There is the Turbo Tax model, where you plug that information into something on the other side, and we're going to look at one model of that.

But more recently there has interest in a model which other people call the Visa model. I'm sorry that Claudener(?) isn't here. I think you are familiar. I think the government people are looking at that. But not only are the government patient record people looking at that, but there is a group out in the private sector that is trying to look at what they call the Visa model, which is a non-medical entity that acquires credibility, and that maintains the warehousing and clearinghouse efforts on behalf of individuals.

So they are exploring that model. I'm hoping that perhaps if you are interested, we can get greater information on that on February 2.

So some of the models that I wanted to show you to start with are -- this comes from the AMA. It is probably one of the most comprehensive. Jeff, I can certainly send this to you. It's on the AMA site, and one way or another we'll get you the URL. The important point here is that it follows not surprisingly that the classic patient entered health history format.

It is entirely generated online, everything from: name; place of birth; medical history; conditions; previous operations; previous injuries; and medical conditions; including medical illness; prescriptions; non-prescription medications; social history; exercise; risk factor elements; vaccinations; family history; lists of doctors; and health insurance.

Now while this one in some ways is one of the more comprehensive of the forms that is online, interestingly enough the AMA doesn't try to keep this for you. What you do is you enter it, and then you print it out. You keep it online and you can make corrections that way, but to the best of my knowledge, it is not actually maintained in a repository at the AMA.

PARTICIPANT: What do you mean when you say "keep it online?"

DR. DEERING: In this case you don't keep it online. You enter it online, and they do the processing for you. Then when you are done it says push print. Then it prints out what you have.

DR. DETMER: That's the only option?


DR. DETMER: You can't download and file to your own?

DR. DEERING: Right. I checked it out. I wasn't able to. I don't know if it will let us do this without it filled in. It says, "Do not print. When done press send to summary button to verify data."

Now undoubtedly since this is an online form, of course you can save to file, and then you would save it on your own hard drive or disk. Now you can see everything is blank here, but this is what the summary ends up looking like. Again, since I filled out nothing, it's whatever was in there.

This is what would print out, or this is what you could keep. Then you would edit it offline at home, to keep it updated. So this is the model. Are there any questions about this approach?

I'm going to take you to Vital Works, which is one that was given in your packet, which without registering -- I didn't actually go into this one, but I wanted to indicate the services that it is providing, which is the health tracker server, the health tracker patient card, and advanced directives.

What this one will have in common with what I believe will eventually come out of Dr. Koop's site, and which is common to another site that I'm going to show you in a minute is that in many instances that they are trying to do is just get you registered online. That's why they can justify doing some of these services, is that they want to then be able to market directly to you online. I don't have specific background about the Health Tracker Corporation though.

I also have a copy of the Dr. Koop page. Dr. Koop, as you saw in the handout, the intent at this point is to create a secure online medical record. It is not currently offered. But I wanted to just call up the Dr. Koop home page so that you see what the front pages of many of these services do look like. It's a multi-purpose health information, pharmacy registry. You can renew pharmacy prescriptions online. You can participate in chat groups. You can e-mail questions to a doctor online. So this will be your all purpose personal health site.

There is something else called health central.com that I thought she had put in favorites, but it must not have gotten there, so let me see if I can just go directly into it. I can tell you what Health Central -- its home page is very similar to Dr. Koop. Again, it's trying to be a multi-purpose source of information and transactions. These sites are increasingly noticing that people value transactions like pharmacy refills.

For that matter, the Kaiser Permanente group has begun a fairly multi-functional Web service for its beneficiaries, which to date does not offer the health record, but does offer a variety of transactions. Again, the business case behind these models is one of customer service and marketing, as opposed to any policy or clinical purposes at this point.

What the Health Central site includes is the same information as the AMA, but it is divided into groups of what your medical history is, of what your doctors are, of what your social history is, and various risk factors. Most of them do include the risk factor profile as well, and will give you a very simple health risk analysis online that says you are at high risk for.

The benefit of this is that many people of course can't probably fill in their cholesterol number. So at least it is a teaching tool, as well as an information repository, and reminding people of the sorts of facts that they have to have available.

DR. DETMER: I might add, I think the first generation of most of these Web sites were essentially trying to get people just specific medical information, almost taking medical textbooks and trying to transpose it into a way that people can just get medical facts.

Now I think you are seeing much more of a second generation or third generation that is starting to try to get far more interactive, and starting to look more at health management and also illness management increasingly.

MR. GELLMAN: Are these sites advertiser supported?

DR. DEERING: Yes, most of them are advertiser supported.

Don, you might like to help us work through this one.

DR. DETMER: I have demonstrated this. This is a company that I had started, that right now is out of financing, so there is no conflict here. I also told her that I didn't suggest it. If she brought it up on the screen, she brought it up.

DR. DEERING: Could you be talked into demonstrating?


DR. DEERING: I'm handing out, Jeff, and I will give to you just a printout that lists the product. Maybe I'll have Don talk us through the product.

DR. DETMER: The concept of this was to actually have a cluster of records that interfaced with one another. So that in fact if you go to the products, the first product is the personal medical record that the patient stores, and they control how securely they want to keep their own data. In other words, they set the standard on what goes out of their record, what the keep at home. They control up front, what they put in it, and what then also goes out.

But it also gives them a fair amount of information on just a general health risk appraisal. Then where they do know specific health problems, being able to get hooked to Websites that relate to their specific illness. In the future of course, you could get into even medical management interface kinds of things with that.

The second product takes then the patient's guidelines on what they are willing to see go, and allows it to go to their physician of use in two formats, one, the patient data going to the physician as the patient has put it in, but in a summarized form. So the report comes out as saying this person hasn't had a pap smear in three years or used to be hypertensive, and now isn't on medication.

So it kind of gives an HMO or a physician getting the patient for the first time, a sense of what are high risks early, or even before they have seen them in an appointment, so that you make sure that at those things are kind of red flagged even before they come in.

The other thing, you allow from the clinician's side, the doctor would say to the patient, look, I need to see you in three weeks or a month. You go to my Website and schedule when is the best time for you that I have open slots. So that it is a matter that you have this interface through the Web that allows the patient to interface not only on data, but essentially on process of care as well.

Then the idea is that you also then can have follow-up outcome data at home. For example, if the patient is on certain medication and you want to track certain things, they can track those things and enter that sort of thing at home. That's essentially is a handshake between the doctor and the patient.

Then if you go to ProFiles, which is the third set of products, it essentially strips off personal identifiers and actually allows the clinician or with personal identifiers that doctor can look at his patient population with identifiers. But if it is going to go beyond that, then a unique identifier for the patient is stripped off, but you could then look at a system level of aggregate covered lives in terms of we have in this population of 5,000 people, 60 of them that are smokers that want to quit, and so many of these, or so and so.

So it's a way of being able to get essentially some denominator information for not only from the perspective of either a buyer or a seller, a physician group for example, trying to decide whether they would want to take a group of patients. What is the risk, exposure if you will, of that? That would also obviously have some financial implications of how healthy, and if you will, how much risk exposure you are getting if you take these on as a company.

So essentially what you are dealing with is a way to try to have in a way, an interface function really for meeting a variety of functionalities. For example, the patient also would have in that a button that would help some of these functions that you were talking about at Kaiser -- how to use it. What is my health plan? How much of my deductible have I already used up? A lot of stuff that frankly, doctors and nurses could care less about, but it's quite important if you are the poor fellow or gal on the other side trying to work their way through this stuff.

So it's a matter of trying to meet the various users, getting the things that they would really like as functionalities, but not knowing anything more than they want to, but essentially giving the patient essentially most of the front end control on who gets to see anything. They can set standards on what even goes.

MS. WILLIAMS: How do you authenticate the patient's release?

DR. DETMER: I'm just saying this is a model I'm talking about. This is not a product out there on the market right now. The idea would be you would use a password or some other kind of encryption kind of thing. But as I'm saying, right now we didn't get the next round of capital financing to get beyond the model we are talking about, but that's the concept.

DR. FRIEDMAN: Are either of you aware of any HMOs for example that make this available based upon medical records rather enrollee input? I get a printout every year when I remember to call my HMO for an -- for example, you go in for a physical once a year, once every two or three years. You get a printout from your doctor that is more or less helpful, saying here is your cholesterol, here is this, here are your risk factors, here is that.

There is also no reason why not only to have personal health records that not only are based upon an individual entering his or her information more or less accurately, but also could be based upon the health plan, the HMO.

DR. DETMER: You mean on the data that is in the plan's health record?

DR. FRIEDMAN: Exactly.

DR. DETMER: Go back bectis.com. One of the biggest challenges of getting to the use of computer-based health records is actually getting physicians to enter the information.

MR. BLAIR: Is this a Website that describes what you are talking about?

DR. DETMER: What it is, is a hand held physician entry, computer-based record on a structured format, that then does in fact create about 80 percent of what a general internist would have in their data record. The problem is that that's been just a nightmare generally in health care to get clinicians to either use a structured entry format, or take the time to do it.

So the limitation of having the patient's data put in by the patient is, of what validity is it? On the other hand, the asset is at least you know what the patient thinks the validity is. That's what they think is their story. That's not inconsequential in its own right.

But it also is true that if you take it from the side you are saying, one of the things that this has, and I noticed that the AMA doesn't talk about, are you on any alternative medicines, or are you taking other kinds of treatments and so forth. You do have the capacity in Vhita model to have a fairly robust kind of -- you can construct this also to put in things you want.

As I say, you control what goes out, but it is the other side. But the point is if patients don't mind, and a lot of patients don't mind sending that kind of information to the physician, you have the capacity to see that a practice plan's clinical record for starters, which actually would be a tremendous asset to try and get computer-based health records into use.

HMOs and ambulatory care, frankly, the Dutch and the Brits are way ahead of the U.S. Most of our computer-based health records have been far more in the institutional site. They are typically those nations are behind on the institutional side, but they are way out on the primary care side.

So I think we're seeing this kind of move along, but there aren't many HMOs right now that have this. But there is a lot of activity, without question. Kaiser has been a huge commitment and investment in that direction. Obviously, that's that what the DOD and VA are talking about too.

MR. GELLMAN: Can I make a couple of comments about these things? I mean my reaction to them is that they are services for stupid people, because anyone who would put any information up on a Website -- I didn't see any privacy notices -- but even if they are there, whatever minimal legal protections you have today with records at a high --

DR. DETMER: That's why we need legislation.

MR. GELLMAN: Legislation won't touch any of these things, because it is voluntarily disclosed information to anybody. You can print your medical record in The New York Times if you want to pay for it. You can put your medical record on a live Website and let everyone look at it if you want. If you want to give it to a third party --

DR. DETMER: Why if there's a law against it, and you have said that you don't want these data going out when you have protections?

MR. GELLMAN: It depends on how you write your law.

DR. DETMER: That's why I'm saying if you had legislation.

MR. GELLMAN: Most of the laws that have been proposed would not touch these sites. You can do anything you want with your own information.

DR. DETMER: That may be, but I think you can write legislation that would cover that.

MR. GELLMAN: I don't, but in any event, what this is suggesting, just to work the other side of this very briefly, is if you pursue these kinds of things, what this says is that we can have all kinds of computerized patient records, and the patients are in total control. You are now setting up a system under which a research wants my record? You can come and ask me whether you can have my record.

DR. DETMER: Well, that doesn't say the doctor doesn't have the record.

MR. GELLMAN: Oh, yes, because now we are talking about giving the patients the mechanism of control over records.

DR. DETMER: No, you're saying the patient also has a record.

MR. GELLMAN: Not necessarily. Once you set up a structure where you are giving patients control, I may be able to say I can control what my doctor can look at my record too.

DR. DETMER: Certainly there are people that want that anyway.

MR. GELLMAN: This is creating a mechanism that will make that practical.

DR. DETMER: Maybe, I don't know. I doubt it. I think you are going to find that if you are legally responsible for the patient's care, you will keep a record.

MR. GELLMAN: That may be the case, but the patient may have a mechanism of control or a practical notice for who is looking at the record, and when it's being looked at.

DR. DETMER: There are a lot of bills that want that right now.

MR. GELLMAN: Well, there aren't any. No one has gone that far, but I'm saying this is creating the mechanism that will allow all of that. And it will undermine and interfere with not only physician use of records, but it will really undermine the ability of researchers and overseers to use records, because you may have a degree of patient control that has been impractical up until now. That may be a good thing or a bad thing, depending on what you happen to see.

DR. DETMER: I think right now it's obviously too early to see.

MR. GELLMAN: Yes, I agree with that.

DR. DETMER: It does empower patients. There is no question, and that is of course what is behind a lot of the stuff, as far as that goes.

DR. DEERING: I think it's a question of whether or not anyone can stop it. Clearly, what this committee does and what other bodies have an influence on it, but I think we put this up to show this as being a minority.

I think those using it would probably not consider themselves stupid necessarily, and would see it as a very useful tool.

DR. DETMER: This is the edge of the edge right now. We're still ahead of really where use is, and utility. I think you are right in terms of predicting what will ultimately shake out of all this.

DR. STEIDEL: There is a tremendous amount of patients controlling their own records going on in the pathology community. You have people going into these storefront labs, asking for a test to be run, walking out with the test themselves, never entering the medical care community, or home purchase testing. So there is a lot of that type of stuff going on.

MR. RILEY: They sit in the control seat anyway. They have the option to tell you or not tell you when you are sitting there with that patient. So to say that the physician has control is just absurd. If I decide to tell you something and I do, or if I don't you have to make your best judgment based on the information that I feed you. But clearly the patient is in control. I don't see this as some new trend. They have always been in control in one form or another. It just validates the fact that this is their intellectual capital, and they need to have due consideration for that.

DR. HARDING: You can follow the money. The money from these sites is going where?

DR. DETMER: Unfortunately, this site doesn't have any money. I think to get to your question in a real term, clearly pharmaceutical companies in the last three to five years are trying to move very much from marketing and interacting with health professionals, to interacting with consumers. So this is a way to interface with those people directly, to cut out any middle person whatsoever.

DR. HARDING: So where in that is a statement that say, okay, you bring this stuff here, and a pharmaceutical company is going to use it for marketing?

DR. DETMER: There are I would say, a set of standards emerging, some which in some instances they put up on the front here are our security policies. Here is what you are getting into if you do this. I would say that's a minority of sites right now, but some do do that. Others also say this meets the standard of X certifying organization, but as Bob would say, we don't have any federal legislation frankly, or state legislation in most instances that really even speaks to this stuff. So it's generally self-regulation, but even then I would say we're on the small percentage.

Maybe you also saw recently that 100,000 or 200,000 women actually sent to a group of fashion magazines, very specific personal data. They just gave it to them, when they didn't even say anything about whether it will be used for commercial purposes. It will be used for marketing and a whole bunch of stuff.

So it's very interesting. The fact of the matter is it's a big country, and some really worry terribly about this; other people don't worry at all about it.

MR. GELLMAN: I think that's true. I think that example was Conde Nast, which put questionnaires in their magazines and they got a couple hundred thousand people to respond. I don't know how many surveys they sent out, because they wouldn't reveal that, but it turns out that a guess is that they probably got a modest response rate.

If you look at the polls, you find that 20 percent of the people have no interest at all in privacy, and they will fill out anything if they think there is something in it for them. That's basically seems to be there.

Of course the other issue is always in all of these things does anyone ever tell the truth? The polls show that people are lying more and more online and in other circumstances.

DR. DETMER: Which gets to your point. At any rate, I say it varies, Richard, from what you would consider at least responsible, to self-regulation, to nothing whatsoever.

DR. DEERING: I think that is where the people who are just beginning to explore the Visa concept are also asking themselves what processes -- what political processes, what educational processes, what marketing processes, what transactional processes -- need to occur for us all to accept putting that information online.

I wouldn't be the one to equate that with medical information, but they are still saying that clearly when that started out, it wasn't given a future at, because people raised what they thought were considerable barriers in terms of its acceptability to the various stakeholders. So they are trying to now tease out where there are similarities or not that might be useful in looking at that kind of an approach.

I think what drives that is it's not so much an ideology of consumer empowerment, although that too, as the sheer issue of portability, consistency, as much as the others.

DR. HARDING: I think to have this brought in at the first point and so forth, would be very helpful. It's just that is there a secondary purpose? Are patients willing or smart enough or just being a fool to allow it.

MR. RILEY: That is probably where legislation is most useful. If you put these sites up like this to gather this information, that you have a full and open disclosure of what the intent is with the information. If you look at consumer behavior with credit agencies for example, there are two things that really tick people off. One is that they have incorrect information in there. The second thing is that these companies are making money off of their information.

Again, it comes back to the issue of intellectual capital. If you have correct information there, and you are able to compensate the patient for the use of their intellectual capital, that is one potential economic model that may drive this forward, where you will be able to get consent from the individuals to use that information.

Compensation doesn't necessarily have to mean money. It may be reduction in your HMO premium, or some other form of compensation. But there is the opportunity there to work from an economic perspective to get the consent that you are looking for, and to get that honest information. But it requires up front full and open disclosure on what your intent is, and to allow the patient to have the option to say no.

DR. DETMER: What I thought we wanted to do -- and we're about at lunch time right now -- was to also have a little bit of a piece about this, because this is sort of almost growing like crazy. Much of legitimate establishment care hardly in a large part even has a touch of what's going on. It's really amazing how it has taken off.

Okay, why don't we break for lunch. We've got ourselves resuming at 1:00 p.m. Why don't we say 1:15 p.m.? How's that?

[Whereupon, the meeting was recessed for lunch at 12:30 p.m., to reconvene at 1:15 p.m.]


Agenda Item: Discussion on Core Elements for Patient, Population, and Personal Records

DR. DETMER: We want to start off first with talking about some of these core elements for the records. I think probably what would be good, we could start maybe first not identifying which set they go in, but just the kinds of considerations we would want to have in there period. Then we can end up putting initials behind them later, if that's the case.

So, Jeff, I'm going to start with the four that you mentioned. The issue of content, communications, representation, and data security. That's at a fairly abstract level, but that's certainly there. There is a lot inside of each of those too. Maybe for a reference at least, one piece of this, you might want to get our paper out.

Dan, there were a number of things in yours that should be kind of a mixed bag at this point.

MS. HAYDOCK: This is going to get to a lower level of detail, but I do have a very simplified version of the Australian model.

DR. DETMER: That would be great. Let's put those up there.

MS. HAYDOCK: It's a different way of looking at information.

DR. DETMER: They have business factors.

MS. HAYDOCK: Which is another look at content, isn't it?

DR. DETMER: Or it's just a way of looking at the players more than anything else. They call them parties instead of -- parties are persons or organizations or whatever. Events, enabling factors, resources and others, environmental factors. It does call them social. Then a set of things they call party characteristics, personal identified, the person's view, state of health and well being.

Maybe what would be good at our February meeting, or one of them would be to have somebody kind of present the Australian model.

DR. FRIEDMAN: Bob Mayes mentioned to me that there was somebody from the -- I forget get the acronyms -- the Australian Institute for Health and blah, blah, blah.

MS. HAYDOCK: She is the lady I mentioned to you guys, Mary Jo.

DR. DEERING: That might be coming to visit?

DR. FRIEDMAN: If you appeal very nicely to them, they will send copies of a nice, big, colored flow chart of the model.

DR. DETMER: What happens if you don't do it nicely?

DR. FRIEDMAN: Then they will charge you $20. The state of Massachusetts will not cut a check to Australia.

MS. HAYDOCK: Would you like me to do that?

DR. DETMER: Yes, I think so. I guess we'll need to look at our agenda for the next meeting, but I think that would be worthwhile. I know Bob Mayes can hardly sleep, he gets so excited about it.

MS. GREENBERG: Is this person in the Washington area?

MS. HAYDOCK: She is stationed right now in the San Francisco area or somewhere in California. We're actually hoping to get her out east, to work with her as well. We had thought perhaps of getting her here in February.

DR. DEERING: I think that what I lacked was the final content. So you have that.

MS. GREENBERG: I think if she would be here in February, it would be nice to have her for the full committee, I think.

DR. DETMER: It probably would. They would probably like to hear her. We'll have to see. We can talk about that tomorrow.

MS. GREENBERG: True, but it relates to the 21st century activity also.

DR. DETMER: What is your pleasure?

DR. DEERING: Would people like to take a look at this?

DR. DETMER: Why don't we let you sit down a moment and read this, and then we'll pick this back up and see if we can deal with it.

DR. DEERING: I can call your attention to what I have heard as the major changes. Moving the first sentence down from the original paragraph, to become the first sentence of paragraph two. The last sentence of paragraph one, last line includes the words, "should seek to."

Under the second paragraph, the second sentence, the work group will monitor relevant activities. Then at the end we added global activities under the pertinent activities.

Item one, we inserted, "at least annually." Item two we modified to reflect the three. That should one item. We inserted a new item five, and inserted the word "appropriate" on item six.

DR. DETMER: Well, I think that certainly is what we had talked about. Do we like it? Okay, Dan you have a motion. Do you want to pick it back up off the table?


DR. DETMER: And a second.

DR. HARDING: Second.

DR. DETMER: Any discussion? All in favor say aye. Opposed? Abstentions?

[Whereupon, the charge statement was unanimously approved as drafted.]

We'll take this into the full committee.

We started putting some things up on the board there: content, communication, security, representation. Some of the things that Jeff mentioned this morning. We added the kinds of language that the Australian model -- Sandy put up. I can tell you, I can read that much better than had I done it.

I just was asking Dan if he wanted to -- what's the best way for us to do this? How are you thinking about this part of our project?

DR. DEERING: I'm trying to think of where we wanted to get. Maybe we should just spend a moment thinking through where we said we wanted to get. We said we wanted to be able to refine and agree on what we saw as the core elements under each of the three separate records. Having done that, we then would be asking ourselves how that related to our matrix. We just added to our charge the fact that we will be looking at other components of the concept paper as well.

So clearly, it would suggest that we were going to take each record one at a time, and try and get our list.

DR. DETMER: These are okay anyway. So which do you want to start with, patient, population, or person? Why don't we do it the way we did it this morning? We'll start with patient. A source that I don't have in front of me today, but would be worth referring to is the IOM 1991 report, because it has a listing of something like 150-some uses.

Those cluster back down into something like 15 or something like that. So there is a tremendous number of uses of these records. Now some of those would probably be double entry at this point, but at any rate, just looking at this list brought that back to mind. So what we'll do is look down through that between now and that meeting, and see if there are things that ought to be on that, that don't get on to our list today.

MR. BLAIR: Don? Terminology. I think in the report right now it refers to it as the individual health record. Is that correct?

DR. DEERING: The table calls it a patient health record.

MR. BLAIR: The only thought that I had was that if our concept includes wellness and maybe eventually environmental information for the individual, that patient is maybe not appropriate, because it's more than a patient. It's whether or not you're a patient. Wellness deals with when you are well, not a patient.

DR. DETMER: So personal health record? Well, personal health record was in this last set we talked about.

MR. BLAIR: I missed that. I'm sorry?

DR. DETMER: I think this is a good discussion. The personal record was really we were talking about this kind of new thing that is sort of emerging out of the Internet.

MR. BLAIR: The consumer health record?

DR. DETMER: Yes, right.

MR. BLAIR: Okay.

DR. DETMER: So I guess we need to talk about it. I think the question is whether the patient record is literally simply a disease record. When you're saying environment, what are you talking about?

MR. BLAIR: Well, maybe my idea was not correct. I thought that the three divisions would be an individual health record as the first one. The second one is a population health record. The third one was a consumer health record. Now maybe that's not right.

DR. DETMER: Then you have the confusion between individual and consumer, it would strike me. One of the reasons -- to go back, why we used, as you will recall, Jeff, in the 1991 study, why we end up going with patient record rather than medical record was just the point you made. We were not interested in just medical in the narrower sense of that word. But I hear you too. We had some that discussion. Octal(?) Barnett as I recall, wasn't too keen on patient.

MR. BLAIR: What you're saying could work. I mean if the first one is the patient health record, and the second one is the population, and the third one is kind of a wellness -- well, I don't know.

DR. DETMER: It's sort of the consumer personal -- it's the one that will ultimately emerge. Historically, people know more about their car when they turn the key in the morning than they know about themselves over five years.

MR. BLAIR: This may have just reflected the fact that I wasn't clear on it. It sounds like your definitions are okay.

DR. DETMER: Mostly probably no single term will capture it, but we do need to make sure when we define it though that we then say this is what we are talking about.

DR. FRIEDMAN: I do think Jeff's question though sort of once again brings us back to the point, since we are talking about the same information being used for different purposes. Somehow or another in the presentation of the matrix and the analysis of it, I think that's going to be a point that we are going to continually need to keep coming back to.

DR. DEERING: One question is just to side step your question, which is very important, for the moment, but it might be salient later, would we envision, as we sit here trying to come up with our little bullet lists, having in fact two to three separate columns under each one? One is uses. One is possibly content, function. And I don't know what the other might be.

MR. BLAIR: Characteristics?

DR. DEERING: Characteristics perhaps, or something like that. In other words, do we want to look at --

DR. DETMER: Sandy, you have been living with the Australian thing quite a bit, right? How do they deal with it?

MS. HAYDOCK: Not nearly as much. I don't think I could really speak to that. I wish I could.

DR. DETMER: All right, so I interrupted.

DR. DEERING: Either we do more than that, or for the time being -- because that might clarify things for us, or it might get us far afield. If we only focus on content, we are clearly going to see slippage.

DR. DETMER: No, I think we need more than content. Content need to be one that goes across all of this. Is that what you are saying?


DR. DETMER: Because there is somewhat different content in each of these. That's clear. There is probably different communications involved also in these various ones. I don't know if security considerations differ uniquely to each or not. That may be somewhat ubiquitous across the board.

MS. HAYDOCK: Doesn't security become a subset of characteristics?

MR. RILEY: It's more than just the characteristic of the data though. It's also characteristic of your hardware computing environment, and your telecommunications environment as well. You have security across all of those models.

DR. DETMER: Well, and the communications things are going to be different too. That's a good point. That's another one.

DR. DEERING: So communications would be four?

DR. DETMER: Yes, that has to do with essentially the hardware and software essentially that is involved in relating to it.

DR. DEERING: So sort of needs technology?

DR. DETMER: Right. So communications I guess involves technology, as well as policy and procedures and terminology.

DR. DEERING: Security will come under something else. So we have content. We have functions. Functions is one. Functions becomes one. Content becomes two. Characteristics becomes three, if I heard correctly, and communications becomes four.

MS. HAYDOCK: Representation is what I think you mean by characteristics isn't it? Whether it is represented in numerical format or text. Is that what you meant by characteristics?

DR. DEERING: Jeff, I think you volunteered that word, or somebody. It wasn't my word.

DR. DETMER: Representations?

MR. BLAIR: Information representations.

DR. DETMER: What does that mean to you, Jeff?

MR. BLAIR: Not only data, text, graphics, image, and audio, but it also gets to the concept representations for medical terminology, which drive us both toward clinical specificity and the convergence of medical terminology.

DR. DETMER: So in a way you are talking about format, as well as terminology kind of things.

MR. RILEY: He's talking about data, text, and meta-data.

MR. BLAIR: The way that you represent the information, in coded form, textual, image, all of this stuff. Now when I talk about concept representation, there you are beginning to get into the idea of getting to the meaning of a medical term, as opposed to just a linguistic word. So that you could begin to have the terminology cross different languages.

You could also have it converge to different code sets, so that when you have different medical domains come up with different code sets, they could begin to reconcile their differences, and reduce the ambiguity.

DR. DEERING: I would just ask whether something else would figure on our list, in which case, where it would come. Who enters the content, who controls, who accesses? Do we want to be listing those? And if so, where how would they fall?

MS. HAYDOCK: It would a subset of functions.

MR. BLAIR: Your voice is fading for me. I don't know whether your question is directed for me or not.

DR. DEERING: That was Sandy picking up on it I think.

MS. HAYDOCK: I was just wondering if that could be a subset of the function. Rather than get so many columns here, if function implies certainly who uses it, does it also imply who creates the data to begin with, and is responsible for maintaining the data?

MR. BLAIR: I did not include in this presentation users and usage. That was a separate piece that you may want to also consider.

DR. DETMER: Well, you have a little bit of uses. Your slide three talks about managing costs, accountable for quality, supporting wellness, patient care.

MR. BLAIR: Oh, from that standpoint? Oh, I see what you mean.

DR. DETMER: A little bit, but I agree in general.

DR. STEIDEL: If you go back to the word "characteristics," to me that encompasses things like who enters the data, who uses the data, how secure the data needs to be. Those are all attributes of the characteristics of the data itself. Characteristics strikes me as a better word for that.

DR. DETMER: All right, let's put it up there. You would have it as a separate category?

DR. STEIDEL: Yes, I like content, functions, and characteristics, where things like how you represent the data would be under content, the formats, et cetera, what type of data it is, image data, et cetera. That's a description of the content of the data.

MS. GREENBERG: And what about the security and communications?

DR. STEIDEL: Security I would look at as a characteristic of the data. If we say the data has to be highly secure, if we just make that as a statement, that implies that it can only used on certain types of computer systems, or communicated through the wires in a certain type of fashion.

MS. GREENBERG: That was under content.

MS. HAYDOCK: So what about communication?

DR. STEAD: Communication is -- I think there are two attributes to the word "communication." We are dealing with this at CDC, because it does have two attributes. One attribute is the hardware attribute of communication. How do you get two systems to talk to each other? I think that's the mega concept over all of this.

Then there is the other area of what are we trying to communicate? What information are we giving to give in the information? To me, that's a content issue.

DR. DETMER: So the word "interoperability" kind of comes to mind with the first one. Is that what you are talking about?

DR. STEIDEL: Yes, linking things together.

DR. DETMER: So make a little sub-bullet for the moment under communications, interoperability. You used different terms, what meta?

DR. STEIDEL: It's sort of like a meta concept. It overlies everything that we are talking about. One thing that we have talked about in our first telephone conference call was the question of is the hardware infrastructure in place to support the national health infrastructure? To me, that's an overlying concept.

MS. HAYDOCK: It's not communication as you are talking about it here?


MR. RILEY: In working for DOD on medical health information, the structure concepts, generally the way I approached it was to have three broad categories that were like that. We looked at communications infrastructure, the computing infrastructure, and the content or media of the structure. Under content or media, most of these things that we are writing up here would fall under those.

If you can envision, think in more than two dimensions, what you end up with is you've got these layers, your communications layer, your computing layer, and your content or media layer. This is the base of the cube. Then going back in this dimension is you have things like administration, security. These are features that are going to be common to all of these things. You're going to have to administer and secure content, the computing infrastructure, and the communications infrastructure.

These actually run out to about five or six different things that run in this back plane layer, but they are a back plane across all of these. If you draw like for example in content, you can divide this up into characteristics, functions, meta-data or what have you. It's a little bit more complex graphic, but it made a lot more sense to me, and it's a way I could chunk it, and get my arms around the whole thing since it was a big project.

DR. DETMER: Are you going to be with us tomorrow?


DR. DETMER: Could you bring some copies of that, or get them to us this afternoon, or whatever?


DR. DETMER: That would be great.

DR. FRIEDMAN: Don, one of the issues that's I'm concerned about as we are talking in terms of the matrix is that in fact with these three types of records, in fact we are talking at three very different levels of generality. Clearly, what Jeff has been discussing and all the work that has been done around the patient record being an extremely -- well, relatively speaking, at quite a specific level, with the patient health record being somewhere out in the stratosphere.

The consumer record, or whatever we are going to be referring to it as being a little bit under the stratosphere, but still not nearly at the level of specificity that we've got the patient record.

What I'm wondering is if it might be useful to try to find a level of generality that can cut across all three of them as a starting point, even if it was the functions, and then maybe circle back around a little.

DR. DETMER: All right, so what would you have us do? I hear what you are saying.

DR. FRIEDMAN: Maybe just starting for a few minutes on functions.

DR. DETMER: Okay. I guess the only question in my mind is whether we could reach closure on whether we need to have A be a number four, or remain A under content. If we could solve that -- because representation, I think comes off, as far as I'm concerned. Then I think we could move on to what you are talking about, and start trying to do that.

MS. GREENBERG: Representation wasn't a characteristic of content, a feature of content?

DR. DETMER: Yes, that's what I mean. I think the question is, is communication part of content, or is it -- because my understanding from the way you had it as a freestanding item. So why don't you make it four? Move communications down there.

MS. HAYDOCK: Doesn't it stand even above three categories?

DR. DETMER: No, I think you have content. The point is -- I don't care. We can renumber them, but I just saying I see it as one of the primary boxes. Then we can talk about whether we want to renumber. How would you have numbered them, Steve?

DR. STEIDEL: I think the numbering is not relevant yet.

DR. DETMER: Yes, I see them as just sides of the square.

DR. STEIDEL: I think we are looking at this type of diagram, where we are really looking at a multi-dimensional space.

DR. DETMER: Now with that, let's now talk about Dan's comment. Do you want to just start on functions?

DR. FRIEDMAN: That's what I would suggest.

DR. DETMER: All right.

MR. BLAIR: Could I ask that you consider as the first place that we start, the users and the usage first? Then from that, once you have kind of got that kind of identified, I think you could then go to the content, the functions, and the characteristics.

DR. DEERING: Users and usage?

MR. BLAIR: Yes, because that tends to drive everything else that you do I would think. If you know who the users are, and you know what they want to use it for, then from that you could identify the content that they need. You could identify the functions that they need, and the characteristics that are required.

DR. DEERING: I'm going to go get a data line for Don, but I would certainly vote somewhere or another to include users and usages.

DR. DETMER: I think that's fine. Well, what I plan to do is, as soon as I can there, the whole report is on the National Academy Press Website, and we'll just go to it.

I can't see writing all those down from one piece of paper to another. The only question is to what extent seeing that list would help us then be able to do in our next step of this process.

MR. RILEY: I think I have lost the focus here. Are we trying to define -- what are we trying to define?

DR. DETMER: What we are trying to do is lay out what are the essential elements of a functional set of records to be used for a set of purposes. So we are talking about what are the uses, who are the users, and what are they using them for. Then once you have that, you also are talking what kind of infrastructure are you going to have to have to allow that to happen.

So in a way it's a fairly grand challenge that we are facing. The point is of course, there is a lot of this ground that has been plowed by nations, as well as other systems. I guess the question in my mind I guess to some extent is does it make that much sense for us to try to wing our way through this in an afternoon or not? I'm beginning to think that's probably not the thing to do.

The question then of course is if that's the case, how do we -- what kind of tasking do we want to give staff so that we then can move in that direction. I guess it would be list where we know some of those kinds of things are so that we can then critique them, and probably ask them also to take a shot at critiquing those. If it looks like some of them are really sufficiently potent, then we probably ought to get somebody to come present to us what some of those look like. So that's one line of kind of thinking as a response.

What's your reaction to that?

MR. RILEY: Well, I was just trying to understand if we get to this process, what is going to come out of that, that this matrix -- I thought the matrix was supposed to give us something that we are going to track activities in.

DR. DETMER: It was trying to allow us to develop that kind of a score card, if you will. So the question of course then is how do you know what score the game is?

On the other hand, it's very interesting. From what Dan said this morning looking at the issue of population-based data, there isn't necessarily per se, a lot out there in some respects.

DR. FRIEDMAN: Not in terms of the population health record. There are certainly bits and pieces in dozens of different systems, but not in terms of a picture of an individual -- not in terms of putting it together as a picture of an individual person.

DR. DETMER: Well, one of the things we do know is when you are talking about the medical record writ large, it's a huge amount of work there. Millions and millions, probably billions of dollars are being spend on dimensions of tracking that thing. So it seems to me like maybe what we should do either now, or we defer it until a little later is to identify some of those sources. Definitely, I think it would probably be inefficient and inadequate what we get done this afternoon.

The other question would be do you want to list some of those places where there are sources now? Or do you want to move to population records, and move to the idea of consumer records, which those are kind of newer boundary areas that I don't think have been thought through in this kind of forum.

What we were saying while you were gone was we were concerned about our capacity this afternoon to do a much better job than nations and groups have been doing working on the medical record or the patient record. That's got a gigantic amount of work as far as listing users and uses.

Probably what we ought to do is identify where we think some of these resources are like this report, and like the Australian and some of the other Websites that are there, and try to pull some of those together. Have staff pull some of those documents together for us for the February meeting.

Then maybe I think the DOD-VA initiative clearly is another one of those. I imagine you have users and uses kind of in place, don't you?

MR. RILEY: Yes, a lot of it I pulled out of that to start with, and then we added to it based on what our experiences were.

DR. DETMER: You have the pages though, representative institutional users, individual users, primary uses, secondary uses, and user requirements. So a lot of that is the reason the book still sells, because it's very groundbreaking.

So are there other sites or sources that happen to come to others of your minds? You mentioned the Australian thing. We have marked it down.

MR. RILEY: Part of my tasking over there is to track all of these things and harmonize all these different models. So the Australian model is one I'm doing, the HL-7 reference information model is another one. Of course the DOD has their family and found-D(?) which are both active data models. The British have the COSMOS clinical process model. There is also the object model for the Good European health records.

When I think of infrastructure, if you identify all these things that you want to do with records in terms of usage and content, there is a set of infrastructure things that have to be dealt with underneath that. So the ability to represent content and reason about content is one, which is one of those things you have there.

Temporal representation and reasoning becomes important. Not just simply being able to record over an entire life time, but the issue of episode of care and encounter and all those arguments that go over that, before you can ever define those, there is an atomic thing that occurs called an event of one sort or another, and I think the Australian model has that there.

Events can be aggregated and treated as episodes of care or health care encounters or what have you. So you need to understand at the fundamental level, the time characteristics of that, and how you can reason about that over time.

Most people don't address the issue of spatial representation and reasoning, but when you begin to do population-based stuff, this becomes very important, because exposures occur within a particular area of free space. So for example in the Persian Gulf, if we bombed a munitions dump that had chemical weapons in it and generated a plume that 25 by 5 miles, that traveled over a given period of time across the landscape, where you were in free space in relationship to that actually determined your risk of exposure, and whether you were exposed or not.

So spatial representation and reasoning is of the same class of problem as temporal representation reasoning. You can probably just say temporal/spatial representation reasoning as a category.

You've got the issues of personal representation and reasoning. How do we identify at a personal level, the individual, and connect that to the content? Then you've got the issues related to how do we represent the sensitivity of the data, which you addressed somewhat in that book there. If we have three levels of sensitivity or five or whatever it ends up being, sensitivity representation and reasoning.

All of these in mind are infrastructure issues, and those are things that I have been tracking over the last five years, looking at the research in each of these areas just for content.

Then under computer and communications, there are a whole set of activities related to that as well; the multimedia communication, multimedia computing. Those are related. They sound the same, but they are not. They are actually different, and there are activities going on in terms of standards that affect each one of those.

So you are right, it's like a thing that is huge, so I don't know how you want to address it.

DR. DETMER: It does sound like you are up to your neck in this stuff.

MR. BLAIR: Who was that who was just speaking, please?

MR. RILEY: Dave Riley.

DR. DETMER: Maybe, Dave, if we can give you a clear enough sense of what we want, you could try to come back to us with something that would try to lay some of this out. Is that fair?

MR. RILEY: Sure. I don't mind doing that.

DR. DETMER: Because it sounds to me as I was just listening to you tick off a number of those things, some of those things would affect all sets of these that we are talking about, and some of those things would not. Actually, some of them are unique to some of the boxes, but not necessarily all of them. On the other hand, it is far more robust than I think we're going to be able to pull up. So I think that would be useful.

Maybe what would be smart would be to ask you to do that for essentially what would be the "medical" record or patient care, the system delivery type of record. We ought to spend our energies maybe starting to look at the issue of population. Kick it around a while and see where we go with that, and then maybe also to look at the idea of the consumer or the personal record, because those are not as obviously not a processed nor as robust and wide and deep.

Does that sound okay, folks?

DR. STEIDEL: I'm just trying to think of what's going to happen today, and what's going to happen a year from now and two years from now. Let's say we enumerate all these sorts of things. I have heard this at two levels of concept. One is we talk about like the HL-7 rim and things like that. That's a very, very detailed level. And are we going to be concerned with that type of detailed level when there are other working groups that are also concerned with that type of detailed level?

DR. DETMER: Clearly, no. Honestly, just in candor, what we really try and focus on are those things -- we ought to take a fairly quick and rough cut at the environment, and try to assess how important are some of these various ones, and then are we likely, with the time and energy that we've got, best to use our bullets.

DR. STEIDEL: And that's basically what I'm asking. I see ourselves as kind of cataloging there is an HL-7 rim and a year from now just cataloging the statement the HL-7 rim has changed.

DR. DETMER: That gets back to the point that you were saying earlier. So I consider that a very helpful question and comment. So I guess with that in mind, how do we best then even do this first rough cut, as far as that goes? I must admit I found Dan's representation -- all of them were interesting -- but as long as we have been talking about statistics and so forth, and having read Dave Kinney's(?) recent book, I don't think we have thought and worked on the issue of population health and population statistics in that context.

DR. STEIDEL: Not at this level.

DR. DETMER: I think that we have lost something by not doing that. I think there is more to be gained. Now the question of course is, is that part of the general working consciousness at this point or not in the policy circles and such? If it is, fine. My guess is it isn't yet.

MS. GREENBERG: Minimally.

DR. DETMER: It's sort of there. But it's almost sort of there as a buzz phrase than it is more something that really in my mind has substance and sort of really uptake per se. So I think there is really some merit to doing. NCVHS has in my view, great reason to do some of that.

Now the question then comes down, is this the right work group for NCVHS to do it? That's the question. But it seems to me like looking at the work plans of the various work groups right now, it probably would be smarter to do it in this group. We've probably got the greatest flexibility.

And I think that if we did some work on that, and also on a personal or consumer record, that could look like leadership. There is some sense where I think you could think through and try to give some direction to that. It does build a better infrastructure I think.

At any rate, what are your reactions to that?

DR. DEERING: One thought that I have is our discussion has clearly pointed out that these things number one, overlap. That they are in motion. It might be too bold to say that they are moving in this direction, but even if they are the DNA or the helix, at least perhaps there are links. So the question then comes in terms of our value added effort. What do we want to get out of the exercise? What would be most useful to our audience?

DR. DETMER: See what I really want to accomplish, what I want to see my government accomplish for the citizenry as a group, I would like to enhance health. I would like to see health enhanced. Not just more services, I would like to impact on health.

Now one of the things I think I know is that if I simply just add up all of the little individual things I currently do to people in the name of doing it for them, I will not necessarily have created as much health for the dollar I spend. So I'm also thinking about how do I measure the health of a population? And to some extent, how can I make both numerator and denominator interface decisions that both help optimize my expenditures at the individual level, and at the population level simultaneously if possible, because I never thought about trying to build that kind of a health care system.

DR. FRIEDMAN: One of the things that the Canadian, 25 million of them, and particularly the folks from Manitoba is they talk about trying to build a system in order to be able to isolate what is the impact of health care on health. For me that's a challenge that resonants.

DR. DETMER: Let's face it, right now for a lot of political reasons we don't care about 42 million of our citizens in an explicit way. But the problem is we are never going to have enough money to throw everything at health services that we would like. The whole GDP could be sucked up, and we wouldn't accomplish it.

So I don't know any alternative but to move to the value-driven health care system that is trying to optimize between the interventions and the populations, as well as the interventions aimed at individuals. Individuals as citizens. Individuals as patients, and individuals as part of the population.

Now with that kind of a model in mind, and my only piece with the vision that WHO had is it talks about emphasizing primary care. That's a decidedly not U.S. kind of frame of mind. I'm not saying that it's bad. I'm just saying it's not the U.S. frame of mind. But if you took that piece out, I think everything else frankly would resonant pretty well within our own backyards.

I guess what I'm really trying to get at is in a way what you would like to have this infrastructure do is move you toward certain ends. I guess in that context maybe what we are talking about in our vision -- at least that's kind of where I was taking us -- if we can build records that allow us to better track population health, and allow people to track their own health. Because I don't think that's not even a value right now, the citizen end.

The idea that I will keep data on my own health status over time as part of what I do by breathing. I make sure my driver's license is up-to-date. I make sure I have a bunch of other stuff I keep up-to-date, but I don't worry about keeping me data up-to-date.

I think part of what we are talking about is actually getting to the time when that becomes a conscious kind of dimension of what citizenship or just living means. I don't want to put it in a national context.

Obviously, this is seeming very much like my own little world construct, and it is. But part of what I see is how do I build a set of functionalities that can allow us to help each of those points of power and authority to synergistically come together that will enhance health at the lowest cost, most effectively. So to some extent, that's my answer.

Now with that kind of a framework, which may or may not make sense to us, although it does to me. Where would go then in trying to move forward, looking at this issue? In other words, I like what Nora is doing, but Nora is also looking at what happens in Manitoba period as a population too. So it isn't just simply the clinical interventions that she is looking at. There is also sort of some look at how that plays out in that population. That's the link I would like to see.

Well, I talked and I got a sigh out of you.

DR. DEERING: I thought that I was hearing -- I wasn't hearing a suggestion, but it suggested to me that perhaps rather than -- and this is to absolutely reverse what I initially suggested -- rather than split apart the records, if we agree that perhaps there is a lot of content overlap, do we start focusing on content only?

DR. DETMER: Clearly, if you really are talking about content per se, obviously when I mentioned the consumer record, that was the first wave, was to just try to throw a tremendous amount of medical knowledge and content out there with the sense that somehow the world would be sunnier for doing that. Not necessarily formatted in such a way that it really is decision support. Behaviorally, we know that doing that honestly is of minimal utility in medical education at least. My sense is that that's going to peak pretty fast.

So increasingly what you are wanting to do is have just in time knowledge service, that is bringing you what you need, when you need it, no more, no less. That's what you give them. Now to some extent you are also wanting to make them thirsty, so they will go to the well to drink whether they need it or not. That's also an issue in this.

I think to start listing all the content also to me, strikes me as not particularly useful.

DR. DEERING: I guess I didn't mean it in that detailed a fashion, but like categories of content that overlap. For example within the three records, just to start looking for commonalities there. There is clearly medical events data in the records. There is clearly personal risk data there again, whether it is entered through the individual or through a BRFSS or whatever, or through the clinical encounter. There are those categories.

DR. DETMER: We could lay out a whole taxonomy. What are we trying to accomplish with it starts coming back.

DR. STEIDEL: Well, I think we do have to lay out some basic requirements that we need in the various areas. Something that comes to mind very quickly, and this involves population records is patient demographics. That's a very basic thing. How do we identify who it is and where they are?

I think we should be at that level, just attribute level. When we are creating a national health information infrastructure, we do want to be able to identify characteristics of the patients, demographics, behavioral risk factors, things like that. These should be defined in better detail somewhere else.

MS. GREENBERG: It's really interesting, and I agree with you completely, but it is clearly not the way people are currently thinking. Certainly when they are thinking about -- let's take something very close to home and near and dear to us, HIPAA.

You've got the major of comments that came in on the transaction NPR saying that the 834, which is the enrollment, which is presumably where you would capture your population, who is covered by this plan, should not be used as a vehicle to collect demographics about people, because it doesn't meet a business need, other than the most basic demographics you need to enroll somebody. I only throw that out as just an indication.

DR. STEIDEL: That's what you need for administrative purposes. When we are talking about gathering data on me, having health records on me is a basic right that I should have, but I don't have it now, because there is no infrastructure in place to get it, which is the way I interpreted what you were saying.

What we are trying to describe I think in this work group is what are the attributes of the national health information infrastructure that allows me to create an environment where I can have that basic right? Part of that basic right is a set of demographic data, behavioral risk factor data, et cetera, that goes beyond what is needed for administrative purposes.

MS. GREENBERG: Well, I hear you. Given the fact that we agree, it's often the same data just used differently. A lot of the content is the same across these different areas. An argument can be made out of a view such as is being discussed here that one population is an enrolled group, and that in fact seeing things as narrowly as they are currently seen or in these kind of compartmentalized ways doesn't advance the information infrastructure.

DR. STEIDEL: Let's take a look at a piece of data as an example that is different, like next of kin. We talk about it in an administrative context, we may be asking for next of kin to figure out who to get the money from in case we are defaulted. Administratively, we really don't care about that, but when we are looking at next of kin from a non-administrative point of view, yes, in your case, who do we call in case they keel over. That's an important piece of information. It is the same piece of data that is being used in two different fashions.

Next of kin of course has a lot of implications from population points of view or surveillance points of view, because we want to know who to contact for medical purposes. Whereas, next of kin from a billing point of view has an entirely different context, but it's the same data. So there are many uses for the same data.

DR. DETMER: You raise another issue. Do we want to have an infrastructure that essentially has somebody asking what is in fact the utility and to whom for collecting such information at the cost that it carries? So you are also are having sort of a standards for standards kind of approach as it relates to some of this.

Because the fact of the matter is if you don't have any absence of some objective, however defined, data, it is totally a political decision on who has the clout to be able to say, yes, you will collect it, or yes, you won't.

To some extent I don't know that the day will come that we will have such robust interactive databases of such utility that in fact you won't still have to do special, specific studies in given instances, depending on what you need. So the question is this intention of how much of this should come out of, if you will, routine business robustly defined, and how much of it frankly you have to do just as special studies either by the agency that does it, federal, state, private, whatever.

I guess from the point of view of policy at the national information infrastructure level is where should you draw the line? If you don't where to draw that line, how should you go about even thinking about it? Because that is a relevant policy consideration.

I guess what I'm really trying to cope with is for the limited kind of time and energy we have got, where is that punch? Which I think is really perhaps -- having gotten this first cut, how much do we need to keep going on laying out more, and where should we try to put our intellectual throw away?

MS. HAYDOCK: I have some thoughts here, just as we have been talking this afternoon, because you make a very elegant and persuasive argument for the need for all of this. A lot of this is kind of next context for me. I have not looked at things from this standpoint before.

In that there are other groups who have done a lot of work in the patient record area, maybe not in population and personal, but maybe so. If there are people out there developing Websites, certainly somebody has given some thought to it. Could we play a more powerful role by -- we are already a consortium, I know. But being the seed for a consortium that included people who are already doing that work in an intensively way, and kind of bring them into the process, rather than sit here in a room and come up with a framework.

DR. DETMER: Right now I don't think that -- this will sound, I'll just say it. It's the way it is. Vhita is only model I know that really has tried to look at how do you transfer meaningful data across these very important domains. Most everybody is roaring out after their one track, and almost their one application. I'm a right elbow specialist guy and that's what I want to do. So they are not necessarily thinking comprehensively across the domain of delivering medicine, let alone prevention.

Now that's just a statement. It's not necessarily a value judgment if that's good or bad. It's just sort of statement of what I think is true.

MS. HAYDOCK: But if they heard your argument, would they buy in?

DR. DETMER: Well, I think that there is some merit to laying out that speak to that, and having people come in and let them interact with one another and hear such things. Ultimately, I think it comes down to some extent what you were saying at the time we broke. What ultimately will happen probably depends on how much utility it ends up having in the market, not how elegant the language may sound, but does the thing get off the ground and fly?

If it does and you want to fly, you'll probably use it. So to some extent, that's true too. But I think the idea of having people become more aware of what others are saying is good. But again, that's what is kind of nice. We have a little bit of a bully pulpit, as much as anything else. By the time we put things up on a Website, a number of people are going to look at it, and some of them, it will influence their thinking. They will be a little bit different after they have been touched than they otherwise would.

DR. DEERING: But to pick up on her point, I think what she is stressing is that one of our value added -- and again, this is an individual observation -- is that we are consistently, persistently, persuasively we hope arguing for linkages among the various areas.

To the extent that we back that up in some ways with either examples of why it work where it works elsewhere, or why just through innovative conceptual analysis and presentation we are able to make linkages that no one has made before, or whether we just diplomatically persuade the parties, gee, this is really a pretty good thing to do. All of those are legitimate ways to accomplish these linkages that I think are central to your vision.

DR. DETMER: I think there is a lot of value that will come out of it frankly. Most of it we currently don't even know, because I just think it's that kind of thing. Once you change your way of thinking, a whole variety of things start happening. It's just like how I will use the Web if I have never used it, as opposed to how I use the Web after I've been using it for about three years, and it's different. It's that kind of thing.

Ultimately, we are also going to be talking about though once you get people sold and saying, okay, I got it, then there is a whole bunch of now how do you do it, and the how to issues. Those kind of issues come very quickly. Once somebody sort of says okay, I'm with you. Now what do we do? So you've got to be kind of ready to start moving to their how question too.

DR. FRIEDMAN: This is in the thinking out loud category, but as we are talking I'm wondering if there might also be some value in trying to lay out an image. Essentially an ideal image of what a national health information infrastructure would do, what it would look like.

In fact, in the concept paper we say it is a set of this, this, and this. But I think that leaves a lot of people with the question of well, it's a set of this, this, and this.

DR. DETMER: It's more anatomy than physiology, I'll admit, in terms of what we have generally written.

DR. FRIEDMAN: I'm not saying it to criticize the concept paper, but perhaps at some point saying not here is how we get there, and not here is what the specifics are, but here. Not at the 25,000 foot level, but at the 25, 50 years from now, here is what we think, if we had a sensible national health information infrastructure, here would be the --

DR. STEIDEL: I found the NHS paper was very good.

DR. DETMER: I did too.

DR. FRIEDMAN: I said the same thing about the Canadian paper.

DR. DETMER: It's a nice piece of work.

DR. FRIEDMAN: It's a beautiful piece of work. It's not directly applicable to us because of the difference in the health care delivery system. But with some tweaks and changes it could be.

DR. DETMER: That's a novel thought. One other option is the best compliment of course is to quote somebody. Maybe that's another thought, would be to actually start trying to rewrite that for American content.

DR. FRIEDMAN: If we just crossed out Canada and put in the U.S. would anybody notice?

DR. DETMER: That's very interesting. As we have also heard, there are some very nice features to the Australian deal. So again, probably if this just American arrogance being what it is, if we took those three models and rewrote it, but wouldn't market it as how we stole three other nations blind, but the American viewpoint for the next millennium type of business, with appreciation to resource blah, blah, blah, that might be the way to go.

DR. FRIEDMAN: The Canadian report that I was thinking of, it's interesting, because it also starts out exactly the way Jeff started out, which is the emphasis is on the what so much as the why, and for whom.

DR. DETMER: The U.K. takes the stakeholder approach. They say who are the stakeholders, and what do they need? What do they need from one another, and what do they need? Now they don't quite say what could they get collectively, although they say it in the Venn diagram.

DR. STEIDEL: They don't ignore it.

DR. DETMER: They don't ignore, but they don't see it in my view. What they have got is not inconsistent with being able to go there as I read it.

DR. STEIDEL: They have few pages.

DR. DETMER: It wasn't sort of core thing, but it was -- that's not critical. I'm very impressed with that work, as you are. So maybe that's what we ought to try to do. We've got two ways of going at this. One would be to go through those nations' documents against our vision template, and pull out any hows and what fors and why fors, because we really didn't talk about why for.

Then think about trying to develop a paper that colleagues(?) that. Then once we have done that, look at it from the context, okay, in the context of the U.S. at this time and this back yard, what should be our work agenda then for the coming year? Is that sounding more productive?

DR. STEIDEL: Yes, it is grounding it into something that is already existing.

DR. DETMER: Candidly, then if we do find some commonalities, you also have the chance of moving the whole world forward, which is really great. I'm also absolutely convinced globalization is here. We just have not caught up with it. If you don't think it's there, you're missing the boat. It's there. It's a question of are we living that way yet, and I don't think we are.

Jeff, have you been tracking this?

MS. GREENBERG: Do we know when this group is meeting in February?

DR. DEERING: That's what I was going to say. My understanding is that some groups have already claimed all of February 2, and some of our members would have conflicts. So we got preempted in a sense.

DR. DETMER: The other question is will we have enough to do too much with anyway in February. If we can hear from this Australian, we might want to hear from her actually in plenary, as far as that goes.

Now it might be good if somebody could give us a rough cut at the NHS too, as far as that goes.

PARTICIPANT: I could do it.

DR. DETMER: Could you? Okay, I think that would be worthwhile. In a year I'll be able to tell you a lot, but right now --

PARTICIPANT: I'm going to actually be seeing one of the co-authors in about two weeks.

DR. DETMER: That would be good. I think that would be useful. Then maybe with the Australian -- are you prepared to give the Canada.

DR. FRIEDMAN: Unfortunately, Don, my wife is out of town those three days. I'll do everything I can to be here February 4, but have to be able to get home in time to pick up my kids.

MS. GREENBERG: The U.K. document, how big is it?

DR. DETMER: It's not that big.

MS. GREENBERG: Is this something that we should provide?

DR. DETMER: Yes. It's a Website too.

MS. GREENBERG: On the Canadian, Peter Waegemann has the experience of -- I was at a meeting where he was on the agenda, and at the last minute he didn't come. Apparently, this has happened to Peter about three other times. I don't know if he's extremely busy and gets overcommitted, what have you, but I don't know whether we'd have much luck. Maybe Jennifer.

DR. FRIEDMAN: Just reading it and having a chance to talk about it, so not necessarily even having a presentation.

DR. DETMER: I think that would be good. I think it would be useful to get the documents ahead of time, let us read them, but then be able to have somebody that could give a very brief presentation, but mostly be able to have an interaction.

DR. FRIEDMAN: Don, did you receive a copy? Because I tried to send the Canadian a couple of times.

DR. DEERING: I got something called, "Health Information Needs in Canada." It's a big 1998 report that I thought was most equivalent to the NHS Website.

DR. DETMER: No, I don't have that. It choked in my pipe.

DR. DEERING: I can track that down.

DR. FRIEDMAN: Three documents, the U.K., there is a Canadian September or October Advisory Council on Health Records, and then there is the Australian.

DR. DETMER: I don't know, Mary Jo if you guys could even do a little bit of crunching against some of what we have been talking about, because at least this conversation, with the last meeting, has helped moved things a little bit.

But it's our assignment to read these things, but also hopefully then be prepared to have a very brief presentation, with an opportunity to dialogue actually with folks a little bit.

MS. GREENBERG: This would be with the full committee?

DR. DETMER: Some full committee; some maybe not, depending on what the agenda is like. In other words, if we can get a short meeting --

MS. GREENBERG: Tomorrow.

DR. DETMER: Exactly. In other words, we have talked maybe about having the Australian anyway. So the point is if we could get a brief presentation on the British and the Canadian.

MS. GREENBERG: Do you think we could give Steve a British accent between now and then?

DR. DETMER: He said he could handle it.

DR. DEERING: Who did you say was going to give the Canadian accent?

MS. GREENBERG: Somebody that we have been working with on this other project. In a way I see that NHII, although -- maybe it makes sense, because it's a work group of the executive subcommittee, which is kind of overarching -- but as at least two other activities that are going on in the committee, if not all of them, but the health statistics work group for the twenty-first century, that's more of the sort of population-based piece. Then the computer-based patient work group is -- in fact, that's one of the things I wanted to ask about Jeff about.

DR. DETMER: That's what I was going to ask Jeff too.

MS. GREENBERG: We don't want to obviously duplicate what they are doing, but in a way sort of provide a template or a framework for those two. And the consumer one, I'm not quite sure. That's an area that the committee has talked about, but has never had a strong focus in.

Then I think some of the things, like what's being done by those work groups. Clearly, because of that overlap, I think you really want to have as much of this to the full committee as possible.

DR. DETMER: We'll want to do it sooner. My guess is like most work groups, you've got to do some work before you take it to the group.

MS. GREENBERG: Well, that's true. But the presentations --

DR. DEERING: The ASMB presentation that we're going to get might be interest to the full committee.

DR. DETMER: I think a number of these things would be. I guess my point is though I'd like to see the work group process this stuff before. It hits a lot of the work groups' areas.

I guess what I was going to ask Jeff was if he had reactions to what he had heard before we found out you got cut off, and also how would you advise us as one of the co-chairs of the CPR group to not essentially doing any duplicative things, but makes sense in an interface way with what is happening. Now to some extent you bring that to us, so it's easy, because I can ask you that question in our meeting.

MR. BLAIR: I'm not sure what I missed before I got cut off. It was a little hard for me to hear. In terms of this last question, I very much feel as if the work that we are doing on our CPR work group, which is -- I just want to repeat the focus on it -- study and recommend uniform data standards for patient medical record information and the electronic exchange of that information.

I feel like that is very synergistic with the health care information infrastructure work effort. We are still going through the process right now in our work group of trying to identify our focus areas. At our meeting on December 8 and 9, we had feedback from a number of witnesses as to whether or not we were focused on the right issues. We had some validation, but there were a number of points that were raised where there were additional topics that we need to focus on.

So we're still at the early stages, but nevertheless, there is a great deal of synergy with the health care information infrastructure.

DR. DETMER: Well, if so, it might be smart then to always keep this committee workshop hat on as well to some extent, so that you don't let that group get into things that you think may be most appropriate for this group at the same time.

MR. BLAIR: Yes, I agree.

DR. DEERING: I understand you have mapped out a full day's work on February 2.

MS. GREENBERG: I hope the CPR group hasn't mapped a full day's work, because actually the privacy committee, which has many overlapping members has pretty much mapped out a 10:30-4:00.

DR. DEERING: My only point was that at some point, if not February -- certain of these meetings could meet jointly.

MR. BLAIR: Is there any possibly by the way -- that's kind of like right in the middle of the day, 10:30-4:00. It would be kind of nice if it was the afternoon or the morning, where the CPR could then be the afternoon or the morning. Otherwise, we are sort of splitting our time.

MS. GREENBERG: I tend to agree with you. I only got the outcome, but Kathleen was kind of given the charge of what time did the subcommittee need for what they were planning on the second, and that was what she came back with. Now this can be discussed at the executive subcommittee meeting tomorrow, but it does leave kind of a fragmented time.

DR. DEERING: Even overlap with that committee, because clearly we overlap significantly with them. I mean I was thinking of the comment earlier that said that there is no work right now being done in the privacy area that takes account of online consumer-driven, market-driven patient records, neither within the executive branch, the legislative branch, or with the standards organizations. If that is so, then that's a major gap that someone needs to be looking at.

DR. DETMER: I tucked that away myself. I need to find out if that's true. I think it probably is. I'll ask Lynnette. I think it probably true, but it is interesting.

DR. DEERING: One of the purposes of what you had seen this committee --

DR. DETMER: Exactly. I hope the rest of you aren't anguished by the non-linearity of our discussion today. I think that goes with getting in to something as grand as this. There is no way you are going to jump out of bed and just lay it out. So it's going to take some forbearance. I don't think there is any question about that.

DR. DEERING: Just because I'm so task oriented, I wanted to understand a little bit about --

DR. DETMER: Never apologize to a surgeon for being task oriented.

DR. DEERING: In terms of getting through these other three reports. I heard you say that you would like to see some effort to look, at nothing else, at the level of tables of contents of these reports. Some cursory overview by some small group or whatever, to begin to identify commonalities, especially in light of our paper, those elements that seem most relevant for us.

DR. DETMER: I think that's a task that is worth doing. What I'm saying though, I don't think we're going to want to expand that into being our reason for being. Now I think we still need to have a little more of this in a library.

DR. DEERING: But you would like this for February.

DR. DETMER: If we can pull some of those together.

MS. GREENBERG: You still want Dave to do what he said he was going to do?

DR. DETMER: If he would, yes.

MS. GREENBERG: That doesn't negate that.

MR. RILEY: I need to make sure that I understand the task that I have here.

DR. DETMER: It's just sort of laying out the terrain map in a categorical kind of way, models that exist out there that relate to robust functional largely computer-based health records, patient record kinds of systems I think. You're not really working on population or consumer records, are you in your thinking?

MR. RILEY: Well, the one project that I mentioned earlier that you and I talked a little bit about that has to do with the Bosnia and Southwest Asia deployments --

DR. DETMER: That would be good then, to the extent that you have that. I think that would be very useful, because I don't think there is going to be a lot of models around.

MR. RILEY: The basic model that we used to build from was what we called the Service Member Life Cycle Model. That's where we identified all the key data points in terms of entry into the military, predeployment, postdeployment activities, collect of serum, collection of certain kinds of baseline data, psychological and physical data, that kind of thing.

So right now there is a significant amount of profile data. That was a fairly large group of people, combined with a lot of environmental data as well. We actually have sensors out in the field that detect biological and chemical substances.

DR. DETMER: I think summaries of that would be very helpful.

MR. RILEY: That was one task. There was a second task.

DR. DETMER: Well, the other thought was whether Sandy and you and Hetty and Maureen, whoever, could get together and kind of take a cut at looking at these U.K., Australian, and Canadian plans and models from the context of how did they relate to what our vision document is. Then I would say in the context of also some of this discussion that we have had this afternoon as well.

Is that clear?

MR. RILEY: Sure. The only thing I don't have is the Canadian document.

DR. DEERING: I think I can get you a Website for it this afternoon.

MR. RILEY: Okay, that would work.

DR. FRIEDMAN: If you haven't got one, I think I can get you one. It's the September or October 1998 report.

DR. DEERING: Yes, I had only the hard copy.

DR. FRIEDMAN: It's interesting, because having looked at all three of those documents, I think that what we are going to find is that there is, as far as I can recall, nothing on the consumer health records. There may be, but I may have missed it.

DR. DETMER: What I said about the U.K. thing, I studied it carefully enough, and it does not. It has an explicit-type thing. There is language that if somebody is thinking that way, you say well, this is how I intend to use their document to make that argument. In fact, I intend to do that, but I don't think that's what they had in mind.

DR. DEERING: The American situation is different in that we are not talking about a policy decision to create consumer health records. What we are reacting to is the de facto emergence of such an animal, and what does it mean.

DR. DETMER: And what would we be desirable things in our view.

DR. FRIEDMAN: Oh, no, I'm just saying in terms of the content of the two documents. There seems to be nothing or almost nothing on that. The U.K. document seems to be primarily medical record/administrative record focused. The Canadian, as far as I can tell, is primarily medical record/administrative focused, but the line between that and the population record gets blurred, which I think is positive.

DR. DETMER: There is planning in management as a key part in the U.K. document that's pretty substantial.

DR. FRIEDMAN: No, you're right, Don.

MR. BLAIR: I had the perception in reading Richard Alvarez's article that in terms of a population record, of population statistics one of the major objectives that they had was to use that to be able to feed back practice guidelines and clinical protocols back to the actual process of care.

DR. FRIEDMAN: That's my assessment too, Jeff. When I said that the line between the patient record and the population record gets blurred, that's what I meant.

MR. BLAIR: That was one of the reasons, getting back to the point I made earlier, that I think it is useful for us when we analyze these things, like for example with the population record, we may say something about who is going to use that information, and then how they will use it. I think that that helps us to lead through this, and try and keep these areas from being so confusing.

MS. GREENBERG: I should mention, although I don't think it precludes or negates the importance of what you are talking about doing, and in fact the more of us that should become really familiar with these documents, the better probably we are able to pursue. But one of the papers actually that was commissioned under the work group on Vision of Health Statistics for the Twenty-first Century is from the Canadian Institute for Health Information. They are going to be looking at these very models that we are talking about. To some degree there are implications for the U.S., but that's not really a focus of that paper.

DR. FRIEDMAN: It will be interesting to see what they come up with too. Looking at it probably from a slightly different lens, but the same people. Jennifer obviously gets involved both in population-based data, as well as the other.

DR. DETMER: See the question I've got, if you took a sort of stakeholder approach, which is what the U.K. did -- they used that diagram. They didn't do it quite this explicitly. But they kind of have persons and patients. What I'm saying is if you have a population at large out there that is actually just keeping their own data, not as a patient, but just as my health data, and what I ought to know and do it. So you've got to create that user need, and create that kind of thing.

You clearly have this for disease management. There is obviously some interface that patients need to be doing with health professionals on that. But there is a sense of data that some of these overlap. The question is, how big is this core that all could benefit from?

To some extent, if you could really see how much cross-functionality you could build in that piece, and make sure that all systems had that, that were in an interchangeable, interoperable kind of mode, and that these other ones had it to the extent they needed it, you could have a much more cost effective system, because all of your transaction costs would be less at any one of those transfer points.

That's the kind of point I was trying to talk about as a management kind of a strategy, and as an infrastructure vision kind of idea. I don't know how big that can be. That's a research issue as much as it is anything else. But I think it's a terrific research question, because there are billions of hours of energy and cost and everything else in there probably suffering to boot.

I guess the question then comes down to what more productively do we see ourselves doing, as far as that goes even this afternoon, let alone tomorrow morning. We had talked about commissioning some other things that we might want to think about in terms of things we know we will want done, thinking along the lines we've been thinking that we would want to see by June say.

Because if so, it would be very smart for us to be thinking through that now, if we expect to see it then, in terms of it takes time to identify people, get some stuff done, in terms of contract kind of work for what we are doing.

DR. DEERING: One thing that comes to mind -- and I wasn't here when you put up that piece --

DR. DETMER: The content?

DR. DEERING: The business, right.

DR. DETMER: That's the Australian.

DR. DEERING: We talked about what is the business model, what are the business elements here. Clearly, in the United States that will be a larger element than it will be in any of these national health system countries. Especially since we are including the person area, the patient area perhaps more actively than they are.

Would there be any value, and do we know of anybody who could speak to what are the market forces that might be driving any of these, that could promote convergence? I know that there has been a great deal written about the business forces behind the electronic patient record, and the administrative needs of this.

There has been a lot written from the research point of view, but no one has ever really addressed the business case for an integrated approach, that I know of, to see if there is any. If you did that with the business interest behind each.

DR. DETMER: This is the U.K. deal. They have planners and managers, which represents business in their system.

MR. RILEY: The POMA effort that was done by Benchmarking Partners a couple of years ago identified this as the value chain for the global health care environment for the U.S. in terms of when you look global value chain, and do value chain analysis. Now they had a more detailed value chain that they developed for specific health care delivery, but this put that other value chain within a context, and those were the major players.

DR. DETMER: I'll write those up here so that people can see them.

MR. RILEY: Probably one of the most important forces right now is this huge shift that is going to occur from a defined benefit to defined contribution in terms of the way health care is procured. I think over the next year or two we are going to see a significant transformation in the industry, because right now in defined benefit, your HMOs have rung everything they can in terms of cost containment and they are not making money.

A lot of them are having difficulties or beginning to. So the shift now is pushing from defined benefit to a defined contribution. So if I'm the employer, I'm going to tell my employee, I'm going to give you $3,607 this year, and you go out and figure out what health care you want to procure.

Basically, what you have done is you have now shifted decision-making into the hands of the consumer. What are the business things that have to occur to be able to support that activity? For example, right now there doesn't really exist a Consumer's Health or a JD Powers equivalent for health care. But if the consumer has to make those decisions, and the consumer is going to demand to know what is the performance of this health care organization.

It also I think, will give an opportunity for health care organizations to provide an incentive or force for patients to take more responsibility for the long-term things like smoking. For example, a health care plan may say if you are non-smoker, we will cut the rate for things that we know affect health. Do you wear your seat belt, are you a non-smoker? These kinds of things may actually affect the price of what you're going to have to pay for health care in that environment.

These will become financial incentives for patients to do better. When you hit somebody in the wallet, you've got their attention. So it goes both ways. I think consumers will demand much more knowledge about the organizations. Does this doctor follow the clinical practice guidelines, or is he known to vary from those? If so, what's the outcome of that. Those kinds of report cards, which is much more detailed than what we get HEDIS or any of those kinds of things right now. I think just that one force, that one driver will be a significant player I think in the next couple of years.

That brings me to the other thing that I was going to bring out earlier, which is I kind of look at this as an eco-system. So the model that I studied has to do with business eco-system model. I have tried to apply that to health care and think in term of -- basically, there are five phases in the life cycle of an eco-system, and what phase are we in? As you evolve from one eco-system to the next, and certain things get replaced.

If you have burned out a forest, what are the first plants that move back in, that then support life for the next group to come in? It's a very different way to look at it from a modeling perspective, but these are some of the major players here in the eco-system. Over time, their roles change and their contributions change.

So that's where I have been wrestling with that over the last six months, is to try to understand from an eco-system, what are the players, and what are the drivers, and what's the life cycle that we're in? As we have these major shifts in the eco-system, life is changed from defined benefit to defined contribution. What does that mean to the life and well being of the eco-system? Because we are in a top-down driven system.

DR. DETMER: You are absolutely right, there is going to be more movement I believe to putting the person more in the decision seat with benefits. It saves the employer having to carry that responsibility and burden. Frankly, it also off shifts some of those costs, so that is also happening with all that. So there is no question, that's a dynamic in the payer side in the U.S.

MS. GREENBERG: How well do we think the individual consumer is in a position to do these things? I would say very minimally.

DR. DETMER: They are clearly not set up.

MR. RILEY: The rapid evolution of these Websites I think, and the fact that they are utilized highly by consumers suggests that the consumer, whether it is explicit or it is just intuitive on their part, understands that they need to be better informed. So they are going out to find that information.

DR. DETMER: But it's the people that go to the Web. It's still a big challenge. But hearing you talk, it does bring to mind that we do need to worry not just about the environment as we would want to create it, but also how does our system act, and what are its future drivers? That's very relevant. At that point then, probably having at some point, Kennedy(?) come in and talk about his book on population health management or somebody thinking on those kinds of terms. He does take a business cut at that question in his book.

The other thought, just also then you are going to be looking at cost/quality kinds of issues. Molly Coy and I just did a summary of an IOM conference that just came out in the Millbank Quarterly this issue. It is on the Web, and you might get that for everybody, because it just sort of summarizes where we are in quality at this point. We call it "Quality of Crossroads," because the issue is in fact are we going to literally start buying and driving and looking to quality seriously, not just sort of the trappings of it. Actually starting to look at it in quantitative terms.

I think that probably also would be worth -- just look at an on log of this kind of emerging environment. We are talking about globalization. We are all over the top of that. That's a driver today. And clearly IT is a driver. That's what we are also talking about.

This whole issue of payment though, and how capitation strategy plays into some of that, because we've been in discount, is what we've been in, and that's rung out. So the issue is are we going to start going to value purchasing and population care, and more shifting the responsibility to the individual is the issue.

I think I've got on one of the slides I've got on one of my talks there is -- the other question of course, do we move to more evidence-based medicine too? Looking at medicine itself, medical technology. Do you know what is happening in terms of medicine and its vitality at the moment, because that's going to play out; how that technology plays in and stuff.

That's good.

DR. DEERING: Besides you, has anyone put that together?

DR. DETMER: What do you mean?

DR. DEERING: All of the things that you have laid out. It sounds like it would be excellent to not only bring in individuals who speak on those, but it would be nice to commission the synthesis.

DR. DETMER: Well, for what that's worth, I have probably given this a few times around the country. I could give a half hour talk as far as that goes.

MS. GREENBERG: It would be a nice thing for you to do before you leave us.

DR. DETMER: I think it could be useful; I don't know for the whole group. I think it would be useful perhaps as an input to this group. Frankly, we're sloshing around in this soup of environment, and what are the essential -- are going to be what you work with. That's what it is.

DR. DEERING: I would disagree that it's only for this group. If you gave that as a valedictory, I think you would fill an auditorium. I think that it also would position the efforts of this group as more future-oriented group. That weren't trying just to state where things are at right now, and cope with that, but we're trying to set an agenda based on an analysis of drivers and trends.

Because by definition the types of recommendations that you had intended to give to the department, and to any other stakeholders were not those that were necessarily for immediate, pragmatic implementation, to use I think Jeff's phrase, but were to set agendas for action.

MS. GREENBERG: As a matter of when I was reading the transcript of the last executive subcommittee meeting it was interesting. If you remember, I was back and forth on the phone, so I hadn't caught everything. There was a discussion about this work group having at least an hour session for the full committee at the February meeting as a possibility. It's on the list of possibilities to discuss tomorrow.

Are you going to be at that meeting tomorrow, Mary Jo?

DR. DEERING: Yes, I am.

MS. GREENBERG: Good. It wasn't real defined, but I actually have with me what you discussed then.

DR. DEERING: I remember one of the things that you had addressed is what can this group do say within approximately 12 months?

DR. DETMER: Right.

DR. DEERING: Not knowing at that point that you would not be there. But I think it encourages us all the more to come up with something specific that we can feel comfortable is set in place within the next month or so, that we can deliver, that is at least robust enough, whatever it is.

DR. DETMER: If we could try to do a paper that would in fact sort of look at roughly where these various international models seem to be pointing things; how that relates to major drivers in the current emerging U.S. situation; and then how would we, with obviously a decentralized system of health care, how would we from a government advisory committee advise both the private sector and the public sector to address some of these things with a short list, maybe no more than five things of some recommendations. Almost just kind of a white paper I guess is what I'm saying could maybe be the product.

MS. GREENBERG: Well, I'm smiling, because I'm thinking in a sense of things kind of coming full circle. When you first interacted with the committee, Judy Jones had you come in as a kind a consultant in a way, but it was pretty much on your own dime. At best we might have paid for your trip or something.

DR. DETMER: It would have been nice.

MS. GREENBERG: Yes, it would have been. Oh, well. You were on sabbatical that year, if I recall. You were doing international work. She had you come in as a sort of a consultant on what are the trends out there internationally.

Now as you leave us, and I'm in sitting here thinking, Don can still be a consultant to this group from the international perspective. I hope we will be able to stay in touch with you, and when you are in the U.S., get you come by, et cetera. As we said, we're a global society anyway, so communicate with you by e-mail, et cetera, because you are clearly going to continue with these issues, just in another venue. It's a loss for the committee, but I'm not quite ready to lose you.

DR. DETMER: I guess the question that I've got -- thank you -- but talking our task, what are your reactions to that as a kind of a product objective? Is that what you guys would want to see us do for this year or not, I guess is my question. Whether we finally just even do another cut, a different cut than obviously this first paper was.

I can see something like that, even if you annually just put out a paper. This was just a set of whats. I think the next one is sort of like much more refined in a way, what we are conceptualizing for this next year. Basically, even if you put up a good concept paper once a year that went up on the Website from this group, with the massage of the full committee, I wouldn't downplay that that could have to make an impact.

I sound like I'm arguing to make a point. I guess the question is what are your thought? Richard, you haven't weighed in much.

DR. HARDING: I think that would be excellent. I think working towards a paper that would kind of amalgamate a lot of the better thoughts of the three countries and the Department of Defense and so forth. And try to through symbiosis or synergy or whatever, have a greater good come out of it. I think that would be a very worthwhile year's work.

MS. GREENBERG: Well, pieces of advice that both the Data Council and to the rest of committee.

DR. DETMER: Exactly. For example, the NCHS, that issue that you are working on should feed this, but just might feed it too.

DR. FRIEDMAN: This may be assumed in what you are laying out for the paper, Don. I would like at either this or the next iteration or the next iteration after that to have some form of what is a national health information infrastructure for the U.S., even if it is one diagram.

DR. DETMER: No, I think it would be great to try to in fact do this three nation analysis from the context of trying to also look at that and rewriting it in a U.S. context. I think that from the quality of those countries' work, I think that would be, in itself, a very worthwhile piece, whether we actually contracted somebody to help do that, or whether the committee did that with the support of staff, I don't know. We'll probably have to see how that plays out.

There are things that are unique to our situation, that those countries don't have the same realities. I think that's terrific, because it's good work. It's good stuff.

MS. GREENBERG: We'll have these two papers, hopefully by the summer, that the other work group has commissioned, one on at least these three systems, and then another one, which goes into more detail on the Manitoba experience or its relevance or lack thereof to the U.S.

But of course none of them will look at say the document that this group put together and related to that. So that doesn't mean that that's not a relevant thing to do. I think it is, but from those, and the work this work group does, and some additional document might come out by the end of year, which you are talking about.

I guess Dan -- we need to think about -- if this group wanted to commission something as well, we'd have to think about how they would fit together, and not be duplicative, et cetera.

DR. FRIEDMAN: The way I see what Jennifer Zelmer(?) is doing from health information, is basically if she is going to look at those three documents. And I think that it's going to be more from the viewpoint of "health statistics," but there is no reason why she couldn't expand it, or do a companion, which obviously makes sense to make it broader.

MS. GREENBERG: Yes, that's what I'm thinking. Let's let her stay on the course she is on.

DR. DETMER: Let's have you guys continue to work with what we have talked about. Let's see what we come out with, and then we may want to do that.

MS. GREENBERG: Yes, it's a possibility.

DR. DETMER: But that's a thought too, if I'm hearing one other thing that the work group could do; would be to do its own white paper. But it could be fairly short, and then it might well want to contract papers to be written, three-four a year on those topic areas that we think are particularly relevant and emerging, that relate to a barrier situation, or a regulation that is needed, or this or that. Are you following?


DR. DETMER: To supplement, so that the committees -- we're all too busy -- to supplement the work. That might be a nice way of doing that, to carry the impact of the committee further than it otherwise would be able to do with the time they've got.


DR. DETMER: Mary Jo, do we want to talk about what we want to do at the February meeting a little more precisely then? I guess that's one question.

DR. DEERING: Could I clarify whether or not we have done away with tomorrow's meeting altogether?

DR. DETMER: Well, I don't know that we have. It looks like we are moving in that direction, but there is still some work we need to do.

DR. DEERING: Because you were going to bring back -- if you could bring things tomorrow, and give a presentation, that's fine by me.

DR. DEERING: Because that might suggest something to us. It might suggest that maybe we are also able to go ahead and map out what I think you said what you wanted to see, which is a three dimension. Go back and see, is that as impossible as we thought it was? Or if we simplify both the number of categories we think we want to include in the model, and the depth to which we want to get, can we at least do something useful that conveys to people our emphasis on the linkages among them? Maybe we'll see that it's more doable than we thought; maybe not.

DR. DETMER: It's a question of what you think you could do. I'm happy to come and hear.

MR. RILEY: I think I've got all the files at home. I can probably pull together a presentation tonight.

DR. DETMER: All right, would you want to start at nine or would you want to start later?

MR. RILEY: It's easier traffic-wise if we start a little bit later.

DR. DETMER: Well, we could start at ten. That's no problem.

DR. FRIEDMAN: I have a one o'clock flight.

DR. DETMER: We'll be finished at 11:30 a.m. anyway.

DR. FRIEDMAN: That would be great. I can't cash in my hotel room anyway. I'd love to hear it.

DR. DETMER: All right, why don't we do that, or 9:30 -- I don't care. Is 9:30 as easy for you, one or the other?

[Discussion regarding time.]

DR. DETMER: We'll be on at nine.

MS. GREENBERG: Now Jeff isn't with us tomorrow.

DR. DETMER: Why don't we meet at ten? Then another thing that is a loose item, we had a letter -- I'm kind of apologetic for this -- to John Eaglehart(?), editor of Health Affairs, actually from me as chair, but also from Mary Jo that was speaking to essentially the paper that has come up. We'll talk about this with the committee.

I don't know if you folks have had a chance to read this or had questions or not. I think the point is it hasn't been published yet, so we could probably still pull it if we wanted to. I see no reason to, but I guess I would like for you to look at that. I did want you to hear me apologize for not actually getting this through people before it actually went to John.

It's fairly direct. I don't think we editorialized outside of our paper.

MS. GREENBERG: We did send this out.

DR. DEERING: I do want to apologize, because the deadline seemed much tighter than it was, and with travel times, it just really didn't seem to --

DR. DETMER: You might look at that tonight too, because we can talk about that tomorrow as well.

MS. GREENBERG: Now on the matrix, that really awaits both the report --

DR. DETMER: I think so. I think we really want to give them the benefit of what shakes out of some of this before we start trying to create our own world.

MS. GREENBERG: I would agree.

DR. DEERING: When this was put together, we really thought ASMB would have presented us with something a lot more detailed than today, and we would at least have their terminology.

MS. GREENBERG: Now see, when she comes back in February, that would clearly be to the work group. So there are some things that you will need. On the 3rd there will be an opportunity for work group and subcommittee working sessions.

DR. DEERING: And we have the opportunity to have a presentation by the people who are working under the auspices of the National Health Foundation and the Progressive Policy Institute looking at this Visa approach to the consumer health record.

DR. DETMER: I like that idea.

DR. DEERING: They are quite willing to come and talk with us.

DR. DETMER: That sounds good. The question is whether we want to do it as a panel? That might be kind of interesting.

DR. DEERING: With whom?

DR. DETMER: The work group.

[Administrative remarks regarding the exchange of documents.]

DR. DETMER: Another thing that I think you also had on your list that I liked very much was this idea of a glossary. I don't want to gloss over it. Now that brings to mind Dan's presentation, which I keep coming back to. It would be nice to get some of those definitions. Is that too much work?

MR. RILEY: I'll be happy to work with somebody on it, sure.

DR. DETMER: Good, because I think we ought to start trying to pull those together to the extent we can, because not all these things are actually that clearly defined and sort of out there. Even in the context of our discussion earlier today with Jeff on what's a patient versus -- that kind of business, what's in, what's out, we ought to start trying to build that set of terms and rotate those through the group for review and comment, so that we start building what we call our dictionary for this stuff.

Hetty, what are we missing?

MS. KAHN: We're postponing the matrix for the moment.

DR. DETMER: While you're kind of thinking of that, let me add another thought. Another thing that we might want to consider doing is start a little Website at our Website of some interesting URLs for folks, because we've knocking them around this table today. I think that could be very helpful too. It wouldn't be a clearinghouse. We wouldn't begin to be comprehensive, and we would say that with a disclaimer.

We're not advocating that people go to these sites, but these are some sites that relate to the things that we have talked about, unless we're prescribed from doing that.

MS. GREENBERG: No. You could even, I suppose, have hot links to them.

DR. DETMER: Well, I think that could be useful in facilitating communications.

MS. GREENBERG: The Health Statistic Work Group -- I don't know how many documents and Websites, et cetera have been mentioned in the large number of calls they have had. I suggested to Gerry Hendershot(?) that he work with Barbara Hettsler(?) in just putting together sort of a library. I was actually thinking of one hard copy of all of those. We could expand it to include this as well.

But then I think a lot of them are available from Websites. I thought that it would just be nice to have in one place even one hard copy of all of these things. But we could put on the --

DR. DETMER: What that will take, though, to do it as I'm talking about it would be somebody maintaining the thing, because they get old. They get shot. So the other question I have, one option would be to start it in the context of our current categories, but we might also then as our model changes, our white papers change, we might also be adding to it in the context of how some of these other things roll out.

MS. GREENBERG: Liza Boa has been our logistical contractor for about the last few years. They are no longer 8A. They grew up while we were working with them. So we are moving to a new contractor. One of the things that I know that Lynnette is negotiating with them, and hoping to put in is some Web maintenance, et cetera. Right now Judy Ball has pretty much been maintaining our Website.

DR. DETMER: I'm not speaking critically of our Website.

MS. GREENBERG: No, it's fabulous.

DR. DETMER: It's fabulous.

MS. GREENBERG: It might be something we could include under the contract.

DR. DEERING: Would you consider linking not only to documents, but to organizations or conferences? I only bring it out as a policy issue, because if you go to the Medical Records Institute, there are some classic conferences.

DR. DETMER: The conference is in February, and you don't get around looking at the thing until April or March.

MS. GREENBERG: I'm a little concerned that it would not be seen as an endorsement by the committee.

DR. DETMER: We have to think through that. Mostly I'm talking about how do you get to National Health Service.

MS. GREENBERG: Oh, right, that's no problem at all. But I'm thinking of conferences for which often are big bucks.

DR. DETMER: But relevant organizations.

DR. FRIEDMAN: Or proceedings. I'm thinking of the Rome conference. It has some terrific sessions, some fairly interesting articles on their Website.

MS. GREENBERG: This was from the European Health Statistics meeting.

DR. DEERING: Certainly one way to build that is to have current staff members nominate documents and proceedings and sites that they feel are appropriate. You need someone to do sort of the filter maybe.

DR. DETMER: Okay, what other things do you have? Anything?

MS. KAHN: I think we have covered the necessary things at the moment. My question is we haven't said anything about the G7.

DR. DETMER: Well, we said around the G7.

MR. BLAIR: There was an article that I got. I think it was Ray Rogers who wrote an article on behalf of the G7 on the global health care information society. My understanding is that the project that he was supposed to work on was not funded. The primary funding was apparently to come from the European Union, and they chose not to fund it.

DR. DETMER: But having raised that issue, Don Lindberg is the U.S. representative to that. Don has been keeping me a little appraised of what's happening with the smart cards over there, which is pretty substantial, among other things. It is probably worth finding out from Don what things are key that are happening on the G7 front. If there are any gee whizzes out of G7, I don't know. But he has been kind of amazed that the smart card has in a way fairly securely and confidentially identifying people in large number.

Clearly, the U.S. is hung up on that issue, without a doubt. So I think that certain has some merits and utility to this. Some of these issues I see as wild cards. For example, will we move toward evidence-based medicine and value-driven health care? I don't know.

Will we in fact pass some legislation that makes it very tough to do public health? It's possible. We worry about it. Or will we never pass legislation, so people don't do things because they are just afraid to. Both of the shadowy downsides of trying to get something done.

DR. DEERING: One of the things that we have down was to invite people from the other work groups, not only some of the other NCVHS works groups, but some of the other areas that we identified, to come and talk with us. One thing we could consider -- it can be very time consuming and it's more educational than productive perhaps, but very useful -- having sessions where we set aside 15-30 minutes on a scheduled basis to hear from some of these people about development in the areas that we have clearly identified.

DR. DETMER: Right. I think that's probably not a bad idea. February will be too tight, probably.

MS. GREENBERG: We may need to have a stand alone meeting, or the day before.

DR. DETMER: I think I feel really pretty good. I think we have laid out a pretty good work plan. We'll hear tomorrow, and still track a little bit of what we initially thought we were going to be doing at this.

Jeff, you won't be here tomorrow. Do you have other things you would like to see? We're going to meet tomorrow, not the full time. We're going to meet from 10 to 11:30 a.m. Do you have other things that you would like to talk about today, because I think it looks like we're about to round it up with what we can accomplish?

MR. BLAIR: No. I want to thank you all very much for allowing me to participate by conference call. While there was some time intermittently when I wasn't able to hear some of the committee members there, and I apologize for not being there personally where I could be more responsive.

DR. DETMER: Well, we were insulted not to see you in person, but it was okay. Actually, I think it worked pretty well. I think in a way it seems like voice is about 80 percent of what you get with voice and actually having people there in the telelink. I don't know how you felt about it.

DR. HARDING: It is pretty good. I think the longer the time, the harder it is. If it's an hour, it's pretty good. Jeff, I felt for you. I know six hours on that phone gets a little bit old, but you were right with us.

DR. DEERING: The only solution is for us to meet in Albuquerque.

DR. DETMER: All right, well, thanks a lot Jeff. We will see you in February though, right?

MR. BLAIR: That is absolutely right.

DR. DETMER: I'll see you later this month.

MR. BLAIR: That's right.

DR. DETMER: Thanks a lot.

Anything else? Thanks very much.

[Whereupon, the meeting was recessed at 3:50 p.m., to reconvene the following day, Thursday, January 7, 1999, at 10:00 a.m.]