Transcript of 2/3/99 Meeting of the Workgroup on National Health Information Infrastructure, NCVHS



National Committee on Vital and Health Statistics

Work Group on National Health Information Infrastructure

February 3, 1999

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

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


Work Group Members:

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


  • Mary Jo Deering, Ph.D.
  • J. Michael Fitzmaurice, Ph.D.
  • Rob Kolodner, M.D.
  • Sandy Haydock
  • Hetty Khan
  • Steve Steindel, Ph.D.


Introductions and Recap of January Meeting

  • Don Detmer

Discussion of Agenda and Glossary Development

  • Mary Jo Deering

Presentation of "VISA" Health Record Model

  • David Kendall

Presentations on Australian, Canadian, U.K. and Department of Defense HII Models

Discussion of Next Steps in Analysis of Country/DOD Models

P R O C E E D I N G S [3:50 p.m.]

Agenda Item: Work Group on National Health Infrastructure Working Session

DR. DETMER: So, if you go to Tab B, the next to third to the last page or the fourth page in, you will see the agenda for this part of our meeting.

Dan, can you hear me?

DR. FRIEDMAN: I can hear you, Don.

DR. DETMER: Good. I can hear you.

Do we need to speak a little louder so you can hear us?

DR. FRIEDMAN: That would be better. Thanks.

DR. DETMER: Where are you at the moment?

DR. FRIEDMAN: I am located at the moment in my house and I will be here for another hour or so when I will have to go pick up my kids.

DR. DETMER: All right. That is fine. I just didn't know what part of the world we might be talking to you from.

DR. FRIEDMAN: No, no. It is just Massachusetts.

DR. DETMER: All right. Well, it is good to hear your voice.

We have visitors from abroad with us today, who gave us some very nice presentations this morning. They will be with us this afternoon as well. We will get into -- back into that issue with some of our colleagues, as well as the presentation from David Kendall. Then we want to summarize a little bit essentially where we have been and what we are going to do next and then have some discussion and then we will adjourn.

That is roughly our agenda as I at least see it.

I would like to let people just, again, go around the room and introduce themselves for you, as much as anyone, Dan, as well as one another in the instance of some new players.

I am Don Detmer and I also chair this working group.


Are there new players that have come in since we have been here all day? It looks like there are. So, why don't we -- let's go around the room as well.

[Further introductions.]

Thank you very much.

Basically what we did at our January meeting, probably as much as anything else was draft the mission statement or work statement. I guess -- I don't know if really, Mary Jo, we need to go through more on that today. It didn't seem to generate too much discussion.

Agenda Item: Discussion of Agenda and Glossary Development

DR. DEERING: I think the draft work statement has been around now for several iterations and it is available on the Web site, I believe, if you would like to look at it and it will be acted on tomorrow.

DR. DETMER: I am going to actually turn the floor over to you at this point and have you add what needs to be added really to our agenda and let's just go ahead and get on into it, including glossary development.

DR. DEERING: Okay. Well, before I do that, I am going to take the liberty of embarrassing Don yet once again. For those of you who are new today, why you may not know that this is his last day as chair of the NCVHS.

DR. DETMER: Tomorrow.

DR. DEERING: His last meeting as chair of the NCVHS, but today is his birthday and I don't think that is a two day event, at least not usually, although it could have started at 11:59 p.m. and ended at 12:05 a.m., for all we know.

I would just like to say I have come to the NCVHS and to this workshop relatively late. It was in August and it has been a sheer delight to come to know Don and to work with him and it has been a pleasure to see the breadth of his vision that has worked out through this group. The staff and the members of the work group have their own little going away present for you.

First of all, it is sort of interesting that I packed it in what is really an old conference bag, but it was a joint conference that our office cosponsored, which was the cosponsorship of the Emerging Health Information Infrastructure and the Partnerships for Network to Consumer Health Information, both of which are represented in Don's vision very much.

First of all, just on behalf of ODPHP, we have a tee shirt with a great cartoon that says -- it is a doctor with a patient, it says, "Take two Web sites and e-mail me in the morning." Wear it in good health in the U.K. I am sure it will liven up Cambridge. And a signed version of the cartoon, signed by all the people here with good wishes and God speed.

DR. DETMER: Well, thank you very much. That is very nice.

DR. DEERING: And your bag. Your first HII bag, I hope.


I think what is interesting about what we are getting through today is that we are trying to look at visions. We are looking at the vision level of a health information infrastructure. What we heard this morning was very robust existing infrastructures and our chore this afternoon will be to get as far as we can in understanding what was our vision can learn from your already existing infrastructures.

Just as sort of background material or background help, I did post up on the wall the bare bones outline of elements from our vision statement just to remind us of what it is and the level at which we were looking at those issues. I stated the what of the concept paper was that it is a set of standards, technologies and applications and the why is both clinical care, public health and consumer and patient information.

The tasks are population-based data, consumer health records that include both personal, patient and population-based records. I can't read that very well myself. What is the third one down there? Knowledge, management and decision support and the fourth is telemedicine.

Under cross-cutting purposes, I know we have standards and -- somebody else read that, who is closer. Come on, Bill, help us with that.

DR. BRAITHWAITE: Research, education and development, universal access to health information resources, international collaboration.

DR. DEERING: Don had set for us a deliverable of about a year from now of a revised white paper and so we have to take what we had and learn from the people who are with us today to see how we can get there.

Would you like me to say a few words about the glossary in passing?

DR. DETMER: Let me break back in before you move to that, but I think it was very worth hitting those again because what I would like to see done at the end of a year's iteration is pick up on some continued dialogue with the nations that are here today, as well as, I hope, a few others.

The international collaboration piece, I think, each of you echoed this morning that you saw ways in which it is really almost dumb if we don't work together, let alone the more important plus side, it is a good thing to do, but I think that would be very useful, some conversation along that line so that that piece is hopefully that our Department of Health and Human Services will really buy in and put some resources behind.

I think beyond just a matter of sentiment, hopefully, getting some actual deliverables that we could see move from that and, obviously, the point that was made also, standards came up this morning, as well as, obviously, the models and so forth. So, I think that to the extent that we can track on these things, it is wonderful.

All of you, I am sure, are aware of the EMLS activity at the National Library of Medicine. I think that we are really motivated to see them say on that and in a collaborative way, as well, with other important libraries around the world, but that is a place where we could probably rest and vest a lot of these nomenclatures and a lot of these paper trails if you will, almost now days, sky trails, but I think that is an element of this as well.

So, it is not just a matter of the work plan but also then tracking it. So, to that end I think part of what we are discovering with the take two Web sites and call me in the morning -- a few weeks ago, I actually had a patient before I did surgery on say are you going to use the technique I saw on the Web last night? You are going to get this surgery according to Detmer. I don't know if it is the Web approach or not.

But the point is is that I think we are also entering the new era as it relates to how health care can be envisioned and delivered. Clearly, in my mind, that is going to have a much more active both personal role for people as individuals caring about their own health and trying to manage it, but also their communities as well.

So that this is the other thing that is sort of interesting. I think if we simply took the world as we found it, it is not the world we are inheriting. So, it is going to be important, I think, to sort of try to stay a couple of iterations, creatively, imaginatively ahead of this because otherwise I think, you know, we really won't reap the ultimate promise of health and empowerment that I think actually is good for health, as far as that goes.

So, anyway, along that line, Dan, as part of this project as well, we thought trying to put together a glossary of terms of that would relate to an enterprise like this -- you added a new word today for us this morning, "infostructure," and I think -- because that is going to be also the nature of some of this as well is finding terminology so we are using a common framework.

So, at any rate, you might gloss over the glossary a little bit and then we will move on.

DR. DEERING: Absolutely. We won't work on it at all.

The discussion had circled around whether or not we were going to create a glossary, how big a glossary we would want to do and to what extent we could lean on glossaries that were already out there. I did go ahead and send out an e-mail to about two dozen people in the broad areas of health infomatics and telemedicine and telehealth. And I got, you know, about six or seven samples, which are both Internet technology terms and telemedicine, telehealth infomatics terms, certainly nothing that is complete.

So, it is just to let you know that we will be looking certainly to what is already out there. The tentative approach is that the glossary is meant to stand with the ultimate revised white paper. It is probably not meant to supplant authoritative glossaries that are already out there and that we could, perhaps, plagiarize shamelessly where necessary to come up with something that was relevant for the universe as our paper cuts it.

I think I will leave it there.

DR. DETMER: Yes, I think the thing that if you haven't -- I don't know how many of you have visited our Web site, but I think it is really well prepared and well maintained and our feeling is that if, in fact, as are others -- I am not saying that we are unique in that, but it is one that I do think is quite nice. And the point is if we have some of these things there, they are at people's fingertips and could be useful.

Betsy Humphries just joined us from the National Library. I just said a nice couple of things about EMLS.

MS. HUMPHRIES: Thank you.

DR. DETMER: In case people did not know her, you ought to.

One presentation that we have on is a model that David Kendall is going to present to us for 10, 15 minutes. He is, as he introduced himself, senior analyst for health policy at the Progressive Policy Institute, which is a think tank associated with the Democratic Leadership Council.

So, the floor is yours. We had at our last meeting talked about a few models. So, this is the last of that iteration.

So, David, welcome.

Agenda Item: Presentation of "VISA" Health Record Model

MR. KENDALL: Thanks. It is a pleasure to be here.

Another cartoon you could have used was the child being born with a bar code stamped on their forehead, the darker side of this whole enterprise.

I wanted to congratulate you all for trying to create a vision for this information infrastructure because I think that is probably the only thing that is going to be able to hold this together. We won't have a massive government program that will cause us to action. It will come into being because a lot of people believe in it and a lot of people are following the same kind of idea and working in the same path even if they are working in very different aspects of it.

So, I think creating sort of a collective vision for this effort is one way to stitch together the wide variety of efforts.

I wanted to talk just briefly this afternoon about a model that we have been commissioned to do, as a think tank. I don't have an operational side of what I do. We wholesale ideas to political leaders. What we have seen the need for in health policy is, given the highly fragmented industry, the lack of emphasis on patients as a center of the system, is a kind of a new kind of governing model for health care systems.

Before I get to that, I just want to align myself with the vision that you put on your Web site as a draft. We think that there has to be a concerted effort to improve the ability for people to exchange information about their health in order to improve health and health care. It is in that information exchange that we think we can have a lot of entrepreneurial and innovate activity around, you know, the use of information and technology.

We premise our thought on the common sense observation that when you leave a doctor's office, you don't leave your health behind, but you leave your health records. So, what is wrong with this picture. We have a problem where our health is always with us 24 hours a day, seven days a week, but our health records and all the things that go into that are just not.

So, we have asked the question, what are the conditions under which a more robust health information network could be funded. And as I am talking, I will pass out a few handouts I have that might be helpful. There are three. So, be sure you take one of each.

Obviously, it has to be a secure system. It has to be one that people feel confident that it won't go to the wrong people. The system has to have respect for privacy. If you get a notice from a company that is selling you something, after giving private information away, you are going to wonder, well, where is this information going to be used.

The information has to be clearly of benefit to the person. There is a great project sponsored by the New England Medical Center and the American Association of Health Plans, which is asking AIDS patients to put on line their personal health information about their treatment.

Obviously, that has a great deal of benefit to the individual because in order to create sort of working database of the results of AIDS treatments, the only way to effectively know what is working is to share that information. So, the AIDS patients in the community are very aware of this and they can see direct benefits for putting in this information, even though it is not going to directly benefit them. They understand that that information can be of help to them.

In fact, this is being done on a personably identifiable basis so that that person can be told how that will affect their lives afterwards. So, I have often said if I have any concerns about confidentiality and security, I would just give that system my proxy.

Another condition we think that is important is collaborations for broad use. Anything that is going to work has to be interoperable and it has to be in sort of an open architecture system and that is going to require a non- proprietary way of doing this. So, there has to be a lot of collaboration in the system.

Now, in our system, given the highly diverse streams of funding and so forth, it is hard to imagine how any single entity could come forward and be a trusted source to pull all these parts together. You know, I personally wouldn't trust any organization existing that would do that. The AMA, would we trust them, all due respect? I wouldn't trust the government.

I wouldn't trust the HMOs. I wouldn't trust -- you know, I can't think of any organization that anybody would trust to be able to pull entities -- at least in our country -- to pull these elements together. That is why we are looking -- we have looked for a new model, which may be not immediately obvious why this works, but the VISA Corporation is a unique institution in our country because it has gained the trust of all the banks that participate in offering VISA cards, but, yet, it is yet -- it is still a voluntary private effort. The VISA Corporation is actually a group of about 4,000 employees that basically operate a trillion dollar economy.

They do this by having all the member banks come together, agree on what the terms are for what constitutes a VISA card and then they go out and intensely compete against each other to offer VISA cards. It is an organization that uniquely blends cooperation and competition in a way that very few other organizations in our country do in a very operational way.

The Internet is another kind of example. It is a self-governing organization that is maybe a little more chaotic than we want for health care, but it is the same kind of model. The founder of VISA Corporation, De Hock(?), has been sort of ever since his retirement advocating for this kind of model for governance of various systems as a way to solve a lot of our society's problems. He is, in fact, organizing the world's religions to apply to religious violence across the world.

He has gotten the in the northwestern fisheries area, the environmentalists to sit down with the fishermen and all these stakeholders to come up with a way of governing that system better. So, we are fortunate that he is personally taking initiative in this area in health care because he feels that if there is a way to do this in health care and to prove how this could work, it would be very instructive.

So, for your glossary, he has coined a new term; CHAORD(?). CHAORD is an organization that combines chaos and order and blends them seamlessly so that you have the effective tension between chaos and order come together in a way that allows people's creative spirits really to come about. So, all this is abstract and theoretical, which is what I am most interested in. But, in fact, there is a very practical side of what I am here to talk to you about and that is there is a group of people that have already started to try to build a system like this for health care.

If you can see on the diagram -- that is the bubble diagram there -- this is sort of the lineage of what we have been able to accomplish so far and there are two sort of lead organizations now that are leading this charge. One is the National Health Foundation, which is an L.A.- based organization devoted to the issues that around the uninsured, led by Rita Moya(?), and the Center for Advancement for Health, which is a D.C.-based center that -- that is a group that is interested in psychosocial behaviors of health and how information sort of gets used by people.

These two organizations have come together to try to start a dialogue amongst a wide variety of organizations, ranging American Hospital Association to the American Association of Health Plans to the College of Emergency Physicians, just to name a few, to define a common vision, much like you have been doing here.

That is the other document that we have passed out here, this revised draft concept paper. It sort of gives you our sense of what our vision would be and what we are pursuing in trying to build a chaordic organization for health care.

Probably the most concrete idea we ever could get our hands around is the idea of a electronic safe deposit box, where you could your information like anything from your living will to your insurance card to your health record that would have, you know, public and private keys that could be used to give access to this information as you wish it to be given out.

And a VISA kind of organization could, in fact, help, you know, either foster the development of these electronic safe deposit boxes or actually create them themselves. We are not so much interested in the technology at this point because we see this issue fundamentally as a governance issue. There are all sorts of technology out there; Smart Cards, everything, you know. I don't even begin to know all of them myself.

What we need is an organization that can come together and agree on a rapid basis and a revolving basis, standards on operational questions that can be quickly adopted in the real world because they are benefiting people on a day-to-day basis. So, that is what we -- and, in fact, along these -- the most recent development of our outfit has been the sort of learning of a -- we have learned of a similar kind of effort at the -- that has been spearheaded by the Department of Veteran Affairs, Bureau of Indian Health Services and the Department of Defense.

I think it is fair to say that what they have learned in trying to build an electronic medical record for their folks is that it wouldn't make any sense for them to have a system that just works for them if it didn't work for everyone because their folks, you know, will move in and out of their own particular systems. So, they have understood and I think, quite correctly that we need to make this a broader societal kind of effort.

So, we have actually had one initial meeting to try to see how we can overlap since we are both very new -- early in our processes and we will anticipate there will be a high degree of convergence there.

The final thing I passed out, it is just a more academic description of what we are imagining that needs to be done here. So, essentially that is it. We are at the beginning of a long process. It is going to be a complicated effort to try to bring everybody to the table.

Probably the biggest thing we struggle with right now is what is the burning platform? What is going to cause people to really, you know, move into this area. Just to throw out a couple suggestions that, you know, I am sure you struggle with the same issue. What is it really going to take to make this thing happen?

You know, one of the crucial things I feel the need for is this system to address the perceived lack of control of their health information. As we move into the privacy debate on Capitol Hill this year, we are going to need to have a better answer than simply the government is going to pass a bunch of rules to protect you.

I mean, unless we operationalize the protection of privacy and the confidential use of health information, all the laws in the world are going to be -- that are necessary to this won't be sufficient. So, I think that is one burning platform.

Another one is just the information -- if we don't have the ability to individually on a personalized basis sort the amount of information that either an individual consumes for their own health purposes or a physician consumes on our behalf as they treat us, we won't be able to deal with the huge amounts of information coming over the transom from the medical research.

So, the ability to have a personal filter with your individual health characteristics identifying what information is relevant to you, I think, is another clear aspect of this.

The third burning platform would be sort of the fragmentation of our health care system. That is a classic perennial problem in the United States, but in given areas

-- and we were just talking about the Indian Health Service -- I mean, where you have a system where all the providers are very loosely affiliated, but still nonetheless affiliated, the need for virtual integration is very strong there. I think that is the area of a non-vertically integrated insurance market, where we used to think that, well, we are just going to have these big HMOs that are going to run everything and solve all our problems. Well, that is not happening.

We have a very disaggregated system and that will continue. So, the need for some kind of virtual integration is critical. So, those are the three things that we are pinning our hopes on now ourselves.


DR. DETMER: Rob, do you want to add anything? I know that you have been thinking about this as well.

DR. KOLODNER: I am Rob Kolodner from the Department of Veterans Affairs.

As David said, we did meet about two, three weeks ago and we really do have two parallel efforts. There is a slightly different focus on each and, so, we see a lot of synergy and basically are involving a number of the same people in both efforts. But I see them as converging. The most important thing is that the focus is to create an organization that everybody can trust. Don't get hung up on trying to get consensus right now because we can't. We don't trust one another. And the technologies also are not the place for the focus yet.

But once an organization is created that people can trust, that will create an environment where we can bring together lots of opportunities. The right technologies can flow and we actually take the barriers out of the way of the people who have the knowledge or the technologies, but there will be a framework around it that will be one we can trust, that we know that the information will be private when it needs to be private and we know that it will be available for the needs of society when that is appropriate.

But trying to come up with those right now in the absence of an organization when we have tried and failed going after the solutions first, that is the essence of this. I think the most important thing about it is that the whole process, just like VISA for banking is open a bank that wants to join and is willing to abide by the rules. They can come in. They are not stuck in. They can get out with 90 days notice.

That is the same thing in health care. It has got to be open to a broad spectrum of health-related entities, including the consumer. Hopefully, we will be able through a number of these efforts to get them started and realize that -- a challenge would be an understatement, but that it appears that the time is right. There is enough consensus that something needs to change, that going along and doing what we are doing now isn't going to solve the problems and we need to approach it in a different manner.

So, we look forward to being able to have enough to share with you and invite everybody in.

DR. DETMER: Questions or comments?

MR. REILLY: I have a question or observation.

It took VISA a number of years to achieve a level of competence in the marketplace. Is there any understanding at this point as to, number one, is it going to take that amount of time for you to achieve that in the health care domain or is there something that can galvanize the health care industry to move to that level of trust quicker?

MR. KENDALL: I think one of the things that is different from when VISA was first formed is the technology. I mean, there are companies out there that almost can do everything that we are talking about right now and the problem is is that those systems don't all work with each other. So, I think the answer is made easier because of the technology, that it is there. I mean, it is never an obstacle.

The unknown is how much force can we garner behind the need to cooperate and that is a -- I think that is just an unanswerable question at this point. While amongst these sort of people that know, you know, this issue -- I am just one of the newcomers here, by the way -- there is a consciousness, but it doesn't extend much beyond, you know, the CIO at an HMO, for instance.

We are plowing new territory and there are probably lots of things we could do to test -- you know, beta test the stuff and, in fact, if you look at like health risk appraisals on line right now, this is a good beta test for the stuff. And there are a lot of people out there putting in a lot of personal health information right on the commuter today without any kind of protection.

So, people are hungry for it.

DR. KOLODNER: Let me take a different crack at it. I agree that the technology is there and the technology has been there, but in terms of working it together so that we are not locked into one proprietary solution or another and so things are really shareable, that is not there yet.

I think it is -- forming the organization is a process. De Hock talks about the fact that it is not a journey because a journey you know exactly where you are going. It is more of an odyssey. We are proceeding and there will be a lot of things that we will discover along the way. And, hopefully, there are lessons that were learned along the way with VISA and its creation and with some of the other efforts that are going on that will allow us to move along quickly and that some of the things we don't have to rediscover or stumble along.

On the other hand, identifying a common purpose and the principles take some time and making sure that it is done in a way that people don't feel that they are left out, but that, in fact, that they can join and be a part of it and really influence what is going on. It will take some time. So, it is not going to happen over night, but over the next few years, I think it very well could be -- the VISA experience was an interesting one because it was four vice presidents of mediocre banks formed it with no major retailer behind it when they first launched, but it was the right idea in a format where people could, in fact, trust it and eventually chose to join in.

So, that is the part we need to get right. The others, the technologies and things will fall into place when the environment is right.

DR. DETMER: Jeff, this is the last question or comment and then we are going to have to move forward on our agenda.

MR. BLAIR: The idea of the VISA model I find very, very exciting. I immediately start to think of how that would be implemented and within the context of the vision that we have in this particular work group, it seems to fit in with the personal record or the consumer record very, very neatly. When I think of an acute care record, I think of it having attributes of tremendous amounts of information gathered, you know, at a local, acute care institution.

I wonder if the VISA model fits very well with that and then the other piece is -- the other model I think was kind of to support research, which was, I guess, clinical and public health record, population-based record, I guess, was the name of it, which also has some different attributes.

I would think that this -- at least I get the impression that this would play one key role within those three types of records, which would be the personal record. Do you see it the same way or do you see that it is applicable beyond that area of the personal record?

MR. KENDALL: I think it moves even well beyond any actual records. I think this could be the basis for, you know, a better transaction, you know, for claims processing, for the other set of putting a living will into place, which is not about any kind of health gathering information. It is about distributing personal health information.

The ability for the -- and the need for cooperation in the health care system is so huge that the limits really are boundless. But, obviously, you have got to start somewhere and we have to start with our high needs and we have to start in an area where there is not a lot of infrastructure already in place because otherwise you will be up against existing systems.

But I think the way you worded it is probably right. I think in the clinical side of things, you know, we have a lot of proprietary data systems that would benefit a lot by having a common way of linking the information. I think the discussion I heard on the Web this morning about the way in which various countries have linked their systems, that -- the ability for our personal health record would be -- the links between the various proprietary systems can be very effective in this area, as an example.

DR. KOLODNER: I think that the most obvious one is the individual record, the electronic health record. I don't know whether it will spread beyond that or how successful it will be there. I think that is what the organization will, in fact, guide and decide and ferret out and debate.

One of the interesting things that comes to mind, though, is if an institution can feed an individual's data vault, Web page, whatever it turns out to be, with that information, then it means they can also provide an aggregate of that data without the identifiers that suddenly acts like a report card and it also turns out that that can then provide some semblance of some public health information, although if the individual is getting care at several places, it may be -- there may be some tricky aspects.

But I think that is where the individual's repository or their data vault or whatever may then -- one of the obligations to society may be that you provide a certain zone of information that doesn't have identifiers, that will allow you to do population things. So, I don't know where it is going to end or how it is going to mix or whether it is going to grow in that direction.

I think that will emerge from the organization and from how the people will interact and how much trust there is, but we will see.

MR. KENDALL: If I could just add, the Health Query, which is mentioned in one of the articles, uses a health risk appraisal system to create a public database for comparison of personal health information with -- that is individual identifiers, but still on a zip code basis, with claims processing information and other arbitrary systems that have been voluntarily donated to the public domain. So, there is a beginning, in L.A. at least, of sort of a rich, morbidity database that we never had in this country.

DR. DETMER: I want to thank you, David. Obviously, I think there are a lot of attractive features to this and it will be interesting to see how it rolls out.

We are particularly fortunate to have colleagues that are also from countries that have been working on things. I don't want to lose the time that we have got to have them here.

So, what I want to do is move us on. I would like to spend at least 45 minutes of our time having a chance to have some of our colleagues assess what we think we have heard and seen out of your models. So, the point is because you are going to go away, we will be in touch, but what we would like to do is have an iterative discussion here seeing to what extent there is convergence and such on some of these things and then, hopefully, after that we can see where it sort of comes together.

So, let me turn the floor over to you, Mary Jo, at this point.

Agenda Item: Presentations on Australian, Canadian, U.K. and Department of Defense HII Models

DR. DEERING: We are moving into our compare and contrast mode or point and counterpoint, perhaps, and, in fact, we have Sandy Haydock from HCFA, who has looked at Australia and Hetty Khan from Canada, who has looked at the Canadian information system and Steve Steindel, who has immersed himself into U.K.'s information strategy.

I think we will go alphabetically by country this time. What we thought we would do is have -- our staff really, except in one instance, took a more broad brush look at your material and your approaches so that they could guide us through sort of the components of your approach and the basic principles of it and to try and tease that out for us.

Then what we would hope is that you would help us refine that for us. Are we on target? I don't know whether we have time to actually move to the charts over there and start taking notes, but we stand ready if we can get that far.

So, I will start with Sandy and Australia.

MS. HAYDOCK: Great. Thank you.

Just to recap and to bring you all up to date, those of you who are visiting from out of town, we as a work group met about a month ago and discussed how we would handle our meeting this month and at that time we were beginning to talk about your three purchase to health care information and agreed that it would be good to discuss that now. Of course, at the time, we didn't realize that we would have the pleasure of your company as well.

So, this adds a whole new dimension to the discussion this afternoon.

What I would like to do with the Australian model is to revisit some of the things that Dr. Reece talked about this morning, some of the things that I thought were appropriate or relevant to the work that we have been doing with our concept paper, maybe point us in some directions that we may or may not have been considering. Give us some maybe thought-provoking ideas, some things to talk about.

I am going to talk not only about the model, which is the topic of the discussion itself, but also the National Health Information -- the development plan, which provides the -- what are the priorities for the information work and also the information agreement, which, as Dr. Reece, said this morning gets those commitments, those necessary commitments.

I guess we are ready to start.

As we looked at the Health Information Plan -- and Laura, please tell me if I am not representing this exactly the way you would -- what I saw in it, again, was that sort of priority setting approach. And I looked at what was it trying to accomplish. Well, one thing it was trying to do was to make sure we got that focus on possibly underrepresented groups. Dr. Reece mentioned this morning the aboriginal and Toreagar(?) Islander peoples, and that is something, of course, that we are interested in as well.

So, again, I am going to try to bring up things that I think are appropriate to what we are also thinking about, the linkage or the relationship of the health record. In fact, Dave mentioned this morning two types of health records, sort of the patient look, the patient type record and the personal record. Of course, we are looking at three sort of views.

We are also looking at the population record, which is an aggregate of that information. A goal of approaching information this way, of course, would be to increase the cost effectiveness of data collection. Now, we have talked in terms of increasing the effectiveness of health care delivery and the cost effectiveness of health care delivery.

Something we might want to look at is looking at the cost effectiveness of the health care collection and dissemination and interpretation as well. Another purpose for the plan, of course, was to ensure that we get the outcomes and effectiveness information, improving the measures and also implementation process for doing the surveys for population data without, you know, a good approach to information, unified approach to information, you lose a lot of the texture and quality of the surveys.

The other thing we looked at and we discussed this morning was the National Health Information Agreement and, as Dr. Reece said this morning, that is the commitment by not only the commonwealth but the states and the territories to participate in this plan, if you will, to agree then on the definitions of standards, the rules, the guidelines to collect and access, interpret and publish data.

DR. DETMER: I would just ask Laura, my understanding this morning was that between actually between essentially what is, I gather, your equivalent of the Department of Health and Human Services or not? Is it the agency in Canberra that has to do with health and welfare?

DR. REECE: That is the federal equivalent. Then there are the state and territories.

DR. DETMER: When you work with the states and territories, is that with their health agencies or with the actual government of those territories per se directly?

DR. REECE: Wherever health is included in their state government.


MS. HAYDOCK: So, really with the goal -- with the objective here of ultimately providing access to better and more timely information for the health professionals, for the community groups, for the government and non-government organizations. And I notice that you didn't mention individuals in that. Was there a reason? Is that not the goal ultimately or is that just not -- in the agreement discussion. I notice we were looking specifically at community groups, but is that the same as individuals or does that translate then to information for individuals?

DR. REECE: The emphasis is on national health data in what we are talking about with the Health Information Agreement. We want to promote access to that and that is what -- part of what the institute and the ABS do are publish aggregated data and interpretations of that. That is published for anyone to look at.

I am not sure if you are talking about individual access to that data or you are talking about information about individuals.

MS. HAYDOCK: No. Individual access to the data.

DR. REECE: Yes. That is rolled into it.

MS. HAYDOCK: Okay. Great.

One thing that I thought was particularly key in looking at the benefits of an agreement such as this is that agreement on the appropriateness of the data collected. That has come up a couple of times in today's discussion. We heard it this afternoon. Dick Anderson mentioned it. It is sort of the flip side, I think, of making information available, making -- we have the technology now. We have the ability to do it. Can we restrain our ability by saying we will collect only what is appropriate? We will manipulate data only as it is appropriate. We have had a lot of discussions about that today.

Moving on -- I do want to try to get this pretty quickly -- what sort of generated the discussion a month ago was actually talking about the information model. So, I am going to spend the rest of this couple minutes talking about the information model itself.

I happen to have a copy -- did you get copies of this made --

DR. REECE: They were supposed to go around to all of this group. But as on my soap box this morning, there is a copy of the model and a copy of the entity definitions that go along.

MS. HAYDOCK: Okay. Great. This might be a good time to just make that available.

DR. REECE: I apologize for the quality of the copy of the model itself. I had to rely on an older one at the last minute and it is not as good a quality as it should be.

MS. HAYDOCK: Yes. I don't want to get too in depth about the model itself or get down to data definitions and things like that, but just to try to talk about some of the major qualities of this model and, again, some of the things that might be pertinent to us.

Obviously, as we said this morning, it is very high level. It is generalized so that it is very flexible. It can be used by any groups. One point, I think, that Dr. Reece made this morning was that making a model, making the representation of information as generic and as high level as possible, allows for buy-in and awareness by many, many different groups, who can say, yes, I can see myself in that model. You don't get so specific.

Another aspect of it, of course, is that it is person centric and I think that is rather intuitive to us and we have been looking at information from the person point of view, too. We talk about information entities and entity groupings and if you look at the model itself, you see the data -- the information types, if you will, as entities. Then these entities are grouped into larger subtypes.

I find processes just as valuable as outcomes. So, I look also to how did Australia put this model together. This sort of top down information modeling, a very formal structured approach, which is easily learned, but requires training, gets everybody on the same page, gets everybody understanding it, probably did help, just the process was very valuable to the groups who were involved, the many people who were involved in putting this together.

Another aspect of this is the fact, of course, that it provides the framework for lower level data modeling, information modeling, information system development and it provides a conceptual basis for their National Data Dictionary, again, at a high enough level that all the complexities of individual systems can be fit within your model.

I have a new toy. Go ahead and put them all up. I think there are four. I just learned to use --


MR. BLAIR: We are referring to this as the Australian National -- I think actually the official name was what, Knowledge Information Model?

DR. REECE: National Health Information Model.

Actually, does it divert from the ISO meta-data model or is it actually an instantiation of that?

DR. REECE: It is an instantiation of it. The two are sort of going forward together.

MR. BLAIR: So, does that give us some guidance in terms of a process? I mean, is that what you use for the process and framework for how you created the Australian information model?

DR. REECE: We did ours based on the ISO 11179 meta-data standard and now the meta-model standard is sort of going forward based on our implementation of the earlier standard, if you get what I mean. We have been feeding back and forth off each other as this has been going forward.

MS. HAYDOCK: Back to just a few of the types of value of having an information model, just a few remarks on that. Obviously, it helps to focus the thinking. At HCFA, we are putting together an information model as well or looking at information just within HCFA and we find it is exceptionally difficult to focus discussion among a broad group of people unless you have some kind of structure in which to start parsing out groups of information.

So, it provides that focus. It gives the community you are trying to deal with information, a common representation, a common language, so they can communicate with one another. Flexibility of use, again, we are getting back to the high level, the conceptual nature of this model being flexible enough so that any group can see itself within that model.

Now, the next few slides I bring up to talk about the types of information that I found really interesting in Australia's model. The process as described was to take -- to do actually workshops with many groups around the country. In those workshops, the participants were thinking about what kinds of things do we have to get information about. What kinds of information do we need to gather? So, again, a way of framing that discussion was to come up with a microcosm or a subset of information in which a large group could convene and say within the concept of well-being and health, we will think about what kinds of information do we need.

Some of the items that were raised -- and, again, I raise these because I think they might provoke some thoughtful discussion -- were the perception, not just well- being and illness -- and, by the way, I do have this in handouts so you don't need to take too many notes -- it is not just well-being and illness, but the perception of well- being and illness.

It is the relationship between outcomes and well- being, the relationships between events and outcomes. You can see how that works if you just look at some different blocks on the model here.

External influences, and I think of things like maybe environmental influences or a leak at a chemical plant could be an external influence. The impact of knowledge in education and research on our health knowledge and our health information and ensuring equity and access and measuring equity and access to health care.

Another of those subsets of information or microcosms is community health, thinking in terms of barriers to health service, thinking in terms of the characteristics of different groups, of privacy and confidentiality, of planning and outcomes and goals and quality, again, a way to structure the way we are thinking about information.

Within the context of mental health, how about looking at target groups or ranges of service types? I am not going to be able to answer specifically what some of these are -- of well-being and status, of the legal status of the mental health aspect.

Finally, well, another one is the institutional care, provider roles and organization, program purchasing, funding, account processes and resources; again, getting to some things going beyond the clinical, going beyond the state of well-being itself. And, finally, access, equity and awareness, the effective knowledge in research again, you can see that come back on service practices.

And then the perceptions of the consumers on changes in the service practices; influences of policy and planning and priority setting coming from the government -- of course, I am always certainly aware of that. And then outcomes in best practices.

So, again, I think there are many things in the Australian model that we can take to the next level and consider as part of our model, but on a real high level -- and, again, getting back to the approach, there were some key things that I thought came out of Australia's experience and one was -- and a couple of these we talked about -- one was clearly getting the participation of stakeholders throughout the industry, throughout the community.

Another is maybe thinking about a broader definition of what health care information is, not just the clinical information, the patient record. Maybe it has to do with the external influences or perceptions; again, taking that model, that disciplined approach to information modeling; the awareness of other activities internationally.

And I think everybody has talked about that throughout the day. Finally, when all of this information is gathered, how do you represent it and promote it in a very clear, intuitive, understandable way so that it will be used by the desired -- you know, by your target audience.

In case, we hadn't gotten your slide, I did put up a forum for more information Web site there so that we do have access to some more information about your model.

I thank you very much for coming today. This helps a lot.

DR. REECE: You are very welcome.

DR. DETMER: That is great. Why don't we then move on -- let's see, we are going alphabetically. Let's move to Canada. Hetty.

DR. DEERING: Do we want to have a counterpoint?

DR. DETMER: I would like to, but I will tell you, I am looking at the clock and I have got to be vicious because at the end of this, I want to make sure I hear from all three countries and I want to make sure we can pull together something out of it. So, I would love to, but --

MS. KHAN: Okay. In the interest of time, I am going to try to be short also.

I am sending around two and those are back and front papers and what my approach was to look at all kinds of paper and the headings that we have over there and try to make a comparison of the Canadian model.

Jennifer, I appreciate your being here and this is very much of just an overview, not a comprehensive look at all. We appreciate your input this morning.

So, basically, to be short, I am not going to go down all the different things that is on the paper. You can read. I will just highlight a few.

The national agencies that I notice in the Canadian model there is a lot of collaboration among basically the three agencies that have come together as an advisory board per se to form this infrastructure. In our model, we have this committee as an advisory committee. We have got the HHS Data Council and different agencies within the Department.

However, I see a strong focus on national mandates within the Canadian system and, of course, our government system is quite different with more of private sector initiatives. I looked the purpose of the Canadian model and information, there is a sentence there that says "Information is power." And I thought that was very powerful, a very powerful statement.

Their focus is a lot on the information that needs to be generated and the focus is quite much on the health information and then what kind of a system do we need to provide that information.

Regarding the benefits of the health system, there is lots of overlap. There are a lot of similarities. The individual was quite well stressed in terms of individual access to the health record, individual empowerment, being able to basically control that information and your health care by that information.

Several of the other things were monitoring health, accountability, improve access to aboriginal communities and there are several tie-ins from our model that pretty much just looks at the same thing and the risk there, of course, was the privacy, confidentiality and security concerns.

You have mentioned this morning about the comprehensive health record and the population surveys and the data that should be extracted from those and we are looking at those also.

What I gathered from this, there is the telemedicine and the -- I am not going to go through these just because of the interest of time, but your model is very much oriented to the user, to being sure that it is user friendly, that the people who are not really experts per se, can handle that model and can gain access to their information.

I think that is one thing that needs to be looked at and to be kept in mind.

The other things we can -- anyone can look at. Now, on the other side of that paper, there are several recommendations that I highlighted of the call for the national mandates to several specific things that would like to be seen. Those are development of common definitions and core held indicators and priority development of standards. You mentioned HL7, which we pretty much involved in here in the U.S. also; development of benchmarks for policy management and service delivery; identification of the determinants of health, information inputs, data elements, collection methodologies, standardized formats for data collection, analysis, dissemination and audit, development of best practices as operating standards; development and dissemination of standard templates for indicators of health.

And it was exciting to me to hear Dr. Reece also mention that and not just for a national infrastructure but for an international collaboration.

Development of national principles, standards and protocols to address the privacy and confidentiality and several specific calls for provision of catalog of reports, data, products and services; elimination of dual or parallel tracks of access or dissemination, increasing visibility and availability of services to the consumers.

And of interest to me was the systems for training, data collection, access, analysis and interpretation to the providers of health care. And I know within our nation, one of the problems we face is being able to have the health care providers really buy into computerized health information infrastructure.

Provision of decision tools, support and training, provision of support and advisory service, leadership in goal setting, promotion of greater use of electronic challenge, which will the Internet and the electronic mail. Reduction in duplication, improvement in coordination and between the national agencies and provincial jurisdiction.

I have gathered that there are several jurisdictions with pretty well-defined health information infrastructure already, which needs to be linked and become interoperable for access. Collaboration with stakeholders in decision-making and consultation with users and front line professionals are needs.

The other handout that I provided was pretty much -- and I think this is what you were referring to this morning with the model where there are the people, environment and the health system. It is on one side.

MS. ZELMER: You need the electronic version. The balls spin.

MS. KHAN: Oh, okay. You pretty much have identified the domains, the person, the environment and the health system and it was interesting to me to go back and look at all the different components of the health data that is necessary for patient care, for improved patient health and improvement of the health of the whole population. That data, having identified the needs of the information, then you get to the point where you told me what kind of an infrastructure and what kind of technologies you need to be able to utilize and to collect that data.

DR. DETMER: Is that the nugget of the strategic kind of approach to it?

MS. ZELMER: Yes. This is basically the overarching health information framework and then it drives down.

MS. KHAN: That is pretty much --

DR. DETMER: What we will do is we will then hear also the U.K. and then I do want to see what you feel are the most salient things that you would maybe see us want to do in the context of trying to move some of this forward here, but also collaboratively.

Yes, Steve.

MR. STEINDEL: I have taken a slightly different approach. As we heard this morning, we heard a great deal about the U.K.'s efforts in modeling and the U.K. also published a year ago, as was mentioned, a booklet on the information for health that is approximately a hundred and some odd pages long, which is more of a policy document, describing how the international health information infrastructure is going to be put into place. The plan now ends about 2005.

What I am going to do is just walk us through the document on the information for health. This is what they

-- a very pretty picture, much prettier than I can draw on the airplane using power point -- was their strategy, their overview on how they view the National Health Information Infrastructure in the U.K.

The first thing they have is a national strategy, but they don't really define that. They define it as a local implementation. So, we are talking about a national effort that is implemented locally.

There are three key components to their local implementation. The first is treatment and care and this is where a lot of the modeling efforts come in. We heard mentioned today the employee health record and then the patient record, the individual for the instance of care and then what carries through at the individual for their life. And these are two key components of their strategy.

Then they have social care records, which also dovetail into that at the local level. And we heard bout the intersection in many of the international models today between the welfare side and the health care side, the social side and the health care side.

This information can be used for analysis purposes, public health analysis, clinical governance, health improvement programs, performance management. These are what they list. We could probably think of others. These provide knowledge for the public, for patients, for health care professionals and for managers.

And notice all of these boxes feed into each other. The information from one side will help develop the information for the other side. Now, what I would like to is now walk you through each one of the chapters and point out the key points from those chapters.

The first chapter is an overview chapter, called an information strategy for the modern national health system and they list a series of objectives. These are my paraphrasing of their objectives; a lifelong EHR for every person. Notice, that is a key component; development of the electronic health record for every person in the U.K.

Then they want round the clock access to patient records by all NHS clinicians so that people will have access to patient care information at any time, any place, seamless care by sharing of information on a national health service information highway. So, part of their strategy is developing a technical infrastructure to make this information available to all the health care providers and it is not going to be just health care providers because they have fast and convenient public access to the information on the national health information infrastructure.

So, we are involving public access and we will see part of this as the development of an electronic library for access by the public. The effective use of these resources by the policy makers, the planners, the managers, both at the national level and at the local, institutional level. So, this information will be used for policy planning. And that is their overview of the system.

The next chapter deals with supporting patient care and here they talk about at length their concept of the electronic patient record and the electronic health record. We heard these mentioned this morning in Dr. Jones' talk. The first one deals with the lifetime information on a particular individual.

The second one deals with the information on that particular incident of care. So, really electronic patient record will be built up from electronic health record. And we have heard this mentioned many, many times, the aggregation of patient information for a lifetime based on encounter or incidence data.

Rob Kolodner mentioned that even just a few minutes ago as a key component. The electronic patient record -- now, this is the lifetime record -- is defined at six levels of information. First is clinical and administrative data and I think we can tell what that is. Then integrated clinical support information, clinical activity support, what has been ordered on the patient, what has been done, knowledge and decision support, specialty specific support. We are talking about when they are seeing mental health providers, for instance, or something like that and then the concept of bringing in telemedicine to make this information available on a more broadly disseminated basis.

Supporting integrated care through NHS-wide standards and infrastructure -- this is a very key component of their plan. Development of clinical information standards, standards that are used throughout the U.K. A very, very key component of their system was privacy, confidentiality and security issues. The information must be maintained in a way that the public is satisfied that it is private, confidential and secure. But they also want to do this in a way that there is improved access to this information.

Don't forget, one of the key components is 24 hour access to this by a care deliverer. So, we have to have improved access, but also maintain privacy, confidentiality and security. A very key component is there must be a hardware infrastructure in place in the U.K. to support this information infrastructure and this includes both the transmission hardware and also the hardware in place at the health care providers so that they can have access to this information.

They also talk about the development of a national electronic library for health. This is a data depository for information concerning health care and aggregate information on the health of the U.K.

Information for improving health and for effective management is the next chapter and here -- and we heard this mentioned this morning as to where it is in the document itself. It happens to be mentioned in 5.17.

This is a key component that they recognized very, very early, the improvement of data quality. If we are going to get all this data in, we want to make sure that it is data that is of equivalent quality and that it is information that can be used for decision-making and that some of what can be used for is to improve the health of the population, to support clinical governance agreement, how health care will be delivered in the U.K., to be used for local health improvement programs. Where we know there is a deficiency in health care delivery in a certain area, we can locate it, improve it, focus on it; a national framework for assessing performance and national service framework for the delivery of health care.

We have heard our equivalence of this being stressed throughout the day and through various parts of the hearing in different words, but very, very similar concepts.

The next chapter was meeting the public and patient needs and this was a relatively short chapter, but I thought it was a very, very important chapter because it was key. What they want to do is reach out to the public and make sure that part of the key component in this is making sure that the information is available to the public.

They want to develop ways to make sure the system helps the public through access through information and a very key component was telemedicine and developing the telehealth structure, which they use the word "telemedicine" -- we would probably use the word "telehealth" as an equivalent here -- the telehealth structure within the U.K. And, finally, again, another mention to the national health electronic library and I think that you would probably be very much in favor of that.

Making it all happen. They talk about -- and I thought this was very important, a very important part of the document, was they stressed there has to be a mechanism coming into place to make this happen. We can't just say all these good things and not talk about how are we going to put it into play.

They talk about developing a national health information partnership, which defines the stakeholders that are involved in this -- I didn't list them. The document does -- and a formation of a national health information authority that brings the stakeholders together to discuss the way they are going to implement the infrastructure. A very, very important part I saw and they did stress it -- I picked up on it as a government worker -- was they understood that you need to redefine the procurement process if you are going to make the national health information infrastructure work.

Having spent the last three or four months trying to acquire some modeling software through the government procurement process, I can assure you this is a key part of what we will have to do to implement this sort of thing.

Finally, they redefine local roles or they define local roles because, remember, we talked about this as a national strategy with local implementation so that the local roles and the local stakeholders are very, very important for making the system happen.

Finally, they give a time line. I am not going to spell out the time line. They did list it. They did have very specific goals for the near term, which was 1998 to the year 2000, a little bit less specific goals, with goals very well spelled out for 2000 to 2002 and then, finally, a set of goals so that it can lead to a finish in 2005. And they give some goals for what they are defining to be -- what is going to be done there. And I think we heard a little bit about what those goals were by Dr. Jones this morning.

That is my summary of what their plan -- and you can see this is a more policy approach document than what we were talking about with the modeling approaches from Australia and Canada and we did hear about the modeling in the U.K. So, it is not without that background.

Thank you.

DR. DETMER: I want to thank all three of our -- Sandy and Hetty and Steve. I think those were excellent presentations.

Now, why don't we open this up. I would like to first let our guests comment and then maybe let's see from the group what we want to do as our next steps on this.

But what would you, I guess, having heard what we have thought through a little bit, seen a little bit of this, what advice, I guess, would you give us on next steps because I have a feeling we will be iteratively relating to you on these things. But where would you have us start?

DR. JONES: I always have views. Can I first say that I think that the three of you did a wonderful job. It is very difficult with representatives from the countries. Thank you very much. You did it very well, indeed.

I would like to pick up on the policy and the vision and the future view. And I think we had the same vision of this what we call the HR, you have this VISA-type organization. The U.K. Government decided that the primary care specialist, the general practitioners, were the custodians and they had sufficient -- they were given sufficient trust by the people to hold that knowledge.

If you don't have it, you do need to strive for another mechanism and perhaps the VISA approach is the right one.

Another thing that, in fact, hasn't been mentioned at all today and that is another little nugget within the information strategy and it is something that has turned the NHS direct and it allows -- it is a mechanism where anyone, the public, can telephone in -- I think you have the same sort of mechanism within the states -- and consult a nurse or a specialist on a problem. That is fine. But let's not stop there. Let's take it further.

Why can't they book appointments for practitioners? Why can't they book people into clinics? If that is the case, why do you need clinic administrators? Your mind can start then rolling and to look to a future, which is defined by a very quick, sharp contact with the patient, a discussion determining the severity of the problem and then dissemination of that and quick response to the patient's requirements.

Maybe that is sort of where I -- you need to take that very lateral view. That is just an observation.

MS. ZELMER: Well, I went first this morning, so I should probably go last this afternoon.


DR. REECE: I am just pondering the difference because at least it seems to me it might not seem the same to other people. There is quite a bit of difference in at least the information that we have presented here today. So, I want to be real careful about overstating or understating a position that Australia has a policy about a particular effective health care or not. I didn't really present that aspect of things.

What I am looking at immediately in the short term to give you some feedback, because that is what you have asked for, is some easy win sorts of things. I think one thing that has emerged is improving the quality of data and data standards is probably a fairly easy target to at least identify and begin to approach because there is a lot of activity going on in that area around the world and that is something that you can start the ball rolling on in some regards without jeopardizing any overall policy, I think, or at least you can start the dialogue on that.

The other thing is something that has come up in all of these is the involvement of the stakeholders and although that happens at a very high level in a committee like this, it is extremely important to get down to that level or a much lower level when you are really trying to sort out some of this.

So, the consultative process that you go through is extremely important. I mentioned to Sandy that we actually did a great deal of documentation of the process we went through and I encourage you to do the same thing because you will learn a great deal from it. In fact, we published papers, which I will send to Sandy for dissemination on what we learned during that consultative process, how far to take it, you know, how far is too far, you know, when you end up coming full circle again and saying the same thing.

Those things ended up being published in more data management circles rather than in health. So, they are probably not things that you have come across. So, there are a lot of extra -- one thing that I have learned is that there is a lot of extra information that is out there that I haven't been aware of because I have either learned -- come into encounters with the Canadian health model or the U.K. model through Health IT or I have come in through Data Management or I have come in through Health Provision.

The picture that you get is very different from all those aspects. So, I would encourage you to take a broad look. I think that has already been said, but make sure that you are looking at an aspect that covers all those areas or get views from all those areas.

I guess I want to agree with things that I think both Steve and Dr. Jones have said about the importance of the process, getting the stakeholders involved in the process but once you get a tool in place the process has to continue. It is not enough just to come up with your framework and come up with this plan that you are going to carry out. You have to carry it out and then you have to set up ways to review it, evaluate it and change it and close the loop and feed it back.

If we have learned nothing else over the decade of formal health information management that we have done in Australia, it is that you have to close that loop, feed it back and keep it on rolling so you get continual improvement because it is not going to be right the first time around and probably not the second time.

It will never actually be totally correct, but you just keep striving.

DR. DETMER: Thank you.


MS. ZELMER: I guess one of the things that struck me throughout today and through the feeding I have done before is that there really are a lot of similarities between the stages that we are at, some of the approaches that we are taking. I mean, we all did broad consultations of one nature or another. We all came up with an overarching strategy that was meant to be complementary to the millions of things going on in our various countries. We are all focusing on standards as a basic building block. We are all looking to some slightly more formal models as a way of getting our heads in sync about what we are doing.

We are all seeing privacy and confidentiality as an important issue, something that we need to move on. We are all looking at electronic health records, the dissemination of health information to consumers, ways that health information can support the development of health policy and the management of our health system.

I mean, we are all doing all those things. I guess in terms of specific, concrete suggestions there has been a G7, now G8 subgroup on enabling mechanisms that -- Dr. Fitzmaurice, were you involved in that?


MS. ZELMER: -- that provided a number of recommendations in terms of things that could happen at the G7 level in G7 countries. That might be a place to look for some ideas. They did a fairly wide consultation process to come up with those recommendations and certainly I think it would be useful if all of us kept in touch. If nothing else, we can crib from each other's notes as we move to the next stage in the process.

DR. FITZMAURICE: Let me just say that within the U.K., Ray Rogers has led the effort on the G8, enabling mechanisms and took a worldwide tour to -- well, first, he did a survey of about 350 worldwide experts. I am sure he missed another 350, all of the people who were in this room who weren't contacted.

Then he put it together, circulated it around the e-mail and took a physical worldwide tour to stop in to gathering places. In the United States, one of the gathering places was the American Medical Infomatics Association meeting in November down in Orlando. He made a presentation and took back then comments from that.

He has crafted and I think it was just last week put out the final version of the document to be altered and is sending that up to the European Commission. He has, I believe, been funded by the European Commission to do this.

DR. DETMER: Is this document available at this point?

DR. FITZMAURICE: I have got to pull it down. I have only got the e-mail from Ray and from Joy Reardon. I can pull it down and make it available.

DR. DETMER: Yes, that would be terrific.

DR. FITZMAURICE: Lots of pages.

DR. DETMER: Well, that is fine.

DR. FITZMAURICE: And it may be worthwhile if Ray is in the country in the future to bring him --

DR. DETMER: Oh, absolutely.

DR. FITZMAURICE: If it coincides with NCVHS to bring him in and talk about it. He is good.

DR. DEERING: When you were talking about bringing him into the country, Betsy may know this, this year's Friends of the NLM Health Information Infrastructure Conference is linking with something from the G7. And I didn't read their conference prospectus all that well, but just as they linked with us a couple of years ago, it looked to me like maybe it might have been in that domain.

So, I only hold that out as a possible opportunity when some people might be in the country. That might be an anchor. I think it is about the 27th and 28th of April is their meeting dates. So, we might look into that.

DR. DETMER: Obviously, we have had a long day and I think all of us are going to need to let some of this filter in, but let's --

DR. JONES: Just one comment. Ray Rogers used to run the Information Management Group within the U.K. Health Service. So, he knows -- he understands the sort of models that we build and he has used them for policy making.

DR. DETMER: Let's hear from folks around the table on the Committee in terms of what you might have us put on the table because as I say I think we are going to have to let this sort of soak into the subsoil, but let's at least go round robin before we adjourn.

DR. FITZMAURICE: I guess I pick up on what Jennifer said and then she echoed the others is that there is a -- the governments are pulling things together. I see a lot of government glue, a lot of government funding involved in this and when I look at our country, it is not essentially government-driven. It is private sector driven and in different parts of the government, they are trying to pull it together to say here is a national information infrastructure. It is an interesting contrast. Is it better directed from on top or undirected from -- on the bottom? I guess that may be the American way of doing things.

But it seems to be working for us. We have a lot of tools available, but it needs some national leadership and some national focus to pull those tools together to sort of weed out some of the unnecessary duplication, to learn lessons from what we have and then to link up with other countries.

Your countries have a focal point and as you look at the United States, you see many different focal points. You see Don. You see Bob Mays. You see a bunch of people running around saying this works for me and the me's are starting to get a larger and larger umbrella. I think there is an opportunity in the next two to four to six years to get a good information umbrella in the United States so we know what is going on and then to link up with your -- your umbrella seems to already exist.

MS. ZELMER: Can I just maybe quickly -- Canada created an umbrella because we had multiple focal points. It was a few years ago, but that is exactly why we did it.

DR. JONES: I think the U.K. did as well, in many respects and you have to go through that pain and it is painful. Congratulations.

DR. FITZMAURICE: Will you share your pain with us?

DR. DETMER: Others?


DR. LUMPKIN: Well, I certainly appreciate the input and will assure you that I know things will work out for the best here. Sir Winston Churchill was once noted to say that you can always count on Americans to do the right thing. Of course, that is usually after trying everything else but I know we will make that correction.

One of the areas of interest and in looking at the diagram and maybe it is kind of based upon my -- the job I do, which I am director of public health in Illinois, which is one of our 50 states and that is to try to put together the population-based events or population-based things.

In trying to look at these systems in a way that there are some things that we do in public health that create non-events and how to look at the interface between a person-based scheme, which I saw in Australia, which talks about person characteristics, the roles and the events and some of the domains in Canada. And I think there are some very interesting approaches that I think we can look at of how to inter-marry some of the events that may not have anything to do with people but may have to do with the global warming and what the hang out is for the Asian tiger mosquito and whether it is an area or not in an area.

Some of the monitoring we do in arbovirus(?) is in birds that we catch and drag the fields for ticks that may or may not carry lyme disease. All of these are kinds of things, I think, that I am just looking at and then I think we will have to delve down a little bit more into your structures to see how they can be integrated into systems, which have a lot of energy and focus on clinical decision- making and all of these factors impact them.

DR. JONES: Can I just come back on that one?

I think the model that the Australians have is an instantiation of, in fact, the U.K. model, which part of it is a higher level yet and because it is at a higher level, it will accommodate populations alongside patients and there is some work being done at the moment at Teas(?) University in Middlesborough in England, which is looking at a patient's record moving that around into a population base and passing it through a geographical interface, just to look at the geography and whether the people live on the wrong side of the river or one thing and another. Perhaps that is the sort of thing that you are looking for in terms of public health.

MS. FRAWLEY: I wanted to thank our three visitors. I kind of came this afternoon for a drop in because I was absolutely fascinated with your presentations this morning.

Just so that you know, the association I work for, we belong to the International Federation of Health Record Organizations and I have worked with Laura's colleagues and Jennifer's colleagues. I have not had the pleasure yet to come to England. So, once Don gets there -- we have been struggling with the issue of who owns the data in this country, besides all the privacy and confidentiality and security ones. So, it was somewhat heartening to hear from your presentation that I am not alone in this morass, that there other colleagues out there struggling with this because we are really -- to me, it is probably one of the biggest barriers and challenges we face right now.

So, it was helpful to hear your experiences.

DR. DETMER: Well, you know, in fact, that kind of is an interesting issue. I think you came out with a nice short list of common challenges that we are all working on. We might as well continue to share or swap experiences on those, but it also looks like we have some common problems.

The other thought might be to try to actually organize some work that actually could take on some future problems, which I think --

DR. JONES: In fact, a lot of the problems that we have identified or talked about today, some work has been done on them. Within the U.K., there is what was called the Caldicott(?) Committee sat on security and confidentiality and it, in fact, didn't feel confident to look at ownership. Talked about a lot of other things, but didn't actually broach that problem, mainly because perhaps they don't see it as a problem in the U.K. I don't know.

But, you know, let us share our problems as well as our -- the things that we have done.

DR. DETMER: Clem, you are probably our number one messenger RNA circling the globe. You have been uncharacteristically quiet through this. What is going on in your mind?

You can tell Clem is a friend as well.

DR. MC DONALD: You are tackling a very complex domain and there are some threads that actually connect --- HL7 is deep in Canada. I think it is actually pretty active in Australia. There is an Australian -- in New Zealand, not so much in Great Britain. There are some ISO activities now that have started up, which may also be relevant, but these things when you get to -- you know, the technical standards are always good with creating enabling things. I actually get very afraid when they get into prescribing things because that is when people try to beat you up and they are happy with enabling things.

So, a lot of what we are wrestling -- some of what we are wrestling with is prescribing a regulatory -- which are much harder. I would rather stay neutral on those.

DR. DETMER: Richard, you have probably raised the word "trust" more than anyone around our table and that came up, obviously, on some of these things. What are your thoughts on that?

DR. HARDING: It did and I was very interested. Yesterday, we had a group of pharmacists here or the pharmaceutical industry actually and they were talking about the fact that they had concerns or -- pardon me -- that clients had concerns about linkage. You mentioned linkage several times or you don't mention linkage, I think, was the word that you used or you don't use that word.

But you said that you got people's permission to look further into their demography or whatever it is that you were doing. I can't remember, that you -- health survey and that you had almost a unanimous approval of that or permission. The pharmaceutical industry said they were lucky to get 20 percent in this country whenever they tried to do something along those lines.

Do you think that is just that Canadians are good -- or that there is an issue of no fear of discrimination because you are stuck in the system, the national system, and you can't get out and nobody is going to get you out or why do you think you have such good opting in to a program like that?

MS. ZELMER: I think one reason why that -- the survey was actually conducted by Statistics Canada and Statistics Canada is known as a bit of a Fort Knox. They have a very well-established reputation in terms of privacy and confidentiality. So, there was a high level of trust in terms of the people who were collecting the data and the rules that they would follow in terms of linking the information.

There was also a case made in terms of this is the kind of thing, these are the purposes for which we will use this information. These are the kind of things that we will do. These are the kind of conditions under which we will do things.

The interesting thing is to me not just that we had a fairly high level of consent, but people actually had to do something. They had to give us their personal health number as part of this process. So, it wasn't just I am not paying attention to this. I am ticking "yes" on the form and signing off so you will get out of my house. You actually had to do something to participate. And it was a very level of consent.

DR. MC DONALD: They live in their house. I mean, right away you have got --


MS. GREENBERG: Wait a minute. They probably are a lot nicer in Canada, but -- we have a similar situation, but what was the sample size of this survey? Was it huge?

MS. ZELMER: It was fairly large.

MS. GREENBERG: Because the National Health Interview Survey, which is an NCHS survey, but it is conducted by the Census Bureau, which has -- which is pretty much trusted here, too. People will tell you, you know, everything from how many bathrooms they have to other things, but -- and then we ask for their Medicare or their social security number so we can link with Medicare files and we get a very high -- very, very high -- and we have strong privacy, confidentiality legislation specifically for NCHS in the Census Bureau that is pretty -- can't be penetrated.

That is different, though, than -- I think that is a much more controlled situation than universal linkage or, you know, getting out beyond into the health care environment.

DR. DETMER: It isn't as though we could never pull that off.


DR. FITZMAURICE: While we are speaking of surveys, the Agency for Health Care Policy and Research goes out with a large national medical expenditure survey. It is called the Medical Expenditure Panel Survey, 10,000 households, about 24 to 30,000 people. We ask for their income tax forms. We ask permission to contact their providers, their physicians, their hospitals so we can follow up on their memory about what services did they receive and how much did it cost. And we get about 88 percent participation rate. That is not a hundred percent, but it is pretty good for that kind of information.

DR. LUMPKIN: I must interject that vulnerable populations are probably overrepresented in that 12 percent who refuse, just as they are in the census.

DR. FITZMAURICE: We have oversampled some of the vulnerable populations to try to take that into account.


DR. MC DONALD: Some of the conversation about like equating this with the VISA problem, I think it is like equating getting to the moon with the social problems in the U.S. I don't think they are in the same universe.

We heard this morning the challenges of getting data from a provider to a payer, who wants to pay them. We are not going to give that stuff because they are going to screw me with it. I mean, there is just so much motivational force on both sides of these things that make it so complicated and not to mention the fact that I am not telling anybody anything because, you know, I live in a small part of Montana and I don't talk to anybody, not to pick on Montanans, but just it is a tough business and it is not that no one has tried. This is not day one and you can name at least five multi-decade corporations, multi-decade money, I mean, billions, AmeriTech is one of them, who has gotten into the health business one way or the other and backed out.

There was a service in about -- I think Searle sold it for keeping your own records on a computer in case you got sick anywhere, 15 years ago and it kind of flopped. I think the guy that is doing Netscape is struggling with his MediCom company. So, it is just a warning that this is -- you know, because I can give you my money and now it is your money. You don't care whose money it is. I give you my sodium level and it doesn't mean anything to you. There are some substantial differences. You can't use my sodium.

DR. DETMER: Dave and, Bill, I want to give you a chance to say something. I think everyone else around the table -- maybe, Simon, you haven't, but I would like to give you a shot because we are going to have to wrap this up fairly soon.

MR. REILLY: It is very informative. I have a question and it is a little bit more technically oriented. Dave on the end down there said that their models were using OMT methodology. On the Australian side, what have you used? Have you used UML or any particular methodology?

DR. REECE: It is currently an NNT(?) relationship type model with our changes -- and I have had some discussions with individuals and mentioned a couple of things that we did differently, which we have documented. But we are looking at translating -- the implications of translating that to UML because that seems to be the emerging standard.

MS. ZELMER: Frankly, there is a modeling tool that we use and I have not a clue what it is, but I can find out for you.

MR. REILLY: The reason why I asked that is because one of the tasks that I have is to take these diverse models and bring them together and do a harmonization on this for DOD.

DR. JONES: In your lifetime?

DR. MC DONALD: I think if you are using a rim(?), the HL7 rim, it is UML. But people actually print it out with special graphics tools because they never -- the spaghetti stuff you get out of UML is even harder to read than the usual --

DR. JONES: Can I just say that we have done a comparison of our model with the HL7, which might be useful.

MS. ZELMER: We have matched the HL7 and the Australian model.

DR. BRAITHWAITE: I just wanted to make sure that Mary Jo includes in our glossary the term that came out of this discussion that I think describes the issues very well and that is "chaortic."

DR. DETMER: Simon? Simon says -- Marjorie, have you said anything, everything?

MS. GREENBERG: I have said my piece.

DR. DETMER: Mary Jo, it looks to me like -- I will give you the last word, but it looks to me like we won't get through our matrix today, but at least what I at least would hope -- we will touch base -- it seems to me like really the fine presentations that we have gotten from our colleagues and helping staff is -- could also try to get together and see what you pull out of this. But that seems to be our next step.

But what else is on your mind as next steps?

Agenda Item: Discussion of Next Steps in Analysis of Country/DOD Models

DR. DEERING: I was going to say I had visions of people's overheads overlaying on top of each other, so we had Australian on top of U.K., to see what slipped outside the edges and then we dropped that off and then we sort of took the commonalities. And I do think, though, that there can be a process that helps us very easily -- not easily, but will work for us to find what those commonalities are.

Just in terms of specific activities, we are -- we will have the obligation of coming up, I think, with a new outline for our paper, based on this effort. I think that is something that we should aim for probably by the June meeting of the full committee.

Other things that we need to have between now and then are we get to take up again our long postponed engagement with the Office of the Assistant Secretary for Management and Budget, who as you all know just delivered how many tons of paper and figures to Congress only two days ago. That is why they are all so exhausted or out at meetings that they are not here with us today, but it is very good in our sense that we hadn't blocked off time for them since we could use it with you.

But our Assistant Secretary for Management and Budget has been interested in this area and they are promising and, in fact, they are eager to work with us on an inventory approach that they have had. In fact, it seems to me that we really need to meet either virtually or in telephone presence between now and then. I will be circularizing all of you to see how we do it and when we do it.

For a variety of reasons, I was sort of hoping for May, but probably before then. So, we will be in touch. I am not sure that we really can get back to the draft matrix yet. We postponed it; we tabled it before because it seemed to be a cart before the horse, a schematic. We have some components on it, thanks to Dave, but I think that that can wait.

We do need, if I understand process right, however, we need to make a list of what it is we do plan to deliver to the Committee as a list of work products and formalize that because I don't think we have ever really laid it out in whatever bureaucratize you may require.

MS. GREENBERG: Well, the ISO standards --

DR. DEERING: So, to me -- and I wanted to mention one other thing and, Betsy, I did finally respond to your e- mail this noon.

MS. HUMPHRIES: [Comment off microphone.]

DR. DEERING: Someone is coming from the National Electronic Library for Health. They have contacted Betsy Humphries, who has contacted me, and my e-mail back to her was that because our work group is interested in their broader information strategy, we would -- some of us would like to be brought in on that. Betsy very kindly has already invited me and I have only ten copies of Betsy's e- mail and your attachment.

So, if you are interested more in the consumer, public information dissemination side of their information vision paper, here is something that talks about the visit that is upcoming in a nutshell. It is February 22nd to 24th. And you are going to work to schedule it to wherever they want to go, I guess?

MS. HUMPHRIES: [Comment off microphone.]

MS. GREENBERG: Do you want to come up or -- I know we are about to end but we are not picking up any of that on the --

MS. HUMPHRIES: They started a little higher toward the top. They contacted Dr. Lenburg, saying that they would like to come over and meet with the Library and they had had some early discussions with Jim Austel(?) of our NCBI group related to some of our information systems.

So, there are seven people coming from the NHS. Muir(?) Gray is leading this group. He is, I guess, the one that is particularly responsible for the National Electronic Library of Health. We have some people who are also more on that -- well, Ann Harding is coming and Dr. Little and Bob Gann and I am blanking on some of the rest of the names.

But at any rate they are interested in a variety of things, including live telemedicine demonstrations in various places and records and also a variety -- talking to a variety of people about a variety of things. So, I will

-- and we are attempting to fit some approximation of all of this into three days and we will do our best.

Fortunately, Bob called me today and said we just want you to understand that we don't all have to go to everything together. This immediately makes it possible to maybe fit in they are interested in.

DR. DETMER: You are talking about one nation in three days. We took on three in half.

MS. HUMPHRIES: I won't make any comment over which of these makes the most sense.

DR. LUMPKIN: I just would like to emphasize the importance of working on this model for the work in the future of the Committee. I think, by and large, and what I have understood today is that the National Committee has really worked on a lot of very important issues, but they have been issues that sort of come to us based upon our own individual experience, what we are interested in, HIPAA, those kind of things.

Once we can all agree upon a model, we can then prioritize the work of the Committee so that we are actually working towards this vision. So, we are going to need to have a very iterative process with the other subcommittees because once we start putting the model together, we need their input because when it is done, a lot of things of new projects and new initiatives that we are going to be talking about will have to be weighed against this vision that we are trying to work towards.

DR. DETMER: I really think that is a wonderful lead to my last hurrah here on this topic.

It was gratifying, in fact, that at the meeting on the computer-based record yesterday, I think it was Kathleen or Jeff who raised the question, how does this fit with what your group is doing. So, I think there is already some sense of that and the interesting thing and really wonderful thing is that I would say that John Eisenberg and Peggy Hamburg response at the Data Council when we first presented our thinking on this in the concept paper was extraordinarily positive.

I mean, the entire day was set aside to really talk strategic vision and start talking about that. So, there is also, I think, a receptor site to the Department on what we do.

The other thing I will have to also say is from what I have seen, unless this committee and the Data Council do it, I am not convinced it will get done. It is not as though of -- we are a many splendored thing and I think it is splendorous as well, but it is very splintered. And I think that the fact is if this group does stay at this, I think it could, in fact, be one of the greatest things actually that this committee could give its country, you know, and possibly, collaboratively, be partners to doing it really for the world, which is really -- you know, that is about as grand as it gets. But that is a really nice thought.

So, on that, let's go eat. No. Let's adjourn and I do really express tremendous appreciation. The idea that three experts would fall out of the sky simultaneously to share this day with us has just been really splendid.

DR. JONES: Not only that, we haven't broken a leg.

On behalf of us, thank you very much, indeed, for inviting us. I thoroughly enjoyed and I am sure Laura and Jennifer have, too.

MS. GREENBERG: We really appreciate your coming.

DR. DEERING: Thank you very much.


[Whereupon, at 5:45 p.m., the work group meeting was concluded.]