[This Transcript is Unedited]
NATIONAL CENTER FOR VITAL AND HEALTH STATISTICS
50TH ANNIVERSARY OF THE
National Committee on Vital and Health Statistics
June 20, 2000
National Academy of Sciences
2101 Constitution Avenue, N.W.
Lecture Room
Washington, D.C.
Proceedings by:
CASET Associates, Ltd.
10201 Lee Highway
Fairfax, Virginia 22030
List of Participants:
- John E. Rolph, Ph.D, Chair
- John R. Lumpkin, M.D.
- The Honorable Donna E. Shalala
- Don E. Detmer, M.D.
- Judith Miller Jones
- Ronald G. Blankenbaker, M.D.
- John M. Eisenberg, M.D.
- Edward Shortliffe, M.D.
- Daniel J. Friedman, Ph.D
- Dorothy P. Rice, Sc.D
- Gary Catlin
TABLE OF CONTENTS
- Opening Welcome
- Reflections and Challenges: John R. Lumpkin
- Remarks
- Don E. Detmer
- Judith Miller Jones
- Ronald G. Blankenbaker
- John M. Eisenberg
- Remarks by The Honorable Donna E. Shalala
- Toward a National Health Information Infrastructure
- Ed Sondik
- John Lumpkin
- Edward Shortliffe
- Shaping a Vision for 21st Century Health Statistics
- Daniel J. Friedman
- Dorothy P. Rice
- Gary Catlin
- Acknowledgements and Closing Remarks: John Lumpkin
PROCEEDINGS (1:05 p.m.)
DR. ROLPH: Good afternoon. On behalf of the Committee on National Statistics, welcome to the National Academy of Sciences on this occasion, the 50th anniversary of the National Committee on Vital and Health Statistics.
I’m John Rolph, and I am chair of the Committee on National Statistics. As you can see from your agenda, we have an interesting afternoon planned. One slight deviation from the agenda is that when Secretary Shalala arrives, she will give her remarks, and they may be slightly out of order, since the exact minute is not known.
This historic building that we are meeting in is a physical representation of a very rich slice of both American history and the history of American science. President Abraham Lincoln signed the Academy into being on March 3, 1863. As mandated in its act of incorporation, the Academy has since 1863, it serves to quote, investigate, examine, experiment and report on any subject of science or art, end quote, whenever asked to do so by any department of the federal government.
The first two questions that the federal government asked the Academy to address are of some historic interest. The Academy was chartered during the Civil War. Ironclad ships were just then being tested. It turns out, the government testers found an unexpected side effected of ironcladding, namely, that it made the ship’s compass inaccurate.
In a war between the North and the South, you definitely want a working compass. So the Academy launched a study of this issue, and they solved it. If you visit any historical sailing ship these days, you can still see the two cannon balls mounted up on a bracket in the compass pinnacle. Indeed, by rotating these iron masses, it is possible to compensate for the ironcladding’s effect on the compass.
The second question the federal government asked the Academy to 9850 Reseda Blvd. APT 316 was whether or not — remember, this is now almost 150 years ago — whether or not the United States should adopt the metric system. This remains a case where the Academy’s positive recommendation is taking a rather long time to implement. However, like many Academy reports, it remains relevant today.
Just to give you a little history, to keep pace with the growing science and technology, the Academy eventually expanded to include the National Research Council in 1916, the National Academy of Education in 1964, the Institute of Medicine in 1970, and these organizations are now collectively known as the National Academies. These private nonprofit organizations provide a public service by working outside the framework of government to insure that the federal government gets independent advise on matters of science, technology and medicine.
The National Academies enlist committees of the nation’s top scientists, engineers and other experts, all of whom serve on a voluntary basis, to study specific public policy issues. The results of their deliberations have inspired some of America’s most significant and lasting efforts to improve the health, education and welfare of the American population.
The Committee on National Statistics, commonly known as CNSTAT, was established by the National Research Council in 1972, at the recommendations of the President’s Commission on Federal Statistics. CNSTAT’s goals has been to contribute to better informed decisions about important national issues, usually by working to improve the statistical methods and information upon which public policy decisions are based.
Our primary mode of operation is to select important public policy issues for in depth study and analysis. Sometimes the committee itself carries out the study, but more often we select and convene an independent panel of experts for this purpose. Topics for panel studies or workshops are either identified by the committee itself, or commonly by federal agencies or by the Congress.
The Committee on National Statistics works closely with the federal statistical system, and to recognize the value of its contributions, many statistical agencies provide us regular core support through funding supplements through the National Science Foundation. The committee held its 81st meeting in this room just last month.
The Committee on National Statistics actually has a long history of working in the area of health statistics. Indeed, we count a former director of the National Center for Health Statistics as a current member, namely, Ed Perrin.
In your symposium materials, you have a list of our health related publications, and you can see there is a remarkable breadth of complexity of the health statistics and measurement issues that the Committee on National Statistics has studied in the past. Indeed, our versatility and flexibility in mounting studies and workshops and our ability to pull in the nation’s experts makes the committee a unique and highly effective national source at the interface between policy and statistics.
Our most recent health related activity was when we were asked by the National Center for Health Statistics to convene a workshop on the future of health statistics. It was called Towards a Health Statistics System for the 21st Century. The funding for the workshop was provided by ASPE, and in designing the workshop the committee sought the help of the National Committee on Vital and Health Statistics, as well as the Institute of Medicine. The workshop took place last November, November 4th and 5th of 1999, was well received by both the sponsors and the participants.
The keynote presentation at the workshop was given by Dorothy Bryce, former director of the National Center for Health Statistics, and who is a speaker here today, and Dorothy’s paper will appear in its entirety in the workshop report.
A general theme that emerged from the workshop was that the U.S. health statistic system, while reasonably comprehensive in content, cannot fully meet the nation’s current and anticipated health information needs. There were a number of suggestions made at the workshop, which I will leave for you to read
when the report comes out.
The report itself is now going into the National Academy’s report review process and is scheduled for release in late August. This workshop report intended for both users and producers of health statistics, will discuss the implications of emerging health needs for the health statistics system, and suggest a research agenda to respond to needs across the health statistics system.
Later this afternoon, you will hear a report from the National Committee on Vital and Health Statistics, their work group on health statistics for the 21st century, and much of what you will hear this afternoon foreshadows parts of the CNSTAT workshop report. We will be happy to send a copy of this report to each of you when it is released in a couple of months.
I’d like to close with the assurance that as in the past, the Committee on National Statistics, in cooperation with the Institute of Medicine where appropriate, is always ready to assist the health and the health statistics community with independent objective advice from the nation’s top scientists and experts on scientific issues that frequently pervade policy decisions.
So on behalf of Committee on National Statistics, I’d like to offer my sincere congratulations to the National Committee on Vital and Health Statistics on this occasion marking your 50th anniversary of service to the federal health and vital statistics communities.
I would now like to introduce Dr. John Lumpkin. Dr. Lumpkin is the director of the Illinois Department of Public Health. He has served on numerous national advisory committees, and is the current chair of the National Committee on Vital and Health Statistics, and he has some remarks and introductions of past chairs. John.
DR. LUMPKIN: Thank you, and good afternoon. The partnership between the National Committee and the National Academy of Sciences Committee on Statistics has certainly been a very important partnership, and one which we hope will continue to grow. The work group on health statistics for the 21st century is just an example of that partnership over the years.
We are here to celebrate 50 years as the National Committee on Vital and Health Statistics. Let’s go back 50 years. What was the world like? Rudolph the Red-Nosed Reindeer by Gene Autry was the number one song, followed closely by Rolling Stone by Muddy Waters. In that year, Truman ordered the development of the hydrogen bomb, and Senator Joe McCarthy had launched his crusade. The Korean War was just on the horizon. Already, the transistor was two years old.
Although people of the time did not realize it, the birth of this committee marked a major epoch in the history of the world. Not only the committee, but certain other events were going on. Historians also note this as a major turning point.
Fifty years ago, we saw the end of what has been known as the Colonial Era, and we have launched into the modern era. Also, the culmination of a process that began in the 1700’s. In the 1700’s, technology had enabled the beginnings of mass production. Energy was the transforming force in the Industrial Revolution that was to begin, and ended just as this committee was coming into birth.
Since then, many people have described that period since the formation of this committee in 1949 as the post-industrial era, but I think it is a different kind of revolution. Just as energy was the transforming force in technology that allowed mass production that launched the Industrial Revolution that changed the face of the world, information and knowledge today has become the transforming forces that affect the world, and information and knowledge management are the technology that is allowing that transformation to occur. It was into this world, this time of change, that the National Committee on Vital and Health Statistics was born.
It is my pleasure as we open up the initial session, to begin some of our reflections on what we have done, where we have been and where we are going as a committee. I have the honor of introducing Dr. Don Detmer, who preceded me as chair, from whom I have learned many of my skills. I learned how to pun from Don.
Don was named recently the Dennis Gillis Professor of Health Management at the Judge Institute of Management, Cambridge University. That is the Cambridge across the ocean — in 1999, which is why he left the chairmanship. At the business school, he directs the Cambridge University Health Management and Policy Research Center. He is professor emeritus and professor of medical education at the University of Virginia, and Professor Detmer has been a member of the Institute of Medicine, the National Academy of Science and chairs its Board on Health Care Services. He was a board member of the Association of Academic Health Centers and the American Medical Informatics Association. Dr. Detmer chaired the committee from 1996 to 1999. Don?
DR. DETMER: Thank you, John for that introduction, and good afternoon. It is wonderful to look across the room and see so many really friendly faces over so many years. I hope I have a chance to talk with a lot of you in more detail today.
This is a wonderful occasion at a great place. Marjorie Greenberg probably will jump out of her seat before the afternoon is over, she is so excited about this. She and Jim Scanlon and I talked about it for a long time, and here we are. So I want to congratulate the current committee and the staff and the department for all of the recent progress that has been taking place.
If I could have the next slide, the NCVHS is a very unique Washington institution. In addition to having an extremely important task, it also has survived 50 years without a pronounceable acronym. Nockaveehus is the closest I can come, but I don’t expect it to take on any time soon.
The period from 1950 to ’95 started obviously with a WHO request, and then for all of those intervening years, the sound advice was offered to the government as it related to health and vital statistics.
I think things took a significant shift for the committee and the nation with the HIPAA legislation, that enacted a nationwide health data agenda for the committee, but also to set standards for the entire country. That was certainly new. At the same time, Secretary Shalala reorganized the committee and also created the Data Council to help integrate these issues.
The health information infrastructure project began in 1997, and by 1999, the concept paper, the work group and the vision for statistics projects were launched, that we will be talking about today.
At the moment, I’d like to talk about what is going to happen in the next 50 years, looking forward. In 2030, I would anticipate that the committee will approve health data standards for computers as smart as the committee members, but not the staff yet, and by 2050 I would expect that they will have improved health data standards for humans, proposed by computers that are smarter than the entire committee and staff in aggregate. But that is looking forward.
Actually, they are gaining on us. This is the touring test results that were reported at least in the UK recently, where people communicated by telephone with a computer, and decided whether — asking any question on any topic they wished, and decided whether they were talking with a human or a machine. In ’95, 17 percent of the time people got it right, ’97, 33 percent. The computers are gaining on us, so I think that is part of the message. John talked about that.
I think the question however is, it is probably better to invent part of the future rather than inheriting all of it, and I think that is the reason that we are here, to talk about a couple of fairly solid interim reports that are here today.
I think that if Clem McDonald were here today, he might consider the titles of them perhaps a little bit Soviet inspired. They don’t quite roll off the tongue actually, toward a national health infrastructure and shaping of vision for 21st century health statistics. This isn’t the way you talk across the fence to a neighbor about what we are doing this afternoon.
I think the question I would put to you today is perhaps, we need to have the committee and the Department think at least about a way of linking these together into a national information for health strategy. I think that is what Dr. Lumpkin was saying. We are in the information knowledge management age. Information for health I think is a term that people can clearly grasp and understand; who doesn’t want information for health? It obviously would have these two components of the health information infrastructure, and obviously this new vision for health statistics. So you may want to consider that.
If that is the case, I can imagine the headlines in the paper tomorrow, the Post, NCVHS 50th Anniversary, Information for Health Will Bring the Right Information To All the Right Places at the Right Time. The article will then say that information for health is the name for the Department’s health information infrastructure in these revitalized statistics.
Another headline I can see coming from this afternoon is, 50th Anniversary Program Opts for Better Living Through Really Vital Statistics. The one that we have been waiting years for, Secretary Shalala Reports Great Health Privacy Law Signed by President.
So what really is information for health, at least as it comes through these documents to me? Personal health records for individuals to attend to their own health, that are linked with patient care team records within the health care delivery system, when folks need help, linked with community health records, with identifiers removed, so that we all enjoy healthy, safe communities and environments.
What are the unifying attributes to care in such an information for health model? These are listed alphabetically. Obviously, the first one, patient centered, is the most important one, and it is in the middle, but that care is accessible, that it is evidence based to the extent feasible, that it focuses on health as well as disease care, that it is safe, that it is secure, has secure data, but also transparent data where there is a need to know that, and that it is timely care.
How will this happen? I think it will happen when trusting, interactive people work together as individuals, patients and communities, but there are other pieces to the strategy that speak directly to this organization, not only the NCVHS and the Department, but the National Academy as well. There has been an extremely important dialogue, as was mentioned, between the two organizations and the Department. But the vocabulary definition code issue still remains an important challenge, although I think clearly, the committee for many years has made real progress in helping the country deal with that.
Increasingly though, globalization is the issue here. Both my concern and point is, I feel that unless the U.S., Australia, Canada, Japan, some of the other G7 countries in other parts of the world, unless we work together to make sure we do have global standards, I worry about whether the rest of the world will have cheap enough and available enough standards to be able to do this for everybody on the planet.
Of course, you need systems monitoring with solid statistical analysis, and that is of course again the piece that Ed Sondig and the group have been struggling with and working on.
All this looks great. The question as far as I’m concerned, I think it is wise to go around the country and get buy-in and dialogue and improve it, but from my perspective, I think it looks great. Of course, to me it is what is holding this up, what do we need to move this along, and from my perspective there are three things we need. We need a road map, so we’ll know how to get there. We need a budget, and we need designated drivers.
If you look at what other countries are doing, there are a few of them that are listed here. The U.K. is using a route map. They are developing a route map. Canada is building what they call a road map, and Australia is building a national plan, and there are others.
I think what we need is an information for health road map. After all, we are the ones who were talking about information superhighways before any of these other countries. I think the road map is not a bad idea.
But we also need a budget. We need a budget that goes with this. I don’t know how much it should be. I don’t know what — the U.K. has about $1.5 billion U.S. new money planned to modernize their information infrastructure. What should be spent on this? $100 million, $750 million, $10 billion? I think it depends on what we want.
The thing that worries me is that unless the federal government invests in this, I worry that we will inherit a very unusual set of loose pieces, much like popcorn, in our vibrant country and economy, but it will be proprietary, and it won’t necessarily be a route map that will lead us anywhere particularly, and that we will all not get the benefit out of this tremendous opportunity that we have.
So what would it be? I don’t know, seven billion new money over seven years, a billion a year? One-time money? I think that would move us a long way.
It is interesting; we do have privacy advocates both in the Department as well as the White House, but we don’t really have a designated driver for the health information strategy. In the U.K., they have a health information policy unit and an information authority. Canada has a national office, Australia has a national office, other countries also have their national offices.
I think the issue is, where is that driver’s location here and where should it be in our government? At the Data Council? Obviously, that is a good port of call, but to my knowledge, unless it has changed, it didn’t have a budget. So it makes it a little tough to run a national strategy with no budget. Should it be at ARC or NLM? Should it be in the Secretary’s office? Should it be in the White House? Where? A new agency?
I don’t have the answer, but I do feel like we need a program that has a clearer name that the public can understand, a road map, a budget, global collaboration, and an agency or a center or a leader, a direction that leads the agenda. If everyone is responsible for it, in a way, no one is. I think that is to some extent part of our issue today. I think there is enormous good intention and tremendous energy, but I think we need to soon channel that and move it forward. I think the committee and the staff and the Department, all these groups, have moved this agenda forward clearly, as well as the statistical underpinnings that go with that, but in a way, the best clearly is here to come.
So a very happy anniversary to the NCVHS, and obviously, all best wishes for continued success. I thank you for the chance to be here again. It is nice to see you.
DR. LUMPKIN: The next former chair of the committee was the chair when I first had my first experience with the committee and had an opportunity to learn how important the work of the committee was, what the role of the national committee was. Currently now, she is the director of the National Health Policy Forum at the George Washington University, and has been since its inception in 1972. She is a lecturer at George Washington University, serves as a mentor at the Wharton School’s Health Care Management program, not to mention a few of us in this room.
In 1988, she joined the committee. She chaired the committee from 1991 to 1996. Welcome, Judith Miller Jones.
DR. JONES: Thank you, John, and good afternoon to all of you. What a wonderful day we have for a reunion. It to me is a reunion, to see a lot of faces of people who have somehow or other been lost on my radar screen of late. But I hope throughout the day and cocktails too, we can renew acquaintances.
That is of course, if my trifocals allow me to read your names. I’m glad that Marjorie produced name tags for us, but like many of us in this room, we are all aging in different ways. I hope that I won’t embarrass myself or let you down by making a mistake.
Let me simply say just a few things. Marjorie asked me to make some comments here today. Had I known I had to follow Don Detmer, I would have said no from the outset. Usually he starts off with wonderful slides. If it is not the two bulls heading like this at each other across the roadway, it is some medical cartoon that leaves me feeling a little chagrined. I don’t know where he finds them, but I have given up trying to compete.
But Marjorie asked me to think back a little bit about my time on the
committee, what the committee was and became, and what have I learned about all of this, particularly from my policy position that I started out with and still hold today.
One of the first things that I learned, and I think back on my time on the committee, is how incredibly bright and hardworking everyone is in the field of health statistics. I really do believe it is among the finest assemblages of people that I ever came in contact with. When I think about Garrison Keillor, I am beginning to think that on average, everyone in this field is above average. It was a delight to work with so many wonderful people, to learn from so many wonderful people, and to see a committee that on its own and was able to reach out to try to draw on the best that we have, and to move it forward.
I would also have to say, and I look back with a little bit of chagrin, that I was a little naive. I am a Washington policy wonk type. I think I’m a little cynical after all these years here, but even I was naive when I entered this. Marjorie asked that I allow my name to go into nomination for this committee, and I thought, my lord, I am not at all qualified for this, other than I am rather opinionated and have strong interests. I actually started out my career at IBM. Larry Weed was my customer on the computerized patient record. But I really was very naive when I said to her, maybe this was one committee that I could serve on that is very apolitical, and it won’t jeopardize my role trying to be neutral in policy. It wasn’t long thereafter that Phil Leek came along and taught us all on the committee that DATA, that I thought was data, really was spelled T-U-R-F. It means that nothing goes forward without a lot of strum and drung.
And this is an area where being neutral gets you nowhere. You have to take the best information that you have, think it through carefully, and then move forward. I for one never appreciated how hard that would be.
I want to simply close by saying that this is a committee with 50 years’ experience. There were many wonderful people who preceded us, who set the stage with data sets and with their analyses of what the problems were and what we needed to do to move forward. It was an honor to work with people in the Department and people on the committee. I think we were the last of a certain era, before the HIPAA regs moved data and statistics and health statistics into the mainstream, which has made your lives and those of you who followed mine much, much harder.
I hope that all of you think about this as I do. Frankly, with only 50 years of experience, the hard part still lies ahead. I now have to worry about privacy concerns and a few other little matters in the policy arena. But data and health statistics are the bedrock of good policy.
I would like to make a promise to you. I will work from where I am on trying to improve better policy regarding data, if each of you will, whatever position you hold, promise that you won’t give up one ounce of your hard work you are already putting forward, and perhaps redouble your efforts, because I think it is truly something that both our country needs, but as Don pointed out, there are global issues involved here. We are still looked upon as a leader, even if we are not funding ourselves as a leader. I’d like to think that through efforts such as this one today, we will actually move forward with a little more zest than we have until now.
Thank you all.
DR. LUMPKIN: Our next former chair I think I just met today. When one looks at the history of this committee and all of the work that they have done, from laying the basis for some of the first surveys for the National Center for Health Statistics, actually its precursor, it is really quite amazing what work has been done.
Ron Blankenbaker is an associate dean of the University of Tennessee College of Medicine in Chattanooga, and chief medical officer at the Ehrlinger Health System in Chattanooga, Tennessee. He is acting chairman of the Department of Family Medicine at Chattanooga from November 1994 to July of 1995.
In 1991, he received the prestigious Louis C. Robbins MD Award by the Society of Prospective Medicine for his contributions to the prospective medicine and the extension of useful life expectancy through research, leadership, education and promotion. He chaired the National Committee from 1986 to 1991. Ron?
DR. BLANKENBAKER: I see there is a little platform here for some of us to stand on. I’ll try not to hit the microphone.
It is so good to be here. Marjorie, it is so good to see you again, and Gail. I haven’t seen Gail in such a long time. We had so much fun, and some difficult times, too.
It is a real honor for me to be here with this prestigious group to talk to you. I’m not going to say very much about the past. All I can say to you is, — well, I will say something. People say I happened to be in the right place at the right time. I was health officer in Indiana when Senator Quayle’s office called and said, we need some more people on committees in Washington, would you be a member of the National Committee on Vital and Health Statistics.
I had to first find out what it was, but then I said, sure. I had gone to the position of state health officer in Indiana by appointment of Governor Bowen. Governor Bowen, a family doctor, later became Secretary of Health and Human Services. Governor Bowen was Secretary when I was on the committee, and just happened to appoint me chairman, the position that I held then for a few years. I just happened to be in the right place at the right time.
But the times that we had with this committee are very momentous ones. We worked hard, we played hard, and I think we really worked hard to make a difference.
I want to show just a few slides to give you an idea with how comfortable I am with technology. These are on Power Point, but I am still using slides. Some people say an expert is a person from out of town with slides. You can also define it as, an expert means a has-been under pressure.
Marjorie, you notice the logo. We created that. I don’t remember how you got it done, but I think it was done fairly cheaply. We thought, is this going to survive, and it sure has to this date. So I added it to all these slides.
I want to talk a little bit about what the committee is doing. This is the first paragraph of the article that you have in front of you, the interim record on the information infrastructure. What I want to point out is the fact that we are using the World Health definition of health, one that I think is very important to what we are trying to do today, and what we are trying to do with the data that we have available to us.
Let me read to you the vision of the regional health council in our area, Southeast Tennessee. This we did in conjunction with Dr. Leland Kaiser, and I’ll mention him in just a minute.
We, the residents of Chattanooga and Hamilton County, will redesign our committee, one person, one neighborhood, one institution and one system at a time, for the cooperation of all people, so that every person will reach his or her economic, social, spiritual, mental, emotional and physical potential.
We have changed our whole attitude towards health to the point where it is living without disease and disability, living without poverty, and being able to meet our potential lifestyle. I am very appreciative of the fact that the committee has included this as its first paragraph.
I’m sure you can’t read this back there. This is a number of people sitting on the beach, and they are all thinking about all the e-mails when they get back home. One of them is saying, I think I’ve got 1,050 e-mails. Obviously, we have become an electronic society.
This is a statistic that boggles my mind. Human information doubles once every five years. Printed conventional information doubles every eight years. In 1951 there were 10,000 journals, now there are over 140,000 journals.
If the trajectory of information generation continues at its present pace, by the year 2020, the information available to us will double every 73 days. It is hard to imagine. If similar progress was in the automobile technology area, you could have produced a Lexus for two dollars that would travel the speed of sound and go 600 miles on a thimble of gas.
I’m going to have to go over here to read this. You are unreachable by e-mail, voice mail or pager, so I am doing the next best thing and delivering this message in person. Hopefully, in this electronic age, we are not going to lose our ability to work face to face.
Dr. Leland Kaiser is a health futurist that we have been bringing to Chattanooga for a number of years, about every six months. It struck me, back in 1996 he said this to us: health care is moving from a service industry to one of information management. I think that is absolutely true. There is so much data that we have available to us today; if we don’t move it around officially, it is going to be very difficult to provide the kind of health care that we are talking about, especially as you look at the definition.
This is a reproduction of the working group charge. Again, I was very impressed with not only the charge, but the work that this group has done. I am pleased to see that the committee is moving forward in many directions that we just thought about back then. We talked about linkages, we talked about unique identifiers, we talked about confidentiality, we talked about how organizations could work together and so forth. Here, we are talking about a national health information infrastructure, something that is absolutely necessary.
I certainly hope that we all recognize — I know that most of you do, and probably you know more about data than I will ever think about — but the data is being thrown at us rapidly, more than we can ever assimilate. Certainly in the future we are going to have to have a means of prioritizing data, being able to pick out what is important and what is not, if we are going to be able to set the stages for providing the kind of health that our public and our communities deserve.
Again, it has been a pleasure being here and seeing a lot of old friends, and hopefully making new friends. Congratulations to the committee for its work. It is obvious that as we tried back in our days, that the committee does make a difference.
Happy Anniversary.
DR. LUMPKIN: We also have a communication in the materials that we handed out from Dr. Kerr White, who was chair of the committee in ’75 through 1980, who was unable to attend today. He has sent his greetings. I would encourage all of you to read them. They are certainly very insightful, and are of the same calibre of the speakers that we had before today.
It was interesting to hear Ron talk about his experiences as a state health official. That is an interesting job for an individual to have. I have the same job in the state of Illinois.
DR. BLANKENBAKER: Something everybody should try once.
DR. LUMPKIN: Yes. The interesting thing about the job is that the average time in office is somewhere about two years, on average across the nation. So it has gotten to the point in my career where I started getting introduced in many places as being the state health official who was appointed by the governor and who has been around the longest in the country.
I used to think about that a lot and say, what are they saying about me? Do I just not know how to move to another job? I really never got any insight into that until actually, just a couple of days ago, when I realized I was going to been having the opportunity to introduce Secretary Shalala.
As I was reading her bio, it said — the first line — Donna Shalala is the longest serving Secretary of Health and Human Services in U.S. history, having served the Department for seven years. It is right up there in front in her bio.
I was thinking about that, and suddenly I realized that we have gotten to the point in our society where, if it is not quick and it is not over, it is not worthwhile. What does that mean? When in fact, when we look at longevity, the ability to start a project and actually see it to an end, is a very important ability, and reflects a certain level of commitment, and commitment and government and leadership is something that is becoming increasingly rare.
Secretary Shalala certainly reflects that. Under her leadership, the Department has enrolled nearly two million children in the states’ health insurance program, and has raised immunization rates to the highest levels in history. She has remained dedicated to preventing young peoples’ use of tobacco, alcohol and illicit drugs. She has also crusaded for improved access to medications and treatment for individuals with HIV and AIDS, not only in this country, but throughout the world.
Throughout her career, Secretary Shalala has been a scholar, a teacher, a public administrator, and as chancellor of the University of Wisconsin-Madison, she was named by Business Week as one of the five best managers in higher education. Through her leadership, the importance of health information and statistics and knowledge management and its transforming role in improving the quality of health care has gained recognition within the Department.
So it is truly an honor to introduce to you the Secretary of the Department of Health and Human Services, Donna Shalala.
HON. SHALALA: Thank you very much, John. First, let me apologize to all of you for being late. As I was walking about the door, I got a phone call, and everyone who could have answered that question easily was here. So will Ed Sondig and John Eisenberg please see me afterwards.
Actually, they will be very proud of me. I actually figured out the answer.
Let me start by thanking John Lumpkin, and apologize. I am pleased that Ronald Blankenbaker, a former chair, went forward. I hate being late for anything. I think it is from years of teaching and having to show up on time for seminars.
I also want to acknowledge the other former chairs, Judith Miller Jones and Don Detmer and John Rolph from the NAS Committee on National Statistics.
I do want to say something about John Lumpkin, though. We were actually 13fortunate to have him. He has been an excellent committee chair. He of course is an exceptional leader in public health. He believes right down to his feet that the power of information technology and its emerging capabilities can actually be used so that everybody has the same opportunity for a healthy life.
And of course, Ed Sondig, who is my tutor in health statistics and has taught me a lot over the last seven and a half years. I do know, Dr. Lumpkin, why Cabinet Secretaries stay an average of 18 months; it is very clear. You get to call yourself Secretary forever, they have to introduce you and write letters to you as the Honorable, and there is no accountability, because you haven’t done anything in the first 18 months, so nothing comes back to haunt you.
What happens after seven and a half years is that things come home to roost. Some of the decisions you made were okay, and some were not okay, and you wonder why you made that decision five years ago. The biggest problem is, you can’t remember why you made that decision five years ago. But you’re pretty sure, given the Department’s commitment to rigorous analysis and to good statistics, that you probably had as much information as you could have had at the time to make that decision. So you are willing to defend it.
I am pleased to be here to help all of you celebrate the 50th anniversary of the committee. The committee has helped to shape this country’s health statistics and data policy for half a century. So I think a golden anniversary is actually very special.
This may be the last year in which — it probably is the last year in which I speak to you as Secretary of the Department. I am not going to push it beyond eight years. I want you to know though, I am proud of what we accomplished together. I think it has been a process of steadily chipping away at the difficult and complex problems through a period of really tremendous change.
Will Rogers once had this to say about nostalgia: Things aren’t what they used to be, and probably never was. There is actually a kernel of truth in that remark. He obviously didn’t know this group, though. If he did, he might have said, no, things aren’t what they used to be, they are a heck of a lot better.
In 1950 in the committee’s first annual report, the report stated that by and large, health statistics are still in the horse and buggy stage of development. We are not driving a buggy anymore. Thanks to all of you, the Department and the country has better health information than ever before in the history of public health. And thanks to all of you, statistical health measures the foundation and the bedrock of the public health system and decision making about the public health system have never been more accurate — whether it is tracking the objectives for Healthy People 2010 or improving minority health statistics or planning disease and injury prevention programs or evaluating the quality of health care or strengthening Medicare or Medicaid, or monitoring the health of Americans, you have provided accurate and I think very critical information.
I don’t think I am overstating things when I say that the committee is one of the hardest-working advisory groups at HHS. I have known many of the former chairs over the years. In the seven and half years that I have actually worked with the committee, I can’t think of a group of people who have worked harder to promote a better and healthier life for the people of this country.
When I expanded the charter in 1996, it was not without thought. When we established the Data Council within the Department, it was really a twofold commitment. First, it was our hope that the committee would provide the basis for a broad-based national health information infrastructure. The Department needed the capacity to retrieve the most recent and accurate information possible, because we believe that the right data makes a difference.
Second, we realized that the Department couldn’t do it alone. We needed to build strong bridges to industry, to the research community, to the public health community and to the American people. I think that is exactly what we have done. Together we have addressed the requirements of the Health Insurance Portability and Accountability Act for national data standards in privacy, and this summer at the end of the summer, I will issue the first privacy regulations in American history, well informed by these databases that have been developed over the years with I think thoughtful input from many people in this room.
Together we have actually improved our population database systems, and together we have promoted a vision for the national health information infrastructure that tries to tie all of these elements together.
So the 50th anniversary symposium is in some ways a victory lap, but we still don’t have a number of things that we need, and all of us need to make a renewed commitment. We have to continue to build a seamless web of information gathering and dissemination that still protects the privacy of individual citizens. There is a huge resistance in the industry now, arguing that many of these requirements are in fact burdensome and expensive, and we ought to eliminate them as a way of streamlining the system. I think it is extremely important that we are wary about some of those charges.
Some of them are reasonable because of the overlapping requests that we make for the health care system, but some of them are in my judgment an attempt to actually avoid giving us information that would allow us to do everything from paying more accurately to being able to describe the gaps in the system and more both efficiency and equality elements that we want to introduce.
We need to find more resources. I am the first one to admit that we don’t have the resources that we need. But I also know as you do that the day is not far off when a patient and a doctor and a pharmacist will routinely communicate electronically. I think that Eugene Hamilton, a former committee member, probably said it best when he wrote about the committee’s responsibility, to seek out brave new worlds to conquer, to point out the problems that exist, and to indicate possible solutions. That is exactly what the committee has been doing for 50 years; it is what we need to continue to do.
For my part, I am going to continue to fight for the resources we need, and to make sure that these huge data systems and the integration of these systems continue to inform public policy decision making in this country to improve the quality of health not only in this country, but all around the world. That after all is exactly what all of us are after.
Thank you very much, and congratulations to all of you.
DR. LUMPKIN: When we were setting up this symposium, we were trying to figure out who to invite to speak. I suggested that we invite the Secretary. The biggest surprise was that her schedule was incredibly tight. But we are thrilled that she made room in her schedule to come here, and we appreciate the time you were able to spend with us.
Our next speaker — unfortunately, the Secretary stole my joke — has served as the director for Agency for Health Care Research and Quality since April of 1997, which the agency has been known that since 1997; he has been there longer. Prior to that it was called AHCPR, Agency for Health Care Policy and Research. So I guess Don’s comments about the NCVHS being able to survive without a cutesy name is significant. He also serves as advisor to the Secretary on quality, and is designated as the lead agency for quality improvement, on issues within the Department of Health and Human Services. He also co-chairs the Department’s Data Council, which is where we have gotten to know him from, and has been opening chair of the federal Quality Interagency Coordination Task Force, fondly known as QUIC, another acronym.
He has served as Deputy Assistant Secretary for Health, and is Acting Assistant Secretary of Health. I want to add that John has been truly an advocate for the committee for health information and data policy within the Department, and it is truly an honor to have an opportunity to hear some comments from him. John.
DR. EISENBERG: Thank you, John. It is true, they are calling me John of AHCRQ these days. Around the agency, people are saying, yes John, the real Joan of Arc, she couldn’t tell the difference between a vision and a hallucination. So as we think about this vision about the future of informatics, let’s just hope that it is a vision and not a hallucination.
The competing title was the National Organizer for the Advancement of Health, the acronym for which is NOAH. So it would have been NOAH and the Arc saving lives two by two.
I want to thank you for letting me come, John. I am really honored on behalf of Peggy Hamburg, my co-chair of the Data Council. It is a pleasure to join you to celebrate this 50th anniversary. NCVHS is one of the few organizations in Washington that has lasted long enough to be a member of AARP. I think that is great. I want to congratulate you. Modern Maturity will be showing up on Jim Scanlon’s desk now, whether it does already or not.
I do want to thank you. Secretary Shalala has already thanked the committee and all who have served it for the past 50 years, and I just want to join her in my role as co-chair of the Data Council, to thank you and express our gratitude for the tremendous amount of work that the NCVHS has put in, and the amount of counsel and advice that we have gotten from you.
It was five years ago that the Secretary established the Data Council and expanded the charter of NCVHS as you have all heard. I think we have had a really close relationship.
I took over as the operating co-chair of the Data Council when Bruce Gladdick left to go back to New York. When I took over, it was very soon after that that Don left. I didn’t take it personally that Don accepted this position in Cambridge, because if my name were Don, I would want to be called a Cambridge Don too. Can you imagine? He is a Don don. That was probably too much for him to pass up. But with John Lumpkin and myself now being co-chair of the Data Council and chair of NCVHS, it takes on a new meaning when people say they have to go to the john.
So now that we can talk about those special kinds of people, I just want you to know that you are also special kinds of people, who are able to bring numeracy to government. You all know there are three kinds of people, right? Those who can count and those who can’t.
There is an old story that I heard not too long ago about a student in Cambridge, Massachusetts. He was walking along, and he went to this little store, one of these little convenience stores. It had several aisles, and one of them had a label, Fast Checkout. He went up to the fast checkout line, and he stood in line and the clerk said, well, I can see you either go to Harvard or MIT. Of course, he was very proud, and he said, yes, how could you tell that? She said, because that sign says eight items or less and you have 10. That means that either you go to Harvard and you can’t count, or you go to MIT and you can’t read.
This is a group that can both count and read, and it is been a pleasure to work with you. NCVHS has provided us with a real framework for bringing in a lot of the benefits of information technology to health and to health care, in a way that does protect the security and privacy of health information, as we have been asked to do through the Health Insurance Portability and Accountability Act or HIPAA, another great acronym.
HIPAA is a vehicle. It is not a great vehicle, but it is a vehicle that we all believe we can use to bring some more sense to the way in which we use data. The question is, can we make this vehicle run.
I hadn’t actually thought about the extended metaphor until Don Detmer’s comments about whether or not we can find a driver for this vehicle. But it is a vehicle nonetheless, and it is one that we are working hard at now, to take those requirements of HIPAA and convert those into, as the Secretary said, some regulations that will be the first privacy regulations that this country has seen.
Doing that has been a collaborative activity, government working together with the private sector, including in a very, very important way the NCVHS, and through the NCVHS bringing together public health and clinical concerns, the concerns of those in the research field and those in the health care industry, to give us some sense and sensibility to putting the HIPAA.
We got — is the most recent count, Jim, 52,000 letters — in response to the proposed regulation. That is a lot of reading. But I want you to know that the letter from NCVHS was the first among equals. It is one that we have paid a lot of attention to. It accounted for at least 50,000 of the others. 52,000 letters in response to a proposal to try to bring some sense in the use of data in the health care system.
In a lot of ways, the relationship between HHS and NCVHS, as we have tried to push the HIPAA regulations into a form that will move us towards more appropriate use of data in this country, reflects on the ways in which a complex federal agency like HHS can with an advisory organization. I think it is exemplary. I think for all the other advisory organizations that we have in HHS or throughout government, looking at the way NCVHS is organized, its commitment, its staff support from the Department, its products, is really an example that we all ought to be trying to fulfil in other places.
If anything, I just want you all to know that it hopefully has made it worthwhile, filling out all those forms that you had to fill out in order to serve on this committee, which I know is a pain. But it also brings politics together with reality. There is a statue, as some of you know, of Albert Einstein in front of the National Academy of Sciences building. There used to be a pamphlet — I’m not sure if this pamphlet still exists, but it had a wonderful quotation from Einstein on it, which is perfectly appropriate for the NCVHS. The quote from Einstein was this: Politics is for the present, but an equation is forever. It does speak to the fact that data and facts can bring us far beyond where the immediate issues of politics will take it, although HIPAA has been a big challenge for all of us, and I think gives us an opportunity to make some steps in the right direction. It is a combination of research and logic, of policy and politics.
It has been interesting, listening to the comments of the former chairs, how that convergence of politics and policy, of numbers and direction for the country have come together. I think it is interesting, in that each of them represent an era in the life of the NCVHS.
I was thinking as I was about to say that, how is Judy Miller Jones going to feel when I call her an era. It is sort of like the Pleistocene Era. But I think what is really impressive about this group is how the ball has been handed from one to the other. As we look back at NCVHS’ distinguished history and its collaboration, it really speaks to the comment that Lord Kelvin once made. Maybe this was even on the minds of those who put NCVHS together 50 years ago, when Kelvin said, if you cannot measure it, your knowledge is of a meager and unsatisfactory kind.
It was the NCVHS that brought us the original proposal for the original proposal for the U.S. national health survey in the 1950s. It has become now a whole family of surveys that the NCVHS directs. The original concept and original content for Health U.S., the annual report card on the health of the nation in the 1970s, the groundbreaking work that has been done on health data standards through the 1980s, and of course, more recently as we have talked about, the issues with HIPAA and health data interchange, health transactions.
Bringing all of this together has really been quite a remarkable feat, one that does bring politics and policy together. I was thinking as I came to this meeting — and those of you who arrived at the same time I did saw that there was a huge motorcade at the State Department of some emissary of some other country. I realized as I was coming in that people like Lumpkin and people like Judy Miller Jones and people like Don Detmer, people like Blankenbaker, all represent that combination that ought to be over at the State Department, frankly. If you think about that ability to bring together policy and diplomacy, to help people come to a consensus and an agreement, it really is a remarkable skill.
Dorothy Rice has always been one of my heroes, or I guess I should say sheroes, and it really is a pleasure to see you, Dorothy. I also want to mention in the spirit of Carl White’s having already been mentioned as one of the chairs, that our agency announced last week the establishment of a Carl White visiting scholars program at the agency, in which five people will be able to come and work for a year or two as a visiting scholar. We couldn’t think of anybody who brought together all of those issues more than Carl White has, and who really established health services research. I didn’t realize that he was going to have a history of health services research in the Journal of Health Services Research in the same week that we announced this visiting scholars program, but Carl like Dorothy I think is a giant in our field.
Having the kind of data that helps us to move the system and to change the system is one that has a long heritage. I had a chance to visit the Florence Nightingale Museum in London about six months ago. If you have never been, I urge you to go to St. Thomas Hospital and look at the museum in honor of Florence Nightingale.
When she went to the Crimean War in 1854 and said we’ve got to do some counting here, and then when she went back to London, she made one comment which is one of my favorites I have to read to you. She said, Uniform and accurate statistics would enable the value of particular methods of treatment and of special operations to be brought to statistical proof.
Now, if you think about what happened to Florence Nightingale after she started saying that, she basically was ignored. She set up the St. Thomas Nursing School, and then people ignored her. The medical community ignored her, partly because she was a nurse, partly because she was a woman, but partly because she was saying, let’s count what we are doing and let’s learn from it.
Florence Nightingale ended her life as a recluse, and in some ways sacrificed herself for all of us who followed, who want to bring numbers to health policy — a very, very important precedent, but also one that showed how frustrating it can be to try to bring numbers to policy.
One of the books that I have read in the past year, one of my favorites, in fact, is a book by Peter Bernstein called Against the Odds. If you haven’t read it, I urge you to take a look at this book. It is called Against the Odds: The Remarkable Story of Risk, in which he talks about how we can come together with numbers to help understand the risk of certain events occurring.
In the book, he comments as follows. He says, the information you have is not the information you want. The information you want is not the information you need. The information you need is not the information you can obtain. The information you can obtain costs more than you want to pay.
So I think it is a sad reflection on the disparity between what we need and what we have and what we can afford, but I also want to urge you tonight or tomorrow to take a look at a report that is going to be released at 6 o’clock tonight, which I think will be one of the most important reports in health statistics that has come out of any agency in the world in a long time. The World Health report embargo is going to be broken at 6 o’clock tonight. I urge you to go to the website — hopefully it is going to be up tonight, if not tomorrow — and look at the World Health Organization’s report, in which they bring together public health and health care, in which the World Health Organization says, we need to be measuring the health of the public. We also need to be looking at how our health systems are doing in responding. It is very symbolic of where we all are right now, bringing together health care and public health, and the leadership by the WHO will give us a chance to do some benchmarking across countries as well as within our own country.
So in closing, let me just thank you again, thank again NCVHS and all of its friends, for five decades of service, for having reached advanced middle age, or as they say, the near elderly status of being over 50. I want to compliment you on having helped us to improve the health data that we have in this country to drive policy.
I want to underscore what the Secretary said about the value and the importance of the work that you have done not just for the Department, but for the entire nation. I want to look forward to a continuing relationship between our Department, the NCVHS and all of those who are here to celebrate and those who celebrate around the country what the NCVHS has brought to us.
Thank you.
DR. LUMPKIN: At this point, we’ll have the next panel come up
DR. SONDIK: I’m Ed Sondik, the Director of the National Center for Health Statistics. I am going to be moderating the next two sessions. I’m not sure what moderate means. I’ll try not to be too moderate.
I just wanted to express my thanks to the committee members, to the past chairs of the committee, particularly Judy Miller Jones and Don Detmer, who with great passion really introduced me to the committee, but especially to Marjorie Greenberg and Gail Fisher — I think Gail is here, right? — who were the first to give me an in-depth precis on the committee and to let me see the tremendous things that the committee had done over its history to that point, and the great potential for the future. I really am indebted to you for that.
I just wanted to mention two things about horse and buggy, since that comment came up. There were two things, one I saw in the paper for the first time, and the other that we are all aware of.
One is the fact that the genome is about to be — at least the rough draft of the genome is about to be published. Think about that compared to where we were 50 years ago, and the implications of that.
The other small article I saw in the paper this morning was on a new disk drive that will be marketed in September that holds a gigabyte’s worth of data in a package the size of a pack of matches. I don’t know the implications of those, and I don’t think that people understood the implications of where we were in technology 50 years ago, but what is clear is, it is going to be driving us and driving the system. The committee is extremely important I think in helping to guide that, support it, and even catalyze efforts.
We have two efforts that Don Detmer has already mentioned, an effort towards a national health information infrastructure, and a vision for a 21st century health statistics system. They are very much related, I think Don is exactly right.
Our first speaker is John Lumpkin, who of course has been introduced. So without any further ado, Dr. Lumpkin.
DR. LUMPKIN: It is a pleasure to have the opportunity to talk about the health information infrastructure. Despite Don’s attempts to change the name, I only will take the following defense: When he was committee chair, he thought it up.
In 1946, Albert Dunn, who was chief of the National Office of Vital Statistics, wrote an article. Actually, he delivered this article in May in Canada, when he had been invited to talk about some of the conceptual ideas they were looking towards as they were developing their health statistics. It is on record linkage. He says that each person in the world creates a book of life. This book starts with birth and ends with death. Its pages are made up of records of the principal events in his life.
Record linkage is the name given to the process of assembling the pages of this book into a volume. This really is the essence of the concept of what we are moving forward on.
I ran across this statement very recently. It was in a paper that hasn’t been published yet, but it was in a very early draft. The author said I could use it. The managed care system is failing. Medical inflation is back, consumer distrust, provider hostility, costly new technologies and political opportunism will no longer allow cost and quality to be controlled by most existing managed care arrangements.
Overall quality has not been deleteriously affected by managed care, but managed care has not substantially reduced unsafe practices, over use, under use and misuse of health care. Paul Ellwood. For those of you who do not know him, he has been called the father of managed care and HMOs.
What has substantially changed? I think we are talking about a transition, because not only have we not met the challenge in our current health care system, which is still undergoing a transformation and will continue to be transformed. Managed care as we know it and all the health care systems we know will continue to transform.
But also, we haven’t addressed inequities, disparities due to race, the fact that if you are African-American, you are less likely to have coronary artery bypass surgery, less likely to have surgery to re-establish flow in a clogged artery, but you are more likely to have an amputation. Inequities with a health care system that provides care, yet over almost 40 million people have no health care insurance, and studies have demonstrated the impact of that.
But our goal is to improve health, health in the broadest sense. The World Health Organization states that health is a state of complete physical, mental and social well-being, not merely the absence of disease and infirmity. Further, health is a state of well-being and the capability to function in the face of changing circumstances.
But we have to recognize that health is more than just this broad concept of health, that the attainment of health is a shared responsibility. It is a shared responsibility of health care, of public health, of the business community, the faith community, of education and volunteer and all sectors of society. Each of them have a role to play. This concept was explored by the Institute of Medicine report in 1997, Improving Health in the Community: The Role for Performance Measurement.
Health is dependent not only on good quality health care, but also on other factors. This issue was recognized as early as 600 B.C. by Hippocrates, who noted that if you want to know about the health of peoples, note the winds and the character of the air, the water and what they drink, the lay of the land and the habits of the people who inhabit the city.
We know that health is a very complex process. It involves a recognition that health is a function, what Evans and Stoddard describe in their field model. Health is dependent not only upon the characteristics of the disease, what is the organism, what is the abnormality in the body, but also the health care, the functioning of the individual before the disease, the social environment, the genetic makeup as we are learning more and more about genetics. It is fascinating that over the years that I have practiced both the clinical medicine and now public health practice, how our concepts have changed.
Remember when we thought peptic ulcer disease was caused by stress and too much stomach acid? Now we know about helicobacter. There are some studies now coming that indicate there may be an infectious component to myocardial infarction, at least perhaps to the process of rupturing the plaque, maybe an infectious process. Patients who are given the influenza vaccine are less likely in the following year to have a myocardial infarction.
Health care, genetic environment, physical environment, well-being and prosperity. We know that poverty has a tremendous impact. In fact, in many diseases, whether it be violence, cardiac disease and other cases, when you adjust for socioeconomic status, many of the racial differences that we are so focused on as a society go away in looking at rates of disease and illness.
This process is driven by the use of data, our understanding of the interconnection of health as a component more than just health care, more than just how we treat an individual who is sick, is driven by the use of data, converted by analysis into information, which is then integrated with rules and generates knowledge. But for what?
Around the 1700s, Percival Pott in the late 1700s identified the link between scrotal cancer and occupation amongst the chimney sweeps in England. He noted that if you were a chimney sweep and you did not bathe on a regular basis, you were more likely to get scrotal cancer.
This information led to a dramatic change in The Netherlands, where the reduction of the rate of scrotal cancer among chimney sweeps dropped dramatically. But the fascinating thing is, there were articles written as late as the 1880s in England, decrying the fact that the rate of scrotal cancer amongst the English chimney sweeps had not changed dramatically since the writings of Percival Potts.
We contrast that with that of Dr. Snow. Dr. Snow used an early graphical information system, a map, and a pencil, and plotted out an epidemic of cholera. As the legend goes, and as we know, the father of modern epidemiology, Dr. Snow, he went to Broad Street and removed the handle of the pump.
Knowledge must lead to action. Information and knowledge and technology enables us to use that knowledge to create better health care outcomes.
Many of us got here today, at least I did, on an airplane. I got here today, about 1 o’clock in the morning. When I walked into the airplane, the cockpit looked a little bit like that. I felt pretty comfortable getting on the airplane, having just flown Mexicana Airlines, which is a little bit different standards, I was quite happy to be on the American Airlines airplane that I flew on.
But what if you were to look in the cockpit and it looked like this? An altimeter, a compass and a speedometer. None of the electronics, none of the computers, none of the guidance systems. Aviation and avionics has gone through a tremendous transformation since the Wright Brothers at Kitty Hawk. But ironically, the health care system, basic delivery of health care, has not changed from the beginning of the 20th century to the end of the 20th century. As we begin the 21st century, we have the opportunity to transform health care through the use of information technology.
What do I mean, it hasn’t changed basically? When I trained in health care, when I trained in emergency medicine, there were three kinds of coats at Northwestern University where I went to medical school. They had the medical students, who had the short white coats, and then they had the house staff, who had the blue coats, and then they had the attendings, who had the gray coats. And then of course they had the research attendings, and they had the long white coats, but we’ll leave them out. You could always tell the hierarchy.
But there is another way that you could tell the hierarchy. The medical students had all sorts of books and notes and things in their pockets. When they went in to see a patient, they would pull out their notes and they would and ask some more questions, and they would look in their notes and ask some more questions. The residents of course would only have a few things in their pockets, and the attendings, my God, they barely even carried a stethoscope.
Our concept of quality in health care has grown up to be that what you bring in between your ears is high quality health care. If that is what you use to provide an outcome, you have given high quality care.
One of my memorable occasions working in an emergency department occurred when I had a lady came in, and she had this funny pain. I was an attending. It was in my early days as being an attending. I used to try to amaze the medical students. But anyway, the medical students and residents both saw the patient. I walked in and I talked to the lady. I always believe if you listen to somebody long enough, they will tell you what is wrong with them. I came to the conclusion this lady had a kidney stone. The resident didn’t believe me, the medical students didn’t believe me, but I was the attending, so we ordered an IVP, and it came back positive. I was a medical god. They were all impressed. That really occurs, as you know, for those of you who have been in teaching episodes. That really occurs with house staff, and the medical students are impressed, especially at the University of Chicago. Medical students there know so much.
We sent the lady home. Now, the statute of limitations is over, so I can tell this story. The next day, one of my colleagues walks up to me and said, remember that lady you saw yesterday? For those of you in clinical practice, you know there is no conversation that ends well that starts off with that phrase, remember the lady you saw yesterday. In fact, what had happened was, she had a fever. I knew she had a fever, the medical students knew she had a fever and the resident knew she had a fever, but we were so focused on making the diagnosis, that once we had the diagnosis, we neglected to remember the fever.
She came back the next day because her pain persisted, and she was admitted to the hospital. She did fine. But that was not a good health care outcome. Why couldn’t my medical record say, oh by the way, Doctor, did you forget that she’s got a stone and she’s got a fever, maybe you need to treat her differently.
Our problem, because we have focused our health care system so much on the conceptual model, and it hasn’t changed since the turn of the last century, that a good clinician is the one that knows every fact, that has it in their brain, and can bring it to mind, not the one who can use information and clinical skills to get the best outcome for the patient. We assume they are the same, and they’re not.
Let us focus on what the business of health is, because we have had too much focus on the finances of health. I would argue that the business of health is to help people stay healthy, helping people get better when they are ill, helping people live with illness when their illness is chronic, and helping people manage a changing lifestyle when their illness impairs their functioning. You know what? When the Foundation of Accountability went to people and asked them, what is important about a health care plan, that is exactly what they said. That is what they wanted to know.
So let’s start with the basics. How can we improve upon this vision? How can we make a change? That is by following through on the vision of the national health information infrastructure, a set of technologies, standards, applications, systems, values and laws that support all facets of individual health care and public health.
We are not talking about a centralized database. That is very important to say at the beginning and at the end and the middle. We are not talking about where data is stored, but how data is used, how it is converted to information, how that information with rules are changed with knowledge and how that knowledge is used for action to improve health and health care.
It connects distributed health information in a framework of a secure network with strict confidentiality provisions. The NHII includes a data model, data standards and connectivity standards.
An example of this, as Don mentioned, is in Australia, the health data model. In this model, this is from the web page, you can see that they have taken many of the health events, health episodes, health data. You click on health events. It is an encounter. You click on encounter and you see all the various variables that make into an encounter very similar to the standards we are setting. You can get all the way down to the level of each individual element of each one of those forms, and there is a standard for that.
That is a piece of what we have to do. But we have to recognize that we operate in a purely unique American environment, and we need to have an American approach. We respect individual privacy.
Albert Dunn talked about the book of life. People in this country believe that they should have a book of life, but it is my book, and I want to know and I want to control who is reading my book, if anybody can read my book, and which pages they are reading. Our system, our approach has to take that into account.
The important thing is that the technology can allow that to occur, certainly to a higher extent than what we do now with current paper records. To respect individual privacy, to insure confidentiality, and engineered for security.
I was shocked, I was amazed when we conducted hearings for security regulations, to hear vendors come in and tell us, oh yes, we’ve got all these things built into our systems, but we have turned them off when we sell them to health care providers. Technology at its finest, decision making at is worst.
The national health information infrastructure also involves data standards such as those under HIPAA, the X-12, all the various letters and number accommodations that we use in this area. It also includes standards for electronic data interchange, the use of the Internet. We had a presentation today, an excellent presentation by the person who is going to comment on my report next, about the future use of the Internet in health, in other connectivity standards. But standards are not enough.
We not only need to improve the quality of data, but we have to understand what the goal of all of this is about. In today’s meeting at the NCVHS, Lisa raised the issue, what are we doing this for? It is not just to have better standards. Administrative simplification is to simplify and to improve the quality of what the business of medicine is. We have for too long thought that the business of medicine is about building hospitals and filling up clinics, when the business of medicine is to help people stay healthy, help people get better when they are sick, help them live with illness when they have a chronic illness, or manage change in their life when they have a catastrophic illness.
If we are going to improve the quality of health care, we need to use technology to push knowledge to the point of service. What does that mean? Did anybody ever buy a computer from an online company? Then you call them pu because something isn’t working right, and the first thing they say is, hm, okay, unplug the keyboard, unplug the mouse, and switch them, and then try that. You try that and that doesn’t work. They give you two or three things to do. Is that technician coming up with that based upon your own personal experience? No. They are sitting in front of another computer and typing in your questions, and the computer lists in priority the things that should be done. Where do those lists come from? They take the best technicians in the nation, they bring them into a room and they run problems through them, and they help them, and they make decisions, and those decisional support information, the knowledge that is taken from the best in the country are at the fingertips of those technicians.
When you tell them whether or not it works, it sometimes moves up and it changes the priority of what they are going to suggest to you. That is moving knowledge at the point of contact with the customer. We need to do that in health care, because not every clinician remembers every single bit, that doubles every five years today. As we heard, it may be doubling every 70 days by the year 2010.
If that happens, then the clinician can focus on the patient interaction, the business of medicine, of helping people stay healthy and people get better.
By using expert systems and decisional support and practice guidelines, we can in fact move knowledge down to the point of service, and that is a lot of what we do when we establish performance measurements. What is an immunization performance measurement? It is moving knowledge that it is a good thing to immunize kids to the point of service, because you know every time you see a kid that somebody is going to be monitoring whether or not you did an immunization. We know when that occurs, behaviors change.
So the goal of the national health information infrastructure is to take health information and create it in a way that it can be applied to — that decisional support systems can be applied to. Those mean that we have to have standards.
So within the context of the health information, we have got the universe of health data. Within that universe, there is provider data, there is administrative data, there is personal data, there is community based data, and there is quality data. Yet we believe that to be successful, the technologies, the standards, applications, systems, the values and laws must work in such a way that in three dimensions, we can provide a workable system, a personal health record, controlled by the individual and family that includes non-clinical information such as health care trackers, health materials, local health and health care services.
It also is a component that can be stored in a household computer. If you read our document, we talk about some scenarios, whether it be in a computer or smart card or some other source, or that matchbook sized one gigabyte drive that somebody can carry around in their pocket. But it allows an individual to have more control over their health information and their health care decisions, because they can get knowledge also.
Think about where our health care system has gone, where we have many patients now who trust an anonymous source on the Internet more than they trust their individual practitioner. We need to move our system back where health care decisions are choices made with the support of a health care provider by individuals in how they will be treated.
The core data within the personal health dimension, patient identification, lifetime health history, treatment plans and instructions that can be fed in from a provider, health risk factor profiles, so they know what they are at risk for and they know their decisions that they have to make, and they know their choices, whether or not to exercise. They know what the risk is. Health care trackers, they remind them to get preventive health care, and the personal liability quality health information resources, those that they consider to be reliable resources. All of these, as well as the history and record of who has authorization to access their health information.
This authorization will change. An individual may see their gynecologist for the first visit, and they are only going to be talking about basic information, maybe their history, how many times they have been pregnant. It will take two or three visits before they are willing to open up to a history of sexually transmitted diseases or depression. That is appropriate, and it is dynamic. The sharing of personal health care information with health providers has to be equally dynamic, and the system has to allow that to occur.
The health care provider dimension is best described by the vision of the caregiver described in the Institute of Medicine study entitled Computer Based Patient Record: An Essential Technology for Health Care, and many of the criteria can be seen in the document that was just being prepared by the committee and will be hopefully voted on tomorrow, on standards for patient medical record information.
Information systems will assist the caregiver by reviewing the information with the encounter record and providing the appropriate warnings, alert and remind us. I don’t believe that the disparity in treatment outcomes and management that occur between African-Americans and non-African-Americans in relationship to cardiovascular disease and so many other ones are conscious every time they see an individual decision. But we do know that these disparities exist, and that outcomes exist, and that clinicians see patients differently, based upon their socioeconomic status, based upon their social class, based upon their race and ethnicity. These kind of information systems can help us overcome these cultural biases that occur in the health care system, by providing the kind of decisional support as well as improving the quality of care.
The health care provider dimension encompasses information to enhance the quality and efficiency of health services for individuals, including information packaging, patient care processes, and integrates clinical guidelines. It allows the system to push knowledge to the point of service at every encounter.
The primary record of care is under the control of the provider, who captures original information. That information populates an individual’s personal health record. They will have the opportunity to have control. They may be stored on site or with a third party health information trusted guardian.
Confidentiality protections include technologies, authorization, authentication, encryption, electronic signature, restricted access by case or location or user. It enables a patient and provider to control joint decision making.
The third dimension is the community health dimension. This dimension supports the identification of health services, but it is interactive with the other dimensions. A person’s health record will draw upon information about community events. How many times have we seen an alert in the newspaper and various other media about temperature heat index? It is an unfortunately fascinating outcome of the heat wave that we had in Chicago in the mid-1990s, when 400 plus people died, excessive deaths.
When we went back and looked at the data, we saw that there were some significant risk factors — being isolated socially, being poor and living in a community with high crime. When that occurred, there was no way to get the message out.
How can we get the information to people at risk when the environment changes? Whether it be too hot or too cold, or a pollution alert, the interaction through the national health information infrastructure will allow information to be given into an individual’s personal health records, individual alerts to be given and decision making to occur.
Population health data and resources statutorily authorize data in public health systems and health care provider and other data. It allows the system to use information in the least identifiable way, to generate the kinds of information that are important to provide health and health care.
No single database of public health information would exist, information would be brought together and aggregated in such a way that individuals cannot be identified. It would allow improved vertical and horizontal integration of federal, state and local systems through electronic data exchange, and data aggregated by geographical area demographic element and health event, et cetera.
The core data would be a standard, core data of health of community members, health care community members, community health resources and community health such as receipt of preventive care. It would provide a continuum of access. On one end, individual identifiable data by authorized public health workers, such as we currently collect for individuals who have sexually transmitted diseases and other communicable diseases, for which there is a purpose for a public health system to have an intervention. You can’t help John Jones who has tuberculosis to make sure they are getting directly observable therapy if you don’t know who John Jones is and where John Jones lives.
Across an interim, view identifiable data for population health research, with institutional review board approval and where appropriate, consent by those involved in those studies, with some data on a population base that is de-identified and will be used for various other kinds of studies.
In public health, we need to know trends amongst individuals with cancer. We don’t need to know who has cancer, but we need to know where there are clusters of cancers to be able to evaluate what is going on in communities. At the farthest end of the spectrum, public access to anonymous aggregated data.
That creates certain challenges within that, privacy and confidentiality, issues requiring resolution such as anonymized linkage variables. We don’t always need to create a database that has identifiers in it to link separate pieces available. You may have a meta database that tells where the various pieces are together, and you create that warehouse, the data that is de-identified, but in fact links from various different databases.
Access, continuum, decision roles, cell size and of course IRB approval and review.
The national health information infrastructure has two requirements. First of all, it is not just a technology. It requires privacy protections and standards and business models, attitudes, and we have to assure equity of access and equity of quality.
Our next steps in moving forward in this vision is a validation of the vision of which this is the first step, to identify the barriers, of which we know privacy protection is a major one. The need for standards, holes in the technology, and the process of development of recommendations.
Over the next six months, we will have four hearings, in Chicago in July, San Francisco in October, and two other locations around the country. This will allow us to pull together a final document which we will make recommendations. Then comes the hard part. As Don says, we’ll be done with the talking and it will be time for the doing.
The time for us to pull this together is now. Every single day, someone is making and developing a new health information system. To the extent that we allow this vision to be brought forth, that we begin to create the standards and the technologies, those investments will be meaningful, and the cost of change will be less. The longer we wait, the more expensive it becomes.
Working together, we can formulate this vision. We can refine it, and by expanding this vision, we can assure the future for health care.
Thank you.
DR. SONDIK: Thank you very much, Dr. Lumpkin. It is a great pleasure to introduce Ted Shortliffe. Ted is professor and chair of medical informatics at Colombia University. He is also professor of medicine and computer science at Columbia. Until recently, he was at Stanford University and probably just smelling all those eucalyptus all the time just got too much for you, I’m sure.
Ted also serves on the Presidential Information Technology Advisory Committee, and is a member of the IOM. He recently chaired the National Research Council’s study committee that led to the report, Networking Health: Prescriptions for the Internet.
I can’t think of a person more appropriate to react to this talk. It is a great pleasure to have you, and I want to thank you actually for also having spoken with the Data Council earlier this year.
DR. SHORTLIFFE: Thank you. If you were expecting in my reaction that I would have a slew of criticisms in the model that you have just been hearing about, I’m afraid I will disappoint. It seems to me that this report and the call for the national health information infrastructure that follows very much along the directions that John has just outlined is exactly right.
But that doesn’t mean I have nothing to say. In fact, there are several things that say that complement the general thrust of what he said, and play to my interests and background a little bit.
The other day, maybe it’s about six months ago, I saw one of those little squares in the bottom of USA Today, where they have the little graphs. Every issue of USA Today has got some composite data summary of something in modern life. This one had to do with the years it had taken for the introduction of technologies to reach 50 million users.
Radio had taken 78 years, and television had taken 25 years, and the World Wide Web had taken four years, the claim, as a way of trying to argue the incredible impact and the rapidity of the impact of the Internet.
Although it is clear to all of us that the Internet over the last 10 years and the Web in particular have changed our way of thinking about the future, — and without emphasizing it, John, I think the Internet is interwoven through so much of what you envision for the future of the national health information infrastructure. You could argue that the next generation Internet’s substantiation for health care is the national health information infrastructure. It would be hard to imagine doing the kind of things that have been described without taking advantage of the Internet and what it will become.
But I think that this four year figure fails to recognize what it really took to get to where we are today. So if you will indulge me, I would like to give a little bit of history, that plays to my own personal experiences and anxieties and frustrations over the last couple of decades.
My knowledge of the Internet, believe it or not, starts in 1969, when I was an undergraduate of Harvard, in the heyday of the protests regarding the Vietnam War. Almost daily articles in the Harvard Crimson about the military-industrial complex. Some of you are nodding your heads, and remembering that era.
There was a story that was in the student newspaper about this outrageous plan that the Department of Defense had to actually link military computers around the United States, using some kind of networking technology. It wasn’t clear from the story, or I’m not sure in anybody’s mind at the time, exactly what would happen when all these computers were linked, but the idea of linking them and developing a technology to allow them to be linked was just coming out.
I remember some protests and some arguments that this was yet another evil plot at the time. It really got quite a bit of local upheaval in Cambridge, for a couple of weeks, and then it died off and other things came along.
So it was with some interest a year or so later that I arrived out at Stanford where I did my MD-Ph.D training, and my Ph.D work was in computer science. I found myself using computers that lo and behold were tied to this network, which by that time was called the ARPA network, and was named for the branch of the Department of Defense, the Advanced Research Projects Agency, which in fact had funded this early work to try to develop new technologies that would allow communication between computers at great distances, so-called wide area networking.
We sometimes forget I think that wide-area networking of this sort actually predated the kinds of local networks we have within our office buildings and medical centers and the like. We started this wide access approach.
So when you say that the World Wide Web took four years to reach 50 million users, you are forgetting that it was totally dependent upon a national infrastructure which had taken a full 30 years to develop. It really was more like 34 years to get 50 million users of the Internet. If you consider the substantial investment by the government, I might add, before the Internet was fully privatized in 1995, when the NSF turned over the last portions of the Internet to the private sector.
In 1973, when I first became a heavy user of the ARPA net, there were 63 hosts on the ARPA net. An address had no dot-com or dot-edu or anything like that; it had a machine name, and most of us knew all 63 machines. There were 63 machines because that is how many you could address with one byte or whatever it was; it was a power of two with a zero on the bottom. Eventually it became clear that they needed to change the way the addressing occurred.
Following over time there has been these wonderful opportunities for innovation and creativity, and then for cooperation of an incredible sort, that is necessary in order to pull off this kind of national network. Why does the Internet work today? No one runs it anymore. It is because standards have been gradually adopted, mechanisms have been put in place that allow consensus to develop and finally for folks to buy in and participate in the standards as they have evolved. Something has worked right in the process that has allowed us to get to the situation we are in now. This infrastructure is almost invisible to us. USA Today could talk about the Web as though it was the phenomenon, whereas it was really only something sitting on top of a much deeper phenomenon that went on for much longer.
Now, host number 56 on the ARPA net was a medical computer actually. It was at Stanford. So the biomedical community has been involved in the wide-area networking world in some form since 1973. So it is not like we weren’t involved in one way or another, the biomedical research community and especially the medical computing research community has been part of the Internet world.
But as one who was using it and sending e-mail in 1973, I can tell you that it was a source of considerable frustration to me personally that as I grew up as a physician and as a medical researcher, that I watched my colleagues in the computer science community begin to really take advantage of this technology. Then it was all of engineering, then it was the main campuses in the universities, and you were seeing it touching on the social sciences, in mathematics, and folks that weren’t computer types at all beginning to utilize the ARPA net for communication and for sharing files and for doing collaborative research, and the health care community just didn’t get it. We were not logging in, we were not installing networks in our health centers.
The academic medical centers did begin to get connected and get involved. The only reason they did was not because they were willing to make investments in connecting to the Internet themselves, but because they were on campuses with universities that saw the importance of such connections, and essentially all the early academic medical center connections to the Internet were through their university connections. In other words, all they had to do was hang a wire out over to the main campus and then they could get on the Internet. They didn’t need to buy their own connections.
All the government research labs were connected to the Internet in those days. I’m talking about the national laboratories, Oak Ridge and Lawrence Livermore and so forth, Rand Corporation, SRI International, all of them connected. But biomedical researchers were not connected, except for the few of us that did computing work. The folks that were doing molecular biology and genetics and so forth — the NIH was very late in getting connected to the Internet.
Why am I telling this story? It is because fundamentally we are talking about cultural change. There is a culture in the biomedical and health world which has led to some resistance to this kind of adoption and involvement in this kind of technology, which we have not seen in all segments of society. We are definitely slower to get it.
So we need to try to ask ourselves what it is about the culture of biomedicine and biomedical research, medicine in general and health in general, that has made it difficult for information technology, by which I mean both computing and communication, to be embraced, to be understood, to be invested in, and therefore to have the impact on these activities with the kind of rapidity that has been achieved in other scientific endeavors.
In 1989, I was asked to give a keynote address at the big medical computing meeting, called the Symposium on Computer Applications in Medical Care. I decided that what I was going to do is talk about networking and what it could do for health care. I tried to talk both about local networking and how you could begin to do distributed computing within hospitals, and connect physicians into centralized computers and hospitals and the like, and then to talk about the Internet, and access to Medline and integration with public health databases and surveillance and so forth could be pursued.
But then when I got to the part about how to actually make it all happen, I realized it is very difficult. Here we were, even by 1989, with an airline reservations system where six or eight or 10 major companies had managed to get together and agree on standards that allowed them to do uniform scheduling of seats on an airplane. They needed it for their business. They knew it was important, and they therefore got together and figured out how to do it. But they weren’t nearly as fragmented as the health care industry then or now. There weren’t as many players. There wasn’t as much fear related to the financial pressures and the changing modes of practice and the like that right now so dominate the thinking of decision makers in health care.
So during the 1990s, we saw a rapid proliferation of the Internet because of the Web, which clearly had a huge impact. But the Web is simply an application that sits on top of the Internet, right? It built on that 30 year old technology. There was no chance that the Internet that the Web was built upon would have existed without government investment. If you talk to folks from the telecommunications industry, they will tell you, there was no way we were ever going to do anything that speculative in 1970. After all, it took 30 years to be ready for commercialization. What company would invest in a 30-year research plan? It took that kind of government role.
So how can we optimally do the kind of things that John was talking about — integrate and use this remarkable technology as the infrastructure to build on, in much the same way the Web has built upon that technology, except now we can go up another level and build on the Web as well.
Well, it has to do as much — you had a list of requirements. There was one that had to do with attitudes and culture. I think that is a part of it. All the other ones are related to that. They will fall if people begin to really understand and appreciate the potential that is here.
How does one communicate a vision, how does one make CEOs of hospitals worry about the fact that they are only spending two percent of their annual budgets on information technology, when they view it as $20 million a year? American Airlines is investing 18 percent. Somehow or another, in an information technology rich field, we haven’t quite become part of the mind set that says information technology is the key to the strategic positioning of organizations in health care.
So today, despite what has happened in the last decade, most of the health care computing that we see on the Internet has to do with consumer health, because the consumers have run with it. The sites have begun to be created that will meet their needs. But when it comes to integration with practice, it is minimal. There is still loads of redundancy in data entry and data management going on in the health care system. There is still lots of paper. It is not all the Internet; a lot of this has to do with the information technology within institutions and organizations, too.
So you heard about this report that we did for the National Research Council, that has to do with the Internet and its future. We were asked in particular to try to address what kind of technological research in the Internet could the health care community best contribute to, to try to make sure that we are on track, and biomedically sensitive kinds of applications are made possible and decisions are made in design to facilitate the proper application of the Internet in health care.
This leads to a broader issue of the role of the health care community in supporting information technology research in general, of which the Internet is one example. But many of the other examples on John’s slides that have to do with data standardization, database issues, wireless computing, we could list a whole panoply of topics where health care has very special interests, security, privacy. How do we get those better integrated into our medical research world?
A good example of this issue came out from the Presidential Information Technology Advisory Committee, which you heard reference to, and it was on Don Detmer’s slides as PITAC; that is the acronym for this organization.
It is advisory to the Office of Science and Technology of the President and to the White House, and it includes academics and business folks from mostly the information technology world, who are being asked explicitly to address issues related to research, and the kind of research that needs to be supported by the federal research community, including all the agencies. There is no single agency that supports computer science or communications research. It is very spread out across multiple agencies. All the agencies that support computer science generally support it because they see that the problems being worked on are potentially important to their long-range mission. It is for that reason that the Department of Defense funded that early ARPA network, for example, in the late ‘60s.
So an issue that is identified as being one of the hot topics for the computer scientist to work on right now is large scale system integration, how to really deal with the scaling problems of large complex systems. The interesting thing is that the problems they work on in the prototype large-scale environments that I have seen don’t come close to being as complicated as the one we live in every day in health care. What could be a more challenging large-scale integration problem than the health care computing environment, where you are dealing with multiple organizations, employees that don’t work for the same organization that provides the computing. That is a big difference from the airlines, right? You work for United, they tell you, use that system. If you are a physician in a community hospital, you decide whether you are going to use the system generally. It is a little different world.
So there is an example of a topic that the biomedical research community has great interest in seeing resolved. At least, the health care practitioner community and therefore that is what biomedical research might well be supporting. But there are essentially no major research programs in that kind of basic research being funded out of any of the health agencies that do research funding.
There is even a little bit of resentment in the other agencies that do research funding about the way in which the health community has tended to ride on the shoulders of research investment by the other agencies that have funded computer science — NASA, DoD, DOE, Department of Commerce to a certain extent, and of course, NSF.
So I would like to make a claim. It has to do with the intersection of research with culture and attitudes. That is, excellent information technology research, some of it basic technology development, if it is motivated by biomedical and health goals, should also be viewed as excellent biomedical research. It is biomedical research. All biomedical research is not test tubes and gels.
The relevance to the biomedical community is not simply whether new technology serves biomedical research. It is what it does for health care, and whether it has substantial challenges associated in the way of innovation.
But the kind of cultural changes that I am talking about occur slowly, and therefore it is an important role for government leadership, it seems to me. For example, in education, the government has a role in all those areas.
So we need mechanisms to be put in place. I therefore support the kind of comments Don Detmer made about the need for both a road map and a designated driver. It seems to me that the activities need to be very broadly viewed. They need to be across the agencies within HHS.
So we need a grand vision. We need an architect of some sort, or at least an architectural mechanism to help make it happen, and to represent information technology and the strategic vision of how it can influence the way in which we care for patients long term and promote health long term, in planning processes within the department.
So in identifying these three groups that John just summarized, the patient, the practitioner and the community dimensions to this future of the health information infrastructure, implied it seemed to me was the role of government in supporting what happens for all three of those groups.
The NCVHS is well positioned to try to help the government figure out what its role is in supporting those activities as seen from all three of those areas. The role of the NIH, the National Library of Medicine, the Agency for Health Care Research and Quality as key research agencies, but other key agencies that have a real interest in this as well, and the need to be coordinated and to have some of the shared vision that is involved, the CDC, FDA, HCFA, HRSA. So this is technology that needs to involve HHS as a whole.
The Data Council is a wonderful move in this direction. Now it is how to keep that process moving. That seems to me is the challenge for the Department and for NCVHS as well.
The report comments on your next stages, the hearings. I think the hearings are laudable, in that they will help build awareness and they will help educate, but I would suggest you won’t hear very much you haven’t already heard at those hearings. You already know what the issues are to a large extent and what the problems are. In parallel, maybe now is the time to start working on getting those well communicated and influencing if not this Administration, the one that is about to appear in a few months.
Now is the time to start the federal process as outlined by Don, it seems to me, Don Detmer, to press for recognition of the role of biomedical research and that community should play in pushing the frontiers of the national health information infrastructure in the next generation Internet, and pushing the frontiers of the information technology research in general, where we have a responsibility that I think we have under emphasized in the past and need to start looking to in the future.
Thank you.
DR. SONDIK: Let me suggest that we take questions after the break, after the second series of talks. We will consider both presentations together at the end. Keep in mind Don Detmer’s words towards action, and where we can go.
So let’s take a break until 3:30.
(Brief recess.)
DR. SONDIK: In this session we turn from the infrastructure to a vision for health statistics. I hope we have some time to discuss how we realize both the infrastructure and the vision.
It is my pleasure to introduce Dan Friedman, who is the Assistant Commission in the Bureau of Health Statistics research and evaluation for Massachusetts. Dan took the reins on this idea, and on behalf of the committee and with the full support and involvement of the committee members, has been instrumental in moving this idea of a vision forward. It is a great pleasure to introduce Dan Friedman.
DR. FRIEDMAN: Thanks, Ed. Ed asked me to speak for 12 minutes, and I am always eager to be cooperative. So I am going to try to speak for 12 and not 11 and not 13.
I have a couple of young kids, a six year old daughter and an eight year old daughter. I have been wanting to bring them to Washington. My notion of bringing them to Washington, my image, is having them stand in front of the White House gate with those big cardboard cutouts of the President and the First Lady. They are not only young kids, they are also really diminutive kids, so they are going to be dwarfed by the cardboard cutouts.
In coming over here today, I was thinking that that is in some ways the way I feel. I feel that I should be accompanied up here by cardboard cutouts of my colleagues on the National Committee, Barbara Starfield and Paul Newachek, Susan Canaan, our writer, our colleagues at the HHS Data Council, who have worked on this, Jim Scanlon and of course Ed Sondik, as well as the host of people at the National Center for Health Statistics, Marjorie Greenberg, Ed Hunter, Jerry Hendershot, Lisa Breuchtman and others, who have also contributed so heavily to this process.
I am going to speak about three things today: what we are doing, what we have learned and what we are going to do next. I am going to try to keep to my 12 minutes.
What we are doing. Starting in 1999, we engaged in a process at Ed Sondik’s request that was a collaborative process between NCVHS, NCHS and the HHS Data Council to develop a 21st century vision for health statistics. We have tried to think of that vision in terms of developing a vision that reflects all manifestations of health and health care, as well as a broad concept of what contributes to health.
We wanted to encompass population health and the health care system and the transactions between the two, and we wanted to address the relationship and the potential synergy between local, state and national data as well as especially between public and private data.
A second goal of our process has been to describe and define the disciplines, components and resources and other elements needed to implement that vision. I completely agree — as Ed Sondik has said, I completely agree with Don Detmer’s comment that in some ways, that is going to be the really difficult part.
A third goal of our process has been to set forth criteria, hopefully clear criteria, for evaluating the health statistics system of today and the health statistic system as it will be later in this century.
Imagine five phases. So much for Power Point. Why don’t we turn it off, and you’ll have to bear with me without Power Point.
The process that we have engaged in consists of roughly five phases. We are in the process today of completing the second phase. The first phase has been learning from the past and the present.
As part of that, we have commissioned a series of papers from American and Canadian experts. We have had a series of consultative sessions here in Washington and at least for me, especially important, consultative sessions in Harrisburg, in New Orleans, in Albuquerque. The local consultative sessions I think have been more than just a question of reaffirming what we already know. I think we have really learned some new emphases and some things that we especially need to focus on.
As was mentioned earlier, we also had a National Academy of Sciences CMSTAD workshop in November that was especially helpful, and we have presented our notions of where we are going throughout the process at a series of forums at professional meetings.
A second phase is this interim report, and the feedback that we are going to receive on the interim report via the Web and via professional meetings.
What we have learned. The first thing that we have learned is what health statistics are. One of the things that has been interesting to me throughout this process is, in looking through the public health literature, with one exception we were unable to find a single definition of health statistics. Interestingly enough, the one exception to that was in John Last’s Dictionary of Epidemiology. John Last teaches at the University of Iowa, and he is Canadian.
One of the reasons why that we have been having some difficulty developing the vision of health statistics, developing a unified notion of where we should be going is that we have not reached anything resembling a consensus or even a common understanding of what health statistics are.
The definition that we have been working from in the committee is not John Last’s definition, but a definition that we have developed, which is, health data that can be generalized to a known population of individuals, events, organizations or institutions that can be used to design, implement, monitor and evaluate specific health programs and policies, and that properly organized and communicated can enable us to assess local or national health in order to mobilize and improve it, and to evaluate the success of those efforts.
In other words, our definition of health statistics is very much health policy and health program based. Essentially what we are saying is that health statistics exist in order to effect health, health policy and health programs.
One of the things that we have heard consistently throughout our consultative process has been that while American health statistics has achieved a great deal and has made a major contribution, American health statistics has also been marked by substantial unrealized potential.
There have been three potential reasons for that unrealized potential that people have pointed to consistently. The first reason has been our lack of having squarely dealt with privacy and confidentiality issues around health statistics. We have started to deal with those issues, obviously around research, we have started to deal with those issues around health care transactions, around medical records privacy, but we have not dealt with them nearly as squarely in terms of unique health statistics issues.
A second reason for unrealized potential is obviously our problem of multiple data silos, public and private data silos, categorical program data silos and local, state and national data silos, without the chutes between those silos to connect them.
A third reason for unrealized potential that we have been hearing quite consistently is the lack of a consistent organizing structure, both conceptually as well as organizationally. These three factors have resulted in unrealized potential, both in terms of a lack of information, a surfeit of data but a dearth of information, as well as multiple unanswered questions, both at the macro level, and one of the things that we consistently hear from across our northern border is for example the importance of trying to determine what is the impact of health care on health.
That is not a question that we have been any more successful in answering than our neighbors in Canada, and we also have had difficulty in answering a host of more micro level questions, ranging from the magnitude of adverse effects to essentially trying to find population-based answers to what makes sense in terms of relative investments in different kinds of primary and specialty care.
Something else that we have learned has been a series of what at this point are very rough-cut principles for developing a much more finely tuned 21st century vision for health statistics. The first part of that principle — the first principle has to focus on privacy, confidentiality and security.
Here, we don’t have so much a principle as a marker, as a bookmark for saying we need to very directly deal with those issues and with the issues that are unique to health statistics, rather than dealing with them only as reflections of medical records privacy or research privacy issues.
A second principle is the need for an overarching conceptual framework encompassing all aspects of the population’s health and interactions between the population and the health care system.
John Lumpkin earlier flashed up the Evanstan Stoddard model. One of the things that strikes me when I talk to folks in Canada about their health information policy approach, when I talk to folks in Australia about theirs, is the fact that regardless of where people are in their respective hearts with those countries’ systems. They returned to common conceptual model for discussing it. That is something that we have not had in this country. That is something that we very much feel that we need to develop.
A third principle is that a health statistics system of the 21st century needs to have the flexibility to identify and address emergent health issues and health needs of the population. Many of our systems have been based upon — that we’re going to have correct information on TB cases and AIDS cases, for example. But our systems are not well suited at this point for trying to identify emergent issues or new issues as they come up.
A fourth principle is usefulness at different levels of aggregation. Right now, we have in many although certainly not all of our data systems what I think of as a series of ping pong balls, which can’t build on each other. The local and the state and the national data systems build on each other to a limited extent, and we really need to reassess that, and for all of the data systems to say where do we need the local data, where do we need the state data and where can we roll it up.
A fifth principle is our need for compatible standards serving multiple purposes, not just standard for health care transactions, not just standards for computerized medical records, but standards that can cut across health statistics, public health surveillance, transactions and electronic medical records.
The sixth principle that we have been working on, that we will be working from is, to the extent possible, unitary Data Council for multiple purposes, not necessarily as we now have, completely separate data streams from the same providers, completely separate data streams, for example, from a hospital for transaction data, for medical records data, for financial data, for surveillance data and for health statistics data, but somehow or another trying to unite those data streams to reduce the health care provider data burden.
A seventh principle is maximum access and ease of use. If health statistics are largely going to be defined by their relevance for public policy and health programs, we need to be able to provide those data easily and in a way where end users at the local community level can actually use the data to identify need of public health interventions and to develop public health polices, not just at the national level, not just at the state level, but at the local level as well.
Eighth, as Don was talking about, adequate and well-managed resources. I think there are substantive reasons for having the kind of public process that we have started engaging in. There are also political reasons. One of the reasons, I believe, why countries like Australia and Canada have been successful in garnering resources, and considering the population sizes of those countries, really dwarfs the resources that we have spent in a unified way on this is that they have had public processes to garner support for health information, that have been careful and they have been inclusive.
A ninth principle obviously is policy relevance. Our tenth principle is broad collaboration, private and public collaboration, as well as collaboration across different levels of government.
What is next? There are three phases that we still think of in terms of the remainder of our process. The first phase is developing the final vision, starting with the regional hearings that John Lumpkin referred to, building upon that through comments on the Web and at professional meetings, mapping the current system against the conceptual framework, so we have some sense of where our data needs are at a much greater level of specificity than we do now, and trying to devote some concerted effort to delineating the health statistics and the population health information specific privacy, security and confidentiality concerns, and dealing with those directly, rather than in some ways trying to dance around them and sidestep them and duck them, as in some ways we have been doing.
Our fourth phase is the final vision report that will hopefully be available in around a year. The fifth and final phase, and one that I am most concerned about, and I share Don and John and Ed Sondik’s concern, is federal, state and local planning around population health information, and trying to move from a broad scaled vision to actual planning in agencies, in public and private agencies that deal with population health information.
Thank you. Next time I’ll come with slides.
DR. SONDIK: I was just going to say, we’re going to try this again. I once had an experience in this very room, I hate to say, with slides, that worked fine out in Hyattsville, and I came in here and there was nothing. But in that case, I think the problem was unfortunately my electronics, so I hate to say that.
It is a great pleasure for me to introduce Dorothy Rice, who as someone mentioned earlier, was a former director of the National Center for Health Statistics from 1976 to 1982. She is professor emeritus for the Institute for Health and Aging, University of California in San Francisco.
Dorothy has done many things. And by the way, all these biographies are in the back of this wonderful program, so you all should read these at your leisure. I just wanted to mention one thing about Dorothy Rice, though. She has done many, many things in her illustrious career, but perhaps the thing you are not known for most is the thing that has probably had the most effect, which has been your work in the analysis of the effects of smoking and the costs of smoking. If there is anything that will have an effect for generations, it is that work. So congratulations to you on that, and thank you for being here.
DR. RICE: Thank you. I want you to know that just in case, I also brought along some overheads. You just never know, but they seem to be coming on, and we’ll hope for the rest.
It really is a pleasure to be here. I want to thank Ed for those kind remarks. You are right, it has been one of the most exciting areas of work that I have been involved in, in estimating the cost of smoking. I think it was a big part of the big settlement and the cases against the tobacco industry. It was great. I am still continuing in that area. I just got two new grants.
I want to thank the National Committee on Vital and Health Statistics for inviting me on this special occasion, the celebration of the 50th anniversary of the National Committee on Vital and Health Statistics. In my day, Carl White was the chair of that committee, that I relied on very, very much. We did some good things then, as the committee is doing now, and has in the interim period. I also want to thank the Committee on National Statistics, of which I have been a member for collaborating with NCVHS. Also, I want to thank Ed Sondik, because he came to me some time ago with this idea of doing something, thinking about this vision for the 21st century, and how to approach this important subject. He really has been dedicated and committed to developing this vision for the health statistics in the 21st century.
Here we are, looking at the interim report that is very, very important. It is the beginning of hopefully a very fruitful effort to get at what we can anticipate in the future.
I have no arguments with any of the interim report. There really is a comprehensive — and the principles are clear, as Dan has set forth, in setting forth what we should be thinking about in anticipation of the 21st century.
But I also want to add — and they have been incorporated, the comments that I have, particularly on the cross-cutting — on the glaring gaps that I think are most important, and the cross-cutting issues. They are all in the report, but I just want to highlight some of the important things that I think need to be addressed, and talk about the next steps.
In terms of the gaps, as you heard, the health statistics for population groups and minorities — and sub-population groups do include children. Barbara criticized me in the past for leaving out kids, so I just wanted to mention them, even though they are not listed there, because I really do think it is important that we get better information about children.
I think that the efforts that continue to — that we have engaged in to reduce the health disparities among the special population groups of low income people, of racial and ethnic minorities. It is recognized that data are needed to monitor our progress toward eliminating these disparities. Except for the data derived from the Decennial census and the vital registration system, the births and deaths, the existing sources of health data really don’t permit examination of socioeconomic differences for any but the three largest race and ethnic categories.
When you come to California and look around, you realize that there are so many different cultures that are subsumed particularly in this Asian and Pacific Islander, that we need to have better data. I know there has been some concern about the Census and I am really annoyed about the criticism of the long form in particular that we have heard about on various TV programs. it really is unconscionable for them to do so. I have written to them, but I don’t get any response. I try.
I think that one of the gaps that I feel that we have is doing more on longitudinal data. The cross-sectional surveys that we have are good, they are important, but I think that there is a need for more longitudinal data.
There are some efforts that we had. We had NHANES-1 epidemiological followup study, the LSOA, the longitudinal study on aging, the NHIS disability supplement, the medical expenditure panel survey, the Medicare current beneficiary survey. Those are really very short term longitudinal data collection efforts. We should start with children, Barbara, and follow them through their life. Really, that is terribly important. I’d like to live to that day when we can start some pilot projects on that and move forward on that.
I think that if you look at — there are some 62 reports that the Institute of Medicine and National Academy of Sciences have produced since 1985. I looked through all those recommendations of those 62 reports, and 10 of them specifically recommended the need for longitudinal studies. I think everyone clearly feels that this is an important area that needs to be addressed further.
The organization of our health delivery system is a pluralistic system, there is no question. The public and private sectors are involved, and the growth has been enormous. If you just go back to 1965, we spent $41 billion on health care, and it was 5.7 percent of GDP, Gross Domestic Product. In 1998, the health care expenditures in the United States totalled $1.1 trillion, trillion dollars. I wonder how many zeroes that is, but it is enormous. It was 13 and a half percent of the nation’s GDP.
So this pluralistic health economy really has presented a challenge, if you will, but also special problems in data collection analysis and dissemination. It has grown so rapidly in the whole industry and the expansion of private health insurance and public health programs. Every one of these programs needs data, without question. So we have just built one data system upon another data system, and each agency in the federal government has to have their own, and there isn’t enough common sharing of the data.
We have private organizations and health professionals, health service providers, health insurance and many others have important interests in collection and use of health data.
I think that the — as pointed out in the vision document, the fragmentation of health care delivery today makes it essential to have integrated, effective statistical systems in order to better understand the health care system and how people face it.
I think an example is HMOs. We now have 78 million people — probably more, because those were figures from 1998 — 78 million people, about 30 percent of the population are now in HMOs. That has resulted in some problems and gaps for health statistics, because encounter and visit data are not uniformly reported for the population covered under these capitation systems.
In the area of outcomes and quality, little information is available on outcomes and quality. I know that AHRQ is now concerned about evidence based data and information on outcomes. I think that we really have to do a better job in measuring the quality of care. We have the HETUS and the QUALY’s and the DALY’s and many variations, if any of you have used these measures. No uniform definitions that we really can rely on that would apply across the board. Maybe we can’t have them, but we should try to do a better job than we have now.
I’m not going to go through all of them, but you can see from the federal point of view a listing of the health status and health utilization and cost surveys that are conducted by the various agencies, many of them from NCHS, but also from the Agency for Health Quality and Research, and Medicare and Bureau of Labor Statistics, the current population survey, from which we get from the marked supplement information on health insurance coverage.
It seems really unbelievable to me that we have to go to a report on unemployment in the United States for our information on health insurance coverage, but we do.
We also have in addition to all the federal surveys that I have listed, we also have the states. The states and private data systems abound in each of the 50 states. The private sector maintains data systems and conduct many surveys of hospitals, health professionals and health care organizations. The private health sector includes organizations also of health professionals and health providers.
But the quality of those data sources is questionable. I think the results of the statistical efforts are duplicative and overlapping data systems in both the public and private sectors.
Take hospital and patient data, for example. You have the American Hospital Association, you have all these private data systems, Blue Cross, that are producing hospital data. In addition, you get data from the national hospital discharge survey, you get data from various provider surveys conducted by NCHS, for example. So we see an overlapping duplicative set of data systems in public and private sectors, and this is a matter of great concern that I think the vision document did address.
I want to give you — it’s not all bad maybe, but here is a good example of some collaborative efforts in our state in California that we are very excited about, the California health interview survey which is modeled after the national health interview survey.
This is a successful effort in process. It is a collaboration of the Department of Health Services, the UCLA Center for Health Policy and the Public Health Institute in Berkeley. The estimated cost is $10.8 million in the first cycle, including data collection by a survey contractor, analysis and dissemination. In the state of California, the National Cancer Institute and CDC have together committed necessary funding for the first cycle of the California health interview survey.
I think there are some things going on in Massachusetts too, and there are other states, I’m sure, of which I am not aware. But there are now instances of collaboration between the states, the academics and the national government that I think is a model that will be used in the future.
I won’t talk about privacy protection and confidentiality. I think that we have covered that quite clearly. In the area of data sharing and data linkage, that also is covered in the vision document that is an aspect of privacy and confidentiality, and hopefully we will finally get some legislation at the national level, where data sharing between the agencies will be accomplished.
It has been a difficult area over the years. It goes way back to when I was involved at the federal level, and hopefully we will be able to move forward in this area of linkages.
I think that the standardization of data elements has been with us for ages, uniform data sets. The National Committee on Vital and Health Statistics has been addressing it for many, many years, and continues to address it, and we have to really continue our progress in this area.
What I really want to emphasize is that we have to invest in health statistics. I think there is a lack of commitment and judicious political commitment, without question, in quality health statistics in the public and private sectors.
Don, you had said that you needed a designated driver, but you know what? You left out NCHS on that list, and that got me mad, because NCHS should be the designated driver in this area. I have no question about it, and there shouldn’t be any question in anyone else’s mind about it. I guess I’m partial.
But I think it really is something that needs to be addressed, and it is part of the next steps. We really do — I think the hearings are good. Hopefully you will pick up some other areas that will help, but mainly getting out to the regions and the areas is very important to make people more conscious of the importance of health statistics. I think that in addition, we have to reassess the core programs of the federal statistical systems and continue to develop and promulgate standards. I think that we have to collaborate, do some more collaboration between the public and private sectors and particularly get them to invest in quality health statistics.
But again, I want to emphasize the resources in considering the future prospects for improved health statistics. To meet the needs of the 21st century, we must recognize the resources will not grow parallel to the demands for data and services. The demands for health data are greater than our ability to produce them, and there are budgetary pressures that require an assessment of current data collection and dissemination procedures, and statistical agencies must make choices often between data collection and research and analysis among the needed data sets.
What I think we need is a constituency. God, I’ve been saying that for years, looking for a constituency that will say — I remember, it was Carl White that said we don’t die of health statistics, we die of a disease. Somehow, we just haven’t been able to get the various constituencies out there to provide the political will to enhance our resources that is so important in this area.
We need a cultural change to invest in the 21st century for the vision for health statistics in the 21 century as we move closer to our objective of a national and systematic approach to meeting the information needs for health policy development and program evaluation. We also need to coordinate our data collection activities, both within the federal establishment and between the government and the private sector.
Although I think progress has been made in coordination, we must continue to avoid unnecessary and costly duplication, encourage comparability of information collected by different systems. One of the principles is, collect data that can be used by multiple sources, and to use the ongoing data collection programs to provide specific information for many organizations. More effort is needed to provide essential data yet reduce the burden on the individual and institutional respondents. We must develop, articulate and implement a 21st century for health statistics.
Thanks to the National Committee on Vital and Health Statistics for the past 50 years. I hope I’m around for a few more of those years, but I know you will continue on into the future.
Thank you.
DR. SONDIK: One slide that Dorothy showed struck me in particular, the slide that showed all of those various surveys. It is interesting to couple that with the Evanston and Stoddard model, if you will, and ask how much of that model are we really filling in. If you do that, it turns out not that much. It points out the importance of a conceptual framework.
It is very important in this effort that we look to other countries to understand how they have met all of these needs. I couldn’t be more pleased that we are able to welcome Gary Catlin here, one of my counterparts if you will from Stat Canada. Gary is the director of the health statistics division at Stat Canada. He assumed that position in 1997. Prior to that he was responsible for the development of the national population health survey.
He is also among other people here today involved in cancer registries, myself included, and I think that must be good training. At least I hope it is, and doesn’t make us too biased in one direction or another. He is co-chair of the Vital Statistics Council for Canada.
Gary, thank you for being here.
DR. CATLIN: I am very honored to be here with the people that have presented earlier in the day. I hope I can share with you a little bit of information from the Canadian perspective on some of the issues.
First of all, Dorothy Rice and I have been given a very difficult task of commenting on this report, which is an excellent report in my mind. Certainly as I read through it over the past couple of weeks, there is a lot of echoes to the very same issues that we were confronting in Canada over the last decade, and naturally are still confronting.
What I am going to try to do in the next 10 or 12 minutes is essentially try to describe the journal that we started on in Canada about 10 years ago and some of the initial stages in that process, and then look at some of the initiatives that we started in order to fill some of the gaps and address some of the issues that both Canada and the U.S. is trying to deal with, and then finally some of the lessons that we have learned in that process.
About 10 years ago, there was a national task force on health information. That task force recommended a health information coordinating council, and listed a lot of the very same issues around standards and person oriented or linked data. It also called for a GDP of health. It lamented the fact that we were spending so much information on health in the country. We had dozens of economic indicators, and yet we had no consistent health indicators that were available to be published on a regular basis.
That was a very influential report. It resulted in the formation of the Canadian Institute for Health Information, which I think could be described as our designated driver in terms of this whole area.
The CIHI is a nonprofit organization, independent organization in Canada that has a board of directors that both CIHI, the organization and the health statistics division at Statistics Canada both report to, their organization on a detailed basis and Statistics Canada for advice on generic direction, in terms of our health statistics program at Stats Canada. So that one recommendation from the national task force has been well received and has had a good influence.
We also started a national population health survey in the early ‘90s that began collection in 1994. That is a longitudinal health survey. It looks at the dynamics of health. It started being both cross sectional and longitudinal and has recently become solely longitudinal.
But prior to that, Canada did not have a regular national health survey. We had just sporadic health surveys that had been done over the years. So that again was another big initiative in those early ‘90s.
After a few years of getting those things in order, there was a series of cross-country consultations that were done. Both CIHI, Canadian Institute for Health Information, and Statistics Canada, we consulted very broadly across the country. That resulted in our health information road map.
In the ‘99 federal budget, we were given $95 million between Stats Canada and CIHI over a three to four year period, to implement or at least begin to implement the vision that was reflected in that road map. I think there are 38 projects that were funded, but it comes down to answering two questions, how healthy are Canadians and how healthy is the health care system. That is really the chorus that we keep going back to in terms of thinking about what we are supposed to be doing with this road map and how we are supposed to move into the future.
In addition to that, there was a health infrastructure report, which is very similar to the work that is taking a look at the technology and the electronic patient record. That work began in 1997. They changed their reference. They stopped calling the analogy from a road map to a blueprint, but they are basically trying to draw a blueprint of what the health information system is out there currently.
That report is supposed to be available by the end of this summer. In addition to that, they will have a two-year strategic plan that is going to move ahead.
We have heard privacy and confidentiality repeated a number of times. I don’t think it can be repeated too often. One of the initiatives out of that advisory committee on health information is to try to produce harmonized privacy legislation across the country. Many provinces in Canada have their own privacy legislation, so what we would like to do is have those all very similar across all the provinces and territories.
Again, while a number of provinces have already introduced some legislation, at least a proposal for harmonized legislation is due out this fall.
I’m not going to go through these priorities here. They are very similar to the priorities that you have listed in the documents that you have. I just put them up here so we could see the parallels, the common definitions across the continuum of care.
Person oriented information refers to creating longitudinal health histories, through the linkage of administrative data, and I’m going to talk a little more about that in a moment.
I wanted to say a few words specifically about privacy. In Canada as in the U.S., we are not suggesting that there will be any megadatabases of all health information held at the federal level. That is certainly contrary to the interests of the public and the privacy advocates in the country. All our linkages that are done — and this is at least at Statistics Canada — are linkages that are project specific. But we do need to build those databases, so that when the important projects are described and funded, we can pull the data together to address the problem that we need to address.
We have started to ask for consent to do linkages with some of our household surveys. The national population health survey back in 1994 was the first survey to attempt that, and it was surprising to most people that 95 percent of people consented to allow us to link our household survey information with their administrative data in the provinces.
We have started to develop some pilot projects using the combined household and survey data, which allows an enormous amount of analysis to be done. There is of course a threat that is perceived from the research community in terms of protecting privacy. There is always that balance between protection of privacy and access to information in the research community.
One thing that we’re doing to counteract that sense is to set up research data centers across the country in universities. These will be funded by the universities, and what they are is a little Stats Can premises on a university campus. So researchers with approved projects can get access to confidential data in the research within an environment that — they become employees of Statistics Canada for the purposes of looking at those data sets.
We also want to make it clear in all we do that the data that is produced through both CIHI and Statistics Canada is strictly for research purposes and statistical analysis and not for administrative data.
I want to mention our new community health survey. This is one of the major initiatives that has come out of the road map. We have heard a number of times today the need for local level or regional data at the state level or below the state level. This community health survey tries to respond to those needs.
It has a couple of characteristics about it that t are different than most national health surveys. It has got for Canada a very large sample size of 130,000 that will be collected over the course of a year.
What we built into this system is an ability for the regions to pick about 10 minutes out of a 35 or 40 minute interview, so that we built up contacts in each of the provinces and within 130 health regions within those provinces, and went out to them with a menu of about 30 or 40 modules that they could choose from to include in our national health survey for their particular region.
It was a little daunting when we started out, but we have ended up with about 25 different surveys, different variations that will begin collection in September. The other is a new area of looking at — in this survey is permission to link to health care records, which will again be included in this survey. That survey starts just a couple of month from now, in September.
The other initiative that according to our chief statistician will be the most influential part of the whole road map is the longitudinal health history or person oriented information, the linking of administrative data from hospitals, from the health surveys to the cancer registry and to the vital statistics information.
The first pilot projects that we have begun have been isolated to four or five provinces within Canada. They are linking things like people who present themselves in the hospital with an acute myocardial infarction, a heart attack. We followed those people for about two years after that, trying to look at both the rapidity with which they get some sort of care, whether or not they get surgery, for example, how quickly they get that surgery, and then we followed them for about two years to find out if that makes any difference in terms of mortality, both in hospital and out of hospital.
It has shown us certainly in the last few months of trying to put that paper together how complex this whole area is. You start to talk about surgery and actual clinical issues. The terrain is very bumpy and very complex, and a lot of care needs to be given before anything is released from this kind of investigation, so that wrong conclusions aren’t reached.
Another of the initiatives is reporting on health out of the road map. Both Statistics Canada and CIHI are committed to producing annual reports together, one led by Statistics Canada and one led by CIHI. The Stats Can report reports on the health of Canadians and the CIHI reports on the health of the health care system. The first reports were both released in March and April of this year, and were both very well received both in the policy sector as well as in the research community.
I think this is an important aspect of the vision for where you want to go with this information, having some regular reporting that needs to be done on the information that is produced.
All our indicators that are produced out of the road map and that will eventually populate these reports are intended to be at the health region level. We have talked today about the definition of health. There is no standard measurement of health status across health surveys anywhere in the world. We all agree on the WHO definition, but operationalizing that definition has been truly an impossible task up to now.
There is a conference that I know Ed and perhaps others from the U.S. are going to be coming to in Ottawa in the fall that we hope will start us on the road to coming up with some standardized approaches to measuring health disability and other aspects of health as well.
Another mention I would like to make on the reporting on health is an insight that was given to us by our former chief statistician about our reports. We tend in our reports to report our focus. The audience that we look at has been government, which isn’t surprising. They usually give us the money to do these kinds of things. And it has been the media and researchers. What we have to do is rethink how we produce some of this information with the public in mind, and what the public wants to know.
To some extent, I think it will make some of our reports easier to do, or at least more simple information presented in it, but it does change the way we have to think about how we produce some of our reports. That is certainly something we are going to strive for in the next couple of years with our reports.
I wanted to say a little bit about international comparability, given that I am here. We have — in the economics statistics field there are lots of internationally accepted economic indicators. It is quite amazing that in the health area, the ICD group of standardized disease and mortality categories are the only ones that are widely used and widely adopted across the world.
We need some of these in health survey. There has been U.S. leadership in these areas in the past through the international collaborative efforts. I think something that certainly Canada would like to work with the NCHS and others on is trying to push that into the area of health survey measures, and maybe this conference in the fall can be the first step towards that.
So a few of the lessons we have learned. It is a very short time; there aren’t a whole lot of lessons that we have learned in the last couple of years. But one of them is, we talk about standards, but the definition and agreement to standards doesn’t necessarily lead to standard collection. Some of my cancer registry experience is certainly coming to the fore in that area. You can agree on them. You have to worry about how they are implemented and when they are implemented, and reporting on them and getting feedback is the only way to do that. But standards certainly are a necessary precursor of getting the good quality data.
We need a cooperation of a vast number of interests. Even in doing our own consultations in Canada with 10 provinces and three territories, we found that after visiting all those provinces and territories several times, we were told that we weren’t meeting people at the right level in the organizations. So it takes almost constant consultation in order to move these issues ahead. It doesn’t take much to move them off the tracks.
I think the public — and something we have to talk to the public about in some fashion is what — they are amazed at what we don’t know, because they know that we collect information. In the health care sector, there are health statistics everywhere, but the fact that we can’t put them together — at least in Canada, we are only beginning to know that somebody has a heart attack in one hospital, and if they go into another hospital three or four months later, there is really not a whole lot of — that that information doesn’t move with them automatically. That amazes the public.
I think there is some public relations that we need to do around that whole area, to get them on site in terms of the vision and where we want to go with this.
I’ll finish up with the same thing that we have heard a few times today. We need investment in this, and we need some long term investment, in order to realize the vision. Certainly the $95 million that we received for three or four years is — we are anticipating that in our next budget we will hear some good news in order for us to carry on some of the work that we have started in Canada. I certainly hope you have the same success in doing that here in the U.S.
Thank you.
DR. SONDIK: I want to thank all the speakers. John, do we have time for a question or two? Yes.
PARTICIPANT: I have a comment. It is brief and I don’t know if it is on target. But I think that if you look at what has happened in the last few years, I think the market value of health has gone up. You look at what has happened with genetically modified foods, you look at what is happening with the genome coming on. You look at a lot of these issues; a lot of the public are getting interested in these matters. I think the environment and these things are really coming into their time.
So part of why I guess I was making the comments earlier in the day about having a road map and a budget and a direction and a plan is because I think we may very quickly find that investment is not the issue. If people are looking at us and saying, what are you going to do, why aren’t you relating this in terms that we can understand and that sort of thing, I think that is a fact. I think multinational corporations are starting to see that.
I may be wrong, but I think it is really quite possible that we could find ourselves on the opposite side of this conversation from what we thought we would be having sooner than maybe we might otherwise think.
I have really enjoyed all of this. Thanks.
DR. SONDIK: Thank you. Anyone else? I was struck in the course of this — I wrote down this little diagram that I think all of you have seen. It is not exactly a diagram, but the progression that we go from data to information to knowledge to wisdom. I was thinking in terms of resources. I think we do a very good job in talking about data and how that leads to information, and that there is value in that. But I don’t think we are doing such a good job, to say the least, in the next phases of that, important in showing how we go from information to knowledge from all of this, and then finally to wisdom.
I think we need to concentrate on that. It may be that this is a good time to do that, because of all of these advances that several of us talked about today. Yes?
PARTICIPANT: Information technology is going to pose dramatic changes for health statistics, in fact, dramatic changes for health care perhaps in the future. Primary physicians will be more focused on diagnoses that they can check with experts through computer technology.
There would be a lot of data generated as new forms are coming in digital data. The private sector will be able to manipulate that I think, far better than our public sector. Do we risk the fear of losing the relevance of our population-based surveys? Will we have a plethora of data in the private sector and be under the Gresham’s law of bad data driving out good?
DR. SONDIK: Which of our panelists would like to take that? Does bad data drive out good?
DR. COLTIN: I agree with your fear. I know when that national health information task force — they mostly talked about the use of administrative data. There was a discussion at least at that point that we didn’t need health surveys, you only needed the administrative data. Fortunately, good sense was successful in arguing that in order to look at the whole population and the people that don’t get care, you need the health survey information, and you need that to get the socioeconomic information and education information, all those pieces. So I think we should be aware of the threat.
DR. LUMPKIN: Myron, I think you raised a very interesting question, but it really gets to the heart of the matter of how we use population-based data. I have heard a description of epidemiology as being described as a science that is looking for a denominator. The difficulty is, when you use administrative data, when you use all sorts of other data, you don’t know what denominator you are trying to associate with your numerator. Particularly since in this country, while I believe we may have a plethora of administrative data, it may be even detailed data on clinical information, we lose track of what we mean when we talk about population data.
HMOs, which have a defined population, have a market; they don’t have a population. Those of us that are trying to define the health of communities find it very difficult to try to use information from any sort of source other than the kinds that we are describing, which are population-based.
So I think that the large potential for data to be collected only intensifies the need for there to be community based on region based or population based sources of collecting data and analyzing it to determine the health of a community.
DR. SONDIK: I would say we probably are going to see more in the way of repackaging of data and in a sense, reselling it, if you will, as the amount of information grows and grows. We are seeing some of that already on the Internet, for that matter, people who are putting sources together of one type or another. So the significant part is who the decision maker, who the user of this data is, is going to be critical.
DR. LUMPKIN: Just to carry that on one step further, because I happen to know how long my summary is going to be, and it is not going to be very long, was anybody shocked to find out that they were selling the Framingham data? Talking about moving towards the proprietary availability of data, I thought that was quite a shocking development.
DR. SONDIK: I must have missed something. I wasn’t aware of that. But the Heart Lung and Blood Institute I guess is short a little bit in its billion dollar budget? Is that the problem?
DR. LUMPKIN: I think they were bitter. I thought that they were bitter. I don’t know who was selling it.
PARTICIPANT: I’d like to make just one observation on my reading of the reports. I thought I noticed there was one group of stakeholders that were not represented, that might be an important group. Those are the basic scientists, particularly those in the genome project. Although we are talking about community based data, as the genome project fulfilled its mission, we then become much more aware that genetic specificity is going to become very important in predicting health and health outcome.
I think we have to start exploring the very complex issues of confidentiality, long term studies, comorbidities and things like that. They ought to be involved fairly early in the process.
DR. SONDIK: Actually, at NCVHS we have had some discussions about that. I think confidentiality issues tend to loom very high in this, as you just pointed out, but I think we have to look at it not as insurmountable barriers, and think about the goal of what we can get from using this data, and think creatively about it.
PARTICIPANT: As more and more patients go to the Web for information, I think they are going to want to know what does the information I am getting mean for me, but also relative to other people, how do I relate, what is my relative risk. If I take this drug, if I follow this protocol. I’m not sure that we are very good at explaining to people how we use individual data versus population data. The only time I hear population data spoken about basically is in scientific meetings and among groups such as this.
I’m not sure the public understands what we are collecting and how it can be used, and how they can relate to it in terms of understanding it in their lives, and the choices that increasingly they will make, because they are the ones who want to be in control more and more of medical and other decisions, including how much to exercise, whether to even tell somebody you are taking supplements, and so on. I’m not sure we have played off that very well just yet.
DR. SONDIK: Any comments? Is there anyone else who would like to get on record?
Well, again, let me thank our speakers on behalf of all of us. I very much appreciate your time, effort and insight into all of this. John?
DR. LUMPKIN: I have been given the task of talking about the future. It is truly a privilege to have the opportunity to be the chair of this very important committee at this particular time.
The reason why I say that is, I don’t know if you have noticed, despite the fact that we asked many of our speakers to talk about the history and the accomplishments of the committee, many of them found it very difficult to state a task, and every time, they always looked to the future. That I think speaks a lot to the quality of the individuals who have served on this committee over the years, that it is not really what we did lat year, it is what we want to do next year, and what needs to be done, and all the tasks.
What we have been talking about today has really been what we need to do in the future. The committee has over the last few years been, I believe, and Secretary Shalala said it, and I thought she was right on point, one of the hardest working committees that I have ever been on. I have been on a number of advisory committees. in fact, my favorite was — I was on the advisory committee for the WIC program, and they met every two years. You go to two meetings and you’re off, it was great.
I think I have been to more meetings of the committee the first day I was on it than I was of that advisory committee. But the productivity can be demonstrated just in what has occurred in one day.
For those of you who weren’t at our meeting earlier, we have a major report required by HIPAA, which is going to the Secretary on standards for patient medical record information. We have delivered an interim report on the national health information infrastructure, and on a vision for the health statistics for the 21st century. That is a lot of work for one day, and it reflects certainly a long period of activity.
But the committee’s plate is even fuller than that. Not only do we have to continue to look at and develop standards under HIPAA, the report for the patient medical record information is just an interim report. Now that we have said what we need to get done and advise the Secretary on what needs to be done, it is time to do it.
The same is true of health statistics for the 21st century, and we continue as an ongoing theme of this meeting, we need to address issues related to privacy. Ted was nice enough — I guess he had to leave — he was in the process of moving. He was in California yesterday and his whole house and everything is moving to New York. So we are thrilled that he was able to come here.
But he raised with his committee’s report on the medical aspects of the Internet a number of issues that we will go to work on in the committee, both in the areas of privacy as well as confidentiality and security, which are handled by two separate committees.
The committee has an ongoing challenge to look forward at how we address issues related to population health. One of the advantages of the reconstitution of the committee is that we did get some additional members. That would have been kind of neat, except they also gave us a lot of additional responsibilities. Part of the challenge has always been to maintain the focus on population issues. We are going to be trying to struggle to continue to do that.
Functional status measurements is a continued issue, issues related to minorities and disadvantaged populations. All of these are things that are on the plate of the committee.
I would be remiss without mentioning at least two individuals who have played a very important role as representatives of the large amount of people who have been working behind the scene. It has never ceased to amaze me, the quality of the work of the members of the committee by those who have agreed to volunteer their time is a reflection of the dedication of the staff of the Department of Health and Human Services who work with this committee. Not only is it the quality, but also the quantity.
I have been on some other committees where I think I would have been much more productive if we had had the same high level of commitment. But as representatives of those staffs, I think Jim Scanlon and Marjorie Greenberg have to be recognized in their role. They have been the movers and the shakers of what has happened on this committee, and I think everything that we have done would have been impossible without them making an effort. I’d like to give them a hand.
Just because sometimes I enjoy bothering people, in case there are a few people who may not know who they are, could you stand?
We have taken an opportunity to pause, to reflect on the dramatic changes that we have seen in our society since the committee has come into being, and the many, many years of hard work of members of the committee and the staff from the Department in creating a new direction in the future. We have outlined the tasks of that future, and raised some goals. We need to rededicate ourselves to our future, to commitment to look for additional funding and as has been mentioned by Don, to get about implementing some of these great tasks that we have before us.
It is always challenging to look forward to the future. In fact, in many ways it reflects the life stage of many of us. The committee is 50, and as I am rapidly approaching that age, I have come to realize that it is just the beginning. So we are as a committee just in the beginning. We have gone through our childhood, we have barely grown up, and we have a lot of work before us.
I appreciate the fact that everyone has come to help us celebrate the end of childhood and the beginning of adulthood, and all the work that we have yet to accomplish.
At this point, we are going to conclude. There will be a reception. There will be some light snacks before that. There will be full snacks at 5:30. We have a little bit of time. I would like to ask two groups of people to come forward, because we do have a camera, and we want to make note of this. We would like to ask members of the current committee to come forward, and also the past chairs and past committee members, because we would like to take some note of this with photos that we can then digitize and put someplace where people can access them. Okay, we’re going to do them outside in the hall.
(Whereupon, the meeting was adjourned at 5:05 p.m.)