[This Transcript is Unedited]
Hubert H Humphrey Building
Room 800
200 Independence Avenue, SW
Washington, D.C.
DR. COHN: Good morning, everyone. I want to call this meeting to order. This is the second day of two days of hearings of the Subcommittee on Standards and Security of the National Committee on Vital and Health Statistics. This Committee is the main public advisory committee to the U.S. Department of Health and Human Services on national health information policy.
I am Simon Cohn. I am a physician and Chair of the Subcommittee. I am also the national director for health information policy for Kaiser Permanente. I want to welcome fellow Subcommittee members, HHS staff and others here in person. I also want to welcome those listening in on the Internet.
We are on the Internet today. I want to remind everyone that there will be testifying, as well as making comments and asking questions. Please identify yourself and make sure to speak into the microphone so that those on the Internet can hear us.
This morning, we will be talking about a number of different federal and private consortia issues of interoperability, PMRI standards and terminologies. In the afternoon, we will be talking about issues related to ICD-10 and recommendations related to next steps, as well as a letter that we are in the process of writing to the Secretary related to the SM-4.
I want to thank for today Steve Steindel, Doug Godetsky, who - I do not know if he is here or not, Donna Pickett, Susie Bebee and others, obviously, for their help in putting this meeting together. I also want to welcome and thank Jared Adair, who is here joining us for, I think, her help and guidance in terms of direction for today=s session.
MS. ADAIR: Thank you.
DR. COHN: With that, let=s have introductions around the table and, then, around the room. For members of the Subcommittee, I would ask if there are any issues coming before the Subcommittee today that you please identify them if there are issues you need to recuse yourselves about and indicate the reason why. Jeff, would you like to do some introductions?
MR. BLAIR: Good morning. I am Jeff Blair. I am vice president of the Medical Records Institute. I am Vice Chair of the Subcommittee on Standards and Security. There is nothing that I feel I need to recuse myself from. However, from a disclosure standpoint, I should let you know that I am a member of HIMSS, HL-7, ASTM, the ANSII Healthcare Informatics Standards Board and, I think, that probably covers it.
MR. AUGUSTINE: Brady Augustine, Gambrill Healthcare, corporate director for special projects, senior biostatistician and medical economist. I have no conflicts or disclosures.
MR. McDONALD: Clem McDonald, Regenstrief Institute, Indiana University. I am Chairman of the LOINC committee. I am a member of HL-7, ASTM - I think all of them - I mean, the usual suspects. If there is votes on LOINC, I will recluse myself from it.
DR. COHN: Okay. Before we go to Mike, Kepa, are you on the line at this point? Well, maybe, maybe not. We actually have one member of the Subcommittee who is at home because of, I think, some - I would describe as back ailments, but, he is calling in and will be participating by phone. When he comes back, I will have him introduce himself. Mike?
DR. FITZMAURICE: I am Michael Fitzmaurice, senior science advisor for information technology to the director of the Agency for Healthcare Research and Quality. I am lead staff on the Secretary=s council on private Secretary initiatives to improve the security, safety and quality of healthcare. I am liaison to the national committee, staff to the Subcommittee on Standards and Security. My agency supports the meetings of the ANSII Health Informatics Standards Board.
MS. BEBEE: I am Suzie Bebee with the National Center for Health Statistics. I staff the Subcommittee.
MS. ADAIR: I am Jared Adair. I am with HHS, the Centers for Medicare and Medicaid Services. I am also the project manager for the Consolidated Health Informatics Initiative, which will be discussed later on this morning.
MR. REESE: I am Bert Reese. I am one of your panelists this morning. I am the vice president and CIO of Sentara Healthcare. I am representing the National Alliance for Health Information Technology.
MS. LOVE: I am Denise Love with the National Association of Health Data Organizations. I am one of the speakers this morning.
MR. KOLODNER: I am Rob Kolodner. I am the associate deputy chief information officer for health in the Department of Veterans Affairs. I am staff to the Subcommittee and, also, I will be testifying this morning, presenting health informatics.
DR. HAMMOND: My name is Ed Hammond from Duke University. I am professor in the Fucal School of Business and Community Family Medicine, School of Medicine, in biomedical engineering, School of Engineering. I am testifying this morning on behalf of the Connecting for Health public/private consortium on behalf of the Markle Foundation. I am president of the American Medical Informatics Association and the Vice Chair of the Technical Steering Committee of HL-7, Co-Chair of the technical committee of the vocabulary of HL-7, Co-Chair of the special interest group on electronic health records and HL-7. I am the convener of OSSO TC-215 working group to messages and communications. I am too old to remember more than that, I think. Thank you.
DR. OVERHAGE: I am Marc Overhage with the Regenstrief Institute at Indiana University. I am here this morning on behalf of the e-Health Initiative and its foundation for e-health.
MR. YASNOFF: Bill Yasnoff, senior advisor, National Health Information Infrastructure of HHS.
DR. STEINDEL: Steve Steindel, Centers for Disease Control and Prevention, staff to the Subcommittee and CDC representative to the comprehensive health initiative.
MS. GREENBERG: I am Marjorie Greenberg of the National Center for Health Statistics, CDC. I am the executive secretary of the committee.
MS. TRUDEL: Karen Trudel, Centers for Medicare and Medicaid Services, staff to the Subcommittee.
MS. SQUIRE: I am Marietta Squire. I am with NCHS and staff to the Subcommittee.
MR. ROBERTS: I am Dave Roberts. I am the director of public policy for the Healthcare Information and Management Systems Society.
MS. LEON-CHE: Nellie Leon-Che, American Hospital Association.
MS. CHULMAN: Roslyn Chulman, American Hospital Association.
MS. KLAUS: Linda Klaus, American Health Information Management Association.
MS. MITCHELL: Cory Mitchell, College of American Pathologists.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics, CDC.
MR. KRAUKAUER: I am Rick Kraukauer, Office of Health Policy, DHHS.
MS. BICKFORD: Carol Bickford, American Nurses Association.
MS. GROVER: Tisha Grover, the e-Health Initiative.
MR. BEEBE: Michael Beebe, American Medical Association.
MR. WINE: Mark Wine, Department of Veterans Affairs.
MR. GODETSKY: Douglas Godetsky, Centers for Medicare and Medicaid Services.
MR. DAWKINS: William Dawkins, Centers for Medicare and Medicaid Services.
MR. SCALLNISS: Mark Scallniss, National Institute of Standards and Technology.
MS. ROSENTHAL: Lynn Rosenthal, NIST, National Institute of Standards and Technology.
MS. FARBERG: Elizabeth Farberg, healthcare consultant.
MS. PICKETT: Donna Pickett, National Center for Health Statistics, CDC.
MS. WILLIAMSON: Michelle Williamson, National Center for Health Statistics, CDC.
MS. TESSIA: Claudia Tessia, Mobile Healthcare Alliance.
MR. COMBS: John Combs, the American Hospital Association.
MR. TEDER: Eric Teder, Health Pathways.
MS. FINGOLD: Helene Fingold, Medicare Payment Advisory Commission.
MS. SOCLOFSKY: Joan Soclofsky, Medicare Payment Advisory Commission.
MS. FERNANDEZ: Bernadette Fernandez, Planning and Evaluation, HHS.
MR. PRICE: Colin Price, NHS Information Authority in the U.K.
MS. HAMBY: Tam Hamby with McKesson.
MR. DOWNS: Steve Downs, Robert Wood Johnson Foundation.
DR. COHN: I want to welcome everyone. This morning, before we start the panel, I think Steve Steindel and Jeff Blair have asked to make a couple of comments related to a little bit of context setting. As many of you who will be testifying are aware, obviously, this issue of interoperability, PMRI standards, has obviously been an issue that many of us have been working on for a long time through many organizations. I think they just have a couple of comments that they want to make about that.
Just to generally reflect on the way that they will go, we will be hearing testimony during the first session. We will be taking a break afterwards. We have asked, then, Colin Price, from the U.K., to make a couple of international observations, knowing that this is not an easy area. Then, I think, we will hopefully spend the rest of the morning in conversations about priorities, issues, potential ways of developing synergies.
With that, Steve and Jeff, did you want to make a couple of comments to lead off the morning? Jeff, okay. We need to, actually, get whatever the presentation you had up on the screen - why don=t you hang on for just a second for them to get it up.
MR. BLAIR: While he is putting it up, I might encourage you to go ahead and look in your folders. There is a very informal brief, a little document, called a quick glimpse. It starts off like that, which we thought might be helpful to many folks to put a lot of things in context. Let me know when B it is up.
DR. COHN: It is actually up. Guide us about when it needs to be moved.
MR. BLAIR: Right. Okay. Steve and I thought that it would be helpful because so many folks work on standards development initiatives, and other healthcare initiatives, from on perspective or another. They may not be familiar with some of the other initiatives that are out there. We thought that we would just pull this together for your use. It is informal. It is not expected or intended to be complete. It is just a quick tool that might help you before we begin today=s session. It is entitled A Quick Glimpse at the Organizations That Have Promoted, Developed and Coordinated Healthcare Information Standards. That is slide 1.
Slide 2 - many folks have wound up encountering different terms for electronic health records, electronic patient records, computer-based patient records, whatever. This is one set of definitions that may help you reconcile that. One of the reasons that we pulled this together is because the health insurance portability and accountability Act, the Administrative simplification provisions of that, uses the words patient medical record information in it. So, this will allow you to kind of, maybe, map - by the way, it never did define it, of course. Some of these definitions are based on usage and experience, rather than any academic specific definitions that we could go to.
The next slide is some of the major - some of the major - contributors to the development of healthcare information standards. These are the private sector major standards development organizations.
The next slide will wind up continuing with that thought, but, indicating those initiatives that have been involved in either facilitating, promoting or coordinating standards in the private sector, and some of the government agencies that are part of that. I am not going to read down the entire list, although I know that some of you really want me to do that.
The next one is a quick - a simple time line. Then, as we begin to learn a little bit more about some of the new standards initiatives, here is some questions that you might consider - at least, it is some that Steve and I thought we would be thinking of in terms of whether the new standards, or healthcare initiatives; whether they are addressing new problems and approaches; whether they are addressing known problems, but, with a new perspective; whether they are addressing known problems with new methods or technologies; or, whether they are addressing known problems, but, with greater industry support or energy.
The second major question that we are beginning with is how do these standards initiatives relate to existing or other standards initiatives. Lastly, how do they fit into the national health information infrastructure? For those of you where the phrase is new, the National Committee on Vital and Health Statistics in January set forth a report to the Secretary which proposed a national health information infrastructure. I think there is many other organizations that have proposed and worked on a similar initiative. It seems like this is an idea whose time is arriving. That is what we are referring to.
That, I think, is - Steve, do you have anything else to add?
DR. STEINDEL: No, Jeff.
DR. COHN: Well, with that, let=s move to the testimony of the day. Ed, do you want to lead off?
DR. HAMMOND: Okay.
DR. COHN: Ed, I will tell you that as I was looking at that list of organizations, I think you Chaired most of them at one time or another.
DR. HAMMOND: First of all, my appreciation for having the chance to talk to this group. I was going to say that those of us that have very, very gray hair have been trying to do this for a long time. I think we are very excited about the fact that standards, and what we can do with standards, has reached the level of national attention that it has gotten. I think this is a delight for us to be able to testify before this group.
Healthcare is a matter of all people and, certainly, a lot of stakeholders. Traditionally, we have dealt with those stakeholders almost as independent beings. We have worried about dealing with this set of people, this set of people, this set of people, without going across. I think driven by the cost of healthcare, driven by patient safety, driven by the need for better quality, driven by reducing the cost of care and, finally, just getting better care says it is time to put this altogether. So, I think this particular organization that I am talking about today, Connecting for Healthcare or Connecting for Health, is particularly well named because that is what we are about.
The vision is that providers - those responsible for population health, researchers, reimbursers, auditors and a very important new player, consumers, will have ready access to timely, relevant, reliable, secure healthcare information through an interconnected, electronic health information system to produce better care at a reduced cost.
The organization, Connecting for Health, consists of these stakeholders that I just mentioned. They are the people that are in the practice of medicine - the clinicians, the hospitals, the health systems, the people that pay for that - the employers and third party payers, the people that help us regulate and do some paying for this - the federal government itself, the people that make the products that we use to do this, the vendors, the academic and research institutes, the national standards groups, the manufacturers of systems and, then, consumer and patient groups. These are the groups that we are trying to bring together to a common cause.
The purpose of Connecting for Health is to catalyze specific actions that let us move toward this concept of an interconnected, electronic health information infrastructure. There are three basic components that we have identified to try to get to those goals. The first of these is to accelerate the rate of adoption of clinical data standards. Being one of the people that were involved early on in the creation of those standards, recognized the importance that creation is only one step. The adoption and the implementation of those standards and the proper use of those standards are absolute necessities for this to be a successful endeavor.
Then, the second, very important component of this, that really got highlighted with HIPAA and still is in the spotlight, is the practices for secure and private transmission of medical information. It is a catchy situation because of privacy concerns and, yet, it is mandatory to have a complete picture of the patient to be able to do that. I think that is a challenge. We have a group that is identified to deal with that.
The third component of this is back to the consumers themselves, the patients and the people that make up our communities and our networks that are employees - all of the participants in this - to understand what they have to say about their health and how they can become more directly involved in their health. So, how do we put this system together to connect to that group, as well?
So, Connecting for Health has organized in three working groups. The first is a data standards working group, of which I am the Chair. The other two working groups are the privacy and security working group that is being defined now, and a personal health working group that is being defined now. Neither of these second two groups have met, yet. There is a steering group, or a leadership group, that is overall - that takes important players from all of these sectors and helps us get access to the rest of the world itself, making sure that we are on the correct track and making sure that all of the pieces fit together. That group will meet for the first time in September.
Focusing on the data standards working group, which is the topic of this meeting, that group basically is organized into three components. There is a consensus committee that, really, deals with the standards per se.
There is an implementation committee that - Marc Overhage is the Chair of that committee. The implementation committee is, really, looking at how we get these standards out into the world and in real use, recognizing the fact that much of the current world is in legacy systems and there have to be pathways that are developed that lets us get there from here. Not only that, but, recognizing the fact that creating standards is a continuous process. It is not we do it and we have finished our work. It means it keeps going. So, there has to be a national pathway in which that gets built in.
Finally, the policy committee that interacts with what we call the consumers of the data itself, largely, the government organizations, the third party payers, and others, that will be using this information and understanding how that all gets put together.
So, looking at this from another perspective, what we have tried to do is to identify the required set of standards. Some of these are obvious. Some of these, actually, have work going on and have reached a level of maturity that these standards are usable today. In other cases, the work is going on, but, has not reached a level of maturity that we can use as standards. In some cases, we continue to discover as we think about this process from front to back - or back to front, I guess - that we need to worry about somebody starting the work and developing these standards. Then, there is an adoption process that comes out of there, looking at the levers for adoption. Finally, implementation, a great deal of importance on the implementation side, how and who will pay for that implementation itself. Then, recognizing that maintenance goes on. The development of these standards, the adoption and the implementation, is surrounded by barriers that Connecting for Health is trying to identify and get the mobilization to overcome those barriers itself.
Connecting for Health had its first meeting on July the 31st and August the 1st of this year. At this meeting, we developed and applied a criteria to a standards set, identifying the standards that had reached the level of maturity that they were usable in today=s systems.
The next thing that we did is to identify the overlapping data needs of the data consumers. To me, the most exciting thing about Connecting for Health is the fact that we do have people that are talking to each other from what used to be independent, vertical silos of activity. Communication is now taking place. Instead of people, like CMS, guessing what information they can get out of the system, or asking because of limited views of what is available, engaging at the very beginning as we are defining the process of producing that data, of collecting that data, of using that data for healthcare, and doing it in such a fashion that it becomes usable for multiple purposes. I think that is a goal of what we are talking about.
Part of this is recognizing the fact that this is a very, very difficult and a huge task, to be able to do this. There has to be some priorities for doing so. What we are doing is crosswalking these priority data needs to the operable standards that we have to get this thing moving immediately. Finally, we have started work on strategies to move standards into play and deliver this required data to all of the stakeholders.
The next steps is to take the results of this first meeting and have it validated by a number of different groups, including the steering group, and other key stakeholders. Then, to identify the next steps to keep this process moving. I am old enough to have a firm belief that anything, if you do not do it very rapidly, the moment of opportunity misses. You are on the front page today and it takes not very many weeks to get you on page 20. So, these are the kinds of things that really have to be kept going.
So, what the next step for us is to accelerate the adoption of required data standards whose development is underway. These are the things that we know we need and we think, with some effort, we can push that to make that happen a little bit faster. We need to articulate the needs for required standards that nobody has, really, started doing very much with. We need to articulate recommendations related to terminology. We need to develop practical strategies that will clear the barriers and accelerate the adoption of the operational set of standards.
Part of that is trying to identify whose responsibility it is to do the next step, recognizing the fact that there are three groups of players. There is the standards producers, the vendors community, the healthcare producers actually in the healthcare consumers as part of this, too. The goal of this is, really, to articulate the road to interoperability itself.
Interaction with the patient medical record information - we are providing a forum where we are bringing all of these people together for connection data users, developers, care providers and trying to engage in a dialogue that helps us understand the art of the possible, the art of the doable and the art of the desirable itself. We are proposing the facilitation of a broad scale adoption and rapid acceleration of some of the standards that we need.
One of the things that I would like to point out is that these are not isolated standards that are created. The standards are all interlocked. Terminology standards have to be very, very carefully locked to a data model, for example. If you look at some of the things that I am talking about here, in terms of the acceleration of those standards, they fit together. It is not just doing one from here, one from here. It is, really, understanding how they merge together that produce the standard that is much more powerful than what we would have with the independence for that.
First, is terminology. For all of my career, I have talked about if we had a terminology that everybody used we would be able to do wondrous kinds of things. I think we now are approaching a point of where the necessity of that may drive a solution to that particular problem.
The other thing that Clem and I were talking about just a couple of minutes ago is the fact that even with standards, people do what they want to do. You have one group that collects this set of data, another group that collects the other set of data. When you try to make these two groups talk to each other, they do not do the same things. There has to be a set of business rules that define, clearly, what data you are going to collect, how you are going to collect it and, then, how you put that into the standard itself. You cannot just shove it in. Text streams are beautiful for flexibility. They are terrible for the fact that you can do what you want to do and I cannot do anything about it. We have to close those kinds of gaps. Clinical templates, clinical documents, guidelines and, finally, the electronic health record itself.
What we are doing is to try to address those gaps by identifying the criteria. I think this is a very interesting piece. Yes?
DR. COHN: Do you have two lessons from clinical documents in your last overhead.
DR. HAMMOND: I am trying to just -
DR. COHN: What?
DR. HAMMOND: Clinical documents.
DR. COHN: Explain what the two plus meant.
DR. HAMMOND: Clinical documents, basically - from an HL-7 perspective - and, you see my connection with that - are at three levels. The first level is simply the document in which - it is an XML document itself that uses the header level only. Level 2 is where there is more detail specification, dealing marginally with templates that are contained in the body of the document itself. Level 3 is a fully structured document in which each of the components of that document are retrievable. Basically, we think - I did not mention the standards that I think are mature enough to be used. Obviously, the reference information model, the data interchange model, and level 1 of clinical documents is part of what we are talking about.
Ways of doing this is to establish the criteria for assessing the options in the gap areas themselves, develop recommendations if there are options. I think one of our problems is the fact that we have not been willing to make choices. We have lived with a world of being nice to everybody. So, many thing co-exist. Too many is worse than none because if there were none, it would be easy to solve. We are recommending and encouraging the undertaking of strategies for developing standards where none exist and try to identify the appropriate body to do that. This requires the expertise, as well as the interest itself. Then, we need to identify resources to accelerate the development of standards.
We need to realize that most of the standards that we are talking about are coming out of volunteer effort. Volunteer effort is not an 8:00 to 5:00 job, five days a week. It is catch when catch can. We are stealing time for the other activities that we are doing. That is what we need to help solve if we are moving some of these things ahead.
Finally, the relationship to other standards that is an issue. The standards groups and experts that we have in Connecting for Health go across all of the different groups that are involved in doing this. We are using existing standards as a basis for the recommendations itself. We are actively collaborating through overlapping involvement, communication, with a lot of other initiatives. There are, at least, 50 initiatives that, if you gave them to me on a piece of paper and I did not see the header, I could not tell you what organization was doing that. The good news is that lots of people have discovered that this is a problem that needs to be dealt with. Reasonable news is that they are moving, mostly, in the same direction. The bad news is if we are not careful, they will go different directions. Somehow, I think this committee may have a responsibility of helping those groups to not be redundant in their activities, to work together and be able to do that.
Finally, I think that our major contribution, or the contribution of this group itself, is a willingness to make decisions and do endorsements for what we think are the solutions to the problem that we are talking about. Thank you, very much.
DR. COHN: Thank you. Rob, you are sitting in for Gary Christopherson?
MR. KOLODNER: Yes, correct. Ed really does not want to give up that table.
DR. HAMMOND: I am weak.
MR. KOLODNER: You are anything but weak, Ed. Good morning, I am Rob Kolodner. As I mentioned, I am from the Department of Veterans Affairs. The slide shows the name Christopherson. Gary was scheduled to be here today. He is the deputy CIO for health. Unfortunately, he called this morning and indicated that he thought he better stay home instead of spreading the wealth of his germs across the country to the members here. So, he is at home trying to get better. I will be filling in for him.
I am presenting on behalf of the partners for the Consolidated Health Informatics Initiative, which is a government-wide initiative that you heard about at the NCVHS meeting, I believe, earlier from Jared Adair. This presents a little bit more about the information.
The particular initiative has to do with focusing on health information interoperability standards in the federal healthcare arena. It is related to GCPR, which this Subcommittee is an advisor to. I will explain just a little bit later where that relationship is. It has its origins in the GCPR.
The CHI vision is a very simple one. It is to enable - a simple, but, powerful one - enable the sharing of health information in a secure environment through improved health. Specifically, it is focused on the federal health information interoperability - the federal agencies and the standards that we use. It is to establish those standards as the basis for electronic data transfer and all of the activities and projects among all federal agencies and departments. The boundaries of CHI are within the federal agencies themselves.
The participation, at this point, includes about 20 partnering agencies, including a number of agencies - permanent agencies - within HHS, as well as VA and DoD, and other departments, such as Department of State, and other government agencies - Social Security and others. We have involved about 100 medical, technical, policy and management subject matter experts in our work, drawing from all of these agencies, and from their appropriate contractors.
The specific answers to the questions that were asked by the committee have been distributed to you. So, I am not going to go through each of those components. They are there for your reference. As you see, I am summarizing some of those.
The four key elements of CHI have to do with adopting those health information interoperability standards in the federal government. Those, specifically, focusing on data and communication standards. As Ed mentioned for the data standards, we are particularly interested in those that are associated with underlying data models.
Obviously, we need to assure that the activities in CHI align with HIPAA. Jared has particular interest in that. That is another hat that she wears, and with the code sets and records that have been part of HIPAA. We, also, need to make sure we align with the HIPAA security and primacy solutions. But, the privacy and security solutions are not in the scope of CHI. We are picking those up and will be using those, but, those are not ones that we are working on within the work that we are doing with CHI.
A key part is to foster a change in this medical sharing culture so that we can move the information that needs to be moved appropriately, but also, making sure that when we do so, if it is information about individuals, that it is done in more aggregate form or protects the privacy of those individuals. The fact is that the federal government has a lot of information in different agencies that could help to improve the healthcare and healthcare delivery in the nation. By being able to coordinate and analyze that and detect patterns, we would be able to contribute to the health of the nation in a way that would be even better than we can now.
The strategy is one that is a voluntary one, outside of federal agencies, but, the idea will be that it will become the standard inside federal agencies. The relationship of this to GCPR is that, if you recall, GCPR had focused on three agencies: VA, DoD and Indian Health Service and moving information between them. There were several components of that, including not only working out the privacy, security and technical interfaces - the business rules for when information is moved - but also, the standardization of that data so it would be computable as it moves back and forth.
That particular component is, in fact, the focus of CHI. It has to do with identifying those standards and adopting them and doing it not just within those three agencies that are required of GCPR, but, in fact, involving every federal agency that is health related. What we are doing is, really, looking to adopt these, to work together, to work with the SDOs, and industry, to identify their standards that are mature. We will be coming out with the first set of standards that we are adopting as we go through this process for the first time and make sure we get sign-off by all of the agencies. Then, we will be releasing it. We will let you know as that is available. We hope that will be in a very short time.
Then, to work through as many of those standards as possible. As Ed Hammond mentioned, some of those are near ready and need a little help getting over the line. Some of those have not been done, yet. They need help getting started. Our particular interest is to look at those that we can immediately use. Putting a standard out that none of us can use for five years, really, does not accomplish anything. What we want to do is to start with those that relate to the information that we have. Then, as the health information systems, for example, in the provider arena for VA, DoD and Indian Health Service, and others, mature and can use those, we want to make sure we pick up some of the standards there. If it is information that is reported to FDA, or to Social Security, or to CMS, that, again, we concentrate on those, as well.
Within the agencies, once we identify those standards, that will become the basis for future acquisitions in government. That will be a requirement for all of the agencies. Therefore, if it becomes standard within the federal government, outside - what we hope is that by taking the risk, by being at the front end and by having enough critical mass that that, then, will attract, in a voluntary fashion, the private sector to consider adopting those standards, knowing that those, in fact, are then going to be around for a while and likely be the ones adopted. Obviously, that also is something that we would bring back to NCVHS, to the Subcommittee, for that to be considered whether, at some time in the future, that becomes part of HIPAA. But, that is a separate related thing. What we really want to do is outside of the federal government look to a very positive, attractive model rather than a mandatory-type of model.
In effect, we are looking to see that where necessary - where it is not already happening - can the contribution of the federal government adopting the standards create a tipping point that, then, helps that particular standard, or standards, to gain widespread adoption.
NCVHS has already indicated that it would be advisory to CHI. But, in terms of specific assistance, there are four specific items that we put up here. Obviously, providing a periodic guidance and feedback regarding directions and plans is one that, I think, is generally agreed upon. Also, in terms of helping to identify new areas needing new standards on which CHI ought to focus are ones that need, again, maturing. Again, CHI is not developing standards, we are looking to see what do we need to do to help those standards mature, working through the usual mechanisms, so that they, then, are ready for us to adopt.
Obviously, marketing the need for the adopted standards to the rest of the public and private sector, NCVHS has been doing it. I think we can continue to do. A number of the efforts here at the table today, I think, are certainly going to be complementary in that regard, as well. Also, we would ask that NCVHS partner be supportive of us and its participating agencies in this effort, as we move forward.
That is CHI. What I wanted to do is, also, spend just a minute or two talking about an example of complementary activity to CHI. This is one that the VA has proposed as a strategy. It is called HealthePeople. Unlike the HHS healthy people, this is spelled with an Ae@ instead of a Ay,@ connecting the electronic aspects of healthy people. Notice it is about health, not about just medical. It is about the individuals, not about the agencies.
The origins of this particular initiative and strategy certainly took a big boost in October of 2001, in a meeting sponsored by the Institute of Medicine and Kaiser Permanente, exploring what it would take to get to accomplish the goal that IOM stated, again recently, of being paperless by 2010. In order to do that, the group identified that what we would need is a good set - a robust set - of data standards, communication standards. Then, as a key part, is a model health information system that demonstrates that it is doable, as opposed to that it is just theoretical. By getting that in place in the next couple of years, then, there would be a period of time for adoption by healthcare organizations, not necessarily of that particular model system, but, of systems over a period of time that would be available in the private sector.
The strategy is both a federal one and a national one. It is public and private. It has to do with having national standards. It has to do with having high performance health information standards - health information systems supporting an ideal health system. It is, really, not only the strategy of CHI of standards, but, it moves into let=s make that, then, have an impact on the health systems that are used by our providers across all places of care.
The concept is one that was collaboratively developed by both public and private sectors. It is a broad spectrum, involving consumers and providers, payers and regulators. It meets the needs of all of those entities and sectors. It is built around achieving a variety of standards, including architectural ones and the security and technical standards resulting in a high performance information systems, at least one of which, we believe, should have a public ownership for entities that are serving the poor and often do not have much in the way of resources.
The graphic that I show, here, shows the existing systems and focusing particularly on the federal, where DoD has its CHCS-1 system, the VA has its VistA system and IHS has its RPMS system, as well as the private sector, Kaiser Permanente - is one entity we have been having active discussion with - as well as other public and private sectors.
We would, then, be moving in the HealthePeople federal - that encompasses the federal agencies that our healthcare providers, or the broader healthy people concept that would be a public-private kind of activities into what each of us is working on now, which is our next generation systems. Those are not the same systems, but, what is remarkable in conversations with all individuals is that the designs, the architecture, the capabilities, are remarkably similar. By moving from those and beginning to converge those, that over time we, in fact, would be able to incorporate the same standards and use the same standards for interoperability. In some cases, we would actually combine activities - certainly, within the federal sector - to move towards very similar, or the same, modules - or, in some cases, the systems. There would be convergence overall toward these national standards in high performance information systems.
As a specific example of the HealthePeople federal, VA and DoD have committed to activities where we will, ultimately, end up with our repositories of our data for our beneficiaries - those repositories will be based on the same standards, the same data models. They will be able to exchange the information, as appropriate. In addition to similarity, or complementarity, at the data level, at the software level, we are looking to identify all of those areas where we can, actually, converge our efforts and develop common software, where that makes sense, that meets our needs, as well as having some separate software.
Again, this could not happen unless we get the data standards. The data standards are the activities that we are doing with NCHI. Thank you, very much. I look forward to questions later.
DR. COHN: Thank you, very much. The next
presenter is Denise Love. Thank you for joining us.
MS. LOVE: I have chosen to take a low tech approach to my presentation. There are copies of what I will speak about on the table. On behalf of NAHDO, the National Association of Health Data Organizations, I am truly pleased to be here and offer our perspective on standards and standards initiatives.
I think I am bringing a different window into looking at standards. My members, I guess, are users of the standards. But, users who can promulgate reporting requirements at the state level and creatively, I might add. A standard is a guideline in some instances, so again, that is my members. You will hear more about it in my testimony. The initiative I am talking about today, really, boils down to, I think, administrative data systems which, I believe, are the workhorses today of state health departments. Also, they are the foundation - the physical foundation - of the National Health Information Infrastructure.
That said, the state of state budgets are not good right now. Data is not a line item that is of high priority in many cases in state budgets. In fact, I would like to say that NAHDO is working to expand and improve state healthcare data. Sadly, in some cases, we are working to assure survival of state health care data collection. On that note, I will provide our perspective on your important work.
Today, more than 40 in-patient hospital discharge data systems are maintained by state health data agencies. Thirty-nine of these are legislatively mandated. About a dozen of these state systems are voluntarily collected. A state health data agency - a government agency - may implement a discharge data system for a private sector organization, such as a hospital system. They may implement the state-wide data collection. The patient=s healthcare encounter triggers the event and the abstract from hospital financial systems includes patient demographic diagnostic procedure and reimbursement information that is reported to the state agencies.
Most states have - because the PMRI or the electronic medical record information has not been readily available over the course of the last 20 years when states have wakened up to the fact that they need healthcare data to govern their - not only their policy making, but, stimulate market decisions and evaluate cost, quality and access, states look to a standard that was out there of data that were readily available. That would include the UB-92 or the Uniform Hospital Discharge Dataset.
The states have found that with the low level of resources they can implement the hospital discharge data system to - it is a population-based source of healthcare utilization data that provides a large number of observations. These data are being increasingly used, again, for quality improvement, research and public health purposes. We all know - and, I do not think we need to debate during the discussion the limitations. We do know that they lack clinical detail. There are diagnostic and procedure coding inconsistencies. I will talk about 50 state formats of a standard. But again, I think that over the last 20 years, and with the help of some of you in this room, the science is catching up with us. We are able to understand and overcome those limitations, certainly, much better than we did in the past.
With that evolution over 20 years, I wanted to acknowledge the Healthcare Cost and Utilization Project, or HCUP, which is led out by the Agency for Healthcare Research and Quality. Twenty-nine states do participate in that initiative in 2002. Again, with their assistance, the science is catching up with understanding how to overcome the limitations of the data and improve the utility of discharge data.
With the advent of HIPAA, I think the state agencies, and others, have the opportunity to improve the timeliness and uniformity of the data they do collect. We will talk about some opportunities that we see to further that catch-up.
About NAHDO. The National Association of Health Data Organizations is a non-profit, membership, educational organization created in the mid-1980s to help states improve the uniformity and comparability of that data and make that data publicly available, timely for more cost, quality and access purposes. NAHDO brings together federal, state, private sector organizations to strengthen and improve those health systems. NAHDO, itself, provides technical assistance to states and advocacy to assure survival, in some cases, work for the survival and improve the healthcare systems their members maintain.
I like to say that NAHDO=s members are not all about data. NAHDO=s members are about information. We are in the information business. To that end, we work hard with our members to mind the data appropriately, to provide information. We are, again - states vary so much because the way that the data are paid for at the state level varies. There are no federal funding sources for healthcare utilization data in the states. So, states will rely on general appropriations which can affect how that data system is implemented.
Again, one of NAHDO=s pushes, right now, a priority for the organization is to expand emergency department data reporting for every state that has in-patient data. We feel that for marginal costs, that state has the capacity to increase their non-in-patient data collection - and emergency department for states that have done so have proven to be not only doable, but, broadly supported by the community without imposing undue burdens on the providers. I won=t go into the - the handout goes into more about NAHDO.
Again, we are about technical assistance to keep the state data systems alive and improve them. So, I commend your work here today because it has implications for these systems.
HIPAA brings folks together. That is, probably, the one good thing, if you can say anything about HIPAA that I have observed. NAHDO is a part of some of the initiatives you have just heard about, including e-health initiatives and Markle Foundation - I have been volunteering some of our field experts to your work groups. We look forward to working with all of you to promote this important work.
What are the data gaps? NAHDO is working actively to fill those data gaps, promote uniformity. Again, there is 50 state versions of a standard out there. States have to negotiate with their providers. Provider systems are not alike. Provider communities differ in their priorities. States have taken a UB-92 format and adapted that to their state needs. We have been working with states to say if a claim does not serve your need, what are the data priorities that you have? What are the data elements that you are collecting that you cannot live without? How can we bring them into the structure of a national standard?
With that, we have identified those data elements that are of high priority for states. Bringing them into a national standard, working with the public health data standards consortium, the New York State Department of Health and states trying to integrate a national standard for public health and research reporting that looks like a claim, but, it is not exactly a claim. It reflects more truly what states are collecting.
Another gap is that states have worked hard to supplement the weaknesses of the UB-92. We have some states that actually have manual abstractions from the medical record into the discharge data abstract. We look forward to the day where we can automate that supplementation of the UB-92. NAHDO is working with - and will be working with - the HL-7 claims attachment to interface - or, we say bridge, perhaps, the EDI and the X12-N with the HL-7 and the medical record.
One thing that stands out with the claims attachment is that it is mandated by HIPAA. That might be our best opportunity in the near term to understand how the clinical data elements can be automated, captured and sent on to a government agency. We will be sending a representative to the HL-7 claims attachment work group to see what connection we can draw there.
Again, national collaborations were involved with the public health data standards consortium. We are working with organizations across research and public health to find out how this national standard for a public health report guide will work and how we can get states to avail themselves of it.
I would like to just say one thing, here, as I listen and as I read through plans for the electronic medical record, I think I would be remiss if I did not say that out in the state world, the hugest gap we find is out-patient data. That is where most of the public health activity is truly taking place. Those of us who have worked in the state field know that physician data systems are not capable of reporting the data that we need right now. The standards do not support, I guess, population-based assessment at this point because we cannot get at it. We feel that there is a continuum. That is where emergency department ambulatory surgery is part of that march, that Medicaid claims encounter, are important to this whole picture and, eventually, physician data. That is an important gap that I would like to just raise to this committee.
What can NAHDO provide to this Subcommittee? As I thought through it, I think we can provide some lessons learned in the front lines of data collection and use. Taking data that once was considered 20 years ago proprietary, and turning it into a public good. That is not an easy process. NAHDO will offer its help to this committee. My states are out there in the front line daily. I think that the lessons learned with administrative data will still apply when you start mining and extracting data from the medical record. I think the magnitude, though, will be greater than what we felt with administrative data.
Privacy and confidentiality is a huge issue in states. All it takes is a reporter on the front page to say that government - which, in NAHDO=s case, might be a state government - wants your medical record. That just raises a frenzy of concern as to why the government might want your medical record and might - members have dealt with that successfully. Sometimes, proprietary concerns get wrapped up in the flag of privacy. So, we have to sort those things out. State laws vary. All of those things are going to have to be dealt with. I believe my members have successfully, or rather creatively, addressed those concerns at their state level.
Data quality and validity will always be with us. There is no perfect data. I think my members have worked closely with each other, and others, to decide how good is good enough to move forward. We can inform you there.
Data collection priorities - it is not without cost. Our provider friends are very concerned about the burdens imposed on them for public reporting. We need to help work together, like we are with the consortium, to make the business case of what those priorities are and not try to get everything just because we can, just because it is there. We need to understand what it is for.
One of the keys for building trust, defining how the data will be collected and how it will be used with the data suppliers, analytic infrastructure - it is not enough to have a wealth of data if you do not have the people that know, understand data, know how to put it together and know how to communicate that data. The game could be lost. That has to be paid attention to. In the days of workforce cuts and other public agency woes, that is a huge concern for all of our data agencies - who pays? Who pays for this? Is it the state government? Is it the providers? Is it all - how is that - it is a common discussion in the states. Again, comparability across provider systems and, then, across states, it can be solved, but, it has to be addressed.
To that end, I thought you might want my opinions on how I think NCVHS leadership can make a difference because in the whole scheme of things, it all boils down to me, leadership, building trust and money to do it. I have been impressed, always, with Dr. Hammond and the work that many of you in this room have done, I have confidence that the technical solutions are there. This is an opportunity to bring this all together.
I think leadership is needed to prioritize data development activities to achieve early success. Work on those key set of clinical data elements easily abstracted from any electronic medical record system and not try to take on the whole elephant at the same bite.
Defining minimum core data needs for external reporting - we need more work and states need more assistance in that what does a cardiovascular encounter look like. What should we be getting from the - and how does that core minimum data set look like in the context of a medical record? I would offer my cadre of state health data agency experts to help with your work groups and subcommittees to assure that public health has a voice in the national standards process and assure that state data agencies who face unique and technical=political-social challenges are brought into the conversation.
I think NCVHS can help us a great deal at the state level by framing the message to the public as to the societal good for their clinical information. That is critical. Then, I would recommend that as a public health data standards consortium brings forth recommendations that we help link those recommendations with the NCVHS, though I am not sure that is a problem. But, I am proud of the work that a group of unfunded and motley crew of volunteers have been able to build national business cases for X12-N priority data development.
Again, I plead to make administrative out-patient data development of standardization a priority across HHS. Again, NAHDO is committed to helping states expand their current data collection efforts. I welcome NCVHS support for this floor, which I believe it is just the floor. But, it needs to be a floor that we have across all states. We are struggling with that. Thank you, very much.
DR. COHN: Denise, thank you, very much. Our next presenter is Bert Reese.
MR. REESE: Thank you, sir. I was, probably, the only CIO on the panel and I am the one with the technical difficulties. I have now moved to my second PC. For all those of you who brought laptops, would you line up on the other side of the room? I plan to go through every laptop here.
DR. COHN: We have a couple of options.
MR. REESE: I can do that from hard copy, if you like.
DR. COHN: Would you want to do that? Are you comfortable?
MR. REESE: Yes. It is the - if John catches up with me, that would be great. I will be working from the presentation of the summary of the National Alliance of Health Information Technology Initiatives, for those of you who have the handouts.
First slide - again, I am Bert Reese. I, actually, serve two roles. I am the vice president and CIO for - we are going to talk about the National Alliance for Health Information Technology which, of course, was spawned through the munitions of the American Hospital Association.
Some of the things I want to go over with you today is, certainly, some more introductions and a case for action for NAHIT. I come, again, from the Sentara environment, which is a large health system. We will talk about that in a minute. Some examples of standards that may be of interest to NAHIT and, then, some of the mission and attributes around NAHIT.
What is interesting about my background - and, actually, preparing for this presentation has been very enlightening to me. I have been a CIO for 30 years, using PowerPoint, now, for less than two, obviously. Twenty-one years at Sentara Healthcare, Albert Einstein Medical Center in Philadelphia and Temple University Hospital in Philadelphia. So, a total of 30 years. Sentara Healthcare, in 1981, was a $125 million per year company. It is now $1.6 billion. It is a full boutique of healthcare services. It is six hospitals, 12 nursing homes, 175 employed physicians. We are also the payer, 14,000 employees, durable medical equipment company, home health, air ambulance services. We are, also, the ground ambulance company for that area. We are down in the Virginia Beach area. If you have ever vacationed there, you have been in our area and, hopefully, you have never had to take use of any of our services.
What makes that important is that I come from what I would say the field. This is my first exposure, both at the AHA level and as a CIO, to any of the discussions around standards. It has been, truly, enlightening. Suzie has been absolutely wonderful in helping me prepare for this and giving me some reading material.
It is the field - what is going on in the field which is, actually, why NAHIT, or the American Hospital Association, decided that NAHIT needed to be formed - or, its members decided it needed to be formed. NAHIT is just in its infancy. Its bylaws have not been approved. They are just in draft form. The Board has not been appointed. Therefore, no subcommittees have been formed. The founding members of NAHIT, which are members of the American Hospital Association - and, they consist of providers, like myself, payers, like myself, the IT technical services companies and the supply chain B all are thinking in terms of now is the time for the field to start to engage in the standards discussion.
The case for action is pretty obvious. This is not news to anybody here. I know that. I just wanted to try to explain to you, I am starting to understand it myself - that is, the complexity of transactions from the healthcare field are tremendous, nothing like what we have in banking, where you have a 125 defined transactions. Therefore, standards are fairly easy to establish. Within the healthcare community, it is amazing the difficulty and complexity of the transactions. There is a lack of standard nomenclature, as you know. IT investment in health care has always lagged that of other industries. There has been a general lack of healthcare leadership in the field to want to fix the standards problem. We think that is changing.
It is the healthcare challenges that are driving the provider and payer leadership to start to go after the standards issues. It is becoming apparent to the CEOs and COOs of the various provider organizations that patient safety cannot be solved without the integration of very complex computer systems, that employee recruiting and retention - which is now a growing epidemic in the country with the lack of clinical job skills - is something that is going to require computerization and workflow management in order to improve the daily work life of our clinicians, physicians, nurses and other key personnel.
The other one that is hitting the leadership right on the end of the nose is around capacity management. That is, that the boomers are coming through the economy. We do not have enough physical plants on-line to treat all the boomers that are coming at us. We have to be more efficient and more precise to get more throughput through the hospitals and to have better integration between the hospitals, long term care, rehab facilities and home care in order to get better throughput to handle the capacity. These are the driving issues that have gotten the healthcare leadership in the field to start to pay attention around the standards issues, which is where, then, the American Hospital Association gets the support in order to help spawn the national alliance.
Let=s talk a little bit about the Sentara experience because the Sentara experience is one of the reasons why Sentara became a founding member of NAHIT. The first thing you have to understand is it is six hospitals. There are only one set of screens at the hospital information system level. There are not the general set of screens, nor the Lee set of screens. There are only the Sentara set of screens. They are non-negotiable. They are done in a collaborative way with members of the medical staff. But, once they are established, they are set in stone. What that means is that we can float nurses from one organization to another and when she walks into the unit, her processes as represented by the computer system are all exactly the same. That gives me better use of labor distribution.
We have, also, standardized on the charge of description master. We call everything the same thing. A chest x-ray is a chest x-ray, it is not a chest bilateral. That means that, then, we have standardized nomenclature within the health system. Now, here is where it gets really interesting. It is down around medical reporting. What we are able to do, because everything is standardized, is that we can pull the data from the actual clinical transactions that deduce the disease management. Once our physicians have agreed on an appropriate protocol of care, and the development guideline, we go to the computer system. It says for pneumonia. We implement it one time in the ordering pathway. It is immediately available to all hospitals, all care givers. The maintenance is done once.
The standardization, then, has given us the ability to, really, aggressively go after our disease management and to improve the quality of care in the Norfolk-Virginia Beach-Williamsburg-Hampton area. So, we are a big proponent of standardization. We are a big proponent of standards. We have great success as it relates to doing it on a very small scale.
Some challenges at the Sentara level is that we think that HL-7 and HIPAA are really good starting points. But, we do not see them as standards. We see them as guidelines. There is too much flexibility in interpretation within those standards that allow us to - that have the discipline that the banking standards would have. We think it is a great step and are moving forward.
The second point is, really, interesting. That is the reason I tote my 30 years worth of experience, not because I am great and wonderful. It, really, just tells me how really, truly ignorant I am. That is, all the work that is being done by these committees and everything that was listed at the start of the day today is not on the radar screen of the CIOs. We sit around and we talk about how it is that we want to do business between one computer system and another. We never have the standards discussion because we do not know about them.
Here is an interesting point for you. I would suggest to you that you follow the money. Once a standard is developed, it is the CIO who is purchasing the multi-million dollar system, if he wants to, in fact, enforce a standard, he turns to his provider of service and says I want this system delivered with the following standard. The marketplace will respond to the market. Once we understand what it is we are to buy, then, my brother and sister CIOs can help enforce the standard at the local level.
Here is an interesting fact. I have about 400 computer systems, major ones - when they work - that drive the health system to do some of the things that I talked about. I spent $600,000 a year for a staff to maintain the interoperability of those systems. They are working on the hand-offs between computer A to computer B on a regular basis. It is a full-time job. If we had the standards, or we as CIOs understood the standards - and, certainly, that would be a major cost reduction and increase in efficiency at the operational level. That is a little bit about the Sentara experience and why we, myself and Sentara and the other founders of NAHIT, are impassioned about trying to participate and learn what we can do to be good partners with the rest of the community.
Here is some of the standards that may be of interest to NAHIT as we start to form the organization. But, I am going to take a different - I say bar coding, patient identification, communication, application systems, you all know what that means. But, I am going to take a different slice at it because I try - because I look at it from the operational standpoint. Let=s talk about bar coding, the idea of positive patient ID for the administration of pharmaceuticals. You all understand what I am talking about. You all understand that is a very difficult thing to do. But, let me tell you what the opportunity that this presents for us as providers. Although, as I tell you about the opportunity, some of you will say, well, what you are proposing is illegal. What I am going to say to you is that is just a barrier.
Let=s take, for instance, an administration where we bar code the name tag of the person administering the drug. We bar code the patient=s ID and we bar code the med. Then, the computer acts as a safety net and says, okay, give the drug. That is how it works. Let=s assume for a minute that it is an oral solid. An aspirin, Tylenol, an oral solid. Why, then, is it that an R.N. has to give that drug? Right? Because another occupation, another job title could, actually, administer the drug if, in fact, the computer systems are sophisticated enough to act as a safety net for that administration of that particular oral solid. What that means, then, that goes back to the recruiting retention issue. That means that no longer, then, do we have to have a nurse push the cart around the floor. She can go back to doing things that she was licensed to do. It is those types of standards that, then, get the leadership of healthcare to start to focus.
Let=s talk about patient identification. I have out there the phrase EMPI, which you probably know stands for Enterprise Master Patient Index. There is a whole industry out there, there is a whole set of companies out there whose primary business is to step in the void where there are no standards. So, companies offering Enterprise-wide Master Patient Index - and, that is the ability to identify Bert Reese when I go to see my primary care doctor, when I get admitted to the ER, it says Bert Reese, again. Or, sometimes, in the hospital, I am known as Bert Reese, Bertram S. Reese, Bertram S. Reese III, because the registrar registered me three different ways. So, the master patient index there, it then is to adjudicate to make sure that when I get registered in any one of the environments that I am the right Bert Reese.
Same with the $600,000 that I have spending on interoperability issues - I use an interface engine that comes from a third party whose primary piece of technology is to act as a broker between one computer system and another. If we had the standards, those types of issues would go away.
The others - EDI, you are familiar with. Of course, what we are talking about is a lot of standards around the application systems which - and, then, nomenclature. These are some of the standards that are on the NAHIT hit list as we start to form that organization.
The mission of NAHIT, then, is pretty straightforward. It is to mobilize the field. The end result of it - as a result of mobilizing the field and getting standard - and helping to assist in the implementation, not necessarily the development of standards, is then to improve the quality of care, reduce the cost, et cetera.
I think in the initial discussions that we are being pretty elementary. We see it as the exchange of data. Ultimately, patient centralized data accessibility, better outcomes, certainly, we can lower costs if our ability to implement standards is successful.
Our direction is pretty straightforward - convene the players, which in our world is the payers, the providers, the IT manufacturers and the supply chain. Then, take out small projects that we think we could be successful at, not to try to solve world hunger, but, see if we can get some traction to get the field to recognize that it has a role and a responsibility in the adoption of standards.
In conclusion, we do not see ourselves as legislative. We see ourselves as strictly an augmentation to other processes, although I have to tell you that we, probably, do not understand all the processes. We are committed to a consensus approach. The strength is in bringing the field together in one particular environment where we can start to educate folks like myself about what it is and where we are on standards and what needs to be done.
Because of the issues that are facing us in terms of patient safety, retention and recruiting and capacity of management, the healthcare leadership is behind it. This is something that they absolutely have to fix as we move forward. We stand ready to work with the committee and whoever you designate in order to be successful.
Thank you for the opportunity to address the committee.
DR. COHN: Bert, thank you, very much. Marc Overhage, our final speaker for this session. Thank you for staying around for a second day.
DR. OVERHAGE: My name is Marc Overhage. I am an investigator at the Regenstrief Institute and associate professor of medicine at the Indiana University School of Medicine.
I am here this morning representing the e-Health initiative and its foundation for e-health, which I will call e-Hi, or EHI, subsequently in the presentation.
The purpose of being here this morning is to help the Subcommittee understand EHI=s efforts to bring together public and private sector stakeholders, to promote and coordinate the national adoption of health data standards and system interoperability, to meet clinical and public health needs. EHI=s primary mission is to improve the quality, safety and cost effectiveness of healthcare through information technology. It has members in a large variety of categories, including healthcare providers, delivery systems, healthcare IT suppliers, payers, non-profit organizations and professional societies, pharmaceutical and medical device manufacturers, physician organizations and research and academic institutions. Its strategic priorities are to increase the awareness of the role of information technology in driving greater quality, safety and cost effectiveness in healthcare. Secondly, to lay the foundation for an interconnected, electronic, health information infrastructure by promoting the adoption of clinical data standards and enhanced connectivity. Finally, to build the case for public and private economic incentives for better quality healthcare enabled by information technology.
As an example of the kinds of activities that EHI has undertaken in order to accomplish those aims, is a public-private sector collaboration for public health. The purpose of this collaboration was to develop and communicate strategies to effectively and securely capture and transmit standardized electronic data of public health importance and to facilitate that transmission using the CDC as a national electronic disease surveillance system projects as a starting point. Then, to leverage existing provider healthcare information systems and data sources to enhance public health surveillance processes to unlock the potential of that data, to achieve results that are applied in six to 12 months.
I think this is a critical point and a good example of where EHI is focusing energy. That is, to take quick hits, undertake projects that yield results that are usable, visible and teach us something in short periods of time. Then, reposition and take the next step. As part of doing this, we hope to lay the foundation for further development of the National Health Information Infrastructure.
A couple of critical success factors. First, the public and private sectors have worked very closely together on this project. Folks from NAHDO, from Denise=s organization, from the CDC, from information system developers, and so on, spent a lot of time. I will come back and talk about that together to make this work.
Second, the support of specific, well-defined, highly visible needs for data in public health. Obviously, the timing worked out well with the events of last year in terms of bringing focus and energy to this. A focus on data for which there is already an installed base. In other words, working from where we can move quickly forward from point A to point B and not trying to start at the beginning and develop data capture and so on. To align efforts with existing funding sources - taking advantage of where the money is. Finally, to leverage the work of experts and those working in the trenches, the people who are developing and implementing information systems and healthcare delivery systems, those in the public sector who are trying to receive, organize and transmit that data.
Next, to involve the key stakeholders, those that produce data, consume data. Then, we use the word, sometimes, data enablers, the folks who are in the middle, transferring the data around - whether those are healthcare information system providers or other kinds of data enablers. To drive usage of standards and strategies through implementation and demonstration projects. Then, to ensure that these efforts are aligned with national incentives, such as the NCVHS.
There is a more extensive list in the back of your written testimony copy. But, there are a broad set of collaborators in this process, including federal entities, state and local public health agencies, healthcare technology providers, healthcare providers, interestingly, holders of large sources of healthcare data - those who aggregate data for other purposes, perhaps - academic and research institutions, standards organizations and various non-profit associations.
The commitment of these members was, actually, remarkable. The collaboration included members and non-members of the EHI itself. But, the work was performed primarily by volunteers divided into seven working groups. These folks put in an incredible amount of time. People invested hours a week in conference calls, meetings, writing things up, reviewing things. It was phenomenal the amount of work that people put in.
Financial support was provided by the Joseph H. Cantor Foundation, the CDC and the e-Health initiative itself.
The products of this initiative in that short time period included proposed additions to the implementation guide for the transmission of laboratory-based reporting of public health reporting information, which is a CDC document. That includes standard formats and specifications of data types that had not already been addressed through the CDC=s efforts. This is an example of where the industry, the private sector, collaboration turned out to be incredibly valuable. These organizations shared data that shocked me in the sense that they - one organization, for example, had spent - it looked like several person years in reviewing microbiology HL-7 messages and looking at how they might need to be standardized, where
the issues and problems were. They summarized those results and they shared that freely with this group. We were able to take advantage of those sorts of efforts that had already happened. It made it a highly leveraged activity.
A second accomplishment was we looked at alternative methods and responsibility models for integration brokering functions. I will say more about that in a moment - as well as a conceptual framework for specifying components of microbiology results reporting. Then, a summary of the scope, methods and findings of sites currently performing syndromic surveillance. Then, it identified some key areas where there seemed to be gaps or issues that we thought could be best addressed through development of demonstration projects that would help us find where the real problems were. We anticipated that there would be problems, but, we needed to sort through where they are.
As an example of one of the work products I mentioned, the integration broker detail - and, if you look in the upper left of this slide, there are CS or clinical systems that hold healthcare data of various kinds. The dotted box outlines the sections of data transformation functions that have to be performed in order to make that data useful for public health functions. On the far right hand side, the various public health data systems that might need to receive that data.
Each of the ovals within the dotted boxes represent a different type of function that was identified that needed to be performed in order to take this data from its native form, if you will, in the healthcare delivery system systems and transform it into useful information for the public health systems. Those include functions such as modifying the format to standardize it so that code could be written once, as was just described recently that would handle the standard message sets dealing with the problem that the healthcare systems data is frequently not coded, in terms of its results. Yet, to do meaningful work on the health side, we need that, and so on.
The next steps that have resulted from this effort are to complete the implementation guides. Those are nearing completion. Also, to promote widespread adoption through communication, education and other vehicles targeted to providers, labs and public health organizations. All of you know there was a huge investment of dollars, for example, into the states and local health departments last year to help address this problem. One of the things that we think is critically important is that those dollars not go into a hundred different diverse projects that do not supplement and build on each other, but, rather some of those get focused into leveraging standards, leveraging health information in a consistent way so that we can get highly leveraged advantage instead of these individual local solutions. Then, to showcase success stories in order to maintain momentum.
The collaboration and EHI support of the PMRI standards and terminology as the standards for implementation - we think one of the things that really helps is the ability to hold up a banner and say to public health, in particular, here are the standards. If you develop those and use those, that will work.
Second, the demonstration projects will enable us to identify issues and provide feedback to this group on the applications of standard terminologies in the trenches, if you will, in operational environments.
Our use of standards is consistent with the NCVHS recommendations to date. For example, H-7, two point access, proposed as the reporting format for not only laboratory results, such as where the CDC was, but, order of microbiology results and chief complaint data. We are in the process, as I said, of completing the implementation guides.
Some of the gaps that we identified during this process were a few areas where accepted vocabularies were lacking, such as chief complaint. The problem that there has been slow adoption at the source systems of accepted terminology, such as LOINC for laboratory results.
Implementation guides have proven to be essential in order to realize the potential benefits of standards in terminology. That point was made in the last presentation. The ability of people to actually use them is highly dependent on having a cookbook. They really want and need that.
Third, a standardized approach to defining and communicating rules associated with data usage will be of great value. An example of this in public health is something called the Dwyer Tables, which is simply a list of what are the laboratory test result identifiers using LOINC codes that are associated with diseases or conditions of public health interest. In other words, a chlamydia DNA probe is something that might indicate an infection with chlamydia if it is positive. Chlamydia is a reportable public health condition in, I think, all if not many states. All this table does is say if you have a positive on this condition, then, it is a reportable case of chlamydia. That kind of highly integrated information is critical to take advantage of the standards.
Fourth, that while full adoption of PMRI standards and terminology may take time to penetrate into the healthcare delivery systems themselves, we think that current data systems, even when they do not today comply with the standards, can be leveraged through integration strategies, as we described.
Some of the key lessons we learned is that we think we are ready to move, at least in the public health arena, to early implementation steps. There are a number of initiatives that are going on today in the environment that provide calls for actions or forums for discussion. It is not so much that we need more of that. We have a model here of rapid short term intensive efforts that are focused on specific problems, in this case, public health, that can produce significant results in a modest time scale.
A partnering of both the public and private sectors leverages complementary skills and knowledge and expertise. EHI has tried to bring together multiple stakeholders. You saw those long lists. The public-private collaboration are committed to working with other organizations through overlapping involvement. You see the same names on many of these activities. Ongoing, open communication and concrete alignment of activities - we think that NCVHS can help our efforts by creating a clear and consistent message that there is a set - when you get there - of operable standards and terminologies that can and should be adopted in the near term by encouraging and supporting demonstration projects that implement those recommendations, including utilizations of the standards and terminologies in order to test them against use scenarios. We think this is particularly important in order to: A) identify where the critical issues are that might not be anticipated, and B) to show the world that you really can do it. It can work and the value can be achieved. Finally, to apply the lessons learned from these demonstrations when shaping future recommendations.
Obviously, ongoing collaboration with other federal agencies - you have heard about some of those other activities going on today. Leveraging the work of EHI and its collaboration for public health, the Connecting for Health effort that Dr. Hammond described. Taking advantage of that information and all of those details, findings, of that, I am sure, can be shared with you through the health initiative. There are not as well developed documents for some of those things, at this point, as we might like. But, there are detailed slide sets and white papers.
Finally, to identify the requirements for migration from the current environment to the PMRI standards-based environment. We think that is particularly important, as we worked with healthcare providers and healthcare information system vendors that migration strategy, really, emerged as a critical factor in how we think about things.
I appreciate the committee=s time this morning and look forward to the questions. Thank you.
DR. COHN: Okay. Thank you. Much to my surprise, we are actually running on schedule. What we are going to do is take a 15 minute break. Then, we will hear from Colin Price. I will ask the panelists to continue in their positions because after that we will just be having discussion.
[BREAK]
DR. COHN: We want to thank you for joining us for a second day. Hopefully, the jet lag is not too extreme. I would also invite you - and recognizing yesterday, I think, you had many important things to say. Unfortunately, with the amount of time we gave you, you went through them so fast that we will probably have to send you e-mails to get clarification on some of them. We would invite you to take enough time to make sure that what you want to share with us is well understood by everyone today.
MR. PRICE: Okay. Thanks for lunch. There are two things. First of all, my jet lag is always worse on day 2. Secondly, I have less slides this time. Being slightly jet lagged, I will undoubtedly be slower than yesterday.
I am Colin Price. I work for the NHS information authority in the U.K. I oversee the management of a range of projects that have to do with implementing personal health related projects, direct care things, such as some of the electronic record demonstrator work, terminology and messaging standards.
I was allotted 15 minutes. Clearly, I cannot give you a full appraisal of everything we are doing. But, I have picked, really, three areas. Firstly, to look at some of the emergent NHS national information technology strategies that we have had over the last 10 years.
MR. BLAIR: Could I just help for the audience - not everyone might know that when you are referring to NHS, you are referring to the National Health System in the United Kingdom, which has had a strategy for implementing an information infrastructure for healthcare for England, now, for what is it? - about three years?
MR. PRICE: The National Health Service has, actually, had at least two major strategies and two revisions of the latest strategy over the last 10 years, in fact. Yes, when I say NHS, I am referring to the U.K. National Health Service, not your Nebraska Health Services.
I will talk about our strategies. Then, a little bit about the structures that we have in place nationally to oversee the delivery of these. Then, something about a fairly new standards approval mechanism that we have put in place.
One of the difficulties I have had coming over here is knowing how best to add value to your deliberations with our NHS experience. I am, actually, following yesterday=s presentation, I did add five extra slides which, actually, use SLINIT CT implementation plans as an example of some of the issues and approaches that we will take to an implementation of a standard. They are, actually, on a separate hand-out, I think, that is being prepared this morning for you.
The first major IT strategy for the National Health Service dates from 1992. It is a pun. It is called Getting Better With Information. This ran for about six years. It was based, as all of our strategies have been, on the patient as the center of operational clinical systems and a focus on using information technology to support clinical care, rather than epidemiology in secondary purposes, but, recognizing that information captured to support care should be derived into secondary uses - trying to optimize cost effectiveness in an environment at that time, when there was a perception that health information technology was a fairly expensive luxury and, also, managing a lot of the expectations about what the outputs and what the benefits would be from IT in the NHS. That was administered by an organization called the Information Management Group that was within our department of health. Many of you may have met, at that stage, members of IMG who attended various international standards initiatives.
Following a lengthy period of development, which ran for about 18 months to two years, a new strategy was published late in 1998. That was called Information for Health, or IFH, as we call it. It was described as an information strategy for the modern NHS. But, definitely, a national strategy to be implemented locally. The game plan was that a lot of the standards would be developed nationally, but actually, the responsibility for implementation for performance managing that and for achieving targets should be met at the local level, at the hospitals and the organizational structures that deliver our family practice primary care.
This was a seven year strategy designed to run through to 2005. That is quite a long time for any IT strategy in any environment, particularly so in healthcare. At the beginning of last year, it was updated with an augmented document called Building the Information Core for Implementing the NHS Plan. Towards the early part of this year, we became concerned. A lot of the targets within the original strategy were not, based on experience to date, being adequately met. This, really, led us to rethink approaches to implementation. There is, now, a new document which some of you may have seen, which is Delivering 21st Century IT in the NHS. All of these documents, including that one, are available on our website. I will give you the URL for that at the end of the presentation.
I want to focus a little bit on our Information for Health strategy because this still forms the basis of most of our intentions. Its aim is for life long electronic health records for every member of the population. Originally, these were envisaged as a longitudinal electronic health record cradle-to-grave, with some episodic electronic patient records for particular episodes of care in acute hospitals, or whatever. We have moved back slightly from that situation, now. I will come on to what that means further downstream.
Secondly, we wanted collections to have round-the-clock on-line access to these records. But, also, to information about best practices to enable decision making and improve clinical care. We also wanted to ensure seamless care for patients as they moved through general practitioners, GPs, which is the community of family doctors, I think, and through hospitals and through community services, including chiropractic podiatry, physical therapy, that sort of thing. We also want to provide fast and convenient public access to information. That might include information about their own health by accessing their own records. It might include access to information resources for particular conditions. It might include access to information about performance of individual hospitals, provider units and even practitioners. Also, of course, providing health planners and managers with the information that they need. Principles where information is person-based and that derives from our 1992 strategy, integration of systems. The fact that management information will be derived from operational systems, the fact that information would be secure and confidential, this is a formidable challenge for the NHS, as it is for yourselves, but, with the provision that we can manage the security and confidentiality issues that information will, wherever possible, be shared across the NHS.
I do not think that there is any difference between this set of principles that I have outlined to you, here, and the principles that seem to come out of your own deliberations around your national intentions for health IT. Moving on to delivery structures, the current high level delivery structure is that of Sir John Patterson, who is head of research, analysis and information at our department of health. He reports directly to our junior minister and oversees a number of bodies, two are particularly concerned with information management. One of them is the information policy unit. The second is the information authority, which is the organization that I work for. That is a special health authority. It is sort of quasi-independent and is established by an act of Parliament.
The IA has about 850 employees and an annual budget of about 115,000,000 pounds, which is about $160 million. That does not sound a huge amount, but, you have to remember that that does not include the local implementation costs of the health IT strategy. The game plan is that the information policy unit commissions items of work from the information authority. We agree on a budget. We agree on a project plan for the piece of work. It is around this that sits the development of a robust business case. These commissions are, generally, within the framework of Information for Health, as augmented by building the information core, which is the update document last year.
All of this sits within a broader framework of a plan for modernizing our National Health Service, which is called the NHS plan. Somewhere in there is EGIF, which is our e-government and tropability framework. Even within the fairly simple structure, there are quite a lot of different aspects that need to be integrated.
John Patterson is, actually, very interested in the U.S. environment, particularly, he has expressed interest in your VA system as an exemplar of what a national electronic record might look like. He may well, himself or one of his representatives, have visited the U.S. to look at this system.
As I said, the original game plan was that Information for Health should be implemented locally. In order to achieve that, local organizations - 300 or 400 of them - produced local implementation strategies, otherwise known as Lizzies. These were, actually, to develop local strategies to suit local priorities. The NHS is quite a diverse organization. I will come on to that in a moment. Not everybody=s requirements are necessarily uniform.
Secondly, there is a lot of legacy infrastructure and systems out there. We did not want to be too prescriptive centrally and make it difficult for people to capitalize on what they had in place already, where it was appropriate. This local strategy approach was seen as a way of ensuring that that would happen.
They were asked to develop plans to show how they would meet the Information for Health targets. They were, also, asked to produce local plans for integration between the different components of the health delivery environment, the acute hospital sector, the community.
Clinical governance, which is an umbrella term for performance management, within the clinical environment has become high profile in the U.K. over the last few years. We have had episodes that you must have heard of, like our Bristol Run infirmary inquiry of pediatric heart surgery. Clinical governance support is seen as crucial to a lot of these initiatives.
Lastly, we have always been slightly concerned about the quality of data that is out there in the systems. Given that centrally required information is derived from these, we wanted some assurance that local steps are being taken to enhance the quality of that locally held data.
In the last year or so, there have been a lot of changes. This is a repeat of yesterday=s slide, but, I think it bears repeating. The NHS is, as I have said, diverse and slightly unstable. On the first of April, 300 new primary care trusts came into being that pull together the 9,000 general practices in the country that, between them, have about 30,000 care physicians. Twenty-eight new strategic health authorities have replaced eight previous regional health authorities. Family doctors have been given new contractual arrangements.
So, there is a lot of uncertainty. There is a lot of turf warring. There is a lot of people trying to decide who owns what problem or what bit of the problem. That is quite an interesting and difficult environment in which to continue to implement a national IT strategy.
As I mentioned yesterday, we have a general election in three to four years. Our government is heavily committed to NHS information technology - a lot of targets around electronic records, electronic booking and clinical messaging, particularly in 2005. This has led us to review our implementation approach. One of the effects of that has been to reconsider the wisdom of leaving local strategy development purely to the local communities and the feeling that we need to, at the very least, provide them stronger guidance, shall we say, to make sure that their strategies are conforming to our central template.
Also, there is a need to be much more ruthless in our approach to standardization. I think a lesson that we may have learned from the U.S. is that you seem to be far better within your health IT environment at engaging with industry and suppliers. We want to take an approach where we have a smallish number of prime service providers who, then, subcontract to a range of other suppliers to deliver, in various domains, the various components of the strategy. We would, obviously, wish to ensure that we specified for the prime service suppliers the standards that we wish them to impose on their subcontractors. So, a shift in approach, much more centralist, much more aggressive.
This has led - and, the basis for this is, really, to be found in the document Delivering 21st Century IT and, I think, a much bigger document that sets out the standards for the right hand vertical column on that, which is the life long health record service. These are both available from the worldwide web.
This is the high level, if you like, macro architecture for our national strategic program. There is a direct-to-general to be recruited, who will sit at the top of this particular sort of global headache. He will oversee the sort of top ring box, which is to do with procurement, the application portfolio, some of the changed management, new working practices.
There is a need to integrate with e-government and the IT industry, a need to deliver compliant open systems and political applications. There is a need to phase the approach and to - one of the first priorities and, again, the basis of this large specification document, is to begin to specify the standards for an electronic patient record system.
Three key targets for 2005 are electronic prescribing, the ability for a patient seeing a primary care physician to book an appointment with a specialist in a hospital there and then, electronically, and thirdly, what we are now calling the integrated core record service, which is the new personification of our life long electronic health record.
Underneath this are some essential foundation services, including consent, confidentiality and authentication. I have to say that, I think, in electronic record demonstration evaluation work, consent, security and confidentiality have been identified as the biggest barriers to making progress. I think one of the particular barriers has been to identify, centrally, who is the owner of this particular problem. I think we may just have cracked that, now. Fortunately, it is not me.
Then, there is some technical things that need to be put in place, like all of these things that are in the vertical column require extra bandwidth in our NHS intranet. We are re-procuring that in the next couple of years. We will be looking at the requirements for that. The bottom thing is data and data interchange standards, based on open technology.
That is our new direction. I think it is very exciting, very ambitious, very challenging, but actually, also very promising. I have been struck, sitting here this morning, about the number of players that you have. I shall always assume that the U.S., because you do not have a National Health Service, as such, would have a lot of players. But, actually, even within our NHS environment, we have a lot of people who have interest in the information agenda. Our organization sits top right, there is the information authority. But, all of these other players are interested in what we do, they want to have input in the standards that we adopt, they need to be engaged as key stakeholders in our projects. So, there is a sort of a heading of managing all these interactions that we have to do.
There has been quite a lot of turf war between these individual agencies, which is not totally surprising. But, I think that a lot of these, now, sit under the control of John Patterson. Through this mechanism, I think we are hoping that they will be a lot closer into working.
Moving on to how we go about agreeing on national clinical data standards - and, there is one line from Information for Health that points out that in the mid-1990s, we did not really have any disciplined approach for agreeing on the national clinical data standards for the NHS. Our solution was to create some committees. The main decision making body is called the Information Standards Board. There are three subsidiary Boards to look at clinical information standards, technical information standards, management information standards. Interestingly, industry, through our computer suppliers association, is represented on all four of those Boards.
There is, also, a fourth component that is not, actually, a Board. It is more of a function. That is to do with knowledge management and making the details of standards available to the Service and to industry and to others. That is the structure.
I have been trying to find a parallel between your committee and one of those. From where I sit, I think it probably relates, perhaps best, to the main ISB. That is just a personal reflection.
Our standards process, really, is what the Board would like to see happen in the approval of any standard. First of all, we would like to see a user requirements document to know what particular purpose the standard is designed to meet. I think this came up yesterday in some of the discussions around terminology.
Secondly, we then need a specification for it. Standards development is undertaken outside the information standards process. That is an approvals process which is kept separate from the development. That is why the Information Standards Board is not actually part of either the information policy unit or the information authority. It reports directly to Sir John Patterson as an independent standards approval process.
There is a stage where a standard is then approved as a draft. Before we will do that, there needs to be at least one successful implementation of this. Then, based on that, we want to see a number of other implementations, at least three, before we will approve it as a full standard. When it becomes a standard, obviously, there is then an ongoing process of, first of all, conformance testing, but also, monitoring. As I think Dr. Hammond said, standards change and need to be updated. That is the standards approval process.
What I am not particular to saying a great deal about is how we might enforce these. We have, traditionally, taken a range of approaches, including linking the incorporation of standards into systems to reimbursement for people buying the systems. We have had organizations to deal with standards enforcement and procurement. We have a new successor organization to deal with that. The reason I do not want to dwell on this is I think it will change as a result of our new, centralized approach to implementation. If you are happy with that, I will pass over at this stage.
That is, really, what has been happening within the NHS. Obviously, there is a lot of international standards offered because when we wish to adopt an NHS standard, we can choose from a range of offerings. It is the European committee, SAN, there is ISO. There is a lot of interest in HL-7. We, generally, see HL-7 as the strategic direction of travel for standard setting within the NHS. The reason for doing that is slightly subjective, but also, based on the fact that there seems to be a lot of industry commitment to what HL-7 is doing. That is regarded as important for the NHS. There are other involvements in IEEE, Corbamed, and other things.
We are keen to continue to be involved with the definitive international standards organizations. We are, also, keen to work with and learn from and share our experience with other organizations interested in implementing standards nationally, such as yourselves.
One of our major headaches is integrating standards and systems. We have a number of targets. For example, implementing electronic booking, electronic messaging, electronic records, which are progressing even as we speak. People are, also, in parallel, developing standards. In a lot of cases, there will be some point downstream when a new set of standards converges with a newly developed set of systems. A logical approach might be to put the systems development on hold until the standards are in place. But, actually, there are a few reasons why we would not want to do that.
But, we have to say that standards need to be interoperable. This is a problem for this particular situation. We need to develop and test standards in a live setting. The only way you can approve something as a draft standard is to have it tested. If we have to either re-engineer the standards or, actually, re-engineer the systems around new standards downstream, we perceive that to be potentially costly. We may also invalidate any evaluation work that has been done either around the performance of the application that uses the standard or the validity of the standard itself. So, this is a big challenge for us. It will be a big challenge for you, I think. I suspect neither party has the answer to this one, at present.
Now, finally, I just want to look, as an example, as CT implementation. This is sort of a bridge between yesterday and today, if you like. We have set some business objectives for the foundation for implementing the standards for the financial year. Our financial year, you realize, runs from April to April, which is why they always seem to run every two years. We want to refine this product, if necessary, to make it suitable for our key initiatives, particularly, these short term strategic requirements.
We want to manage the supply chain around this new standard, which means engaging with our end users, our political groups, our managers who will use the secondary information and, also, engaging with suppliers. In the new environment, that includes our prime service providers.
We want to integrate it with other standards. We want to get it through our standards approval process as a draft standard and, then, as a full standard. Because we are a public sector body, and because standards change, and because the world moves on, we need to establish a fairly minimalist, ongoing evaluation mechanism so we can keep a check that benefits are continuing to be delivered by using the standard.
There are some business issues that we have to address around distributing the standard, the terminology. Also, producing a formal, long term implementation plan - costing it and that sort of thing. Lastly, because - certainly, our primary care community uses an existing de facto standard which is the early version of the systems, we have the issue of providing migration support. That is what issues are established in our business objectives.
Second is looking at the organizational focus for implementing the standard. We have, basically, a range of choices. This is based on, say, implementing it for family doctors. What we need to know is who is responsible for performance managing the implementation of this and who will carry the cannon. Basically, our view at the moment is that probably sits with our 28 new strategic health authorities. But, establishing who owns the problem, who is accountable, I think, is quite important here.
The second I want to pick up on is, really, what do we do about moving from existing systems to new ones. I have a hunch - and, this is merely speculation at the moment - that the cost of upgrading your clinical system so it runs with SNOMED, rather than with read codes maybe only accounts for 20 percent of the implementation costs. I have a feeling that doing something with 15 years worth of legacy data and 50,000,000 primary care records might account for something like the other 80 percent of the costs.
The issue is, really, what are the benefits from dealing with the legacy problem. As part of our planning process, we need to assess the value of all of our primary care data and work out to what extent we should formally migrate it all, to what extent we should archive it and what are the implications of doing this. I think this is a difficult decision for any national initiative to implement a new standard, whether there are implications for the management of legacy systems and legacy data.
Just to show you how we are going about it at the organizational level, this year this is the program that is looking at rolling out SNOMED. EARTICA stands for Evaluating and Rollout of Terminology in Clinical Applications. There are two streams to it. There is an evaluation program that has a multi-professional quality assurance component, a component called ADVENT, which is to do with kicking off implementation in primary care. We will look at the migration issues.
There are two small projects to do with allied health and social terms. Then, a foundation program which picks up some of the business issues around SNOMED - a new implementation business case to augment the development case. Getting it through some of our formal, government gateway processes for large scale IT projects, developing a long term implementation plan and putting in place the long term evaluation mechanism. Lastly, a stream to do with implementing this new - integrating this new terminology. We need to make sure it maps to our 55 data sets currently under development in the NHS. We need to make sure it cross maps to ICD-10 and to our procedure classification, OBCS-4, so that we can derive secondary data from it. We need to put in place some U.K. extensions for U.K.-specific drugs and for administration terms. So, this is a program that costs about $3 million, that runs for this year, just to put in place the implementation of one new standard. It will put in place the foundation for the implementation of one new standard across our National Health Service.
In summary, then, we have had over 10 years worth of national health information technology strategies and, still, we are coming up with new ones as the world moves on. Our current focus is on implementing a number of key targets within the next three years with centralized management, ruthless standardization and recognizing that industry has a key role in this.
We have set up a separate standards approval mechanism that needs evidence of live implementation. It, actually, also needs plans for how the human behavioral change management aspects of the new standard are going to be managed. It, also, needs to show evidence of stakeholder engagement.
Lastly, I think we will recognize that standards integration remains a significant challenge around this. Thank you, very much.
DR. COHN: Colin, thank you. It was very interesting and useful. I think we have a fair amount to ponder about. Now, before we go into open discussion, I invite you to stay seated and participate in the discussion. I think we have a brief presentation from David Roberts.
MR. ROBERTS: Thank you, Mr. Chairman and members of the committee. My name is Dave Roberts. I am the new public policy director at the Healthcare Information and Management Systems Society. I would like to thank you for this opportunity today to describe several of our national initiatives that we believe will help promote and coordinate health data standards in support of achieving systems interoperability, data comparability in order to meet clinical and public health needs.
I would like to ask your permission to include my entire written statement in your transcript. I am just going to summarize briefly.
We have, basically, five ongoing national initiatives right now. I wanted to begin by stating that HIMSS has a strong Board of Directors and member commitment to our initiatives. We have over 13,000 individual members and 90 corporate members, now, of HIMSS. As recently as August 16th, our Board of Directors met and reaffirmed their commitment to our activities. We believe that our members=extensive involvement in all of our activities through SIGS or task force or task groups or committees has made HIMSS a leader in the healthcare technology arena for over a decade. Basically, our five major ongoing activities - first, we have a very successful initiative called Integrating the Healthcare Enterprise. It is a five year project. It is done in coordination with the Radiological Society of North America. We are in year 4 of this project, right now. Its basic purpose is to improve systems interoperability for sharing patient data by promoting coordination of established communication standards, such as HL-7 and DICOM to address specific integration needs.
Secondly, we are working closely with other associations, such as the Coalition for Health Information Policy, which is made up of HIMSS, the American Medical Information Association and the American Health Information Management Association, who is represented here today.
We are, also, working with the electronic health initiative. We recently co-signed a letter with them for promoting some legislation on Capitol Hill. We are, also, a founding member and on the coordinating committee for the National Alliance of Health Information Technology.
Thirdly, on August 16th, HIMSS successfully completed a merger with CPRI Host. We merged its organization into HIMSS with the goal to make the electronic medical record ubiquitous in healthcare.
Fourth, we are planning on sponsoring a congressional-type field hearing next February 11th in Seneca, California, in conjunction with our annual conference. It is going to be made up of industry experts and congressional staff. We are going to take testimony from three panels. I am sure Rob will be pleased. It is going to have the provocative title Should the Federal Government Mandate an Electronic Health Record? We are going to provide this information back to your committee. Also, the Senate Health Committee has asked for a copy of it. Also, the House Ways and Means Committee.
However, because of the time constraints today, I would like to just highlight our fifth and top initiative. At our June Board of Directors meeting, HIMSS unanimously adopted the NHII as the society=s long term public policy initiative. We established an NHII task force. I am pleased that one of the members sitting at the table here today is a member of that task force. We believe that this initiative will, really, offer an opportunity for our members to play a key industry role in shaping the direction of the NHII.
The purpose of this task force is going to be to take on activities that drive the country towards the creation and universal adoption of NHII. Initial projects that we are looking at include creating an inventory of existing technologies and practices, developing a gap analysis on what needs to be created and creating a HIMSS version of the NHII. We do not want to duplicate what others are doing. We want to add to what others are doing.
Most importantly, HIMSS believes that a key role that we can play with our membership in this area is to help assist to standardize the terminology used throughout this area and help identify or reaffirm the relative attributes of each term.
In summary, HIMSS is committed to working with you and others in our industry to establish standards and systems interoperability. We thank you, again, for the opportunity to present this written and verbal update. HIMSS wants to use the expertise and knowledge of our membership to help you meet your goals. We welcome any opportunities to be of assistance. Thank you.
DR. COHN: Thank you, very much. I think it is time for general discussion, at this point, questions and otherwise. I think that maybe I will start out. This is very much of a level setting. I heard during our earlier presentations that there seemed to be sort of a plethora of e-health entitles and, certainly, if I were on the Internet, I might be a little confused about what initiatives are really related, what initiatives are not. I would, specifically, I guess, ask Rob Kolodner or Ed and Marc to comment if there are interactions, since all of you seem to have e-health as some part of your main, to sort of indicate what the interactions or relationships are. Rob, is there any relationship between HealthePeople and any of the other initiatives around the table?
MR. KOLODNER: The HealthePeople is a strategy that builds on a number of the things that are going on in VA. In fact, VA has been very active in a wide variety of the activities that are mentioned today, those things that are public kinds of entities. We end up being involved in a lot of those and bringing to those efforts a perspective and involvement in bringing back to our system an attempt to actually implement, on a day-to-day basis, the kinds of things that might be ready to be implemented. There is an ongoing dialogue, at least, in those activities we are part of, whether that is HL-7 or whether that is Connecting for Health, or a number of things like that.
I think that what you hear is almost a web of connections across these efforts. The approach that each of us is using is to see how that particular niche where we are working in, how does that advance the whole effort. It is not a single, monolithic thing that one group, or one initiative, can push over the finish line. It is something that covers a wide arena. We heard, for example, a group that is, really, represented less well at this table and this conversation, is the individual practitioner. And, yet, that is where the majority of information, key information, about individuals is. That is, probably, the place - at least, in America - I think the U.K. has done a better job than we have in this area - this is a place where there is almost no penetration of electronic support.
At least, those of us at the table who, for the most part, are representative of the institutional level, a lot in the health provider arena, but also, in some others, we tend to interact a fair amount in a variety of areas and switch one hat after another, as we heard from Ed and Clem=s list - and, even Jeff=s list of activities.
DR. COHN: Well, maybe then, I can ask Ed and Marc to identify is there a relationship between - either formal or financial or others - between the Markle Foundation activities and e-health initiatives? Just so we understand.
DR. HAMMOND: I want to go on record as saying that I am now sort of speaking for myself with some of this because I think the level of cooperation between the groups is the highest it has ever been. Rob and I, and lots of other people, are working together in, probably, four or five different settings. In an ideal world, we would be working together in one setting. I think that is the very, very big difference. The problem we have to overcome, and I do not know how to do that, but maybe you all could help lead us in that direction, is the fact that each of these organizations starts off with a constituency with funding with a purpose. You meet up another organization that is similarly inclined. The conversation is, yes, let=s work together. Come over and join us. We will do this together. So, we need some neutral way of satisfying the needs of our constituencies. The individual needs of the group itself in putting these together so that we do the work once in one place.
The Connecting for Health has brought together - each of these bring together a larger set. Part of the problem is communication. Somebody pointed out that knowledge - I guess you had the bullet that said that we really need to inform. Most people do not know what is really happening in this world. In fact, people that are involved in doing some of this are not aware of everything that is happening in this world. This is why it is necessary to almost participate in each of the activities as it takes place.
The good news is, I think, that this meeting, itself, we have had some side communications already that is promoting a level of working together with a group beyond this. I think the intent is absolutely there. I think it is process that we need. I think that the U.K. has solved the problem by having - I love the word information authority.
We, basically, need an authority that, really, helps us put this stuff together. We get to do what we want to do. We get to do what we are doing, but, we are doing it in a slightly different environment.
DR. OVERHAGE: As Ed said, it gets difficult to know what hat you are wearing when you are talking. I won=t try to wear any. I will probably get sunburned. But, the - I think that the communication, I like the description Rob had as a web of communication because that is really how things are working today. For example, in the public-private collaboration in public health that I described, the e-health initiative undertook with the CDC, one of the products that came out of the working group was we are, really, struggling with what to use for a drug vocabulary, a very important issue. I do not know that that necessarily spawned directly - but, what was fortunate is that it was identified as a high need and all of a sudden, like magic, appeared what it looks to be a very viable, powerful solution to that problem.
I think one thing that e-health is trying to do is, number one, not duplicate what somebody else is doing. They want to take advantage of leverage and enhance those communications so that we do not do that. Number two, and I think this is the other thing that e-health is focused on - and, I will probably get wet lashes later for getting this wrong - but, is to try to leverage things that are going on in industry that, sometimes, might be viewed as being the evil empire. But, they actually have a lot of value to bring to the activities that we are doing.
I shared the specific example of work that had been done in the organizations in message structure, for example. What I routinely hear from the industry groups that are working with e-health are we would like to help. We do not know exactly what to do. We have people. We have brains. We have a variety of things. Ask us when you have a need. We will try to contribute because we want to contribute because, A, we are good citizens and, B, in the long run, it helps us. Not in the short run, maybe, but in the long run it helps us. I think what e-health tries to do is leverage those kinds of assets that we have by building collaborations that we might not be able to accomplish things otherwise.
MR. McDONALD: I think what - you were a question and what I - I do not want to put words in your mouth, but, what I was thinking, too, was, boy, we have a lot of these all of a sudden. You are, maybe, wondering out loud whether that is good or bad. I, actually, think - I wonder that, too.
I have, actually, put it together in my own head in a way that made a lot of sense. But, it maybe just mutated some thoughts. That is, I think that the e-health thing has got a strong public health volunteerism sort of slant. The governmental agencies - there is a lot of - sort of Rob=s - the one he described, there is a lot of threads of crossover for humans in these different things.
HL-7 is, actually, sort of a mixer and allows this because the same people are going back and forth between HL-7. I liked how HIMSS described their role. In my life, I have always tried to merge organizations because we do not need that many. You cannot go to that many meetings. I liked how they described their not taking over this or supporting something that already exists. That, I thought, fit kind of nice.
I, frankly, was not quite as sure about the alliance, which seemed to be mostly focused on bar codes, which is also sort of non-overlapping. But, it is brand new. I was not able to articulate - I am not sure there is as much human crossover, which would be a good thing if there was, between some of these other groups, as I know there is and these are very highly incestual. That was a bad word. Actually, I just had some recent training in what you can say in the office and not. I think that was, probably, a bad word. So, I will take it back.
DR. COHN: Clem, like you, I try to put order into all of this. It is obvious from my view it is sort of hard to know what the organization represents versus what the people represent. I personally want to thank everyone because I think it is a wonderful panel. We are seeing almost all aspects of - not the food chain, but, it is the chain of standard development to getting things in. We sort of see - I heard a lot from Marc about the basic elements of how we move forward. I heard from Ed, there are certain things that if we just get them accelerated and, then, implemented - what is it going to take to do that? Obviously, CHI has a whole set of issues about really making things real.
I was struck with Bert=s comments about - as a CIO and, once again, I do not know if this represents the alliance or his own views - but, I was struck and reminded that in some ways we are really successful with all of this stuff. We are going to make standards really, really boring. The HL-7 is only going to have to meet once a year, believe it or not. Sometime in our -
SPEAKER: Standards are boring.
DR. COHN: Yes, well - what I am saying is that it is - and, we are not going to have much to do in terms of this. Obviously, we will all have other things to do, but, in a sense, we would like to get to the day and going, once again, into a nice simple analogy about the electrical plug and the electrical sockets. I do not know that it is really Bert=s responsibility to know that he needs version 2.X of HL-7 and X implementation, along with mechanisms to get NCPDP transactions and X-12 transactions and on and on and on. What he needs to be almost assured is that when he is buying something, it does all the things that he needs and that it is interoperable. If I were in his situation, I would be sort of demanding that of the marketplace.
Obviously, it is everybody else=s responsibility to move that forward and fill in all the gaps that we have been talking about to that point. I guess I also reflect on my own organization, which is Kaiser Permanente, where our CIO is not only in charge of all the IT, but, he also deals with all the administrative transactions in his spare time and the administrative systems and processes. So, he does not - probably like you, Bert - does not sit around and spend a lot of time thinking about exactly what version of a standard he needs. But, he would dearly love to have it already in the systems.
MR. REESE: I have tried to put my arms around whose fault that is, whose fault this is because that is - I am a manager, that is what we do. I finally found the target. Your electrical plug outlet is a great analogy. The fault of this lies at the feet of innovation. I think when technology or processes or electrical circuits, or whatever it is, or when they are in the innovative phases of development, there will be a lack of standards and a lot of trying and error, trying to figure out how to get it right. I think the reason why the marketplace has not gotten there is because innovation has reined healthcare, both in the clinical diagnostic computing equipment - MRIs and CT scans - back into the administrative support systems. What we think is some of the areas within healthcare no longer require innovation and, therefore, there is a good place where standards could be. Then, part of it is like an ADT transactions, admissions discharge transactions. Some parts of healthcare no longer are being - are at a point that require innovation.
Up in the clinical areas, where you start to use computerization to help out the patient safety agenda, there is still innovation occurring up there. So, standards, if put in place, may curb some innovation. So, it is a balance. But, at some point in time, somebody had to say let=s make that electrical socket that way and let=s all innovate around that particular agenda. Where that line is, I have not figured out, yet.
MR. McDONALD: In response to that, I think you are under shooting. We have gone to all five major hospitals in our city. We can get everything we want clinically in an HL-7 message. The only problem, really, is on the coding side. I am, really - I will bet you money that you are more HL-7 standardized in your organization than you are thinking. But, you have too many other things to worry about.
MR. REESE: You are right.
DR. COHN: The final comment, of course, I would make in all this is that Denise - hopefully, one of the great beneficiaries if we can all do this right, bringing up the point, very rightly, that we can have all of the data in the world and if we do not have a plan about how to use it, people to help, people willing to share the data, we will have created a great edifice which will have not great public health value and value for public policy.
MS. LOVE: For about 20 years, NAHDO has believed in the trickle down theory of standards. We are hoping that you can solve some of the problems. I heard about a healthcare authority in the U.K. I am just counting in my head. I think the states have about 39 healthcare authorities, or commissions, who make data policy for that state. That is just another wrinkle in this conversation - how to get them on board and how to get the message to those authorities that help make data policy and inform the Department of Health about new data sets that they collect.
MR. KOLODNER: I think when we identify a metaphor, we need to be careful about how far we can push it. I think there are - we talked about a variety of standards. The question is can we get a messaging standard that stabilizes so that - it is like the plug and the socket. The problem that we have when we get into terminology is we are changing the electricity, or the forms, or the things that are going across. That is the nature and I would expect into the far distant future, be the nature of medicine because we are advancing medical knowledge. The need to update terms as the explanation of disease changes from peptic ulcer as hyperacidic condition to an infectious condition, as knowledge changes, there are things that are being communicated that are not going to become stable and, therefore, you stick with it. But, there are parts of it that may allow us to put something in place that does not have to change as rapidly.
I am not sure that I have seen a conception of the world of standards to see which things might be getting to a point where once we are sure they are working, they can change much more slowly. I think it is almost a difference between information models, data models, which can change more slowly once you get it where it is serving its purpose. Then, the content and the terminology, which is still going to need updating.
DR. COHN: Yes. I will make a comment on that. In the HIPAA world, obviously, they talk about not changing the message standards very often. On the other hand, you can change code sets very frequently. So, it is, maybe, that sort of distinction. Mike, go ahead.
DR. FITZMAURICE: As I look around the room, I see a lot of - I won=t say old friends. Any one of you could give the vision, anyone sitting around the table could give the vision. While I frame my questions for CHI and, probably, for Rob, it is really a question that can be addressed to any one of your organizations.
You have done a remarkable job educating your constituency, your partners. CHI has done a great job of focusing attention on the need for federal government to be a follower in the adoption of private sector standards. Indeed, one goal is for the CHI standards choices to be an acquisition standard for the government. I heard that statement. I am not making that up.
I would ask are there plans for CHI, or plans for your organization, to be a partner in the development and use of healthcare data standards beyond education? - first, by participating in standard development organizations and the working groups, to get the standards you want and your constituents can use. By joining in the standards coordination efforts of others, I see a lot of you interchanging like interlocking directorates. I think that is good across these organizations. We need commonality of education.
Thirdly, by funding jointly among the constituents, among the agencies in the case of CHI, specific standards development and limitation activities. Then, finally, by demanding standards of interoperability in which a purchase - I think, if you push that, then, you will help the market to get where we all want it to get.
I would ask you are your organizations prepared to move in that direction, or is it mostly an educational organization? I will start with CHI. Anybody else can join in if they want to.
MS. ADAIR: This is Jared Adair. Rob Kolodner kind of looked at me and asked me if I could respond to your question. Of course, he did that in midstream and I noticed you had multiple questions in there. So, I may not get them all, but, let me try to answer what I can.
I believe your first question had to do with whether or not members of the CHI council would participate - or our staff would participate in standard development organizations. The answer to that is yes, they have and we would hope that they would continue. That is an effective way for us to participate in the development - you know, have our needs known at the very beginning, at the very threshold. I believe that that would continue.
I believe the second question you asked had to do with coordination between not only the wonderful panelists today, but, between other groups. I will be forthcoming and tell you that that is, really, one of the reasons why we came to the NCVHS and why we, I think, kind of - one of the things that we would hope to get out of something like this is how is it that we, as a group of federal sector, what we kind of refer to ourselves as the federal healthcare sector, representatives, those of us that pay for, provide, touch, use healthcare information. How is it that we can have good conversations with these other groups? How can we have a forum to do that in that is all appropriate and that many people can understand what those conversations are? I think the we are kind of, really, anxious to know how we could do that. I mean, it is a very important element for us.
I lost you on the next question. I am sorry.
DR. FITZMAURICE: [OFF MIC]
MS. ADAIR: Funding joint -
DR. FITZMAURICE: [OFF MIC]
MS. ADAIR: Well, sure. I need to, now, push back just a little bit. By that, do you mean standards, or do you meant that in systems applications?
DR. FITZMAURICE: I think I mean systems applications [OFF MIC]
MS. ADAIR: Okay, sure. We have been - we have tried to be, and may not have been, very clear about what we believe the scope of our initiative is. You can call it that we may have moved the line too far inward, which is a fair one, which is that we believe that our function is to adopt, for this federal healthcare sector, what the standards are. And, stop at that point. Do we believe that the follow-on that will happen are those types of things? Yes, I do. But, it is not within the context of the initiative we have put on the table. Do I believe that things will happen as a result? Yes, I do.
MR. KOLODNER: I think that is where the allusion to a related federal activity comes in because, obviously, if, in the context of CHI, there is a federal adoption of a certain standard - say, LOINC - then, VA and DoD, and any health service in the provider organizations, can work together at that systems application level - the HealthePeople federal, to actually collaborate and leverage the investment that each of us is doing, to move it forward.
MS. ADAIR: Because - let me just finish to -
DR. FITZMAURICE: [OFF MIC]
MS. ADAIR: Right. But, I think that I have to go back because when you talk about the sharing of applications, one has to take a look at what an application is. It is the personification of one=s business rules. So, one has to take a look at where the business rules align. Standards, vocabulary, messaging, and those types of things, are not necessarily business rules, but how you move. So, there is - in my mind, there is a difference.
DR. FITZMAURICE: [OFF MIC] that will support different functions. CHI has given some very good answers to that. In each of your own organizations, these are some of the questions that the public, and others, are going to be asking, to measure your successes. We have never had a time in the country when there has been so much vision, so much common vision. [OFF MIC]. So, we have a pretty good base of education to move on.
MS. ADAIR: Our intent is to - just from a CHI perspective - take a look at what is out there, what NCH has brought forward to us and take a look at the types of things and to not recreate any wheels, but, to try to build off what has happened.
MR. REESE: From the national alliance perspective, I think we see - even though we are in the formation period, I think our view is a lot - is more finite. We are the traction part of the process. That is, we will represent the market. We will represent the purchasers of the systems. We think we can participate in standard development, as appropriate, whatever that means. We can participate in standard testing because we have the environments to do that. But, more importantly, we have the providers, the payers, the technology companies, the supply chain, all in the room. When we agree on this is the direction for a particular standard, it is the providers, then, who say we want to buy the systems based on the following standard. Then, the market will speak. It is a limited view.
DR. FITZMAURICE: It is the producers and the purchasers and the vendors of systems, they are going to be driving the standards. The standard development organizations want to have the demand for what they produce. [OFF MIC]
MR. REESE: I will move from the national alliance to a personal comment. Sometimes, we require adult supervision. So, I think as the various organizations have developed standards that are generally acceptable, and we as the purchasers can understand that, then, we can push the providers of those services, the vendors, to meet those particular standards.
DR. FITZMAURICE: [OFF MIC]
DR. HAMMOND: I think that is a multi-part question. I am going to wear multi-hats in giving you my idea of answering that. From the perspective of Connecting for Health - and, I think I would answer many of your questions, yes, I think the intent here, really, is to try to cover that spectrum, and try to identify, influence and endorse, and be very, very specific in terms of what we are trying to accomplish, and recognizing the fact that - I mean, I object to the words that we should prioritize this. For the most part, the truth is that if you are missing a link, you have a gap that won=t let you do what you want to do. So, it is not that you have the choice of doing a few of these, you have to do enough of it that you have - the car will run on the road. That is part of the problem.
The second part, I think, is that most people misunderstand how to create a standard. There are multiple ways of creating a standard. If you want a standard quickly, MicroSoft will give you as many as you want. They do it without a problem. You see what you get with that particular process. The government can create standards a lot faster. If you go to the consumer world, to the open world, to the consensus world, there is a process. That process involves the users of systems, providers of care. It involves the vendors that have an investment with what they have, willing to change, but, are willing to change only connected with the user community and the demands of the user community, as somebody said here. That has got to be staged. There has to be a motivation for doing so.
If you have something that works, you are not going to change it because somebody came out with new technology. Those are pieces that, really, have to built into the system. I do not think interface engines, for example, are bad things. I think they are good things that permit us to live in a bracket of time instead of a line of time. That is a piece of what we are doing.
Finally, I think nobody fully understands how much money it takes to create standards. It is a free process and I firmly believe that anything that is to be used by the public - either by requirement or by encouragement - has to be free. So, I think that these standards that we are producing, really, do have to be available and used for the marketplace itself. So, there has to be a stage process in which open systems, I think, in terms of the contribution - one of the things that keep occurring is a new set of people will discover a new need. If the current standard does not do it, their interpretation of that is not to go back and fix the current standard. It is to create a new standard to be different. We have to change that mind set for what we are doing, as well.
The other thing is, simply, the misinterpretation of what the standard means. The comment, in terms of HL-7 - I mean, Clem and I both, probably, have used HL-7 for a number of years to transmit entire patient records, in both in-patient and out-patient care settings. I will admit that I control both sides of that. That is a critical piece for making that work. If I control both pieces and I know how to interpret that and if I do not control both pieces, I have problems. What that really does is go beyond the messaging itself to some type of conformance agreement on the part of the players. So, creating the perfect messaging standard does not guarantee interoperability. There are other pieces that have to be there. Hence, my comment. We cannot just do one of these things and expect it to do everything. It has got to be the suite of standards that we use to make all of this work. But, we are getting there.
DR. COHN: Okay. Jim and Jeff, then, Bill.
MR. SCANLON: I want to thank everyone for a very interesting and insightful presentations. Just a couple of things. I think, to the extent that each of the initiatives is focusing on standards, they are more or less, probably, the same standards, or hopefully will be. It is the same people. So, hopefully, we will have agreement. You can, at least, agree with yourself the previous day. The basic suite, presumably, will come out the same.
There were a couple that, actually, were focusing on technology rather than standards. I think Bert=s - the bar coding and other, they were, actually, information technology applications which, presumably, reflect the hospital=s set of priorities in terms of where they would like to put technology investments over the years. Again, strategically, I am sort of looking at what is the basic model for carrying through - I mean, this is a developmental and an identifying consensus standards. And, then, an education kind of component, as well, and then, using your consistencies to sort of move this through to actual buying it.
The basic format is still the voluntary consensus, business case framework in the United States. Where do you see the capital coming from to, actually - where do you see the capital coming from in hospitals, for example, Bert, in actually making these transitions and changes?
MR. REESE: A couple of comments. The capital for standard development, I am not sure. The capital to have a vendor adopt a standard will come from the purchase agreements that we sign with them. The beauty of the alliance and association with HIMSS is if we can get the education right - I say that we have an education problem, I would like to use another phrase which is more of a Bertism, it is that we have a sales problem. A lot of people have been doing a lot of good work. I think one of the responsibilities of the alliance and, I think, one of the responsibilities of HIMSS, is to make sure that we can understand it and get it down to the membership so we know what to ask for. I think once we have a consensus that it is good for our communities, then, the money will follow with our purchases.
When I buy a computer system now, I have to put so much money in for interfaces and implementation costs. If I am going to buy a particular piece of software, or PAC system, or something different, I collaborate with at least 10 or 15 CIOs before I make that purchase to figure out what version of what interface works well. I spend as much time evaluating the interoperability of the systems than I do the actual application. So, I think - that is something that is typical of HIMSS and CHIMES and HI-C, and all of these other professional organizations. Now, we are thinking within the alliance, we will be able to take it with a more - CHIMES and HIMSS have other agenda items. We are hoping that the alliance only has one agenda item. That is, the adoption standards in the field.
DR. COHN: Jeff, I think you have a question? Bill, then -
MR. BLAIR: Ed, I am thinking way back to the time, it must have been 10 years ago, when you wound up coming up with an initiative to try to accelerate the development and implementation of healthcare information standards. I think many of us who have followed your lead on that are still trying to do that. It sounds like you are still trying to do that, as well. So, I think we all are sharing that vision and that direction.
The experience that I have had in the last five years on the NCVHS with respect to our responsibility, NCVHS=responsibility, with respect to evaluating, selecting and recommending standards that would become part of HIPAA has been sobering. I think we have continued to have the same sense of urgency. But, of course, it has taken much longer than we had thought. The NCHVS, also, wound up trying to pull together some kind of a direction with the national health information infrastructure. We are still in the mode, now, of selecting HIPAA PMRI standards. We have done the message format standards. We are beginning the process, now, of doing the same thing with vocabulary standards.
One of the things you may have noticed, that in our last recommendation on that, because we noticed how long it took to get things through the cycle with an NPRM and final rules, we thought it might be better to go with the idea of recommending standards where HHS would indicate that their guidelines, along with incentives to accelerate standards - that is one part of the preface. Now, I am Tevia from Fiddler on the Roof.
On the other hand, we hear Colin Price=s presentation to us. After many years of trying to be very open in the United Kingdom, they have decided to become more aggressive. I am sensing, with your presentations - all of your presentations - and the fact that there is a Bill in the House of Representatives and one or more Bills that are about to be introduced in the Senate, and efforts within the Executive Branch, and all of your initiatives, a new sense of urgency to move towards - whether we want to call it a national health information infrastructure, or something else, but to move healthcare into the information age and get the benefits of the paradigm shift.
Now, here is my question. Does this new sense of urgency translate into a desire on your parts - I would like to hear from all of you - that the federal government should be more aggressive, like Colin Price has indicated they are in England, in terms of not only accelerating things faster, but maybe, going towards mandates rather than guidelines, or should we leave that acceleration to the private sector? In short, does this sense of urgency mean that the federal government needs to move faster and take more of a leadership role or not? May I hear from you?
DR. HAMMOND: I think that my answer today would be different than my answer, perhaps, two years ago and, certainly, four years ago. But, part of this, in answer to your question, is, well - and, I am purely speaking for myself. To some extent, the government has gotten a free ride in standards. You, basically, have sat back, watched the private community develop those standards, made comments about the slowness with which we do this, but, have not gotten engaged. Standards you have produced, you have produced on your own. I think we, now, truthfully, are at a point of public-private collaboration here and engagement. So, I think that what I would like to see in terms of accelerating this is, really, a partnership. A partnership not only in funding, but also, a partnership in expertise, a partnership in defining need in doing this.
The second comment is creating standards is a multi-staged job. The first step of that is getting the technical expertise that is necessary to create the standard. The second step is to broaden that to a slightly larger community, to cut off the sharp edges, to polish the process, to close the gaps, and so forth. That, with funding, can happen rather rapidly. Then, it is less important - but still, the time it takes is less important - but, still an important piece of this is the buy-in and the ownership of the community-at-large. That community-at-large includes the vendor population, in general. It includes the government and the regulations. Ultimately, it includes the providers of care and the consumers. So, I do think that we know enough to accelerate this process without giving away any critical pieces of what we are talking about.
MR. BLAIR: What do you mean by giving away?
DR. HAMMOND: I think we can still retain the openness that I think is extremely important for what we are doing. It is one thing to create a standard and hand it down as a mandate to everybody and say go do it. It is another thing when an open process is creating that standard. Truthfully, I think that the product community, absolutely, does not object to the public community being involved in this. In fact, we welcome it with open arms as a partnership because much of healthcare is a partnership between those two groups.
MR. BLAIR: So, the balance is - you still feel - is fine?
DR. HAMMOND: In fact, I will tell you that my experience has been the minute you start working, you forget who you represent and where you come from. It is a group of people solving a problem.
DR. COHN: It is all the same people, anyway. Other comments on this one?
DR. OVERHAGE: This is Marc Overhage. Just briefly, one of the senses that I get, and I have not been deeply engaged in the standards development, as Ed and Clem, and others have been over the years, is that there - and, Rob used the phrase Atipping point@ in his presentation. I feel like we are, at least, close to that in the sense that what I hear is - and Bert said it - it is that the vendors are saying we will implement whatever the market tells us to implement. The people who are paying their bills are saying we will ask for whatever we ought to ask for, but, we do not know exactly what to ask for. Federal agencies are saying, golly, we would love to get stuff in a standardized way. People are all kind of poised and I think the challenge, right now, is how to make the shoe drop or make somebody blink, or whatever metaphor you want to use. If there were a clear - you know, this is the thing that we are going to use. It does not have to be a mandate, I do not think. I think if there was a clear, yes, this is it, the vendors will build it because people will ask for it. The federal agencies will get it - but, the question is how to tip it over the edge. I do not have that answer. But, I think that is the real thing that is needed. It is that nudge and not so much somebody to say you have to do it. I think if there is clarity, that if you use this, this and this, you are going to have 90 percent of the problem solved and we can give you some confidence about that. It is going happen. We are on that precipice.
DR. COHN: Denise, comment?
MS. LOVE: Having mandated at the state level, I agree that that is one tool, but, only one tool in the tool kit. I was involved in two different data development aspects at a state level, one with a mandate. So, the hammer. It got people at the table and it got the attention. But, then, a health information network that came around voluntarily around a specific transaction, they went rapidly. They standardized. It was linked to savings. The bottom line, it was in their best self-interest to get their standards across the board together so they could save money. I think mandates are a tool as long as we have some incrementalism. I do not know what we - when you say mandate, I do not know what you are mandating. Is it everything in the medical record being standardized? Or, is it certain transactions or business case information needs mandated?
MR. BLAIR: Specifically, some of the standards from HIPAA, Congress directed that they be mandated. In executing that process, that is a very powerful tool, but, would the process to get it to the final rule just is very, very lengthy. So, one of the trade-offs is you can, maybe, go faster with guidelines, but, it has less authority. That is part of what I was trying to get a flavor of, is where is the balance.
MS. LOVE: Then, I think as long as there is that dialogue of the users and the collectors, then, the mandate is a tool.
DR. COHN: Just to clarify, with HIPAA, it was not just the standards, it was the capabilities. Remember, this would be like mandating over entry systems everywhere in the country using a standard which is, I think, a very different thing than I think any of us had thought about in terms of this one. Other comments before Bill gets a chance to ask his question. Colin?
MR. PRICE: Colin Price. This is a sort of philosophical and, perhaps, slightly provocative comment. Most of the benefits from adopting standards come from the fact that we are adopting a standard. I think there is a perception that some of these standards development activities have, actually, been detracted a lot of the potential benefits from standardization have been lost due to the lead time. That is just a thought for you.
There is an instance that I won=t quote in detail, but, we are pressing ahead with a particular project that is using edifact technology that we know is, effectively, obsolete within the NHS. But, we want to pursue the change managements and benefit realization program in that particular project using standards. Then, we can migrate further downstream. So, just an observation for you.
DR. COHN: Thank you.
MR. YASNOFF: I think it is clear from the testimony that there is an increased recognition of the need to coordinate the standards initiatives. I want to ask the question that Mike usually asks and I am surprised he did not ask. It is what should the government do to facilitate this coordination, either in general or among these organizations? In other words, given that there is this widespread recognition of the need for coordination, what is the government=s role? What should the government=s role be?
SPEAKER: Good question, Bill.
MR. YASNOFF: Thank you. Do I get a prize for stumping the panel?
DR. COHN: Jared, would you like to answer that question?
MS. ADAIR: I thought it was on a voluntary basis. Let me drop back for just a second. This is Jared Adair. Let me drop back for just a second. And, a little bit in response to the earlier question. The CHI project - what we are looking to do is to build into - each of our federal agencies is required to have an architecture. In that architecture, you stipulate either standards that you are going to be building to, et cetera, et cetera. Our plan is to have the federal agencies build that into their target architectures so that people would know that that is what we are moving towards.
We did not envision, as Colin put up on his chart, what is the implication to your legacy systems, which is your money and questionable benefit. I wanted to clarify that our approach had been and that, as Rob had put on the slide, it was kind of the tipping approach.
I think that that kind of - your question - just make sure I am answering the right question. Could you just, in a short word, put it out there again? I want to make sure I have it right.
MR. YASNOFF: I am, actually, referencing the point that Ed Hammond made earlier, that there are all these various initiatives going on and they all, in their own way, are attempting to perform the coordination function. So, the question I have is - and I am interested in the opinion of the panelists, obviously - is in coordinating these coordination efforts, or as Ed, I believe, put it, in making sure all the energy is focused in one direction, is there a role for the government? If so, what is it?
MS. ADAIR: Albeit not a panelist, Rob and I work together on the CHI. Let me, at least, try to respond with what we have been doing from a CHI perspective. I do not know that it answers your question about what the government should be doing. It does say what we have been trying to do. That is that in a number of these projects, as you saw up there, there are, in fact, federal participants. Many of those federal participants are the same people that sit on the CHI council, either us in person, or staff that we have. Our hope is that as these many activities are going on, some of them have very focused - not very focused, but, specific activities. They are not quite as broad as others. But, we are, through this kind of - and I am going to use a term that is probably not appropriate, but, that is okay - it is cost fertilization type of that we, in fact, will be getting messages out as to what is going on and the feedback loop as to what is going on. Hopefully, we will begin to realize that we will all be working in a similar direction in making sure that our business needs are, in fact, reflected there.
I think that it is, really, one of trying to make sure there are not - as somebody said - the same faces that you see over and over again at these meetings, that we will, in fact, be getting to the same place at the end. We may not be making all of the same choices or decisions, but, in fact, getting to a place where there is interoperability, which I think is truly the goal. Maybe not a specific answer, but -
MR. REESE: Given that my knowledge is, probably, more imperfect on this subject than anyone else in the room, what I would look for as an operator is that someone should have an inventory of all the standards that we are trying to do. Some entity should, then, know the status of the development of those standards and who is working on it. Then, participate in breaking down any bottlenecks where that group had - that volunteer group is having problems getting through it and making sure that groups have been formed to work on all of the standards. Sort of the adult supervision model. Not necessarily mandate the standard, but, to make sure that we know what we are working from and that work is being done. If that is the role of the government, so be it.
DR. COHN: I think Clem has both a comment and a question.
MR. McDONALD: I do. I will bet large amounts of money, or maybe bottles of wine, that there is going to be a tremendous impact from this project. Just the government saying this is what we are using, that is the tipping force. If you want to have something, a rallying cry, that is it. It is beautiful. It is brilliant because you could not do any kind of mandate in America. We cannot - you, at least, have an idea of patients. We do not even know who the hell our patients are nationally. Then, we talk about all standardization problems, that is just how we are. That is not - I think we eat our children on this. Oh, we are doing so badly. This is so crummy. There is a major industry and interface engine business just because everybody is using standards, but, not quite right. There has got to be $300,000,000 or $400,000,000 a year business.
If you look back on how hard it is to standardize, as I recall, we still are not doing the metric system in this country 40 years later, although, that was mandated by law. It took 50 years to get half the homes with color TVs from the time they were invented. It took 50 years to build an interface system, which is no invention. It is just kind of getting some men out and doing it. I think we are doing pretty well.
MR. PRICE: I feel better.
MS. LOVE: Can you solve the unique identifier problem? That is what I would like government to do.
DR. COHN: You can follow-up and, then, I have a comment. Then, hopefully -
MR. McDONALD: Actually, I did have a question. It is a small environment and you have a different way to manage it. So, it is a magnificent experiment to see how this all plays. We can take lessons from it because you will be ahead of us. It would be really interesting - do the physicians in the U.K. know those read code numbers by heart like they do the CPA numbers? I am amazed. The surgeons go, that is a 35297. I say, geez, how the hell do you know? They do not name them by names.
MR. PRICE: The answer is that they know some of the five character alphanumeric read codes. I would be very surprised if they can remember any of the SNOWMED - 64-bit, unsigned integer. It will be an interesting one to watch.
MR. SCANLON: Well, again, I go back to basic strategy questions. We assume, in the U.S., that HIPAA administrators - the suite of standards involved there, including privacy and security, was more or less a platform for what we are now talking about as the next wave of standards. Again, the mandate was kind of a double edged sword because not all of your colleagues are all that motivated to, actually, implement some of the HIPAA standards, even though it is an industry - presumably, an industry standard.
What did we learn? Is HIPAA - is the HIPAA experience instructive, albeit it was a mandated, regulatory kind of approach for what is being envisioned here in terms of strategies. It does not - even with HIPAA, where we thought we were adopting your own standards, it turned out that, well, maybe they were not really standards as much as we thought. We got a lot of push back from your colleagues in industry. How do you - who makes this assessment about where we really are and what the likelihood is?
DR. COHN: Well, Bert, it seems like you should be the one answering this question.
MR. REESE: I am trying to figure out whether I want to do an editorial on HIPAA or not. I was trying to figure out if I am more than 100 miles from home.
I think, officially - I think HIPAA is a good thing, if it would work. Unfortunately, it gets, like everything else, it gets ground down. It looks like it has had a root canal, at this point. The teeth are missing from it. Some of the advantages that were originally in the original legislation are being watered down as it goes through the iterations. From an operator=s perspective, there is - especially, around EDI transactions, there is a whole industry around clearinghouses. That was the beauty of the clearinghouse. We pushed our data upstream to a clearinghouse. We told it to target organizations. It worked like a big interface engine and pushed the data out to the target payer. We did not have the problem, we had the cost.
When we started down the road with HIPAA, we figured, well, there would be some financial advantages. If HIPAA was strong enough in its standards, we, then - this intermediate, the clearinghouse and, therefore, that reduces our cost. That was a good thing.
I am not aware of how the standards were developed. I - this, again, my knowledge is imperfect. I felt more like I was on the receiving end of that one. I was constantly in a battle of trying to learn. Then, I went into a mode of do nothing because there was so much public uproar that if I started investing millions and millions of dollars to remediate the Sentara environment to reflect the original version of the HIPAA legislation, as it has turned out now, I would have wasted a lot of money. So, I think - I applaud HIPAA. I thought it was a good idea. I thought it would make a difference. It would reduce our costs. It would do the things that the original writers would have intended. But, I think, now, it is probably not going to be as strong. We won=t get the benefit that we originally thought.
I do not know if that answers your question. But, it is part editorial.
MR. YASNOFF: This is a question for Marc. You mentioned the issue of standardizing what I would interpret as business rules, such as the Dwyer Tables. This, in my mind, relates to the issue of standardizing guidelines. There is an HL-7 group working on standardizing guidelines. My question, here, relates to identifying a potential gap in standards. Do you feel that the HL-7 guidelines group is addressing this issue of business rule sufficiently, or is that a, in your mind, gap in standards?
DR. OVERHAGE: This is Marc Overhage. I am not sure I can answer that adequately because I am not involved in the HL-7 guidelines group and have not seen a lot of the work product. I am not sure that I can answer it well.
What we did find in this public-private collaborative is that there were several examples of those kinds of business rule or business data maintenance that does not, really, have a very good home right now. It seems like you might be able to create good homes for some of them. But, there was not a place to turn to say these are the guys that are doing that, as distinct from a lot of the other types of standards or terminology - for terminologies, in particular, that we have been talking about where there are homes for many of it.
One of the challenges, I think, is figuring out - there may be a standard for representing it. I do not think HL-7 is going to manage populating the standard with content. That is not their usual modus operandi. I may be wrong and others can correct me here. So, even with the standard, you have to figure out who owns the content that is of broad interest and national value. Denise=s organization has a lot of these kinds of issues, as well.
MR. YASNOFF: As a note for the committee, I think, based on what we heard yesterday and today, this is an area we may want to investigate further sometime in the future.
DR. HAMMOND: Just a quick comment on that. One is that I think there are all kinds of business rules. When we talk about business rules, we talk about everything that is a business rule that says what you collect in an out-patient encounter, for example. You, really, are defining that. It goes all the way up to where it includes, as part of CPOE, in terms of the ordering the set of rules for doing that. Those are both clinical rules, as well as reimbursement rules. For example, one of the things that I think we should push for is for CMS to, as part of its care plan and reimbursement strategy, those should be in a standard format. They should be machine readable, machine understandable so that we can build programs to interact with those.
I think the thing that is important to recognize is that the thing that is making standards, and everything that is involved with that, so important is that we have almost pushed the healthcare system as far as we can push it. I think the next advantage is the cost containment that we are going to do is going to be from the automation of some of these processes. What we are going to do is replace human resources. We saw some examples of that already. There is a shortage of some of those resources and we need some ways of doing that. I think business rules may become one of the more important things that we are doing in the standard.
The other point is that not only business rules, but, some of the clinical guidelines where they have to be populated. One of the things that none of the standards community, to a large extent, has engaged is really the clinical specialty groups to begin to get them to populate. HL-7 has some activities. Clinical trials is doing some of that, now. Pediatrics has come in. But, this community itself really needs to begin to identify, bring those people together and get them engaged in providing content to some of the standards.
DR. COHN: Okay. I want to begin to put this all together. We do need to finish up. I want to make a comment. I want to offer to Jared if she has an final comments to make for herself.
Certainly, I am very impressed with all of the activity. When I walked in, as you could tell, not being quite - a lot of the names sounded the same. I was trying to figure out who is doing what. Unlike Bill, who described who is doing coordination, what I heard is that there is a lot of groups trying to deal with acceleration, trying to deal with the big issue of implementation. How do we get things from an HL-7 committee, actually, out and being used. I think that is all in our interest. I think there is a great commonality of interest.
In addition, I am hearing that there is a lot of agreement about many of the areas that really need to be worked on, or that need to be decided. I think the government can be very helpful in terms of being in the position of being the tipping point, providing some direction and some guidance to the industry. Of course, there also needs to be industry input to make sure that we do not wind up with the government having one standard and private industry having another, which is probably the worst of all worlds.
I am left sort of, also - even though Bill did not bring this up since he is now in charge of NHII stuff for HHS, or at least staff on that - that in some ways, when you - Jared is, at this point, in charge of - not in charge, but providing leadership with what the government is going to be doing. As we begin to look towards the overall issue of a health information infrastructure, it may be too far to state that the national health information infrastructure is only as good as its weakest link. But, we may not be far from the truth if we say something like that. Knowing that Jared, obviously, and Karen, both work for CMS, which is a big player in all of this stuff - one of the things that we need to be all thinking about and not to be decided today, certainly, is the issue of, yes, it is great to identify standards, it is great to make them voluntary.
Jeff has sort of commented about this issue of how do we get them out there. To me, as much as it is important that we have standards out there, it is important that we have functionality out there using the standards. How are we going to encourage the healthcare industry, which is already in a relatively dire financial straits - how are we going to encourage the diffusion of all of this out to the country? So, the question gets, then, to be it is not just identifying the standards, but, it is also providing some encouragement to the industry, some incentives, changes in payment policy, what have you, to begin to encourage everyone out there to implement functionality using the standards.
This is just from my view how it all comes together. I will let Jared and Ken make a comment. Marjorie has made a comment that - I guess we will talk about it off-line. She talked about whether there need to be some additional information on this shared CBHS data, some additional conversations. I do not know how much of it will be at the meeting or not. We will talk about it off-line.
MS. GREENBERG: Well, I want the input of Jared, but also, of other members of the panel. We originally, at the June meeting, when Jared reported on the CHI activity, and it was - I thought that happened the day after the AHA, the June meeting was the day after the AHA launched its activity. We started with the Markle Foundation, et cetera. The full committee had asked to have all of these groups at the meeting in September, which is about a month from now. Then, I think, we accelerated that - but, at the same time, looked at the Subcommittee to kind of organize things and to be the contact. So, we accelerated that by having this panel here today.
My feeling is that, probably, we do not need to repeat this in a month for the full committee, but, again, it - we need to inform them about it, report to them, about this discussion and what is going on. I think this needs to be monitored or we need to be partnering with these various activities throughout the coming year.
DR. COHN: My own view on this as I listened is there is good activity going on. The good news is that the next meeting after September is in November. It might be appropriate to have you all come to the full committee at that time to talk about progress.
MS. ADAIR: I think that I feel some responsibility at the September meeting. I felt like I got an assignment out of the last one. They said come back in September and I said yes. What we could say at that session is that we have had a conversation that has brought people together that are interested in the same direction.
DR. COHN: Well, thank you.
(Break for lunch)
AFTERNOON SESSION 1:25 PM
DR. COHN: Okay, we are going to get started in just a second. Please be seated. This afternoon we are obviously changing topics. This morning we had discussions about consortia and intra-operability primarily focused on clinical systems.
This afternoon we move back to some of the major work of the Subcommittee and the full Committee in its responsibilities under HIPAA to look at code set issues, specifically in this case hearings that we have held earlier related to the possible migration to ICD-10 and then a little later on this afternoon we will talk about the position papers that we have been working on related to DSM-IV and some issues from previous conversations.
Now, we want to thank Jim Daley and James Cross for joining us. As I understood it Blue Cross as well as AHP, are one of the opportunities we would have for their testimony on the issues related to ICD-9 and so we are obviously asking you to present to us information and your perspectives on these issues.
After that we will begin to discuss I think possible letters going to the full Committee for September. So, with that would you both introduce yourselves and then we will ask Jim Daily to start with testimony.
MR. DALEY: I am Jim Daley. I have a fuller introduction as part of my testimony.
DR. CROSS: I am Dr. Jim Cross. I am the National Medical Director for Aetna, Incorporated.
DR. COHN: Okay, you know, I should, also, comment before we start this session that even though there was nothing about this morning that I needed to recuse myself about, clearly if there are issues related to CPT that come up during this session I will need to recuse myself from any votes related to that since I sit on the CPT Editorial Panel.
With that, Jim, do you want to start with the testimony?
MR. DALEY: Good afternoon. Can everybody hear me okay in the back? As I said before, I am Jim Daley. I am the HIPAA Program Director for Blue Cross Blue Shield of South Carolina, BCBSSC, speaking on behalf of the Blue Cross and Blue Shield Association or BSBSA.
The Blue Cross and Blue Shield Association is comprised of 42 independent locally operated Blue Cross and Blue Shield Companies that collectively provide health care coverage for 84.4 million, nearly 30 percent of all Americans. BCBSSC provides innovative health benefit plans, dental and vision benefits, pharmacy benefits, life insurance and worker's comp benefit management.
We are, also, the nation's largest Medicare and Tricare administrator and we provide Medicaid services to the State of South Carolina.
A subsidiary offers software products and clearinghouse services to providers. Because of our span of interests we are concerned with different requirements from a variety of perspectives.
On behalf of the association I would like to thank you for the opportunity to offer our comments on the possible replacement of ICD-9-CM with ICD-10-CM and ICD-10-PCS.
Our purpose today is to describe the potential implications of a possible migration to ICD-10-CM and PCS code as HIPAA standards and to describe the possible industry impacts of such a migration. These are by no means meant to be the only impacts. However, it is a preliminary review of the kinds of impacts that we feel could occur and to provide recommendations that we will need to successfully implement the initial HIPAA mandates and conduct a detailed impact study of ICD-10 before proceeding.
The health care industry continues to focus on the need to improve the quality and availability of health care within the US while controlling the ever-increasing costs of these services. The new national standards for health care transactions and code sets were adopted to simplify benefits administration, to introduce greater savings and fewer hassles to consumers, health plans and health care providers. BCBSA fully supports these goals.
As you can see from this slide there are different perspectives within the health care industry. Each of these needs to be assessed to determine the impact of ICD-10 on that portion of the industry and ultimately on a consumer.
As payers we are primarily concerned with providing consumers with the best health care products and services at a reasonable cost. Our function is to reduce the financial impact of an illness via the coverage we offer to our members.
The cornerstones of health care administrative systems are the diagnosis and procedure codes. This slide shows some of the uses of those codes.
While the current transactions and code sets rule could be viewed as the equivalent of resurfacing the exterior of a building, move from ICD-9 to ICD-10 diagnosis and procedure coding could constitute a change equivalent to replacing the internal framework within that building. Such a wholesale change needs to be approached cautiously such that the potential impact can be evaluated and the code changes do not result in unintended adverse consequences.
You can see from this slide the use of bench marking, decision making, health care policy, public health tracking, billing and reimbursement and research and many other uses.
According to a report aired in March 2000 on migration to ICD-10 it was viewed as the most significant overhaul of the medical coding system since the advent of computers.
A recent report from the General Accounting Office raises concerns about the ICD-10 implementation challenges and the associated financial burden on the industry, and a quote from that report is on this slide.
One example of the complexity of migrating to ICD-10, the report references an example where under ICD-9 a single code would be used to designate coronary vessel aneurysm repair, code 36.91 while under ICD-10-PCS 180 different codes could be used. What are the implementation challenges? Diagnosis and procedure codes are used throughout the health care industry to classify symptoms, classify treatments or services, perform reimbursement functions and to conduct statistical analysis for health care training or rating purposes.
Since clinical codes are the underpinning of virtually everything in the health care system change to ICD-10 would impact payment, medical policies, quality improvement programs, benefit design and fraud and abuse detection among other things.
To look at the vast differences between ICD-10-CM, ICD-10-PCS and the currently used code sets the implementation of ICD-10 code sets would present tremendous challenges to the health care community. Impacts would be of two varieties, first, the direct impact associated with being able to use the codes and second, the impact associated with transitioning from one coding scheme to the other.
As you can see we anticipate the cost could approach Y2K for transactions in code sets costs. These costs and benefits are not fully demonstrated. There are significant transitional issues which I will address later, and this is coming on the heels or actually in the midst of the initial HIPAA standards implementation.
This slide shows the players in the industry that we feel would be impacted in some way by ICD-10. Of course, some would be impacted more than others. You can see payers and providers are the obvious choices, researchers, software vendors, clearinghouses and the list goes on and on.
What would the impacts be? Software, obviously the software built in house as well as packaged software would be impacted. Reimbursement and contracting procedures might need to change, care management policies, training forms and I have slides to go into a bit more detail on each of these as well. You can see they have very far reaching impact.
In terms of the payer software this slide shows some of the things that would occur. This would again impact both that built in house as well the purchased software and including any interfaces between the software build in house and those vendor packages would need updates as well. You need to carefully coordinate the timing of those updates. Also, over the years many applications have built logic that directly interrogates specific code values. This would add to the complexity of accommodating ICD-10.
A fully functioning crosswalk must be available in order to consider such a change and changes of this magnitude will require very extensive testing. You can see some of the impacts, screens, databases, files, reports, etc., data warehouses that are used for reporting, all sorts of impacts from there, including actuarial, etc.
At the bottom I list other changes. OCR optical character recognition programs would need to be adjusted in the event there are some paper forms you want to scan and determine what the content of the fields are. VRUs, the voice response units, some of them scrape actual screens that have the codes contained on there in order to reply to a caller.
If those screens change as a result of the codes then the way the VRU picks up that information would also need to change, but there is a whole bunch of downstream impacts that may not have been considered at this point.
In terms of provider software this is used for a variety of things, scheduling, billing, claim submission, financial performance and it is, also, even in intensive care ER activity that does use some of the codes. They would all be impacted by ICD-10.
Software vendors have these considerations as well as some others. Packaged software would require those changes to the screens, logic reports, databases, etc., but also the vendors would need sufficient lead time to update the applications, negotiate with their clients and to roll out the new releases to support its sites.
Reimbursement is based on knowing what condition the patient had and what was done to treat the patient. Modifying diagnosis and procedure coding would impact the very heart of the reimbursement process and we noticed the GRGs the APCs, line pricing, etc.
In addition, the sheer volume of fee schedules that would need to be revised and renegotiated is staggering. According to AMA numbers in the year 2000 practicing physicians count was 647,000 with an average of 13 contracts per physician. That would mean just on the physician side 8.4 million physicians' contracts and fee schedules would need to be renegotiated.
On the hospital side using South Carolina as an example we contract with 70 hospitals. Using the 2000 payers across the nation and that 70 count as an average there would be approximately 140,000 contracts that would need to be renegotiated.
On the provider side, again, the impacts would include fee schedule and contracts and new software, being able to use the new coding changes, also, more extensive documentation. If there is more specificity in the codes you might need more documentation in order to determine what was the appropriate code to use.
These codes are, also, used for practice costs and projections. So, that could be impacted by ICD-10.
Even employers and members may see some kind of an impact as a result. As far as the consumer is concerned they have coinsurance payments. If the new ICD-10 codes impact the amount that is billed at a particular visit, the 80/20 amount could change.
Administrative services only contracts ASO where the customer is self-insured, self-funded, those could see some impacts as well in terms of the payment amounts that could then be coming off of their bottom line.
In terms of procedures, procedures would need to be reviewed and potentially modified as a result of the new coding scheme. For certain documentation within payer organizations use of hard copy how that is processed, error correction and there are several listed on this slide that show the kinds of things we need to review for current procedures and determine how those might be impacted by use of the new diagnosis and procedure coding.
Extensive training would be required and during this training period there is potential for incorrect coding to occur while becoming accustomed to using the new codes, and it is my understanding that within the practices, the physicians' practices there is a constant need for training because the turnover rate is significant. So, this would not be a one-time event. It would be an ongoing event of the training and potential for miscoding.
Even though HIPAA does not apply to paper, paper would be impacted by ICD-10. Provider visit sheets list some of the specific codes and check them off as to what the condition was, what was done to the patient.
The actual claim forms themselves, the paper claim forms would need adjustment as well. Other forms, there are forms used to requisition tests within hospital. Those use some of the specific codes. Those might need some changes as well.
As far as statistics goes there could be significant impacts within that area, trend analysis utilization management, a whole bunch of areas where the codes are used. We will need to resolve how to relate statistics coded under ICD-9 to those coded under ICD-10.
It appears evident that payers and providers would need systems and processes to manage both the ICD-9 and ICD-10 code structures during the transition. Even with a suitable crosswalk maintaining dual coding structures would be at best cumbersome. The ability to capture historical data or review utilization information would be difficult.
The ability to perform retrospective audits and other reviews relies on the comparison between historical claims and current claims. In particular the payers need accurate utilization data in order to determine rates for their customer accounts.
There are many concerns that would need to be resolved. Would information systems be required to support dual standards during the transition or would there be a high cut over data. How would archive data be treated? Some of this archive data may be needed for audit trails.
If HIPAA doesn't apply to papers but providers continue to use ICD-9 coding for an indefinite period there could be ongoing need to accommodate dual coding standards. We feel this would be unacceptable. It would, also, impact certain electronic transactions such as claims status if the claim came in on paper using an ICD-9 code but the transaction standards say that you must use an ICD-10 code.
Other considerations, there is a potential for increased fraud during this cut over period as everyone is a little bit unfamiliar with the codes or even if it is honest it would be very difficult to figure out if fraud was occurring for another reason. Patient treatment hopefully would not be hindered by this, but with any confusion in what the coding structure is there could be some potential implications in terms of care management.
Costs and benefits have not been fully studied. We feel an industry-wide evaluation of all of the potential downstream impacts needs to be done.
In conclusion our recommendation is there should be no proposed migration until the initial HIPAA mandates have been successfully implemented, there has been a thorough impact analysis including the cost/benefit of migrating and of not migrating and transitional considerations have been thoroughly analyzed and other alternatives have been assessed.
In conclusion, we believe this is a very big change with high impact and potentially high risk, with high costs and the implications have not been fully defined. Therefore we believe this should be approached with caution.
Thank you for the opportunity to testify. The Blue Cross Blue Shield Association stands ready to assist you in analysis of the ICD-10 impact.
This concludes my statement, and I realize I forgot to slip through a couple of these slides.
DR. CROSS; Good afternoon. My name is Dr.Jim Cross. I am the National Medical Director at Aetna and I am appearing today on behalf of Aetna and the American Association of Health Plans.
I would like to discuss the potential impact on our health care system and on our health plan operations if an entity were to change from the ICD-9 to the ICD-10 coding system for medical diagnosis and procedure codes.
A few comments about the two organization, AAHP is the principal national organization representing HMOs, EPOs and other network plans. AAHP and its member plans arrange health care services for approximately 160 million members nationwide. Aetna provides health care and related group benefits to approximately 14 million health care members, 11 million dental members and 12 million group insurance members. Aetna's PPO networks have over 500,000 health care providers and our HMO networks have over 380,000 providers.
Aetna's products include a full range of health insurance including HMOs, PPOs and indemnity. It, also, includes dental and pharmacy benefits as well as group insurance products such as life, disability and long-term care insurance.
As the National Medical Director for Aetna I am responsible for our claim medical management operations which provide clinical review of claims, and I, also, oversee the development and maintenance of our coding logic, clinical and coverage policies and case management and customer service efforts related to transplant and special case care as well as all of our reimbursement policies.
I will skip my credentials since I think you can read those if you need to.
Now, the comments around ICD-10. The use of ICD-10 codes for medical diagnosis and procedure coding has been debated for a number of years. The primary reasons given for changing to the new coding system are concerns that the current ICD-9 codes may be inadequate and do not provide sufficiently detailed information needed for health care research and statistical analysis.
What has not been fully assessed, however, is the potential cost and administrative burden on the health care system for moving to a new and complicated coding source and structure such as ICD-10. A thorough consideration of all the costs and complexity associated with the proposed migration to ICD-10 should be performed by this Subcommittee before any action regarding a recommendation for such an important change is made.
Currently ICD-9-CM is used for diagnosis coding inmost inpatient and outpatient settings. These codes are contained in Volumes 1 and 2 of ICD-9-CM.
Volume 3 of ICD-9-CM includes procedure codes and is used in inpatient institutional settings, primarily hospitals. There are separate procedure coding systems for other health care settings, primarily the current procedural terminology fourth edition CPT-4 developed by the American Medical Association that is used in physicians' offices and other outpatient institutional settings such as ambulatory surgical centers.
While CPT-4 procedure codes would continue to be used in many outpatient settings the proposed migration of ICD-9 to ICD-10 would require the use of ICD-10 for procedure coding for inpatient institutional settings.
Additionally this Committee is considering recommending the use of ICD-10 for all diagnosis coding. However, a change to ICD-10 for diagnosis and inpatient procedure codes would result in a tremendous burden especially for physicians because of the cost and complexity associated with such a major change.
The migration to ICD-10 diagnosis and procedure codes would result in a substantial increase in the number of fields used for the coding process and a significant change to the system that requires both numbers and letters.
For example, the current diagnosis codes are a numeric system with some supplementary letter codes using a minimum of three digits and a maximum of five digits.
In contrast ICD-10 diagnosis codes combine letters and numbers and use a minimum of three digits and a maximum of six digits. Likewise current ICD-9 procedure codes are numeric with a minimum of three digits and a maximum of four digits, while the ICD-10 procedure codes are alphanumeric with seven required digits.
As a result the number of possible codes increases dramatically. As the Center for Medicare and Medicaid Services explained during hearings before this Subcommittee earlier this year, whereas ICD-9-CM procedure contains less than 4000 codes the current draft of ICD-10-PCS contains just under 200,000 codes.
CMS, also, reported that the number of procedure codes could be expanded as new medical technologies are developed. The changes to ICD-10 will, also, dramatically increase the complexity of the coding process. This migration will require substantial physician and hospital staff training and will significantly add to the administrative costs of coding.
In addition, switching to ICD-10 will increase the amount of time that both physicians and administrative support staff will need to devote to the coding process.
Currently physicians and administrative support staff utilize a four-digit numeric code to designate procedures. If a change is made to use ICD-10 there will be more fields in the code. All of these fields will have to be completed for an accurate and useful code. Physicians will have to provide additional and more specific information in the medical records on what procedures were used and will likely become more hands on in coding processes.
In addition administrative staff will have to spend more time coding and may require more frequent consultation with medical staff to determine exactly which procedures were performed.
A coding change would, also, have significant impact on health care payers such as insurers, managed care plans, employers and government programs as well as their business partners such as billing and claims services, vendors and health care clearinghouses.
The migration to ICD-10 would impact multiple systems used by these entities. Existing hardware systems such as computer main frames and optical scanning systems would have to be updated or literally replaced in order to be able to process these added fields.
New claims and other transactions forms would have to be developed. Extensive software revisions would have to be changed and would be required to incorporate the ICD-10 codes into payment and claims systems.
Payment methodologies would have to be modified to take into account the new codes. For example, ICD-10-PCS is a tool for reporting procedures and not a reimbursement methodology because it does not include any global codes that are necessary for payment adjudication systems. While ICD-10-PCS provides detailed and separate codes for every step or part of a particular medical procedure such as heart surgery, it does not have a mechanism that allows for the combination of related codes for reimbursement purposes.
New payment related codes have to be created and agreed upon using a combination of the individual ICD-10-PCS codes. Health plans are already devoting considerable resources to implement the HIPAA privacy and electronic transaction standard regulations.
Additional HIPAA administrative simplification regulations, security, health care provider identifiers, health plan identifiers, claim attachments and first injury reports have yet to be published by the Department of Health and Human Services.
A change in the current coding system for medical diagnosis and procedure codes will significantly add to this burden and could greatly exceed what is currently being spent on HIPAA implementation.
Recently the United States General Accounting Office issued a report analyzing whether a single coding system should be adopted for all inpatient and outpatient medical procedures. This report includes a discussion of the merits, the disadvantages of changing to ICD-10-PCS.
While the GAO report notes that ICD-10-PCS provides greater coding specificity which I don't think anybody argues about the report also makes the following observation about the difficulties to switching to ICD-10-PCS and I quote the General Accounting Office's comments which states, "However, the design and logic of 10-PCS raise concerns about potential challenges in its implementation."
For example, there are some cases where 10-PCS's specificity creates a significantly greater number of codes for certain sets of similar procedures which may reduce coding accuracy. In addition 10-PCS is a distinct departure from the design and logic of ICD-9-CM, Volume 3. The existing health care administration system would need to be changed significantly to accommodate a new code set imposing additional financial costs and administrative burden on members of the health care industry that are currently undertaking changes to comply with the adopted standard code sets under HIPAA.
We support the conclusions outlined in the GAO report on the potential administrative and financial burdens of changing to ICD-10 from the current ICD-9 coding system.
Before a decision is made to approve a migration to ICD-10 for either diagnosis coding or procedure coding or both it is crucial that a study be conducted of the potential cost of this change.
It is unclear what the financial impact will be on government programs such as Medicare and Medicaid and on private sector payers. Finally, the ultimate consideration must be that in the end all of these costs will be passed on to the consumers who are already experiencing a significant rise in health care costs.
Aetna and AHP support the goals of administrative simplification. We believe that all of the participants in the health care system, patients, doctors, hospitals, health plans, employers will benefit from the development of systems and processes that provide efficient and effective data interchange.
We, also, believe that any significant changes to that system such as a potential migration to ICD-10 must be thoroughly assessed with its benefits weighed against the potential costs.
While the change to ICD-10 may seem attractive in some respects, particularly its specificity and its ability to be used by statisticians it is premature to act without adequate information.
We urge NVCHS make a more detailed analysis of this issue.
Thank you.
DR. COHN: Thank you very much for your testimony.
Are there any questions before we go into session discussing our --
DR. MC DONALD: Do you know of any organization that uses ICD-10-PCS in the US today or has experience with it?
DR.CROSS: I don't believe ICD-10 is out at all in terms of practical use within at least the commercial industry, and I think ICD-10 international is a totally different system than ICD-10 for the US.
DR. MC DONALD: The question I really had was that there seems to be a lot of interest in having some more specificity. Is it available, and can we get it?
DR. COHN: I don't mean to break in but we would hope that Pat Brooks might have the answer to some of these questions.
Pat, would you introduce yourself?
MS. BROOKS: ICD-10-PCS is available on the CMS home page and its various versions have been there since 1998. So, it is available in draft. We will have another draft this year. As far as anybody using it, this is in development. No one should use it yet. It is a developing system like the diagnosis system that we have developed, stress system. Germany may make a decision to implement it, and a number of other countries have translated it into their own languages and they may implement it before we make a decision.
DR. MC DONALD: Would it be permissible to use it today?
MS. BROOKS: Anyone could use the system for anything they wanted to. I don't know that they would but if they wanted to take it and use it it is in the public domain. Anybody could use it for anything they wanted. That is why other countries have called and said, "Can we use it?" and they are evaluating it for their own use, and we said, "Sure."
DR. CROSS; Before anybody would use it, I think there definitely would have to be agreement that we are all going to use it because it would just be a one-way street.
DR. MC DONALD: There would seem to be a very strong interest on the part of hospitals within their own institutions to kind of analyze things, and I thought there might have been a couple of them using it.
DR. CROSS: You are just saying that they could sort of double code for what the industry needs to function and code for their own interest.
DR. MC DONALD: Actually I was assuming that. The other question was --
MS. BROOKS: There is a conversion table on the home page right now. So, yes, between ICD-9 and 10-PCS.
DR. MC DONALD: So, they could have done it once and then --
MS. BROOKS: Nellie Leon-de-Saint would probably be in a better perspective to see how many would willingly take a draft system and double code. She is here.
MS. LEON-DE-SAINT: Nellie Leon-de-Saint, American Hospital Association. I can't say how many hospitals are currently double coding. Obviously with a shortage of coders, a larger amount of work I don't think a lot of hospitals will be doing it, but I can tell you that when ICD-10-PCS was made available for testing we had a number of hospitals who volunteered to do double coding for a short period of time to see how it would work, and then they submitted their comments to 3M and to CMS and the majority of them had said that it was very easy to use and easy to learn, and they really liked it.
MS. TRUDEL: I have a question for the two testifiers. We have been hearing a lot of testimony about ICD-10 and ICD-9 over the past 3 or 4 months, I think, and what you have been saying appears to be flying in the face of what we have heard.
A number of people have said to us that they believe that ICD-9 is and I am quoting, broken and that there is a finite amount of time before ICD-9 ceases to be specific enough to meet some of these needs that you have mentioned, but it sounds to me from your testimony that what you are saying is that ICD-9 will be just fine for the foreseeable future, and I am wondering if you could comment on that?
MR. DALEY: My perspective today we reimburse based on ICD-9. We haven't had a lot of issues associated with that. Obviously you have been tuning up the ICD-9 by putting codes in other sections, things of that nature, but we have been doing okay with that system.
Again, for reimbursement purposes we don't necessarily go down to the granular level of detail that you might want for research and other statistics.
So, our needs are slightly different from the other parts of the industry.
DR. CROSS: I think the challenge with ICD-9 has been that it really probably hasn't been utilized by the practicing providers or the provider community as much as it could in the kind of detail that it already provides. In other words, with ICD-9 you can use a relatively generic code and not have to go down to the specificity for it to be a valid code. You can use three digits and not necessarily have to go to the last two.
So, many of the submissions of ICD-9 are not as specific as they could be if they wanted to be coded that way, but most providers, at least physicians who don't have the coding expertise or take the time to do the coding code at a relatively higher level of coding which is adequate for reimbursement and adequate for communication with payers but it is not as specific as it could be.
Many times we have trouble just getting the diagnosis at all on the claims submission particularly with laboratories, but I think we have a wealth of data at Aetna.
Our challenge has been to use that in information that can actually impact health care and health care outcomes. We have more data than we know what to do with, but we have a challenge to make it into information and change our administration and impact health care outcomes.
So, I don't think we are short on data. We may not have all the specificity we want, but ICD-9 gives us more than enough specificity if it was appropriately used by all the practicing community.
MS. TRUDEL: Okay, so, you are saying that there is no need to change anything.
DR. CROSS: The only thing that I would say is that at some point in time we will run out of code combinations or of numbers that we would like to have in ICD-9. So, I am not saying that eventually we might need more room to have more things identified. My big concern is not so much around ICD-10 diagnosis coding. My major concern is around ICD-10 procedure coding.
I think that if you are going to implement ICD-10 I would implement it in phases. I would do the ICD-10 diagnosis piece first because I think that the industry and practicing physicians would get used to the methodology, get used to ICD-10 and it wouldn't be the complicated part of procedure coding which is more related to reimbursement logic. I think the challenge will be how you merge or migrate from a CPT-4 environment into an ICD-10 not the statistical reporting environment but a reimbursement and coding logic that is really sort of at odds with each other in terms of what their goal is.
MS. BROOKS: I don't know if you are aware but ICD-10-PCS was developed not to replace CPT for physicians' office use. It was intended to replace ICD-9-CM procedures in hospitals. I noticed at several points you mentioned about physicians having difficulty using 10-PCS but the system wasn't developed or proposed for that setting. Does that change your views any?
DR. CROSS: I think if we were going to use ICD-10-PCS and still allow for CPT-4 coding to take place that that is possible. It depends on what we pick for HIPAA standards in terms of the coding logic.
MR. DALEY: One of the other considerations is that it was originally proposed to move to a single procedure coding system and that is one of our concerns as well. I just feel that right now I am speaking to some of the potential impacts of ICD-10 and the people I have talked to, most of them when I approach them they say, "What on earth is it?" They have no idea. So, to say that it has been researched; the hospitals have tested it, that is one part of the equation, but as far as the reimbursement and the payer community there is a lack of knowledge as to what ICD-10 is all about and not an understanding of what the impact might be. With some further study maybe it will turn out it is not as bad as we think, but today all I want to do is raise the point of we don't really know. We think it is very big. The GAO report seems to support that concept but we need to study this to determine just what it is. I am very fearful of going into some kind of a mandate that says, "Here is what we need to do. Here is a time frame," and when the rubber hits the road everybody looks at it and says, "Oh, my God, what do we do now?"
We need to make sure we know what we are doing if we make a recommendation.
MS. PICKETT: I just wanted to clarify something. The ICD-9-CM diagnosis codes are up to five digits and those are mandatory. You cannot use less than the full five digits if a five-digit code exists, and it is required by Medicare and other payers as well. So, it is not optional that they can use the three-digit or four-digit code when in fact a five-digit code exists.
MR. DALEY: That is correct. However, in terms of usage within a payer organization you don't necessarily need to interrogate all of those. You may get the general idea of what the condition is and then pay based on the procedure.
MR. BLAIR: Jim Daley, when you wound up suggesting that we do cost/benefit studies, and I think both of you did, Jim you also suggested that we consider alternatives, but you didn't mention what alternatives you might have had in mind. Can you tell us what you are thinking?
MR. DALEY: We haven't really discussed what those alternatives are. I have tossed around amongst our plans possibilities. One of the things that we need to look at is ICD-9-CM, if you are talking about changing to alphanumerics ICD-9's right now don't even have alphas in there except for the V&E codes. Has anybody even pursued modifying ICD-9 for a new technology or things of that nature? There is a whole variety I am sure that could be investigated. I am not prepared to suggest what those might be at this point.
DR. COHN: Any other questions of our speakers?
Thank you. We will continue the discussion. Feel free to as we have questions or whatever, we may pull you back to the microphones here.
Before we get into the options we have at this point I would do a little introduction on that. I just actually had a question for you about PCS and it occurred to me after I think all of our hearings and all of this, and this actually has to do with the fact that PCS has about 180,000 codes for hospital and surgical and whatever procedures, but in the real world people are paid typically based on DRGs which are heavily diagnosis with a little bit of procedure attached with them or maybe I am overstating this, but some combination but it involves basically taking the data and collapsing what already exists into fewer categories.
Is there a human issue that you see sort of coming down the way that use of something that is this much more granular would allow CMS to do that we need to know about?
MS. BROOKS: I just talk about the new technology requirements. I am not an expert at it, but we did have in last year's proposed notice of the DRG, we talked about changes people wanted us to make and Congress asked us to be more responsive to new technology and do a quicker job of updating the code system and be more specific at it, and so we laid out a number of issues and new technology issues like will we keep ICD-9 and do the best we can or we could make up a new system or we could move to 10-PCS, and the bottom line is without enough room for the DRGs we can't be responsive to that mandate to Congress.
DR. COHN: How many DRGs are there currently?
MS. BROOKS: I think it is about 111.
DR. COHN: Are you going to go up to 100,000 in the next several years?
MS. BROOKS: What we think we would do --
DR. COHN: I don't mean to be --
MS. BROOKS: No, but if we went to ICD-10-PCS I think it would be in the transition year. I, also do DRG analysis. I work on changing them every year. We would take the ICD-10-PCS codes with our conversion table that now exist and we would take the codes that affect the DRG and put them right back where they were. So, in essence the very first year I would think what we probably would do but I can never say never is that the new DRGs would not have a payment impact per se. We would simply map the new coding system where the old one was, and we have a contractor that we already have in existence. He has been paid to do that if we decide to go with ICD-10-PCS. So, no, we are not planning to double the number of DRGs but as people in the industry, hospitals say, "I think you need to be more specific," in order to do that sometimes you need more codes.
On the issue as brought up previously about the aneurysm repair I kind of laughed about that one because yes, we have one code for aneurysm repair but if you use it you can't tell what part of the body it is in. So, that is sort of ludicrous. You have an aneurysm somewhere, you know, guess where, but this procedure when you repair an aneurysm, ICD-10, there are a number of codes because they tell what you are doing.
Sometimes there are more codes but you still only select one. One goes on the bill. You pick the right one and put it down there. So, this 10-PCS code in our experience with our testing didn't lead to more codes on the bill. It just lead to more information.
MS. GREENBERG: I just wanted a clarification, too, which I think you were maybe making before, Pat, but the speakers mentioned about all the impacts on fee schedules, but it is my understanding that ICD-9-CM, Volume 3 for which ICD-10-PCS would be a replacement is not used in physician fee schedules.
Is that correct?
MS. BROOKS: There would be no impact at all on fee schedules because it wouldn't be used on fee schedules.
DR. MC DONALD: She said, "Fee schedule." Did you mean that?
MS. BROOKS: It is the same thing.
DR. MC DONALD: So, the ICD-9-CM is not used by any private insurance carriers?
MS. BROOKS: No, for fee schedules, for Part B kind of --
DR. MC DONALD: But I mean that is the point you are trying to make is it wouldn't apply to physicians but insurance companies. The question is does it apply to insurance companies?
MS. BROOKS: Not for inpatients. Most of the fee schedules are for payments like lab fee schedules and durable medical equipment and physician fee schedules. All those use CPT and since that wouldn't be replaced it wouldn't affect that.
MS. TRUDEL: I can answer your question, Clem, I think. Because the HIPAA standards are going into effect before any of these decisions are made the only use that would be possible of ICD-9 procedure codes is inpatient hospital. They can't be used for outpatient. They can't be used for physicians. So, the answer is yes, it is not going to have any impact on any fee schedules.
DR. MC DONALD: Do insurance companies use ICD-9 procedure codes to decide anything but what they pay the hospitals?
MS. GREENBERG: It affects the DRGs inpatient.
DR. MC DONALD: Some insurance companies would have to do something different. Otherwise they wouldn't pay for care..
MS. GREENBERG: For inpatients.
DR. MC DONALD: That is what we are talking about. This is only an inpatient discussion.
MS. GREENBERG: I think fee schedules don't really apply to inpatient hospital costs. I mean they apply to your physician services in the hospital but those are based on CPT.
DR. COHN: I am not sure I know how to introduce this particular piece but let me try and obviously we need everybody's thoughts. We have obviously had a number of hearings on this issue. We know it continues to be a relatively complex and contentious issue, not made any easier by the fact that we have caught two different pieces that we are playing with, one diagnosis and the other procedures and so sometimes even in our comments that have been made to us it is hard to sort of tease them apart.
Now, we actually do have a range of possibilities for what we want to recommend and let me just sort of lay them out for a second just to remind everybody that we have a lot of options here.
We currently have a letter that talks about putting them both together and talks about NPRM or a notice of intent and probably at one extreme in terms of things that we might recommend is moving forward to an NPRM. Probably the step before that is an NOI. There is also of course the issue about and we did hear testimony that allowed the industry would prefer to basically implement things together but we do have the option that we just for example heard one speaker who said, "Geez, staggering them might make some conceptual sense." So, that is another dimension, but before the NPRM and the NOI there is also, we have heard a fair amount of testimony that says, "Geez, there need to be some additional studies done looking at the whole issue of implementation costs and benefits and we will get a better understanding of the costs."
So, that is obviously another piece which can either be with or without the NOI or NPRM. We could recommend that separately or just recommend that.
Other options include recommending nothing. Another option includes further Subcommittee deliberation which I will tell you would not be in the next 2 or 3 months. We have already got a full list and yet there probably are other options but as far as I can tell that is the spectrum that we are dealing with right now.
Now, I guess I would ask the Subcommittee, you know, you have all been here for most, if not all the testimony, and, Kepa, you are on the phone, too. Kepa, I apologize. Somehow the volume is very low. Can we turn it up?
DR. ZUBELDIA: The question is do they see any advantages to implementing ICD-10.
DR. COHN: I think we heard from one that he felt that the diagnosis was important. The other felt that eventually, so, we should ask them, please?
MR. DALEY: Okay, we have a couple of things that might be advantageous, of course, the increased specificity might help us in certain conditions to know a little bit better exactly what the condition is which could help in expediting payment, etc. It might help in care management and also for fraud and abuse, again, it is a double-edged sword. There could be incorrect inaccuracies in the coding. However, assuming it is coded accurately that would help us to detect who was up coding who was using certain codes in not the most up and up fashion. So, there are some advantages there but again we are just very concerned about the complexity and the degree of impact it would have and we want to make sure we fully understand what we are recommending before we go in and recommend something that turns out to be more than we intended.
DR. COHN: I don't know if Dr. Cross had a comment. I heard him say that diagnoses were valuable.
MR. WILDER(?): Good afternoon. My name is Tom Wilder with AHP. I apologize, Dr. Cross had another commitment. So, he had to take off but I think we would agree with those comments. Obviously going to ICD-10 gives you more specificity but it also increases your complexity, and we are worried about the administrative and financial costs of having to deal with that complexity, and we are worried about making a recommendation to do that change without really sitting down and thinking about those costs and who ultimately pays for them.
MR. BLAIR: I think it sounds like the recommendation that we do a cost/benefit study, I think, Simon you have asked for that for quite some time, and we are hearing it again in testimony now, and I think that that is an appropriate thing to do. I just want to make sure that our recommendation is broad, that it includes the providers from the standpoint of improving patient outcomes, that it includes clinical research that could benefit by this, that it includes information that our hospital administrators can use to be able to improve the cost structure. In short I would like to have a global picture of the cost/benefits because it may turn out that the cost/benefit for the provider sector, for the payer sector is as our testifiers are indicating not as generous a cost/benefit as it might be for providers or clinical researchers but we need to take a look at it as a whole.
DR. COHN: Other comments? Clem?
DR. MC DONALD: At the request of the Committee I reviewed all the material and I don't know, I mean I would like to at least mention what I came up with and it left me cynical, I should say, and I was the sort of odd guy out on the Committee. So, I am happy to remain that way, and the main guts of it was as I read the actual written comments it was two and one-half to three and one-half, in terms of for and against. There were six testifiers and at that time AHA had a very strong comment about not doing it until, and this was all in bold print, I think it is until -- no decision should be made or discussions entertained regarding a potential single procedure classification system until all contending systems have been thoroughly tested for compatibility with existing payment systems and also, all bold AHA believes that before the implementation should take place the new system should be tested for all services and settings.
So, there were some strong caveats. That didn't appear in their letter. So, they may have changed their mind about it. The second thing that was in there was they discussed the, I think the rationale really boils down to running out of space and I understand that. So, when they want to add things they don't have a place to add them and I think if the current count is right there is room for 10,000 ICD-9-CM codes, and they have used up 3000. So, then it boils down to that they have got a problem with the tree which is a problem because you shouldn't make, nowadays people think you shouldn't make coding systems with fixed positions in them and of course the PCS has that same kind of limitation but finally the part that I am really most worried about is the fact that when I looked in the guts of the reports and all the things that came out there was a detailed description about studies that were and it boiled down to 100 charts that were looked at by contractors of CMS and without any numbers reported of any kind except to say that one contractor thought it was a little slower and the other contractor thought it was all right.
I hear these other stories. I actually thought I heard a figure of 1.7, but it wasn't in this particular -- 1.7 times slower. It is just that we should have some idea what happens when we go to 3000 codes to 170,000 codes for the coding costs and it may be out there but it wasn't in any of the published data and it should be an independent, and now, CMS led this development. CMS gave it to their paid consultants or paid contractors and said, "See how it works." It doesn't have the same feel as if it was an independent study that did it. I would like to sure see somehow whatever we agree on somehow we get some actual numeric by an independent reviewer who looks at this stuff and writes it up in enough detail that you can understand it.
MR. AUGUSTINE: GAO reported, in their report they refer to this.
DR. MC DONALD: I am giving you the numbers. It is a nice report. It is a CMS report but when you get down to the data it is 100 cases.
MS. BROOKS: No, that was the final point.
DR. MC DONALD: Twenty-five hundred cases were just for finding out what was missing. That was studied in terms of time. I can read it.
MS. BROOKS: They did summarize what we did, but we presented the last couple of years the detailed studies. What we did is we began with 5000 records side by side where could they be coded; was it workable; was all the information there? That was a 6-month, 8-month project and we did a second part study. Before we did all that we did have an independent study. AHA and AHIMA, I don't know if they put in their report, did their own independent study.
DR. MC DONALD: But the published studies, I mean something written down so an independent person cam read it and see what it meant.
MS. BROOKS: We did publish it through our ICD-9 Coordination and Maintenance Committee which we gave to the Committee.
DR. MC DONALD: When I sent this letter and I actually pleaded for is there any more data, and this was about 3 weeks ago, but let me go back to the 100. It says in your report it was 100 cases that were actually measured in some fashion.
MS. BROOKS: That was the final part. Okay, the final 100 just did one small thing. We already knew that they could code as quickly with it. That was to say that if you have all of the record here and you were coding with ICD-9 or with ICD-10-PCS how fast could you do each one in that time period and they found out they were about the same. The number of codes were about the same number of codes. It was just a small additional study that we looked at.
DR. MC DONALD: But that is the only one that spoke to the question as I read the report and I can pull it out. You said that the first 2500, the big one was,you kept changing the code system. It was really a proof of concept. You weren't measuring timing. You didn't mention anything about measuring timing. When you measured timing you didn't report the times and 100 is hardly any kind of sample.
MS. LEON-DE-SAINT: Nellie Leon-de-Saint, AHA. I just wanted to clarify for Dr. McDonald in terms of the statement you just read when we requested additional testing, that is when we were talking about a single procedure classification system because we felt that before it would move across to the outpatient setting that the outpatient setting needed to be further tested, but as far as inpatient hospital reporting we felt that ICD-10-PCS was sufficient.
DR. MC DONALD: But could I ask her about did you actually summarize, I mean report in any formal way the studies done by the hospitals that tried it out?
MS. LEON-DE-SAINT: No, that was an informal volunteer testing and so basically they didn't keep numbers or submit any documentation to us. Basically they were just calling us with sort of a gut feeling. This was of course before the CETA(?) conducted their testing. I don't know if AHIMA kept any -- I see Lynn shaking her head.
DR. MC DONALD: Do you have an idea, was it 1000 charts or 10,000 or 200 or 10?
MS. LEON-DE-SAINT: We can give you our detailed report from each of the CDACs(?) that they did over a couple of years. We can submit that. You probably have some summary reports where I did follow-up briefings where I gave real synopses.
DR. MC DONALD: The thing you submitted at the last thing it is like a 20-page report.
MS. LEON-DE-SAINT: Yes, it was very short, abbreviated but I will give you the more extensive one that we had published several times.
MS. SERT: Mine is a general comment not in response specifically if you are finished. Katherine Sert speaking for the Association of American Physicians and Surgeons and what I would like though is to emphasize Dr. McDonald's comment at the June meeting which is we don't hear physicians clamoring for more codes. What we hear physicians clamoring for is more guidance on the existing codes and I think what we are specifically looking at, let us address the issue of audits and prosecutions and the fraud issue more so than the payment issue and what we hear from our members all the time is that they can't, and I am specifically thinking now of let us address specifically CMS, that they can't get guidance from the Medicare carriers on the correct code to use because the typical response is you contact the carrier and the response is it is in your guidance. Look at your bulletins, and then you roll the dice and if you pick wrong then you risk a prosecution. So, our members who are the physicians on the ground trying to fill out the master forms and code these things feel that by increasing the universe of codes that really what you have done is increase their odds of losing at this regulatory roulette, that the more codes there are the more chance they have of picking the wrong one and then having that come back to them in a prosecution.
Now, that I have said that though I wanted to also confirm that we have no position on whether to support any of the codes either way. I just want to address the general issue of increasing this universe of codes and the numbers.
So, we need to take that into consideration.
Thank you.
DR. COHN: I think what we are trying to look for is direction of where the Subcommittee wants to go with all of this recognizing that there is this, as I said spectrum of choices. Kepa, do you have a comment? You asked a question before. Do you have a comment? No, Steve doesn't actually. He is pointing to Michael Beebe.
DR. BEEBE: Michael Beebe, American Medical Association. On the issue of cost/benefit analysis and additional research the AMA has contracted with Price Waterhouse Coopers(?) to do a thorough cost/benefit analysis of implementing ICD-10-PCS in the inpatient setting only, also, analysis of implementing CPT in the inpatient setting and then also moving ICD-10-PCS into the outpatient setting, so looking at the different scenarios for achieving a single procedure coding system and also just keeping it in the inpatient setting.
We have contracted. They are working. I expect that we will have something back from them around November maybe.
DR. COHN: Do others have comments? Kepa, do you have a comment?
We have lost, I guess one of our members on the phone.
DR. MC DONALD: The last discussion reminded me of some things in my own practice life that really don't have to be directed to this discussion but maybe explains my prejudice. The number of specific codes does drive you crazy. It absolutely is not helpful in practice to have to pick between 27 codes for depression. Fortunately, that is one I think that does allow the three-digit code because of all these fine boundaries especially when you are following a problem, and you are trying to keep that problem tracked under some kind of a code, and there is a sort of extension of codes into what one could conclude from the clinical data like diabetes, uncontrolled or out of control which we can tell very easily from the hemoglobin A1C and in the long run I would rather see fewer codes and more data sent along in the data for the support of these things which could be automated and could be more continuous data but that is not part of this discussion.
DR. COHN: Do you have a comment?
MR. AUGUSTINE: First, I was going to respond to something that was said earlier in the response about the benefits to the providers, to the payers, excuse me. I used to work as a director of employee management and a director of medical data analysis in a managed care plan and if there is anybody in the world in that is good at collecting and analyzing data that is pretty much ahead of the curve it is managed care and payers. They are moving in utilization of medical management and disease management. They have been kind of on the forefront of a lot of these activities, and they would definitely be a beneficiary of this. I don't know if they really want to admit that right now because they are really looking in the face of all the costs, especially in light of HIPAA, on the heels of HIPAA.
Now, it has been interesting. With Clem's letter and all the testimony it seems like we really kind of still have yet to get a good handle on this, and I would like to see some more information or have information kind of put into a better format for me to see going forward.
DR. MC DONALD: I wouldn't be opposed to doing an ROI letter if we had all these caveats built into the letter so that these kind of data would be something additional we collected and published, and I actually know there is some additional data because I heard this number 1.7 from somewhere, but I just really worry about it as this Committee's responsibility is it should have something to say that on the basis of 100 samples maybe viewed by two people or five people or 10 people, that would never get published in my field with any confidence about the data and if really you had two super coders or they were highly incented because they are your employees indirectly, you know versus having some indirect researcher doing it, and we might come up with numbers. If it is 1.5 it wouldn't be bad, but if it is three or five or 10 times more that could be pretty bad to institutions.
DR. COHN: Let me make a comment here, and I do think that we have a responsibility to try to move the process forward and you know we talked about all these various options, but we do have I think a couple of sort of middle options. I don't think we should just sort of not do anything.
Obviously one option is writing a letter to the Secretary saying, "Gee, you need to do cost/benefit. You need to collect data. You need to do more evaluation. We would like to hear back from this in 6 months and they should be also collecting similar data so that we could understand the cost and benefit to this," and we can do that alone. We could, also, if we think that it would encourage the industry, I mean one would observe that when the industry heard that we were beginning to think about a letter suddenly there are additional offers to give some testimony. The question is is that better handled by something like an NOI or even an NPRM, and so once again I am not saying what the answer is. I am just beginning to try to, what I think is to see something moving forward. The question is what is the best tool for us to use to move things forward. Is it just requesting at this point more information in terms of cost/benefit or do we really want to do something like you do an NOI or an NPRM?
Yes, Steve?
MR. STEINDEL: Simon, from the point of view of getting information, doing an ROI is the way, one of the ways that government requests information from the private sector.
PARTICIPANT: NOI.
MR. STEINDEL: NOI. So, that would be equivalent. Notice of intent is the equivalent to asking for further cost/benefit analysis.
MR. BLAIR: I think I heard you just say, Clem that you would not object to an NOI as long as it had the appropriate caveats.
DR. MC DONALD: Actually is there a mechanism available to actually get a substantial enough study comparing the time to code across a wide enough sample of cases that we really have and in different institutions not set up so narrowly on the CMS, some contractors but on real people who are out there?
MS. BROOKS: Could I comment on that? If one did an NOI or NPRN where you actually said, "Well, we are really thinking about this," I think what you would see is that hospitals on their own would do their own independent because they know we are know we are serious. They are not taking us seriously and they are waiting, and they are not going to put any effort into it until they actually think it might happen. What we would have to do is an impact analysis. If we write an NPRN then we have to go through and analyze the cost and it forces everybody in the government to sit down and figure out a projection. Everybody can write in and say, "You are wrong, because now that we have really studied it we know what is going on."
We would have to write out like I said that we think it would take X amount of hours to train people. We would have to lay out that we guess it would take this many more codes. Then the hospital industry would sit there and read it very carefully and check themselves and tell us if we are correct or not.
After a comment period we would have to write back, and we could be in several iterations but that is how they would take it seriously and do their own studies.
DR. COHN: This is tough to say about the difference between an NPRN and an NOI because NOI is sort of a new tool, and I don't actually understand that. It is new to us.
DR. GREENBERG: We talked about doing an NOI. We considered doing an NOI for the unique patient identifier and that was why we had the hearing in Chicago as we kick off to exploring doing an NOI, and that was as far as we got. So, it has been discussed in the past.
DR. COHN: So, the question here is what is the difference between an NOI and NPRM which is what I think the Subcommittee needs to understand and what the full Committee will need to understand, too, because it is not something that we have done before, at least in our period here.
MS. TRUDEL: Right. A notice of intent is just that. It is the Secretary saying, "I am thinking about doing something, and I am soliciting comments and input from the general public." It has the advantage obviously because it is in the Federal Register of getting the widest possible amount of input. It gives people a really good idea of what the potential issues are. It can lay out options. What happens after that then is after assessing what input you get you then would need to go through a notice of proposed rule making which essentially if it is written correctly does the same thing because a notice of proposed rule making says that we are thinking about doing something in particular.
Actually a notice of proposed rule making can be written in such a way that you bring forth several options for doing something and specifically solicit comment on it. We did that in the proposed rule for the national provider identifier. We had a number of different options as to how enumeration would work. We didn't recommend one. We simply said that we could do it this way. We could do it this way or we could do it this way. We solicited information.
A notice of intent doesn't require us to do a full-fledged impact analysis because this would be a major rule. It would have more than I forget the threshold right now, but there is no question it would be a major rule. It is $10 million worth of impact.
The requirements for doing an impact analysis under those circumstances are rather structured and so if it were done as an NPRM the department would have to do a fairly significant amount of the impact analysis before the NPRM were even issued. So, those are basically the two options and then again after publishing an NPRM then we have to go through the whole process of analyzing all the comments and we have to respond to each and every one of them that is significant.
DR. COHN: That is for NOI as well as NPRM?
MS. TRUDEL: With an NOI I believe what would happen is we would simply take the input and then the Secretary would make a decision as to where he wants to go from there.
DR. COHN: For an NPRM?
MS. TRUDEL: For an NOI you assess the comments and then decide whether you want to do an NPRM.
DR. COHN: Okay, great.
MS. TRUDEL: But there is no requirement to provide public responses to comments that you receive through an NOI.
MS. GREENBERG: I think the impact of doing an NOI would inevitably, as opposed to an NPRM, would inevitably delay the process and there is no way around that and some of you have been able to watch the rule making process.
MR. BLAIR: It is not clear for me the way you expressed it whether you are saying the NOI would delay it or the --
MS. GREENBERG: The NOI because an NOI must, if you are going to proceed, an NOI must be followed by an NPRM before you could do a final rule.
MR. BLAIR: So, it would delay the process.
MS. GREENBERG: It adds another step to the layer. So, it delays the process. I mean I think that is the main thing particularly when you are really unclear about something like it was in the case with the unique identifier, whether they really should even consider going forward. Then delaying the process was not a negative really and in fact in that case even discussing the NOI led to a decision to, well, actually a mandate not to proceed.
Today we heard from some organizations that had a lot of concerns. We heard from organizations in two previous hearings who expressed concerns but also who expressed very serious concerns about how much longer we can continue with the current code sets and sort of following up on what Karen said in her question, I do think there is a disconnect between statements that our whole health care system is based on these codes, but it is fine to stay with ones that are 25 years old, are not being used by the rest of the world, can't be updated and basically reflect old terminology and don't reflect current medical thinking because if our system is based on codes it would seem that we would want it to be based on up-to-date codes, but you know that is just, I mean as I hear that there just really seems to me to be a disconnect between if the position were that codes are not important, these codes are not important then I guess it wouldn't matter whether we use old codes, new codes, whatever, but given that everybody seems to agree that they are important, then not at least looking towards implementing newer more world-wide accepted codes in some cases or certainly more up-to-date and specific codes seems to me to be shortsighted, but we, also, did hear that some people felt that we really needed to move this process along.
From what Karen said you can accomplish everything in an NPRM that you can accomplish in an NOI, plus it forces the department to do the kind of cost/benefit or impact analysis that you were suggesting you might want to ask the Secretary to do in any event which an NOI does not. So, I think that if you went with an NOI it would delay the process and it probably wouldn't even provide as much information to the public and for decision making as an NPRM would.
On the other hand, it would be the appropriate choice if you are really unclear that the country should ever replace 9CM.
DR. MC DONALD: Just a couple of clarifications. Again, in the testimony last time or the published testimony one of the persons who was supportive of going to ICD-10-PCS said that however, they would want the newer codes that were in ICD-9-CM that were not in ICD, that is there is some newer stuff in the ICD-9 now than there is in the 10 that would have to be carried over.
MS. GREENBERG: Everything is --
DR. MC DONALD: I mean to say the point about being old or new and one other clarification, too, is it true that the only codes in ICD-10-PCS that would ever be used are those codes for surgical and medical procedures?
MS. BROOKS: It is true that hospitals today only report things for the most part that affect the DRG assignment. Sometimes they report others. I imagine they will do the same thing in 10-PCS even though there are a lot more codes that they could use for internal purposes. They might not bother to send them in on their bill.
DR. MC DONALD: And those other codes weren't tested at all. Is that, also, true?
MS. BROOKS: We did not ask hospitals since we were not going to mandate that they report them for billing purposes. We tried to make it the same environment. We said, "Code the kind of things that you would code in today's world under DRGs," and they did that. They know exactly what affects the DRG, and they are very astute.
DR. MC DONALD: The question I am leading to is why would we send out the whole package. I mean why deal with the whole package if it is only a subset that we really looked at and tested and people will use?
MS. BROOKS: We felt like so many people, so many physician specialty groups said that these things were important for their own use that we developed them, and if we want to have them for all users, obviously we care about hospitals getting paid but if people have need to capture data better for their own internal purposes we wouldn't want to withhold that section just because we know it will never get on the bill.
DR. MC DONALD: Even if never tested?
MS. BROOKS: Even if it is never tested because we assume that for instance the radiologists said that this was important to them and they would use it for their internal purposes. We believe that they are probably correct and so they believe they have tested those codes. So, yes, we would let them use them internally if they chose to. We wouldn't be opposed to that. We would have to discuss that part if somebody wanted to test them and found major flaws. They certainly could write an NPRM that said that this chapter and this other one are really inappropriate. We think you should delete them. That is the kind of thing we would have to respond to.
DR. MC DONALD: I don't think I would be willing to go along with a NPRM with the knowledge we have now.
MR. AUGUSTINE: As it was stated earlier an NPRM with options so that we would do the analysis of the impact, either we adopt both or adopt one or the other or stay as it, and maybe develop for the future.
MS. TRUDEL: I am just thinking about this in terms of wearing my HHS hat and likely having to work on this project.
It may be that the recommendation that the Subcommittee wants to pass forward perhaps should concentrate more on raising the questions and the concerns and specifying some of the issues that the Subcommittee members think that need some more information, in other words, we have concerns. Here are some of the specific things that we think should be addressed and input that should be solicited and it would include things like Dr.McDonald just mentioned.
As far as recommending to the Secretary a notice of intent versus the notice of proposed rule making I think that ties the hands of the General Counsel who will need to look at the Administrative Procedure Act and decide what he thinks is most appropriate.
Perhaps rather than directing the department in that respect you might want to specify the concerns that need to be raised and the issues that need to be addressed and that includes cost/benefit analysis, documented testing of actually coding using real-life charts, whatever.
DR. COHN: What to other members of the Subcommittee think? This is obviously a slightly different approach.
Oh, Jeff, I am sorry.
MR. BLAIR: I think, Clem has valid concerns and with respect to broader testing, more comprehensive testing I think that our testifiers today reiterated the need for the business case. I just want to make sure that it is broader than just the cost/benefit for payers. I think it has to be across the board. I think that Marjorie's --
PARTICIPANT: With respect to broader testing --
MR. BLAIR: Is that Kepa? I think that Marjorie's observation that if we went with an NOI it would be a further delay and with our histories of going through this process that is of concern to me because if we do have consensus on the fact that this would not be implemented prior to 2 years after the October 2003 date, in other words it would be somewhere after October 2005, before we would ask for compliance and adoption, I am concerned that if we go the NOI route then we have a vulnerability to having that date slip maybe uncontrollably later than that. So, now that I have sort of said this what I am basically saying is I think we should go the NPRM route. It sounds like that will allow for the cost/benefit issues and the additional testing and i guess I am kind of looking for Clem, your reactions to this because I think that we are accommodating all of the valid issues that you raised. Have we failed to do that with an NPRM approach to include all these things?
DR. MC DONALD: I don't see that if we are making the wrong decision doing it fast isn't helpful, that is i am not confident we know the right decision.
MR. BLAIR: But wouldn't we still get negative feedback?
DR. MC DONALD: Everywhere I look on that it is 7000 more codes by the numbers. They are just not down to the right part of the tree. We have got 23 more letters to put in front of it. There are ways to keep on coding for the next year. So, making codes up with a little bit of inventiveness, it is not -- we have used 3000 of 10,000 codes and that is in 10 years, right? Isn't that the numbers.
MS. BROOKS: Actually there is an extensive write up on this issue about extending the existing codes like you talked about in that new technology where I got, I probably to you that we have laid out, looked at all these options for how to take what we have got and expand it and just trust me the way we have laid out the tree adding a lot of alphanumeric digits in there is not getting at the root of the problem. The basic problem is the structure and pulling it apart we did have some public meetings on this with our ICD-9 coordination maintenance committee about 10 years ago where we asked this question and we looked at ways to take the current system, the current four digits and expand and work with it, and people were adamantly against it. It would have been worse they felt. We did explore this in depth.
DR. MC DONALD: But I am just still on the surface, I mean I have to trust and you are saying, "Trust me," but we are deciding this for the whole country and on the surface you are saying that it is broken because there are no more codes and I do the numbers and there are 7000 more. I understand the tree problem. There are letters that haven't been touched and I don't know if all -- one of the things it takes -- I don't know whether it is Q hasn't been touched. Under Q we just make that same tree and we have repeated codes. I don't know what is possible. You have
got letters. You have got E and V.
MS. PICKETT: The diagnosis part has V and E, but the procedure part has not letters. It only has four digits.
DR. COHN: Clem is saying that there are things that could be considered.
DR. MC DONALD: I mean it is dumb you are facing is it is dumb to put structure and codes. That is what everybody says nowadays.
DR. COHN: I want to sort of bring it back to what we are going to do about all of this stuff. I just heard Jeff propose an NPRM and obviously Clem, I think you proposed maybe --
DR. MC DONALD: I would go for an ROI with the right set of caveats.
DR. COHN: And obviously I think that the key issue here and I may be a little off in how I am describing this but the differences between things like NPRMs and NOIs and just expressions of concern which I think is what sort of Karen, you know, these sort of open issues that need to get resolved is how certain we are of a solution and of course as I said and I will apologize to Marjorie, but there are obviously multiple different things that we are talking about at once here. If we were somehow talking about diagnoses separately from procedures we might come up with different answers but the fact is that we are sort of putting them all together. Now, I am not sure that I feel that I should be pushing a particular solution here but what is it that we want to do? Do we want to just come up with a letter expressing concerns and just next steps? Do we want to propose an NOI? Jeff obviously just recommended an NPRM. Brady, you haven't commented.
MR. AUGUSTINE: My comment earlier was the possibility of an NPRM with the options listed either implementing one before the other or both of them internally.
MR. BLAIR: And all the caveats and all the testing.
DR. FITZMAURICE: Just a couple of thoughts. I wouldn't like to see things personally stretched out beyond the time that many of us are no longer around the table. Maybe it would solve some of the problems and yet I understand Clem's concern about wanting to put some brakes at different points to make sure we are making the right decision. From that standpoint a notice of intent sounds appealing. What I find appealing about an NPRM is that it forces the government to come up with a benefit/cost analysis essentially an impact analysis that you get to see before you make your comments and it kind of frames the issues and you can disagree with how the government has done it in comments and then there is a certain amount of trust. You trust that if the government takes these comments with an NPRM that it will turn out a good final rule and if the final rule is good there will be public response that might cause changes. Is there a process for change? So, I don't have a recommendation to make. I just am sensitive to the time that it takes to get something through, the fact that the system seems to have worked with an NPRM up to this point with all the other HIPAA standards and that I think everybody is agreeing we need more information on benefit/cost for payment and for other uses as well. I tend to agree with was it Marjorie or Karen who said, "Let us not tie the hands of the department, but encourage them to give us enough information to make the better recommendation at some future time."
MS. TRUDEL: It seems to me that the Subcommittee and the Committee have four alternatives that they can follow. One is to do nothing right now. Another is to make a specific recommendation to the Secretary with respect to going forward with a specific code set and this does not appear to be what people are comfortable doing. Another possibility would be for the Subcommittee to hold onto the issue, do more studying, have more hearings.
Another possibility is for the Subcommittee to say to the Secretary that we have looked at this all we can. There is a need for broader involvement and input and therefore we suggest that you use the regulatory process to do this and in doing that say that we don't feel comfortable making a recommendation to you. There needs to be broader study. There need to be these particular issues looked at and that is the recommendation. The Secretary then implements as he decides he wants to.
DR. MC DONALD: Is that tantamount to an NPRM?
MS. TRUDEL: No. I mean that could be anything. That could be an NOI. It could be an NPRM. It could be an open town hall meeting, an industry forum that is advertised in the Federal Register. I mean there is a whole broad range of issues and that is not something that would be decided at CMS.
DR. MC DONALD: I would like to stay one step sort of an NPRM.
MS. GREENBERG: I just want to clarify. I mean I am not advocating doing these separately and I think we heard from the industry that they, although we have heard mixed things; today we heard maybe go with diagnosis first. At the last hearing we heard it is preferable to go with the two of them together, but I really am unclear, Clem as to whether you are talking about both 10-PCS and 10-CM or whether you are primarily --
DR. MC DONALD: Well, 10-CM I think is not that much different. Really it has got a lot more accident codes or something in it and it really isn't much richer as I understand it. I haven't studied that, but that is not my -- my concern is this other one, the PCS because I mean we have got 3 years of use in Europe with it, with the diagnosis code. So, there is a lot of experience, maybe more.
MS. GREENBERG: More like 7 years in some countries, but a lot of countries have implemented 10-CM. Because it hasn't really been available to them, at this point countries haven't implement 10-PCS. So, they have replaced 9-CM, Volume 3 or whatever they were using with new procedure classification systems and I don't know if any of them are quite as detailed as 10-PCS but some of them have characteristics similar to 10-PCS but in any event I do think if you are going to raise all these issues even, I mean I think it is important that you differentiate what your comments are about.
MR. AUGUSTINE: You know of the options that we heard the last one that Karen gave was tantalizing but I was wondering how much follow up it would receive, whether or not in that type of rule cost/benefit analysis would happen if we just followed that last option.
MS. GREENBERG: It has to if an NPRM is pursued.
MR. AUGUSTINE: I mean that was the option without the NPRM.
MS. GREENBERG: I thought your last option said, "Pursue the rule-making process." It did say the rule- making process.
MR. AUGUSTINE: Oh, okay.
MS. GREENBERG: The last one included the rule-making process, but it didn't say which technique in the rule-making process.
MR. BLAIR: I am always very hesitant to have a position where Clem has a different view. So I want to narrow the gap and clarify something. Here is the piece where, Clem help me clarify this, okay, because it is my understanding that the pluses and minuses for a notice of intent, the plus is that it doesn't set the process in a time line that might be as easily bound. On the other hand the negative is that it might cause some delays. When you talk about an NPRM you wind up having less vulnerability to an open-ended delay but my perception is, and I guess maybe I am too close, but my perception is that we will still with an NPRM have the opportunity to identify if the direction that we have set is misguided, incomplete, incorrect and either corrected or aborted. So, I don't see a down side and tell me if I am wrong. I am not seeing the same down side you seem to be seeing for an NPRM, but I do see the up side that it protects us from an open-ended delay.
DR. MC DONALD: I see tremendous enthusiasm on the side of the CMS people for going straight ahead as fast as we can with ICD and PCS. I think that that question has been answered in our minds. If I had this kind of data an I as asking to change a procedure in my own little hospital based on this data I would not, forget specifics, I would not make any changes in procedure because I couldn't believe anything in this data. That is how I look at data. I mean this is, as data this is useless. One hundred cases, and the numbers say how many people, and now maybe it is. It doesn't tell you anything with direct comparisons on 100 cases. You don't know how many, but the thing I am interested in is the comparison not the continuous improvement, the comparison data with the final one and you might have had three reviewers and two of them might have been the best stars and out of 100 cases you probably get 80 percent of common disease diagnoses which are easy to code. So, I am just saying that I wouldn't make any decisions on any kind of process change in my little hospital based on that data. It seems to me with 280 million people in this country and a trillion dollar budget we should put more investment in the solidity of the data we use to make these decisions. That is really the crux of my worry.
DR.COHN: You know I think the real question is what are we going to do about this. We are obviously in a situation where Kepa is not on the line or if he is he can't get through.
DR. ZUBELDIA: I can't hear on the phone and so I have to listen on the Internet.
DR. COHN: Kepa, can you hear me now?
DR. ZUBELDIA: Yes. I would like to make a comment. I have been trying for a while, but --
DR. COHN: We would like you to make a comment.
DR. ZUBELDIA: Being on the Internet puts me 1 minute behind. So, when you say, "Kepa, do you want to make a comment?" by the time I hear that you are already talking about something else.
(Laughter.)
DR. ZUBELDIA: I think that the issues about ICD-10 diagnosis codes and I think in PCS for institutional and ICD-10 for professional are very, very different issues and whether it is a notice of intent or notice of proposed rule making I think that they need to be separated. There is an additional issue as to whether the adoption should be synchronized or not but the issues, the problems across the impact are different for all three and after everything that I have heard I am more inclined to go with the notice of proposed rule making because it forces the government to make an impact analysis. But I think that it has to be a separate impact analysis for procedure codes, for diagnosis codes and for procedure codes for professional services, too. That needs to be looked at separately.
DR. COHN: Okay, so, Kepa, I am taking your comments because I actually was going to ask people what their vote was, and it sounds to me like your vote is in favor of an NPRM and you obviously want to work on the wording of our letter. Is that correct?
DR. ZUBELDIA: Yes, I am in favor of the NPRM but with the caveat that it has to be separate, addressed separately.
DR. MC DONALD: More than one NPRM?
DR. ZUBELDIA: No, perhaps one NPRM with different sections or the three different types of things that we are talking about because the impact of the diagnosis, even the testifiers today they were not as concerned about implementing diagnosis codes as they are implementing procedure codes.
DR. COHN: Okay, so we have heard one vote.
Jeff, would you like to express your vote?
MR. BLAIR: Again, I still have that same reluctance because I don't want to miss an issue that you see that I don't, Clem, but it seems to me like the NPRM will allow us to get at all of these issues and get them surfaced and if we are misguided we could get them corrected. If I didn't see that then I would be hesitant to do that, but I --
DR. MC DONALD: You have got to vote your conscience.
MR.BLAIR: And I go for the NPRM with a caveat that Clem has set forth for testing. I would like to see him specifically contribute to defining those and where the thresholds are and I would like to add Kepa's suggestion that we have separate sections in the NPRM for PCS and for ICD-10-CM and the rest of the other pieces I think could be pretty much similar to the first draft that we already have with the deadline of October 2005, no sooner than.
MS. DALMER(?): I am Theresa Dalmer with Blue Cross and Blue Shield Association. I just want to remind everyone that the NPRM process did not identify the problems with the NDC code as we have begun to understand as we really started getting more into it. Also, with the 2005 time frame if you look at what has to be done over the next couple of years 2005 for an operational date is really very ambitious in that over the next 3 years we are going to be doing the identifiers as well as the modification to the transactions and code sets, and when you put it all together and you start looking at a time line 2005 is on us pretty quickly, and at least the people that we have talked to if we ever began to really start to implement this we would need a considerable amount of lead time not to mention the transition time going from one to the next.
So, I would just offer that, and an NPRM and I understand it does provide the cost/benefit analysis, but at the same time it is a very defined period of time, Potentially what 60 days? And in that period can you really get the kind of information that you need to be able to accurately anticipate what the impacts are, and when I hear the comments about the impact on the hospitals what we try to emphasize is that it should be the impact across the entire industry, that we should be looking at all of the players, and I am saying not just the payers and not just the providers but really looking at all the players, and you need to reach out to them. You don't often, not often the ones that are looking at the Federal Register and the NPRMs, but rather really reaching out and bringing that input into play.
MR. BLAIR: Could I add that I think maybe one of the things we should put in our recommendations is that we have more than a 60-day period for response, maybe 90 or 120 days? That is one piece. The other piece is at least the way I am envisioning it is the NPRM as much as putting us on a time frame which could be changed by public testimony and critiques of that time frame, it also puts in place a vehicle by which it is a mechanism. It is a mechanism that formally solicits public comments and critiques. It is a mechanism where we wind up getting the business case. It is a mechanism for us to get all of these other pieces in place. So, my thinking on it is that it is the beginning of a process and we have to do something, and I feel like --
DR. COHN: So, your vote is unchanged, okay.
DR. ZUBELDIA: I have a comment.
DR. COHN: Kepa, what I am actually trying to do is, well, Kepa, go ahead.
DR. ZUBELDIA: One concern that I have is that the impact analysis may not necessarily reflect private industry the way private industry may want to be reflected. From the impression I got today Michael Beebe is talking about the AMA doing its own impact analysis and I wouldn't be surprised if the association does their own impact analysis. Is there a mechanism by which those private industry impact analyses could be reflected in the process?
DR. FITZMAURICE: I think if they were submitted as comments to the NPRM they would have to be considered.
DR. ZUBELDIA: But the comments don't have to be published in the final rule. I think that perhaps some sort of -- none of the above and maybe a white paper before the NPRM in which the government publishes in a very informal way its own impact analysis and perhaps side by side with some of the private industry impact analysis and then put that in the NPRM.
MS. GREENBERG: If there are other impact analyses that have been performed just like WEDI(?) did one and it was highly referenced I know because it was supportive of the approach that was being recommended, but it was highly referenced I know in the NPRM on transactions and code sets, if other impact analyses have been done and submitted to the department I think they could also be referenced and they could be you know, where there was agreement it could be stated and where there wasn't agreement it could be stated, but I mean I think probably the people preparing the NPRM would look as widely as possible for input on this. I mean I think that could be accomplished. Am I correct, in the NPRM?
DR. ZUBELDIA: Would it be possible for the government to publish its impact analysis independent of the NPRM prior to the NPRM?
DR. MC DONALD: I would like to suggest doing a notice of proposed rule making and include all these ideas in that.
DR. COHN: Proposed or NOI?
DR. MC DONALD: NOI, whatever that is.
PARTICIPANT: Sometimes the best action is no action.
DR. COHN: Be aware, everyone that this will be an interesting discussion with the Committee in September whatever it is we decide. Kepa, I don't know what to do with your comments other than just listen to them and note them. You voted in favor of an NPRM. I don't know whether you are now modifying that with an NPRM with three or four other things, but I would certainly encourage that maybe there is some way for us in whatever letter we begin to further define when we get some of these additional thoughts in. Is that okay?
DR. MC DONALD: Could we get you to change your vote, Kepa to an NOI? Then we could do all this stuff.
MS. GREENBERG: I think you have to understand about the -- can I just give a little clarification here about the rule-making process, that you know it is started by the writers but then it goes through the department. It goes to OMB. It goes through a lot of people, a lot of parts of the government before it actually gets published the NPRM. So, to ask that pieces of it get published prior to the NPRM really doesn't make sense because you can't go through that whole vetting process on a non-NPRM. I mean it just isn't going to happen. As bad as bureaucracy may seem it does push things to some kind of conclusion that just informal exchangedoesn't.
MR. AUGUSTINE: I would say that I agree with Kepa's original proposal of an NPRM but separate out the different aspects.
DR. COHN: Kepa, are you still in favor of that?
DR. ZUBELDIA: Yes, I am still on the line, but the problem is that I cannot hear your conversation. So, I have to turn on the Internet and that lags about 1 minute behind.
So, if you would repeat the question loud, I can hear you on the phone.
DR. MC DONALD: Kepa, do you want to change your vote to an NOI instead of an NPRM?
DR. ZUBELDIA: What would be the advantage of changing to NOI?
DR. MC DONALD: You could get all that extra stuff done.
MS. GREENBERG: I just explained why you couldn't.
DR. ZUBELDIA: If we can get all that extra stuff done, yes, let us go for the NOI. I think it is important that the industry sees the government impact analysis and other impact analysis before they evaluate the NPRM. Otherwise there is not going to be enough time to compare more than one impact analysis side by side.
DR. COHN: You know, Kepa, I do apologize because I think there is sort of lobbying going on in terms of the voting and I actually think we need to finish up the vote, but I have heard two in favor of an NPRM. I am now confused about what your vote is, whether you are in favor of an NPRM or something else.
DR.ZUBELDIA: My vote is in favor of having a full impact analysis done by the government that can be compared with private impact analysis and then go as expediently as possible to a rule making and the reason why I am putting it that way is that if the government needs to do an NPRM in order to do the impact analysis then let us do the NPRM. If the impact analysis can be done as part of the NPRM process and pre-published before the NPRM is published and let the industry do their homework, also I think that is positive. If the impact analysis cannot be done unless the NPRM is in the works then let us go for the NPRM. That is why I am voting NPRM but pre-publish the work that has to be done for the preparation of the NPRM, if that is possible.
If it is not possible then I think it is more important to have an impact analysis than to not have an impact analysis. So, I still vote for the NPRM.
DR. COHN: Okay, and I guess people from our staff are saying that is not possible. Is that correct?
MR. BLAIR: Maybe we could extend the public comment period so that they --
DR. COHN: We are not in charge of anything. We are an advisory committee to HHS. We are a subcommittee of an advisory body to HHS. So, we have our limits in terms of how many good ideas we can --
DR. BEEBE: A question regarding the limits. Can the Committee contract to do an impact analysis?
DR. ZUBELDIA: That would be great.
DR. MC DONALD: I would vote for that.
DR. COHN: I am abstaining actually. I won't say that I am more confused than when we started, but if you remember when I started off the conversation I sort of said, "Gee, we can go for an NPRM. We can go for an NOI. We can send a letter to the Secretary saying that there needs to be cost/benefit which is a little bit different than an impact analysis and all these things before we proceed forward. We can do nothing. We can hold further hearings," and I am at this point, Kepa once again a little confused about whether or not you want an NPRM or what you really want is an impact study and this is I guess my own confusion just because you seem to be saying multiple things.
Brady, do you have a comment?
MR. AUGUSTINE: My comment is the same. If we could actually do it ourselves and make sure that it happens in a timely fashion then I would rather, if it is possible for us to do our own impact analysis before we take that next step because that next step even though it kind of forces certain things to happen in a certain time frame there is a lot of resources and expense that go into that that may not be necessary if we can find out beforehand what route we might take or recommend.
DR. FITZMAURICE: I suggest that undertaking a study with hard coders and run through 1000, a couple of 1000, 5000 forms, double code them and compare them. That is probably beyond the resources of NCVHS and you would have to have some monitoring continually of it to make sure it follows the methodology that NCVHS would agree upon even if the resources are available. So it depends on the resources and depends upon willingness to --
MS. TRUDEL: I just want to reiterate the four possible things that the Subcommittee can decide to do. The Subcommittee can make a specific recommendation that the Secretary go forth with a standard. The Subcommittee can do nothing. The Subcommittee can agree to study it more or the Subcommittee can recommend that the Secretary do additional study, analysis, whatever using the Administrative Procedure Act process.
Now, I thought I was hearing that last one. Now, I think I am hearing the Subcommittee wants to do more study, but again, there are only those four things that are possible to do.
MR. AUGUSTINE: Who of us vote around this table?
DR. COHN: There is four with Kepa there. I am abstaining. I believe I heard three but I keep getting confused by what Kepa is voting for and that was really the clarification I was --
MS. TRUDEL: I think Kepa and Brady are both at this point voting that the Subcommittee should study it more.
MR. AUGUSTINE: Not hear more testimony. I mean if it is was possible but apparently it is not possible for us to commission our own study --
MS. GREENBERG: You could commission your own study but it would not accomplish what the NPRM process accomplishes and then you could hold hearings on your study because then you would, also want to hear what other people's studies were and it is basically, you know, if you want to do a study while the department because you don't trust the department's impact analysis, you can do that parallel through the department starting to develop its impact analysis but I don't really think it is an alternative because it is really then no different than delaying for another few years and doing a study and never getting that out of the woodwork saying to the people out in the field that this is something we are seriously thinking about. It is really not any different than that. I mean you can do a study, but --
DR. MC DONALD: That is why I think the notice of proposed rule making is the most sensible because we will get studies cooking either our own or industry.
MS. GREENBERG: You don't mean that actually.
DR. MC DONALD: NOI? No. I don't know the letters. It starts with notice and I don't know that, I mean depending on what the time frame is whether it is feasible in any circumstance. I don't know that we could just shorten and we have given this notice now we have shortened the time until the final. I don't know that that would have to change the total time.
MS. GREENBERG: It has to.
DR. MC DONALD: Is it always 2 years before you say something can be done or 3 years?
MS. GREENBERG: Trust me. It will add a few years to the process if you do an NOI.
MR. AUGUSTINE: With that understanding I keep my original vote.
DR. COHN: Okay, I think what I am hearing then is a vote in favor of going forward with an NPRM, with one negative.
DR. ZUBELDIA: Let me add something. Would a letter from the Committee to the Secretary act in fact as a notice to the industry of intention of going in this direction and would a letter like that put the industry in gear to do their own impact analysis in parallel with what the department is doing?
DR. COHN: Kepa, I think it might be possible to put a sentence in sort of saying that we should ask the Secretary to inform the public of this direction and urge industry to perform its own impact and cost/benefit studies in advance of release of any NPRM.
DR. ZUBELDIA; I think I would like that.
MS. GREENBERG: Frankly, also, in reflecting on what Jeff said about wanting a full industry appraisal it may be that the Committee would like to commission some type of study that tries to quantify to some degree the benefits as well as, I mean we know that we have people out there who are definitely going to be coming forward with quantifying the down side, I mean all the things that have to be changed. I mean that is inevitable or to do both. I mean I don't think it is an alternative. I think it could be a complementary activity and it would probably be helpful maybe to the department.
MR. AUGUSTINE; And we have to remember that quantifying, being an econometrician --
MS. GREENBERG: It is in quotes.
MR. AUGUSTINE: Being able to quantify the benefits is --
MS. GREENBERG: Very difficult.
MR. AUGUSTINE -- extremely difficult. It is a lot easier to quantify costs.
MS. GREENBERG: Exactly as Dr. Sondik I think testified back in May.
MR. AUGUSTINE: Yes, when you are looking at the public good some of the things are so spread out they are impossible to measure.
DR. COHN: Donna and Karen, have you gotten all this?
I mean I think we are basically approving the original document.
DR. MC DONALD: No, wait a minute. I didn't hear we were approving the original document.
DR. COHN: What I am saying is effectively what we are approving is the direction of the document. What we will do is to basically revise it further. I think we will need to send it out to everybody in advance.
DR. MC DONALD; Do minority opinions count? I mean what does that let me do?
DR. COHN: You are certainly allowed to give a minority opinion. Since I abstained, I am not even going to be -- make sure I am not going to do one in presenting this to the full Committee.
MS. TRUDEL: It goes to the full Committee and you would have the opportunity to discuss your minority view and the reasons behind it at that point.
DR. COHN: Remember at the end of the day we are just a Subcommittee.
I am actually abstaining only because it has to do with procedural codes as well as diagnosis codes. I will say that even regardless of that I do have some inclinations about the date of implementation identified in the document which appears to me to be impossible. So, that is just something that we should think about as we move forward but that is small compared to I think most of the discussions that we have all had.
DR. COHN: Now, the question is do we have time to talk about DSM or should we hold that for the September meeting?
DR. MC DONALD: I have a flight at Dulles and I want to say --
DR. REGIER: Could I speak to the DSM? We have been waiting for 2 hours to speak to the DSM issue with my attorney as well. So, we would really like to address that.
DR. COHN: Should we take a minute or two to discuss this? I think we have a letter presenting your perspective but we have developed as a white paper that is in front of you, three white papers.
Let us do a short break.
(Brief recess.)
DR. COHN: I think in this last session we are going to try to do is to talk some about DMS. Obviously we have received letters from the American Psychiatric Association as well as I think the Secretary has received letters requesting that DSM be named a code standard for use in psychiatry and mental health basically instead of the mental health sections of ICD as I understand it. There has, also, been a request really to the Secretary really not to us with relationship to as a code set and criteria developer that there should be a designated standards maintenance organization and be part of that group. We actually asked Donna and other staff to prepare some white papers, not that they are official white papers but just generally to sort of lay out the issues and come up potentially with some recommendations in terms of approaching it to use as a basis potentially for letters to the Secretary. These are still draft. I think we wanted to get some of your views on them. We would obviously invite views from the American Psychiatric Association who are here to obviously critique some of our approaches.
Now, basically there are, I guess we have three background papers, and, Jeff, I apologize, you have not seen them but I just want to sort of generally go over the content and approach just to see if it makes sense for the basis for a letter.
Now, first of all the relationship to the designation as a data standards maintenance organization I think the view was that people developing, that basic data standards maintenance organizations are typically maintainers of transactions of one sort or another as opposed to how did we describe this here? We have a code set or what was the, or diagnostic criteria maintainers and just in the same way that for example any of the other code sets were really not part of the DSMO organization for example, X12 or HO7 or other groups are, we felt that they really fit more into activities that were really not DSMO activities. So, our recommendation was that they probably should not be part of the DSMOs in the same way that NCVHS or other groups are not really part of that. So, that was I think the initial view that we had to sort of go forward with.
Now, we had also received concern that mental health as I said would not be able to use the diagnostic criteria in DSM and I think that this is actually a much more substantive issue and I am hoping that I have the right --
MR. BLAIR: I didn't understand your last sentence.
DR. COHN: Diagnostic criteria?
MR. BLAIR: Yes, diagnostic criteria and mental health would not be able to use the diagnostic criteria.
DR. COHN: In DSM.
MS. GREENBERG: Once ICD-9-CM was adopted under HIPAA as the standard for diagnosis reporting.
DR. COHN: I will say sort of in shorthand that obviously our interest would be that mental health professionals and psychiatrists would be able to continue using DSM, the diagnostic criteria in DSM as well as the crosswalks that exist between DSM to ICD. In most cases they are using things that are basically ICD on bills and other transactions, and I thought I would just read through this jus because I think it says it better than I might just to sort of make sure you understand where we are coming from and then we will obviously ask the American Psychiatric Association for input and feedback regarding this.
Now, basically this will just take me a minute or two. It says, "The American Psychiatric Association in testimony to our Subcommittee and in a letter to the Secretary of HHS requested one, that DSM-IV be designated as the official code set for purposes of reporting and transmitting information about mental disorders and two, APA be named as a DSMO." This paper summarizes issues related, having DMS-IV adopted as a HIPAA code set. The APA states that the request is a short-term solution until such time as ICD-10-CM is implemented.
In the August 17, 2000, final regulations ICD-9, Volumes 1 and 2 including the official ICD-9-CM guidelines for coding and reporting as maintained and distributed by HHS for the following conditions, diseases, injuries, impairments, other health problems and their manifestations and causes of injury, disease, impairment or other health problems are adopted basically.
During the 60-day comment period for the transactions and code sets NPRM the department received no comments from the health care industry recommending the adoption of DSM-IV. There is then a description of ICD with various parts and appendices. Unlike ICD-9 published by WHO where the glossary is actually part of Chapter 5, the mental disorders glossary in ICD-9-CM was moved to the appendices along with other classifications that were not used by health information management professionals or other users of ICD-9-CM classification and as such are not part of the main classification. This appearing in the appendices had not been updated since ICD-9-CM was introduced in 1979, which is an awful long time ago. The appearance of archaic terms in ICD-9-CM is not unique to mental disorders.
To accommodate the inclusion of newer language and terminology the terms and concepts are incorporated into ICD-9-CM through modifications to the alphabetic index to facilitate the coder's ability to assign the appropriate code in ICD-9-CM.
New codes may also be added to the tabular list when space is available. Recommendations for modifications are presented to the Coordination and Maintenance Committee which are held twice every year. Then there is a description of DSM-IV. The DSM-IV is divided into several sections, instructions concerning the use of the manual, the DSM-IV classification, descriptions of the multi-actional system for diagnosis and diagnostic criteria for each of the DSM-IV disorders accompanied by descriptive text.
At the time DSM-IV was published APA acknowledged in Appendix G that the official coding system in use was ICD-9-CM and that DSM-IV had harmonized its criteria with ICD-9-CM codes.
As stated in the DSM-IV manual the criteria and descriptions are intended to be used by individuals with appropriate clinical training and experience in diagnosis. The manual notes that it is important that DSM-IV not be applied by untrained individuals. The specific diagnostic criteria including in DSM-IV are meant to serve as references to be informed by clinical judgment. The diagnostic categories, criteria and textual descriptions are meant to be used by individuals with appropriate clinical training and experience in diagnosis.
The next title is the DSM-IV overlap with ICD-9-CM. The codes in the DSM-IV correspond to the codes in Chapter 5, Mental Disorders of the ICD-9-CM and in most cases the codes and code titles in ICD-9-CM and DSM-IV match. The ICD-9-CM as currently used in hospitals, hospital-based outpatient departments and other health care settings for a variety of purposes including use in administrating financial transactions. It is only the terms, definitions and diagnostic criteria that differ between the two classifications. The actual numeric codes remain identical to ICD-9-CM.
The APA has expressed concern about mental health practitioners, agencies and other institutions abandoning their use of DSM-IV diagnostic criteria because DSM has not been adopted as a HIPAA code set. HIPAA does not invalidate the use of diagnostic criteria that exist outside of the code sets adopted for the purposes of establishing diagnoses or for use in other areas such as quality of care reviews and other activities including fraud and abuse investigations. Given that ICD-9-CM has since 1979, been the official code set used to report diagnoses for government purposes including reimbursement mechanisms such as DRGs and because the DSM-IV diagnostic criteria have been crosswalked to the appropriate ICD-9-CM code numbers the claims submission process has not been impacted.
As noted by APA practitioners have been able to meet government reporting requirements because the codes and terms provided in DSM-IV have been crosswalked to ICD-9-CM.
Recommendation. If there is agreement with the information presented in this paper, the subcommittee will draft or our staff will draft a letter to the Secretary of HHS for our review at the next meeting that states one, that the continued use of DSM-IV diagnostic criteria that crosswalks to ICD-9-CM codes does not pose a conflict with code sets adopted under HIPAA and two, adoption of diagnostic criteria used to establish diagnosis is outside the scope of HIPAA.
So, that was sort of long winded, but, Jeff I thought you needed to see it. I thought probably also the APA since we just handed them this draft document probably also needed to hear it and think about it a little bit.
As I said, I think the intent is to allow use of both as well as to acknowledge that the crosswalk is a legal crosswalk between these. Obviously I would like to hear comments from APA and their views on this.
DR. REGIER: Thank you very much. I am Darrel Regier and I am the Director of Research for the American Psychiatric Association and also the Executive Director of the American Psychiatric Institute for Research and Education and we appreciate very much the summary and analysis that you just read.
Unfortunately, we think there are some concerns in the legal interpretation of what a code set is that will need to be addressed by the Office of General Counsel of HHS because basically our counsel who is representing us today, Mr. Robert Charrow has evaluated this at our request after we received a number of concerns from the Managed Behavioral Health Association from American Academy of Pediatrics that also publishes DSM primary care, from the State of California uses DSM for all of their reporting for substance abuse and mental disorders. A number of groups have been approaching this concern that the DSM is not a designated code set.
Now, the reason for the concern is the specific statements within this 45 CFR 162.103 which says that implementing the regulation, it defines the code set to mean any set of codes used to encode data elements such as tables of terms, medical concepts, medical diagnostic codes or medical procedure codes and a code set includes the codes and the descriptors of the codes. It goes further in defining, The regulation defines descriptor to mean the text defining a code.
Now, the problem is that without DSM-IV being identified as a code set the fact that it is an informal, it is not a legal, and that is what is one of the things that is wrong with the background paper, it is an informal crosswalk. It is not a legal crosswalk. It is in no place encoded in the statute in HIPAA that this would be a descriptor for the code set and the legal opinion that we have is that in the absence of that the default descriptor is in fact the ICD-9-CM glossary and definitions and that basically what you would do by not having DSM designated as a code set is that you take a giant step backward in terms of the criteria that are used officially, legally in this country whether it is for clinical practice or for fraud and abuse or for any other monitoring of clinical practice.
So, in the past before HIPAA there was this informal arrangement that we have had for years where we worked very closely with the National Center for Health Statistics to make sure that in fact as ICD-10-CM was coming along that the names will match between DSM-IV and the ICD-10-CM. We, also, work very closely with the World Health Organization in terms of getting a whole reorganization of the Chapter 5 section to actually focus on explicit diagnostic criteria which was an effort of over a decade, but even here once you get to ICD-10 the actual criteria that are in the ICD-10 clinical guidelines and in the guidelines for research are different than what is in the DSM-IV and what you have now in the draft of the ICD-10-CM is an introductory phrase that says, "Please refer to the DSM-IV for the definitions of these codes."
Now, the thing that we have no basis for in law right now is a determination that that type of informal referral without DSM being a designated code set, a descriptor of the actual number and name will have the force of law, and so, with that I would turn it to Mr. Charrow to perhaps comment on his interpretation of the law.
MS. TRUDEL: It seems that we have a legal argument here and since there are no lawyers in the room other than yours it therefore seems to me that it makes sense for us to raise this issue with the department's general counsel and I guess my question is if the department's general counsel feels that the concept of diagnostic criteria is indeed separate and distinct from the HIPAA code sets does that not make a difference in what the Committee's view would be.
DR. COHN: I guess the question is whether the legal counsel would support this approach. It seems to me basically I agree with you and I am just thinking in the conversation and I am unfortunately not a JD. I am an MD obviously, but I guess the question I would have for you is if this construct actually was supported and felt to be appropriate by HHS does this meet your needs recognizing that we trying to think to get between here and ICD-10 and obviously beyond these comments.
DR. REGIER: I think if in fact the law would be explicit in acknowledging DSM-IV criteria as the descriptor for the code numbers that that does advance things considerably. There are as we had mentioned in our testimony and in our letter some names that are archaic and actually are pejorative in the way that they are used at the present time. Instead of major depression for example which does not appear in ICD-9-CM it is defined as affective psychosis and we have had complaints of patients who have had their code read out that they are psychotic when they have major depression and it has caused a great deal of consternation on the part of people who have to deal with that code.
Now, that is more than just a little archaic. I mean that can have implications for their employment. It could have implications for stigma and a whole range of things for people in these settings. So, continuing to use the ICD-9-CM names is not insignificant because of the advances that have taken place over the last 25 years since ICD-9-CM was implemented.
MR. BLAIR: Since we will have to not be able to vote on this at this time because apparently legal counsel for CMS isn't present to finally resolve this my question would be is there another issue which is not a legal issue whereby CMS or anyone else --
PARTICIPANT: HHS.
MR. BLAIR: Thank you, HHS or anyone else objects to DSM becoming a designated status maintenance organization, any non-legal reasons why they shouldn't be?
MS. TRUDEL; Jeff, this is Karen. The designated standards maintenance organizations as Simon indicated are very specific to the transaction standards. They are not clinical code set maintainers. The AMA's CPT Editorial Panel is not a designated standards maintenance organization. CMS is not designated standards maintenance organization although we are the maintainers of HCPCS. So, the concept of the DSMO title is just not a fit. It doesn't fit. That is not the purpose of that organization. The people who participate in it do not have code set expertise. They don't pretend to have clinical code set expertise and they are specifically working on modifications to the transaction standards. So, I think that is really a threshold issue. If that could be gotten beyond then the issue has to do with clinical code sets and ICD-9 versus DSM and the discussion appears to be that the numeric codes in ICD-9 are fine but that the English language descriptions of the codes in some cases are considered to be inappropriate and then the issue of the diagnostic criteria. So, let us keep all the issues separated.
DR. REGIER: And I think we certainly don't have an interest in being an out-of-step DSMO. So, that is not the issue. The issue is the code set and whatever mechanism is necessary to have a publicly approved code maintenance organization is what we were interested in.
DR. FITZMAURICE: I have a question. You are never sure that you understand everything. Is one of the major issues that whoever maintains the appendix, I guess it is Appendix B does not replace the diagnostic criteria with the current diagnostic criteria proposed by APA and is there some reason why they don't do that?
MS. PICKETT: The appendix as written was originally part of ITCHED-9 as published by WHO. We cannot change that but we did not update it either and that is why it is in the appendix and ITCHED-9 itself as it is in ITCHED-10 their glossary is actually embedded within the mental health chapter. In 9-CM the decision was made a long time ago to pull those out and put them in a separate appendix and not have them embedded as part of Chapter 5 so that one would not be guided by the glossary that is in ITCHED-9-CM in the appendix and I think either AHA or AHIMA could speak to the fact that when coders are trained in the use of ITCHED-9-CM they are no trained to use any of the appendices in the glossary and in fact for the morphology section that is part of ITCHED-9, but the tumor registries are using, have progressed and now are using ITCHED-02. So, again those things were placed in the back because they were originally part of ITCHED-9 which we did not change but I think it is understood by the users of the classification that those are not updated. They are antiquated and that there are more relevant descriptions of conditions and diagnostic criteria much as the same as has occurred over time for diabetes. There are new diagnostic criteria for diabetes but again they are not part and parcel of the classification. Mental health is the only section where WHO actually included diagnostic criteria.
MR. FITZMAURICE: If they are outdated and everybody recognizes that they are outdated is there any objection to having a clinical modification that replaces that appendix with an updated appendix since they are not supposed to use it anyway?
MS.TRUDEL: Mike, what I think we are saying here is that we do not believe that we adopted those appendices at all and that therefore there is no standard for diagnostic criteria in ICD. So, why would we want to update it if it is not a standard?
MR. FITZMAURICE; Why even publish the appendices? I assume there was a reason for publishing the appendices, that they came over from WHO and --
MS. TRUDEL: They came over from WHO.
MR. FITZMAURICE: So, is one way to get it changed to go to WHO and say, "Would you readopt this and send it over to the US and ICD-10 and ICD-9?"
MS. PICKETT: The only difficulty with that is that WHO no longer supports ICD-9. The rest of the world as you know has moved on to ICD-10.
MR. FITZMAURICE: So, even they don't care.
MS. TRUDEL: So, what we are saying here is there is the need to clarify that those appendices are not part of the standard, that they are not relevant, that they are not required nor even recommended.
MR. FITZMAURICE: But it is a waste of paper to put something in there that is so archaically out of date. Would we have an objection if coders did use the current diagnostic criteria?
MS. PICKETT: Coders do not use diagnostic criteria. That is the domain of physicians. The coders are not using it.
MR. FITZMAURICE: But it does seem to me confusing that you have got an out-of-date appendix. It could be replaced by an up-to-date appendix which they still shouldn't use or delete all the appendices.
MR. BLAIR: Karen, is there a text to this?
MS. TRUDEL; I think the solution that was proposed, Jeff in the paper I think is --
DR. COHN: I think we need to have it reviewed by HHS lawyers and probably have a discussion with you.
MR. CHARROW: There are two issues actually and I think one of them is the scope of the regulatory provision that Dr. Regier just read and the second is the legal impact of having a single official code set that has no diagnostic criteria. The statement in the white paper by the way that ICD-9-CM displaced DSM-4 prior to HIPAA as the official code set is only partially correct.
MS. TRUDEL: We did not adopt diagnostic criteria for CPT either. So, it is perfectly legal.
MR. CHARROW: I am ont saying perfectly legal. What I am saying is that CMS in its various regulatory and guidance issuances, for example, instructs surveyors who inspect hospitals to use DSM-IV when reviewing patient charts because there are diagnostic criteria. The Inspector General's Office uses DSM-IV when determining whether or not to prosecute under the fraud and abuse statutes and 675 state laws have incorporated DSM-IV criteria either for payment decisions with respect to their Medicaid programs, eligibility with respect to welfare benefits or in fact with respect to Sienter(?) requirements when it comes time for prosecuting folks for murder one in seeking capital punishment. Eliminating DSM-IV as an official code set would have profound implications with respect to those 675 state laws. It certainly would have implications with respect to those state laws that affect Medicaid because there are now only a single code set under HIPAA. The word "single" is what is important.
Obviously if there weren't a single code set then the crosswalk, that informal crosswalk which has been used up until now certainly would be viable. It no longer will be viable once HIPAA becomes fully effective.
MS. TRUDEL: Okay, so what you are really suggesting is not that the APA be a code set maintainer or that we adopt the diagnostic criteria. It is that DSM replace ICD-9 for this range of disorders. That is what is being proposed. Let us be clear on what the proposal is.
MR. CHARROW: I think the proposal is since ICD-9 effectively has no diagnostic criteria the government has been using diagnostic criteria in its various functions, that those diagnostic criteria should be retained and the only way of doing that is adopted DSM-IV.
MS. TRUDEL: And what I think this recommendation is saying is that while diagnostic criteria are very appropriate in many uses we aren't adopting any diagnostic criteria as HIPAA standards.
DR.COHN: Our main concern is that people continue to be able to use the diagnostic criteria to make the diagnoses, that the diagnoses, the crosswalk be appropriate. I mean once again I am not sure that I have a solution to this one in terms of us all talking together. I feel that this needs to be turned over to the lawyers to sort of see if this provides a viable option.
I am struggling under the sense that we are now getting close to HIPAA implementation. Obviously at the end of the day we want to make sure that what we are doing does not disrupt practice for therapists and I think the full Committee, and I can't speak for them but I am sure that they would feel much the same way and so I am sort of searching for what I think is a practical, reasonable, easily implementable and speedily implementable solution and knowing how quickly the NPRM process has been occurring recently I wouldn't necessarily hold tremendous hopes that an NPRM could be handled in a way where this could be done before HIPAA is implemented and therefore I think what I was doing was searching for what I thought was something that made sense, gave the country and the health care professionals what needed to be happening but certainly if you have better things that can be done recognizing that we are dealing with a time frame I would be happy to hear it.
We are not against you on this one.
MR. CHARROW: There are obvious solutions. First of all I think if you just read the definition of what a code set is the text, the diagnostic criteria do fit within that definition. So, the code set is something intended to have extant diagnostic criteria. That is what the regulation says and that is what the preamble to the rule says.
MR. FITZMAURICE: Excuse me, I have got the definition of a code set here. It says that the term code set means any set of codes used for including data elements such as tables of terms, medical concepts, medical diagnostic codes or medical procedure codes. Now, that is not the preamble, that is the law. I think it could fit or it could not fit. I mean lawyers can argue about whether it fits. It does seem to make some sense to me.
MR. CHARROW: It does say though a code set includes a code and the descriptors of the codes, 45 CFR 162.103, definition of code set. The regulation further defines descriptor to mean the text defining a code. That is, also, 162.103, but there are ways other than an NPRM by the way of accomplishing the goal quickly and that is why I sort of sat back and was chuckling at your prior discussion of ICD-10.
Obviously the Secretary has the authority to do a variety of things including engaging in reg neg which is fairly quick.
MS. GREENBERG: Obviously one solution is to go immediately to 10-CM where the descriptors are the same but I think from the previous discussion I don't think that is going to happen.
DR. COHN: I don't know that that solves the problem either.
MS. GREENBERG: No, but if the concern is descriptors because the law does mention descriptors, this is true. It doesn't seem to mention diagnostic criteria, but I mean something else that could be done though I don't think Donna finds this particularly desirable because it would take some work, etc., but is you could align the descriptors in 9-CM with DSM-IV since the codes are the same.
DR. PICKETT: But it wouldn't include the diagnostic criteria. I just want to mention in terms of diagnostic criteria any time at a coordination and maintenance committee meeting the issue of including diagnostic criteria into 9-CM has been raised we have received a lot of negative response about that. They felt the diagnostic criteria used by physicians to establish a diagnosis should remain outside of ICD-9-CM as it does in every other instance with the exception of mental health. So, the inclusion of diagnostic criteria for diabetes, lupus or any of those things has always been very much not on the radar screen.
MR. BLAIR: Does this mean there won't be any vote?
DR. COHN: I don't believe so.
MR. BLAIR: If there is not would it be okay for Brady and I to go down to --
DR. COHN: Why don't we just sort of move that into the next step. I sort of agree that the next step is to have this reviewed by legal counsel within CMS. I am sure I will probably be talking to you on the phone. I do apologize that we have run so late already, but we don't have a solution anyway. We will be again meeting during the breakout session in about 3-1/2 to 4 weeks at this point. I still sense that this is sort of a lawyer-to-lawyer set of discussions or at least there is a piece of it that involves that. So, let us get that going and see if then we can further discuss it hopefully but that it all gets resolved but we will sort of see where we are at that point. We want to make sure that this turns out right for the practitioner and we thought we had a solution. We need to see. I certainly wouldn't advise probably your attorney speaking against something that HHS is willing to adopt that actually might solve the problem but then I am not a lawyer. I would defer to you on that one.
DR. REGIER: What we have to anticipate is that in fact there will be a court challenge to that interpretation. If in fact DMS criteria is certainly accepted then that is what our objective is.
DR. COHN: Okay, sure, and as you can tell I am not a lawyer but let us continue the conversation and hopefully you probably have Karen's card but she will be doing some work within CMS and as I said we will just be keeping a close eye on this.
Now, having said that I think we are losing our quorum very rapidly. As we adjourn the meeting thank you all very much and I am sorry we ran so far over.
(Thereupon, at 4:15 p.m., the meeting was adjourned.)